279 10 3494 https://omeka.ibu.edu.ba/files/original/6e4bc501fb64bd81ce39913735223a1c.pdf 6d56c35b29faee3357ea0083e621d28d PDF Text Text Journal of Economic and Social Studies R&D Investment, Governance and Management Entrenchment in French Companies Listed in SBF250 Abderrazak DHAOUI Department of Econometrics and Management University of Sousse, Faculty of Law Economics and Political Sciences, Tunisia abderrazak.dhaoui@fsegs.rnu.tn Fathi JOUINI Department of Econometrics and Management University of Sousse, Faculty of Law Economics and Political Sciences,Tunisia fathi.jouini@fdseps.rnu.tn Abstr ct This study seeks to explain the management entrenchment by investment of free cash flow (FCF) in research and development (R&D), debt, market structure (internal or external), the multinational nature of firms and the characteristics of the board of directors using a sample of 128 groups of French companies listed on the SBF250 between 2003 and 2008. The results show that investment in R&D helps the managers to enhance their authority with respect to the shareholders. The multinational nature of the firm exerts a significant effect on the entrenchment strategy. Manager replaces the internal capital market to the outside market to avoid scrutiny by creditors. We also find an insignificant effect exerted by the debt on the management entrenchment. Finally, we find the absence of a significant relationship between management entrenchment, as measured by discretionary accruals and seniority of the officers, and the characteristics of the board of directors. Keywords: Management entrenchment, R&D, internal market capital, governance mechanisms. Jel odes: G32, M14 Volume 1 Number 2 July 2011 5 �Abderrazak DHAOUI & Fathi JOUINI Introduction The management entrenchment is a deliberated behavior realized by the manager considered as more informed actor which consists of serving own interests at the expense of the shareholders as less informed actors. It takes different forms. The first consists of influencing the accounting results by increasing or decreasing them according to individual needs. The second consists of increasing the specific investment to let the information asymmetry between shareholders and managers more complex, which helps these later to maintain their stations long term. The agency theory (Fama, 1980; Jensen and Meckling, 1976; Jensen and Murphy, 1990) examines this subject on the level of the agency conflicts characterizing the relationships between managers, shareholders and creditors. It insists on several control systems limiting the differences of opinions and interests. The confrontation of this theory to the entrenchment one explains why some control systems are ineffective when managers serve their own interests at the expense of shareholders. This is to say that the integration of the entrenchment theory hypotheses contributes to determine the limits of the mechanisms of control exerted on the managers to incite them to act in the interest of their principal (Alexandre and Paquerot, 2000). The analysis of the control systems exerted to influence the management behavior is essential to the comprehension of the organization’s function and its performance (Alexandre and Paquerot, 2000). However, in spite of the importance of studies realized within the framework of the independent firms, this topic remains, according to our knowledge, little explored within the framework of the companies of group such as the particular case of the multinational companies having a strong internationalized activities. The aim of this study is to identify and analyze the factors, being able to reinforce or to attenuate the discretionary behavior of the managers in order to discover the differences or the similarities of the behaviors of the managers in multinational and domestic companies as regards the entrenchment target. To reach this objective, we try to confront various theories to argue the entrenchment strategies. The diversity of ideas developed in theoretical and empirical studies shows the absence of agreement on the matter. It leads to question on the strategy and trajectories of entrenchment and on the effectiveness of the control systems imposed on the managers. Accordingly, the central question of our research: 6 Journal of Economic and Social Studies �R&D Investment, Governance and Management Entrenchment in French Companies Listed in SBF250 Up to what point can the entrenchment systems influence the effectiveness of the control imposed on the managers? To bring answers to this question, we try to investigate the managers’ discretionary behavior of 128 multinational and domestic French firms over the 2003-2009 period to identify the influences which they had to undergo by the mechanisms of control. This paper is organized as follows: in the first section we present the management entrenchment strategy and the factors influencing positively or negatively the opportunist behavior. We present in the second section the methodology and the estimated model. Results are presented and discussed in the third section. Finally, the fourth section is devoted to the conclusion. Entrenchment strategy and influences of the control systems imposed on the managers Trajectories of the management entrenchment The concept of entrenchment was developed by Shleifer and Vishny (1989). It is a strategy which focuses on the directors to increase their own utilities in their organization by increasing their private expenditure and/or the cost of their replacement. Profiting from a situation of entrenchment, the manager may, likely, decide according to his situation to benefit from his capacity in pecuniary or different other form. He can also increase his capacity to be maintained in the station longest possible or to transmit his capacity to a successor whom will have chosen (Paquerot and Chapuis, 2003). For these reasons, he develops various strategies. S/He uses the resources of his organization to invest in specific activities which increase the firm’s risk and generate a significant informational asymmetry. Accordingly, s/he increases his capacity and different advantages s/ he perceives such as the good remuneration and the security of his job (Alexandre and Paquerot 2000). Volume 1 Number 2 July 2011 7 �Abderrazak DHAOUI & Fathi JOUINI Compared to the whole of the firm’s partners, the managers have a better access to the specific information. This latter constitutes an essential resource for the organization. It represents for the controllers a source of power (Pfeffer 1881, 1882, Pfeffer and Salancik, 1978). However, the strategic state of the managers enables them to control the access to the information and to restrict its availability for the other actors in the organization. Their investment and finance policies depend on the nature of their objectives. They act in increasing the informational asymmetry towards the controllers to increase their discretionary behavior. To spur their opportunism by preparing a suitable land, they maintain various transactions with the subsidiaries such as specific investments (transactions in physical flow) or use their internal capital market (ICM) (transactions in financial flows) by transferring internal resources between the parent companies and their subsidiaries or between the subsidies themselves to finance specific and (geographically) diversified investments. These strategies help them to limit the access of the other actors in the firm to the information. This indicates that the innovation and the decentralization constitute a means to avoid the control exerted by the shareholders and other stakeholders. Stiglitz and Edlin (1992) explain how the managers can benefit from the informational asymmetry to restrict the shareholders’ control and to dissuade potential directors to postulate for the firm management. The investment policy constitutes, in this way, a notable entrenchment tool. According to Alexandre and Paquerot (2000) the increase in firm risks through a particular investment policy in the specific sectors but wellknown to the managers can eliminate potential competitors’ teams without necessary skills to a good management of the firm. The aim of the strategies adopted by the managers is to increase their discretionary behavior “using the means at their disposal, i.e. their human skills but also the firm’s assets, to neutralize the control systems and to increase the dependence of the stakeholders towards the resources which they control” (Alexandre and Paquerot, 2000; p. 9). They can also increase the informational asymmetry towards the stakeholders by investing in assets they have good know-how. This makes information more complex to apprehend for the stakeholders and the potential directors. Thus, it would be beneficial for the manager to increase the dependence of the shareholders to them in order to increase their discretionary behavior (Shleifer and Vishny, 1989; Morck, Shleifer and Vishny, 1990). They make specific transactions with the subsidies and decentralize the specific investments even this can be against the objective of maximization of the shareholders wealth. 8 Journal of Economic and Social Studies �R&D Investment, Governance and Management Entrenchment in French Companies Listed in SBF250 The management entrenchment becomes easier once the informational asymmetry between shareholders and managers increases. Thus, the latter find advantages in investing in assets raising their discretionary. Innovation and decentralization constitute strategic tools helping managers to increase their informational asymmetry towards the shareholders. Particularly, the decentralization of the specific investment makes more difficult the control of the managers’ behavior. Geographical, linguistic and cultural disparities induce – at least - two effects: the increase in the cost of information and the decrease of their pertinence and quality. Moreover, the R&D includes large part of tacit knowledge (Grant, 1996a and b; Nonaka and Takeuchi, 1997). Thus, it would be so difficult to transfer this information, which constrains the shareholders to exert their control on the managers. These investment policies (innovation and decentralization) decrease the effectiveness of the control exerted on the managers. The comprehension of their effects on the effectiveness of the modes of control and consequently on managerial behavior can be given starting from the confrontation of the contributions of the entrenchment and the agency theories. H1a: The increase in R&D facilitates the management entrenchment. H1b : The decentralization of the specific investments reinforces the managerial discretionary behavior. The managers use the free cash-flows to finance the R&D programs. This financial mode allows them to avoid the debt finance which constitutes an effective system of control. In this sense, Jensen (1986) supposes that managers can increase their wealth at the expense of the shareholders by investing the free cash-flow in specific assets and limiting their distribution as dividend. In the same idea, the entrenchment theory (Shleifer and Vishny, 1989), argues that the managers invest these funds in specific investments to increase their compensation and their private expenditure since they are related to the increase in the firm size. Thus, they take advantages using the free cash-flows to avoid the control exerted by the external market and to increase discretionary behavior in making decisions, which enables them to increase their authority towards the shareholders. Indeed, investment on the free cash-flows even in non-profitable projects increases the firm size over its optimal limit. This gives the manager more ability to increase the value of assets under their control and to constrain the control exerted by the shareholders. Volume 1 Number 2 July 2011 9 �Abderrazak DHAOUI & Fathi JOUINI Taking the predictions of the agency theory as a starting point, several studies suppose, in opposition to the pecking order theory of Myers (1984), which excessive use of internal financing is due to the agency conflicts between the managers and the creditors. Seeking to limit external control, the managers prefer internal finance compared to the debt. This helps them to protect information relating to the strategies of development of their organizations (Gertner, Robert, Scharfstein and Stein, 1994). It reinforces, also, their discretionary power and limits the control exerted by the creditors. In the absence of a bank control, the managers can make decisions serving their own interests. They benefit from the stability of the cash-flows to increase their investment in R&D. The decentralization of these investments offers them additional possibilities to improve their wealth at the expense of the shareholders. It reinforces informational asymmetry between managers and shareholders by increasing the knowledge dispersion which induces several difficulties to evaluate present and future value of the firm. H2: The presence of free-cash-flows helps manager to invest in R&D in entrenchment targets While many existing studies report that diversified firms can rely on internal capital markets that enable them to pool and reallocate corporate resources more efficiently than external market (Williamson 1975), several recent studies challenge these findings. Anxious to increase informational asymmetries towards the shareholders, the managers invest in R&D and diversify them. They benefit from the presence of the internal capital market (ICM) to transfer the financial resources from the subsidiary with excess financial resources to those having important investments in R&D. Thus, the ICM helps managers to finance the specific investments, which consequently support their entrenchment. It constitutes a fundamental financing instrument for risky investments which are rationed on the external market. Its presence reinforces the managers’ opportunism and decreases the shareholders gain. Accordingly, the transfer of resources to the subsidiary with high R&D, through the ICM is considered as induced by the objective of maximization of the managers’ wealth at the expense of the shareholders (Jian and Wong, 2003; Liu and Lu, 2004; Thomas, Herrmann and Inoue, 2004; Chang, 2003; Friedman, Johnson and Mitton, 2003). The managers can, particularly, make special transfer of resources serving their own needs (Jian and Wong, 2003 and Thomas et al., 2004). They manage to divert the firm resources to specific projects offering them more independence on the share- 10 Journal of Economic and Social Studies �R&D Investment, Governance and Management Entrenchment in French Companies Listed in SBF250 holders and other external controllers using internal transfers of capital between parent companies and their subsidiaries and/or between the subsidiaries themselves (Chang, 2003; Friedman et al., 2003; Liu and Lu, 2004). This transfer of resources to the specific and geographically diversified investments helps managers to paralyze the control systems by increasing the informational asymmetry within the organization. This asymmetry contributes largely to affect the effectiveness of the control systems and prevents the controllers from applying a sanction. H3: Managers substitute their internal capital market to external market in order to avoid the control system. The agency relationship: conflict of interest which incites to develop different mechanisms of control The financial theory supposes that various modes of control can be used to force the managers to manage the firm in accordance with shareholders’ interests. The shareholders’ structure, the composition of the board of directors, the presence of institutional investors, the incentive compensation, the debt… constitute direct or indirect control systems influencing the managers’ behavior. Thus, the shareholders concentration and/or the presence of financial or institutional shareholders are supposed to have a positive influence on the firm performance. In the same way, the presence of certain administrators (financial or institutional), the part of capital held by the member of the board of directors and the recourse to external administrators more independent and more qualified than the internal ones should exert an effective control on the managers (Alexandre and Paquerot 2000). At the same, the use of debt and incentive compensation can dissuade the manager to manage the firm so as to improve its performance because they must pay future engagements towards their creditors and they have to improve their remuneration since it is indexed on the performance. However, in opposition to the agency theory which proposes various mechanisms to control the managers and to incite them to make decision improving the firm profitability, the entrenchment theory relativizes the role of these mechanisms. It supposes that they will not always be sufficient to limit the opportunistic behaviors of the managerial teams (Alexandre and Paquerot, 2000). Volume 1 Number 2 July 2011 11 �Abderrazak DHAOUI & Fathi JOUINI Thus, the potential investors find in the development of the agency theory a whole of control systems allowing, on one hand, to discipline the managers and on the other hand, to incite them to manage the firm in accordance with their principal interests. In this design, the board of directors constitutes, according to the agency theory, the principal internal mechanism of control. The board of directors presents a specific influence on the other control systems exerted on the managers and has a significant disciplinary role dissuading the managers to act in the shareholders interest. This conception was supported theoretically by Hermalin and Weisbach (2003). They announce that the board of directors contributes to reduce the agency conflicts between the shareholders and the managers. This prediction is confirmed recently by Lefort and Urzua (2008). These authors confirm the fact that the board of directors plays a pivotal role in the management control. It constitutes a principal mechanism of control which seems to reduce the agency costs between the shareholders and the managers. Particularly three significant dimensions of the board of directors are frequently discussed in previous theoretical and empirical studies. The first is related to the size of the board (Jensen, 1993; Yermack, 1996). A big size increases the effectiveness of the control exerted by the board of directors because in such case there is high possibility to be composed by more experiments and competent members. However, the difficulties in coordinating the individual contributions, the conflicts at the time of the decision-making and the difficulties in maintaining good relations between the members as well as the high costs of communications between them seem to reduce these advantages and the effectiveness of the control exerted by the board on the managers (Lipton and Lorsh, 1992; Jensen, 1993). H4a : The larger the size of board of directors, the higher the effect on the managers’ activities. Moreover, the entrenchment theory insists particularly on a pivotal dimension: the independence of the administrators to the managers (Alexandre and Paquerot, 2000). According to Weisbach (1988) and Rosenstein and Wyatt (1997) internal administrators have more capacity to be opposed to the most contestable decisions that make the managers than it is the case of internal administrators. Their presence increases the shareholders wealth rather than reinforce the management entrenchment (Cotter, Shivdasani and Zenner, 1997; Black, Jang, and Him, 2006). 12 Journal of Economic and Social Studies �R&D Investment, Governance and Management Entrenchment in French Companies Listed in SBF250 Dahya, Dimitrov and McConnell. (2008) investigated the effectiveness of the control exerted by the board of directors. Using a panel of 799 companies in 22 countries, they conclude that the independence of the board of directors ensures an effective control on the managers. Kor and Misangyi (2008) confirmed the same result using a sample of 78 firms over the 1990-1995 period and by Lefort and Urzua (2008) using a sample of 160 Chilean Companies. More recently, several studies such as Lin, Ma and Su (2009) and Lau, Sinnadurai and Wright (2009) have confirmed the same results that independence of the boards of directors improves the effectiveness of the controls of the managers. Particularly, Chen, Dyball and Wright. (2009) confirmed this relationship using a sample of 101 Australian firms and conclude that external administrators have more capacity to control the manager when compared to internal ones. H4b : The presence of external administrators reinforces the effectiveness of the board of directors The distinction between the function of chief executive officer (COE) and the chairman of the board of directors constitutes the third dimension is considered as very important. Few studies support the idea that the duality of functions improves the firm performance (Godard and Schatt, 2000). They consider that duality facilitates the management and avoid divergence in making decisions and strategies. It leads consequently, a higher performance (Godard, 1998). Oppositely, several studies consider that duality limits the separation of the functions of decision and control. It plays against the principle of independence of the board of directors through the manager influences (Mizruchi, 1983; Patton and Baker, 1987; Daily and Dalton, 1993). However, the separation of both the function of CEO and chairman of the board limits the capacity of the manager to influence the control exerted by the administrators (Beasley and Salterio, 2001). Thus, the separation of the two functions seems to limit the discretionary behavior of the manager and to ensure the effectiveness control exerted by the board of directors (Jensen, 1993). H4c: The separation of both the function of CEO and chairman of the board of directors improves the effectiveness of the board of directors. Regarding the influence of institutional investors, schools of thought seem to be opposed. The first, represented by the holding of the agency theory, confirms the Volume 1 Number 2 July 2011 13 �Abderrazak DHAOUI & Fathi JOUINI hypothesis of institutional investors controllers. They contribute effectively to the control of the managers (Brickley, Lease and Smith, 1988; Barclay and Holderness, 1991; Bethel and Liebeskind, 1993; McConnell and Servaes, 1990; Mallette and Fowler (1992); Chaganti and Damanpour, 1991; Agrawal and Mandelker, 1992; like Bathala, Moon, and Rao, 1994). The importance of capital they hold gives them more authority toward the managers (Brickley and al, 1988; Pound, 1992). It incites them, in addition, to invest in manager control because they will not have the capacity to liquidate their situation easily. This control helps them to avoid the losses associated to the managers’ discretionary they can support. Moreover, the importance and the diversity of investments they carry out give them the advantage of easy access to information, which facilitated their control of managers. In this line, Alexandre and Paquerot (2000; p. 15) suppose that “the resources they hold help them to exert their control at a weaker cost than the other stakeholders. In fact, the nature of their activities and the importance of investments they carry out allow them a better access to information, which implies simultaneously a better knowledge on the performance of the companies of the sector, abundant information on the environment, a better knowledge of the management market… Moreover, they have particular skills to analyze available information about the firm and its environment. These various advantages enable them to exert their control at a weaker cost compared to individual shareholders “. The second current of thought, represented by the holding of the [Les tenants de la théorie] entrenchment theory, supports the hypothesis of institutional investors serving the managers interests. Pound (1988), Wruck (1989), Shivdasani (1993) and Slovin and Sushka (1993) argue that institutional investors have the capacity to collaborate with managers at the expense of the ordinary shareholders. Their presence limits, consequently, the effectiveness of the other mechanisms of control (Neumann and Voetmann, 1998) and encourages the management entrenchment. In fact, they can act as speculative shareholders and privilege the short-term return on the long-term (Ben M’Barek, 2003; Coffee, 1991; Stapledon, 1996; Bushee, 2001). Having such qualities, they find more advantages in collaborating with the managers rather than in investing in their control. This increases their own wealth on the shirt-term level even at the expense of the other shareholders. H5 : Institutional investors contribute to an effective control on the managers In addition, the use of the stock-options as incentive compensation is considered by the financial theory intended to incite the managers to invest in the more profitable 14 Journal of Economic and Social Studies �R&D Investment, Governance and Management Entrenchment in French Companies Listed in SBF250 projects (Baber, Janakiraman and Kang, 1996; Kole, 1997; Hutchinson and Gul, 2004). They are intended to solve the agency conflicts by indexing the managers’ compensation on the firm performance (Caby and Hirigoyen, 2005). Thus, several studies show that the stock-options play a pivotal role in aligning the managers’ interests on those of the shareholders which reduce significantly the agency conflicts (Core and Guay, 2001; Hartzell and Starks, 2003; Yermack, 1995; Mehran, 1995; Palia, 2001). They reduce the divergence of interests between the shareholders and the managers and incite these later to make more profitable decisions. Taking the agency theory as a starting point, these studies consider that incentive compensation contributes to align the managers’ interests on those of the shareholders. It influences positively the performance since it incites the managers to make more profitable decisions (Jensen and Murphy, 1990; Murphy, 1986; Hall and Liebman, 1998). Oppositely, several studies support the predictions of the entrenchment theory and reject the assumption that incentive compensation reduces the conflicts of interest between managers and shareholders. According to Chen, Steiner and Whyte (2006) and Sullivan and Spong (2007), the stock-option can induce more risk for the shareholders on the long-term and can cause damages to their wealth. In fact, the managers try to increase the value of the stocks they hold in order to improve their compensation. This incites them to manipulate the accounting results to enhance them or to smooth their volatility in order to influence the way the potential investors perceive the future profitability and risk of the firm. This masks the real profitability and affects the firm growth since the shareholders make their decisions using bad information about accounting results. They can invest in projects increasing the failure risk or reject others more profitable considering bad information about the performance and the risk of their company. Oppositely, the managers take advantages of these manipulations when the firm profitability is low and their remuneration is based on stock-options. The increase on the firm value at a shirt-term improves their remuneration even if it affects negatively the firm value at a long-term. H6 : The remuneration by stock-options serves to align the interests of the mangers on those of the shareholders. The debt is considered in the financial literature as an external mechanism being able to dissuade the managers to make decisions maximizing the firm value (Jensen and Meckling, 1976; Jensen 1986; Denis and Denis, 1995). It contributes to reduce the Volume 1 Number 2 July 2011 15 �Abderrazak DHAOUI & Fathi JOUINI free cash-flows problem (Jensen, 1986) since the managers are asked to pay their engagement towards their bank which limits their discretionary power (Stulz, 1990). In fact, the creditors accept to finance only profitable projects to guarantee the refunding of their debt. They refuse, consequently, to finance specific investments since their value decreases in case of financial distress (Nekhili and Poincelot, 2000). The debt incites, therefore, the managers to invest in more profitable projects in order to avoid the disciplinary effect of the external market. However, the managers can use their investment policy to influence the capacity of the creditors to evaluate the profitability and the risk of their projects. They invest in R&D in diversified subsidiaries in order to increase informational asymmetry. In such situation, the creditors encounter serious difficulties to obtain necessary information to evaluate the profitability and risk of the investment and to control the managers. H 7: The debt exerts a significant disciplinary effect on the managers Methodology and results Sample Our sample includes 128 French firms with dimensions to index SBF 250. Since they present an atypical financial operation or that their economic operation is difficult to conceive in the reason of insufficiency of available data, certain companies such as banks, the insurance companies… are withdrawn from the initial sample. Firms are classified in multinationals and domestics. To distinguish between them, we refer to two criteria used in Doukas and Pantzalis (2003). A firm is defined as multinational when this firm reports foreign assets and foreign sales ratios of 10% or more. On the other hand, the firm is defined as domestic firm only if it reports any foreign assets and foreign sales. Using this classification rule, two groups of companies are identified: the first includes 56 domestic firms and the second includes 72 multinational firms. 16 Journal of Economic and Social Studies �R&D Investment, Governance and Management Entrenchment in French Companies Listed in SBF250 The financial and managerial data are collected using annual reports. Collected data covers the 2003-2009 period. Our final sample consists of 128 groups over a period of 7 years (896 observations). The use of the panel data give the advantage to benefit from the both, individual and temporal dimension of the available information. Model The aim of this section is to present the relation between the management entrenchment, the investment strategies and the governance. The model giving these relationships is shown below: ENTRit =+ α 0 α1R & Dit + α 2 DECit + α 3CFit + α 4 ICM it + α 5 NADM it + α 6 EXTADit + α 7 SEPARATit + α 8 INSTITit + α 9 STOKOPit + α10 DEBTit + α11SIZEit + α12 INTit + ε it with ENTRit : the management entrenchment measured by both the discretionary accruals and the seniority of the managers of the firm i in the year t, NADMit: size of the board of directors of the firm i in the year t measured by the number of administrators, EXTADit: Independence of the board of directors measured by the number of external administrators divided by the number of all the administrators, SEPARATit: Boolean variable having the value 1 if there is separation of the function of chief executive officer and chairman of the board of directors of the firm i in the year t, and 0 otherwise, INSTITit: Boolean variable having the value 1 if there is presence of institutional investors holding more than 5% of assets of the firm i in the year t, and 0 otherwise, STOKOPit: Boolean variable having the value 1 if firm i use the stock-option as incentive compensation in the year t, and 0 otherwise, DEBTit: the debt used by the firm i in the year t measured by the financial debt divide by the total liabilities, Volume 1 Number 2 July 2011 17 �Abderrazak DHAOUI & Fathi JOUINI R&Dit: the R&D expenditures divided by the total net sales, DECit: Boolean variable having the value 1 if there is R&D decentralization in the firm i in the year t, and 0 otherwise, CF it: The current cash-flows are used as a indicator of the capacity of the firm to generate future cash-flows. The selected cash-flows correspond to the result before depreciation, expenses and taxes, ICMit : internal capital market measured by the volume of transactions between headquarters and their subsidiaries or between subsidiaries themselves, INTit : Boolean variable having the value 1 if the firm i is a multinational company, and 0 otherwise, SIZEit : the firm size measured by the logarithm of total assets, εit : the error term. Results and discussion The estimation of multiple regression models requires the absence of multicolinearity between the independent variables. This problem refers to a situation in which two or more explanatory variables are highly correlated. A problem of bi-variable multicolinearity arises when two independent variables are strongly correlated. Kervin (1992) estimates that a serious problem of multicolinearity arises starting from a limit of 0,7. Table 1 presents the Pearson correlation between exogenous variables appearing in our model. 18 Journal of Economic and Social Studies �R&D Investment, Governance and Management Entrenchment in French Companies Listed in SBF250 Table 1: Pearson correlation between independent variables size Dec Mic ebt int size 1.00 Dec 0.35 1.00 Mic -0.10 0.12 1.00 0.20 -0.15 0.00 1.00 int -0.40 -0.67 -0.04 0.18 1.00 ebt f tkopt nstit R& f -0.13 0.16 0.21 -0.28 -0.14 1.00 tkopt 0.15 0.14 0.02 -0.04 -0.06 0.05 1.00 nstit 0.21 0.44 0.06 -0.06 -0.36 0.10 0.01 1.00 R& -0.01 0.34 0.17 -0.10 -0.17 0.36 0.08 0.25 1.00 a ext a 0.02 0.10 -0.02 -0.03 -0.09 0.00 0.00 -0.00 -0.09 1.00 ext 0.02 0.03 0.03 -0.03 -0.01 0.02 0.07 0.05 -0.03 0.07 1.00 sepa at -0.00 0.04 -0.00 -0.05 -0.03 -0.01 -0.01 -0.01 0.01 0.11 0.08 sepa at 1.00 Results in table 1 indicate that all correlation coefficients are lower than 0,7. Consequently, we conclude the absence of bi-variable multi-colinearity. In addition, the sample combines both individual and time series data. This seems generate a risk of homogeneity on the sample which leads to bad estimators using the MCO regression. This requires some tests to identify if there is a presence of individual effects in the data and to specify in such case whether it is a fixed or a random effect. Two tests are used. The first is the test of presence of individual effect. The result is an “F-Statistic”. There is individual effect if the “p-value” is lower than the significance level (here: 10%). The second is the “Hausman” test. This later specifies the type of effect. The result is a “Chi-2” statistic which indicates that there is a random effect if the “p-value” is higher than 10% and a fixed effect otherwise. The results of the two tests are presented in table 2. Volume 1 Number 2 July 2011 19 �Abderrazak DHAOUI & Fathi JOUINI Table 2: Homogeneity and Hausman tests Models Model 1: ccruals Model 2: eniority of the manager Homogeneity (127, 752) 5.90 5.78 Hausman test Estimation Method rob > chi2(12) rob>chi2 0.0000 18.23 0.1090 GL 0.0000 19.72 0.0727 Within Results in table 2 indicate that all «p-value» of the statistics “F” are lower than 10%. Thus, we reject the hypothesis of homogeneity of the data. Moreover the Hausman test indicates for the first model (Accruals) the effectiveness of the random effect estimator. However, the estimator gives bad results if there is a strongly correlation between the errors and the explanatory variables. For this reason, it would be better to use the GLS estimator. Oppositely, the results of the “Hausman” test indicates for the second model (seniority of the manager) the effectiveness of the within operator. Table 3 presents the results of the multi-variable estimate regression for entrenchment measured by both the accruals and the seniority of the managers. Table 3: Multi-variable estimation regression result Variables endogenous ccruals (GL ) Variables exogenous oef. . R&D D M DM D R U D _cons z-statistic 0.1689 0.1091 0.6690 0.0725 -0.0008 0.0035 -0.0189 -0.4872 -0.2170 -0.0108 -0.0008 -0.0369 0.7440 Wald chi2(12) rob > chi2 Log likelihood umber of obs umber of groups bs per group: seniority of the manager (W H ) oef. . z-statistic 1.75** 0.2415 6.33*** 0.0455 31.15*** 0.6526 4.40*** 0.0424 -0.54 0.0008 0.13 -0.0340 -1.62 -0.0016 -7.23*** -0.4945 -2.45** -0.4224 -0.71 0.0038 -0.07 0.0313 -10.99*** -0.0408 14.12*** 0.8160 1499.46 R-sq: within 0.0000 (12,806) 316.5322 rob > 892 umber of obs 128 umber of groups min = 6 avg = 6.97 max = 7 bs per group: 2.29** 2.53** 26.82*** 2.09** 0.57 -1.20 -0.13 -5.63*** -5.68*** 0.24 2.45** -9.64*** 13.18*** 0.5922 97.55 0.0000 892 128 min = 6 avg = 6.95 max = 7 Significant at the level:: (***) 1% ; (**) 5% and (*) 10%. 20 Journal of Economic and Social Studies �R&D Investment, Governance and Management Entrenchment in French Companies Listed in SBF250 Results in table 3 indicate a positive influence of the R&D on the management entrenchment. The specific investment helps managers to improve their own returns at the expense of the shareholders and to maintain their position at long-term. They invest in R&D to escape from the control through the increase of informational asymmetry. The specificity of these investments is that they are in dependency to the managers’ private knowledge and competence. They influence, consequently, their presence in the firm at the long-term. In addition, managers may profit from the increase on R&D associate to this investment to influence the effectiveness of the control they supported by limiting the capacity of the controllers to specific information. This gives them more authority towards the shareholders and the other stakeholders. The financial intermediaries have generally different means to dissuade the managers. They have easy access to private information which offers them more authority towards the manager. Consequently, they exert an effective control. However, the managers influence the quality of the control using their investment strategy. They increase their investment in R&D to avoid the debt finance since creditors refuse to finance specific investment. In fact, there are intangible assets which can’t serve as guarantee in case of financial distress. They, also, increase the firm risk and help managers to transfer incomes from creditors to shareholders. Managers profit from this situation of less debt finance to serve their own interests at the expense of the shareholders. Giving that they are less controlled they increase their private expenditures in specific assets allowing them to be in the firm at a long-term. These results seem to confirm the prediction of Nekhili and Poincelot (2000) considering that the R&D, as risky and intangible investments, cannot be easily financed by debts. Taken together these arguments allow us to confirm our first hypothesis according to which the R&D reinforces the management entrenchment. Given that the R&D considered separately influences positively the management entrenchment, his decentralization reinforces the opportunist behavior since it generates several problem of informational asymmetry. Results in table 3, indicate a positive and significant impact of the decentralization of the R&D on both the accruals and the seniority of the managers. Managers decentralize their specific investment to decrease the effectiveness of the control. They increase the difficulty to access to private information serving to a good control because the external environment constitutes a major resource of uncertainty. Volume 1 Number 2 July 2011 21 �Abderrazak DHAOUI & Fathi JOUINI Thus, innovation and decentralization constitute essential factors encouraging the emergence of the favorable conditions for the management entrenchment. Indeed, diversification increases organizational complexity which affects negatively the quality of available information. This indicates that innovation and diversification help manager to maximize their own interests at the expense of the development of the firm. They increase, particularly, the informational opacity which enhance the authority of the managers towards the shareholders and others controllers. Particularly, decentralization leads to more uncertainty by increasing the cultural and linguistics distance between the actors. Environmental uncertainty influences the control system and helps managers to make decisions serving their entrenchment. These arguments seem to confirm our hypothesis H1b according to which the decentralization of R&D reinforces the management entrenchment. We notice, in addition, a positive influence of the internal finance on both the accruals and the seniority of the managers. This seems to confirm our second hypothesis and the predictions of the theory of free cash-flows developed by Jensen (1986). That is to say that managers use the excess of internal resources to finance specific investment serving their own interests at the expense of the shareholders. They invest in intangible assets even they are non profitable in order to increase their discretionary behavior. The R&D generates serious problems of informational asymmetry between shareholders and managers and helps these later to entrench largely in the firm. The dependence of these investments on the managers’ knowledge and competences helps them to be maintained in their station at a long-term and/or to maximize their private expenditure. Internal finance is used by managers to avoid the disciplinary effect of the debt. This indicates a positive relationship between the internal resources and the management entrenchment. The excess in cash-flows helps managers to finance specific investment generally constrained on the external market. This helps them to increase informational asymmetry by investing in specific assets and to avoid the control of external market. We also note that the presence of the ICM reinforces the opportunistic behavior of the managers. It presents a positive impact on management entrenchment. In fact, the managers substitute their internal market to the external market in order to avoid the control exerted by this later. Secondly, the flexibility of transaction within 22 Journal of Economic and Social Studies �R&D Investment, Governance and Management Entrenchment in French Companies Listed in SBF250 the ICM helps them to transfer the excess of financial resources from subsidiaries with less investment in R&D to those with high investment in order to enhance their opportunistic behavior. In other words, the internal finance of the risky investment by ICM allows the managers to improve enhance their gain at the expense of the shareholders. This result seems to be in contradiction with the prediction of Williamson (1975) according to which the IMC serves to finance the well profitable projects and to exert a significant control on the managers. Taken together, these arguments allow us to confirm the third hypothesis according to which the managers prefer the ICM to the external market in order to increase their discretionary behavior by investing in specific investment and eliminating the control of the debt. We note, Moreover, a negative relationship between the management entrenchment and the presence of institutional investors. These later, exert by comparison to their individual competitors, more effective control on the managers. They have necessary skills allowing them a better evaluation ex ante of risk and return associated to new investments and to ensure an effective control ex post on the managers. In fact, given the importance of assets they hold, the institutional investors are incited to invest in the control of the manager instead of liquidating their portfolio of assets because the sale of the blocks of stocks affects negatively their value. The high assets they hold give them, also, more authority towards the managers. These later have to manage the firm so as to improve its performance in order to avoid the possibility of massive sale of assets held by the institutional investors. In fact, the massive sale of assets decreases the stock price and affects negatively the firm performance what reduces, significantly, the managers’ gain particularly if their compensation is based on the performance. This confirms our hypothesis H 5 according to which the presence of institutional investors exerts an effective control on the managers. We note, in addition, a negative relation between the management entrenchment and the incentive compensation. This result is in accordance with our sixth hypothesis. The stock-option as incentive compensation serves to align the interest of the managers on that of their principal (shareholders). It decreases the agency conflict between shareholders and managers and incites the later to manage the firm to improve the performance. These results confirm the predictions of Caby and Hirigoyen (2005), Core and Guay (2001), Hartzell and Starks (2003) and Palia (2001). Incentive compensation constitutes, in fact, a very significant tool Volume 1 Number 2 July 2011 23 �Abderrazak DHAOUI & Fathi JOUINI serving to resolve the agency conflicts by punishing the deviating behavior of the managers. The shareholders are the first victims of the discretionary behaviors of the managers. Since they have not necessary skills to control the managers, the shareholders choose to align their compensation to the performance. In fact, the stock-options incite the managers to invest in more profitable and less risky project what improve the firm performance and consequently their own gain since their compensation is indexed on the performance. Results in table 3 indicate, oppositely, a non-significant relationship between the debt and the management entrenchment measured by both the discretionary accruals and the seniority of the managers. The creditors have, naturally, several means to discipline the managers. They have an easier access to private information enabling them to better control the manager. Moreover, the obligation of refunding of the debt and the interests is supposed to reduce the manager autonomy compared to the shareholders. In particular, the creditors agree to finance only profitable projects to guarantee the refunding of their debt at the date of payment. They require fixed assets as guarantees and refuse to finance the R&D investments since they are in major part intangible. Their net value asset (NVA) is very weak even null in the event of discontinuity of the business. Consequently, the creditors incur high risk when they finance these investments. Moreover, the increase in the expenditure in R&D helps the manager to increase the risk of the activity and to ensure a transfer of wealth from the creditors to the shareholders. Thus, to avoid the debt and to neutralize its disciplinary effect, the managers try to increase their investments in R&D. They also decentralize these investments to increase informational asymmetry and to incite the creditors to refuse the financing of their organizations. This strategy influences the effectiveness of the control exerted by the debt on the manager. For these reasons, the debt constitutes only marginally part in the financing of the R&D. The decentralization of these investments reinforces informational asymmetry and pushes the creditors to minimize the debt amounts. Taken together, these arguments seem to explain why the debt exerts a significant disciplinary effect on the managers. We note, moreover, the absence of a significant relation between the management entrenchment and the characteristics of the board of directors. Several arguments 24 Journal of Economic and Social Studies �R&D Investment, Governance and Management Entrenchment in French Companies Listed in SBF250 can be proposed to interpret this result. Firstly, the increase in the investments in R&D restricts the access of the administrators to information and gives the managers more authority vis-à-vis the shareholders. Secondly, the high number of administrators can generate conflicts and poses a problem of coordination and several difficulties to maintain good relations between the members, which affect the quality of the control exerted by the board of directors on the managers. This leads to reject the hypotheses H4a, H4b and H4c, according to which the characteristics of the board of directors (size, independence and separation of the functions of CEO and chairman) exert an effective control on the managers. We also note a positive relationship between the multinational character of the firm and the management entrenchment as measured by the seniority of the managers. However, the impact of the multinational character on the accruals is non-significant. The leaders diversify their investment to benefit from uncertainties characterizing the external environment of their organization. The investments which they maintain abroad are difficult to control because of the cultural differences and the linguistic difficulties that may face the controllers. Moreover, the costs of transfer of knowledge increase the difficulty to access to information about foreign activities. This seems to reinforce the capacity of the managers with respect to these actors and to increase their discretionary behavior. Conclusion This study examines the effectiveness of the control system imposed on the managers by confronting the assumptions of the theories of the agency, the incentives, and the management entrenchment. The effectiveness of the control systems seems to be influenced by the deviating behavior of the managers. Their statute offers them the capacity to make decisions affecting the shareholders wealth and the effectiveness of the control exerted by these later. The innovation and the (geographically) diversification help them to influence the quality of the control exerted by the board of directors and the financial intermediaries. The investment of the additional resources in R&D in geographically diversified units increases the investment risks and the informational asymmetry toward the partners of the company. It increases, consequently, the discretionary of the manager at the expense of the shareholders and prepares to a favorable ground to their opportunistic behavior. Volume 1 Number 2 July 2011 25 �Abderrazak DHAOUI & Fathi JOUINI However, the institutional investors have necessary competences to exert a more effective control on the managers. The diversity of the investments they carry out helps them to evaluate more efficiency the position and to ensure a more effective control. The alignment of the compensation to the performance may solve the agency problems between shareholders and managers. It incites the later to make decisions which do not affect the shareholders wealth and which create more value. References Agrawal, A. and Mandelker, G. (1992). Shark Repellents and the Role of Institutional Investors in Corporate Governance. Managerial and decisions economics, 13, 15-22. Alexandre, H. and Paquérot, M. (2000). Efficacité des structures de contrôle et enracinement des dirigeants. Finance Contrôle Stratégie, 3 (2), juin, 5-29. Baber, W. Janakiraman, S. and Kang, S. (1996). Investment opportunities and the structure of executive compensation., Journal of Accounting and Economics, 21, 297-318. Barclay, M. and Holderness, C. (1991), Negotiated block trades and corporate control. Journal of Finance, 46 (3), 861-878. Bathala, B. Moon, K.P. and Rao, R.P. (1994). Managerial Ownership, Debt Policy, and the Impact of Institutional Holdings: An Agency Perspective. Financial Management, 23 (3), 38-50. Beasley, M. and Salterio, S. (2001). The relationship between board characteristics and voluntary improvements in audit committee composition and experience. Contemporary Accounting Research, 18 (4), 539-570. Ben M’Barek, N. (2003). Le comportement des investisseurs institutionnels dans le contrôle des entreprises de leurs portefeuilles : une contribution à l’étude des facteurs d’influence. thèse de doctorat, Université de Nice. Bethel, J.E. and Liebeskind, J. (1993). The Effects of Ownership Structure on Corporate Restructuring. Strategic Management Journal, 14, 15-31. 26 Journal of Economic and Social Studies �Black, B. Jang, H. and Him, W. (2006). Predicting firms’ corporate governance choices: evidence from Korea. Journal of Corporate Finance, 12, 660-691. Brickley, J.A. Lease, R.C. and Smith, C.W. (1988). Ownership Structure and Voting on Anti takeover amendments. Journal of Financial Economics, 20, 267-291. Bushee, B.J. (2001). Do institutional investors prefer near-term earnings over longrun value?. Contemporary Accounting Research, 18 (2), 207-246. Caby, J. and Hirigoyen, G. (2005). Création de Valeur et Gouvernance de l’Entreprise. Economica, 3ème édition, Paris. Chaganti, R. and Damanpour, F. (1991). Institutional Ownership, Capital Structure, and Firm Performance. Strategic Management Journal, 12, 479-491. Chang, S.J. (2003). Ownership structure, expropriation, and performance of group-affiliated companies in Korea. Academy of Management Journal, 46 (2), 238-276. Chen, C.C. Steiner, T.L. and Whyte, A.M. (2006). Does Stock Option-based Executive Compensation Induce Risk-Taking? An Analysis of the Banking Industry. Journal of Banking and Finance, 30, 915-945. Chen, R. Dyball, M.C. and Wright, S. (2009). The link between board composition and corporate diversification in Australian corporations. Corporate Governance: An International Review, 17 (2), 208-223. Coffee, J.C. (1991). Liquidity versus control: the institutional investor as a corporate monitor. Columbia Law Review, 91 (6), 1277-1367. Core, J.E. and Guay, W. (2001). Stock option plans for Non Executive Employees. Journal of Financial Economics, 69, 253-287. Cotter, J.F. Shivdasani, A. and Zenner, M. (1997). Do independent directors enhance target shareholder wealth during tender offers?. Journal of Financial Economics, 43 (2), 195-218. Dahya, J. Dimitrov, O. and McConnell, J.J. (2008). Dominant shareholders, corporate boards, and corporate value: A cross-country analysis. Journal of Financial Economics, 87, 73-100. Daily, C. and Dalton, D. (1993) Board of directors leadership and structure: Control and performance implications. Entrepreneurship Theory and Practice, 17 (3), 65-81. �Denis, D.J. and Denis, D.K. (1995) Performance changes following top management dismissals. Journal of Finance, 50 (4), 1029-1057. Doukas, J.A. and Pantzalis, C. (2003). Geographic diversification and agency costs of debt of multinational firms. Journal of Corporate Finance, 9, 59-92. Fama, E. (1980). Agency problems and the theory of the firm. Journal of Political Economy, 88, 288-307. Fox, J. (1991). Regression diagnostics: an introduction. Newbury Park, CA: Sage. Friedman, E. Johnson, S. and Mitton, T. (2003). Propping and tunneling. Journal of Comparative Economics, 31 (4), 732-796. Gertner, Robert, Scharfstein, D. and Stein, J. (1994). Internal Versus External Capital Markets. The Quarterly Journal of Economics, 109 (4), 1211-1228. Godard, L. (1998), Les déterminants du choix entre un conseil d’administration et un conseil de surveillance, Finance Contrôle Stratégie,1 (4), 39-61. Godard, L. and Schatt, A. (2000) Faut-il limiter le cumul des fonctions dans les conseils d’administration ?. La Revue du Financier, 127, 36-47. Grant, R.M. (1996a). Prospering in dynamically-competitive environments: organisational capability as knowledge integration. Organisation science, 7, 375-387. Grant, R.M. (1996b). Toward a knowledge-based theory of the firm. Strategic Management Journal, 17, winter, Special Issue, 109-112. Grossman, S.J. and Hart, O.D. (1982). Corporate financial structure and management incentives. J.J. Mc Call, ed: Economics Information and uncertainty (University of Chicago Press, Chicago, IL). Hall, B. and Liebman, J. (1998). Are CEOs Really Paid like Bureaucrats?. The Quarterly Journal of Economics, 103, 653-691. Hartzell, J.C. and Starks, L.T. (2003). Institutional Investors and Executive Compensation, The Journal of Finance, 58 (6), 2351-2374. Hausman, J.A. (1978). Specification Tests in Econometrics. Econometrica, 46, 1251-1271 Hermalin, B. and Weisbach, M. (2003). Boards of directors as an endogenously determined institution: A survey of economic literature. Economic Policy Review, 9, 7-26. �Hutchinson, M. and Gul, F.A. (2004). Investment opportunity set, corporate governance practices and firm performance. Journal of Corporate Finance, 10, 595-614. Jensen, M. and Meckling, W. (1976). Theory of the Firm: Managerial Behavior, Agency Costs and Ownership Structure. Journal of Financial Economics, 3 (4), 305-360. Jensen, M.C. and Murphy, K.J. (1990). Performance Pay and Top Management Incentives. Journal of Political Economy, 98 (2), April, 225-265. Jensen, M.C. (1986). Agency Costs of Free Cash Flow, Corporate Finance and Takeovers. American Economic Review, 76 (2), 323-329. Jensen, M.C. (1993). The Modern Industrial Revolution, exit and the failure of internal control systems. Journal of Finance, 48 (1), 831-880. Jian, M. and Wong, T.J. (2003). Earnings management and tunnelling through related party transactions: Evidence from Chinese corporate groups. Working paper, the Chinese University of Hong Kong and Singapore Nanyang Technological University. Kervin, J.B. (1992). Methods for business research. New York: Harpet Collins. Kole, S.R. (1997). The Complexity of Compensation Contracts. Journal of Financial Economics, 43, 79-104. Kor, Y.Y. and Misangyi, V.F. (2008). Outside directors industry-specific experience and firm’s liability of newness. Strategic Management Journal, 29 (12), 13451355. Lau, J. Sinnadurai, P. and Wright, S. (2009). Corporate governance and Chief Executive Officer dismissal following poor performance: Australian evidence. Accounting and Finance, 49, 161-182. Lefort, F. and Urzua, F. (2008). Board Independence, firm performance and ownership concentration: Evidence from Chile. Journal of Business Research, 61, 615-622. Lin, C. Ma Y. and Su, D. (2009). Corporate governance and firm efficiency: Evidence from China’s publicly listed firms. Managerial and Decision Economics, 30, 193-209. Lipton, M. and Lorsch, J. (1992). A modest proposal for improved corporate governance. Business Lawyer, 48 (1), 59-77. �Liu, Q. and Lu, Z. (2004). Earnings Management to Tunnel: Evidence from China’s Listed Companies. Working paper, http://ssrn.com/abstract=349880. Mallette, P. and Fowler, K.L. (1992). Effects of board composition and stock ownership on the adoption of “poison pills”. Academy of Management Journal, 35 (5), December, 1010-1035. McConnell, J. and Servaes, H. (1990). Additional evidence on equity ownership and corporate value. Journal of Financial Economics, 27, 595-612. Mehran, H. (1995). Executive compensation structure, ownership and firm performance. Journal of Financial Economics, 38 (2), 163-84. Mizruchi, M. (1983). Who controls whom? An examination of the relation between management and boards of directors in large American Corporations. Academy of Management Review, 8 (3), 426- 435. Morck, R. Shleifer, A. and Vishny, R.W. (1990) Do managerial objectives drive bad acquisitions?. Journal of Finance, 45, 31-48. Murphy, K. (1986). Incentives, Learning and Compensation: A theoretical and Empirical Investigation of Managerial Labor contacts. Rand Journal of Economic, 17, 59-76. Myers, S. (1984). The Capital Structure Puzzle. Journal of Finance, 39, 575-592. Nekhili, M. and Poincelot, E. (2000). La fonction R&D et la latitude managériale : une analyse théorique. Finance Contrôle Stratégie, 3 (1), 5-28. Neumann, R. and Voetmann, T. (1998). CEO turnovers and corporate governance: Some evidence from the Copenhagen Stock Exchange. Working papers, Financial Department, Copenhagen Business School. Nonaka, I. and Takeuchi, H. (1997). La connaissance créatrice. La dynamique de l’entreprise apparentée, DeBoeck Université, Bruxelles. Palia, D. (2001). The endogeneity of Managerial compensation in Firm Valuation: A Solution. The Review of Financial Studies, 14 (3), 735-764. Paquerot, M. and Chapuis, J.M. (2003). Politique d’endettement et réseau d’appartenance des dirigeants. Cahier de recherche du FARGO Patton, A. and Baker, J. (1987). Why do not directors rock the boat?. Harvard Business Review, 65, 10-12. �Pfeffer, J. and Salancik, G.R. (1978). The external control of organisations: A resource dependence perspective. Harper & Row, Publishers. Pfeffer, J. (1981). Power in Organisations. Harper Business. Pfeffer, J. (1982). Organisations and Organisation theory, Ballinger Publishing Company. Pound, J., 1988, Proxy contests and the efficiency of shareholder oversight. Journal of Financial Economics, 20, 237-265. Pound, J. (1992). Beyond takeovers: politics come to corporate control. Harvard Business Review, 70 (2), 83-93. Rosenstein, S. and Wyatt, J.G. (1997). Inside Directors, Board Effectiveness, and Shareholder Wealth. Journal of Financial Economics, 44, 229-250. Shivdasani, A. (1993). Board composition, ownership structure, and hostile takeovers. Journal of Accounting and Economics, 16, Issue 1-3, 167-198. Shleifer, A. and Vishny, R.W. (1989). Management Entrenchment: the Case of Manager-Specific Investments. Journal of Financial Economics, 25, 123-140. Slovin, M. and Sushka, M. (1993). Ownership concentration, corporate control activity, and firm value: evidence from the death of inside blockholders. Journal of Finance, 48 (4), 1293-1321. Stapledon, G.P. (1996). Disincentives to activism by institutional investors in listed Australian companies. Sydney Law Review, 18, 152-192. Stiglitz, J.E. and Edlin, A.S. (1992). Discouraging Rivals : Managerial Rent Seeking and Economic Insufficiencies. NBER Working Paper, n° 4145, August. Stulz, R.M. (1990). Managerial discretion and optimal financing policies. Journal of Financial Economics, 26, 3-27. Sullivan, R.J. and Spong, K.R. (2007). Manager Wealth Concentration, Ownership Structure and Risk in Commercial Banks. Journal of Financial Intermediation, 16, 229-248. Thomas, W.B. Herrmann, D.R. and Inoue, T. (2004). Earnings Management through Affiliated Transactions. Journal of International Accounting Research, 3 (2), 1-26. �Weisbach, M.S. (1988). Outside directors and CEO turnover. Journal of Financial Economics, 20 (1-2), 431-460. Williamson, O.E. (1975). Markets and hierarchies: Analysis and antitrust implications. Collier McMillan Publishers, Inc., New York. Wruck, K.H. (1989). Equity ownership concentration and firm value. Journal of Financial Economics, 23 (1), 3-28. Yermack, D. (1995). Do Corporation Award CEO Stock Options Effectively?. Journal of Financial Economics, 39, 237-269. Yermack, D. (1996). Higher market valuation of companies with a small board of directors. Journal of Financial Economics, 40 (2), 185-211. � Dublin Core The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/. Extent The size or duration of the resource. 1062 Title A name given to the resource R&D Investment, Governance and Management Entrenchment in French Companies Listed in SBF250 Author Author DHAOUI, Abderrazak JOUINI, Fathi Abstract A summary of the resource. This study seeks to explain the management entrenchment by investment of free cash flow (FCF) in research and development (R&D), debt, market structure (internal or external), the multinational nature of firms and the characteristics of the board of directors using a sample of 128 groups of French companies listed on the SBF250 between 2003 and 2008. The results show that investment in R&D helps the managers to enhance their authority with respect to the shareholders. The multinational nature of the firm exerts a significant effect on the entrenchment strategy. Manager replaces the internal capital market to the outside market to avoid scrutiny by creditors. We also find an insignificant effect exerted by the debt on the management entrenchment. Finally, we find the absence of a significant relationship between management entrenchment, as measured by discretionary accruals and seniority of the officers, and the characteristics of the board of directors. Publisher An entity responsible for making the resource available International Burch University Date A point or period of time associated with an event in the lifecycle of the resource 2011-01 Keywords Keywords. Article PeerReviewed H Social Sciences (General),HB Economic Theory,HG Finance https://omeka.ibu.edu.ba/files/original/89301abe1f55d2831bbf7d67dae22b5f.pdf 8dd8701bdfb32c846bc50b02535be554 PDF Text Text Journal of Economic and Social Studies Measuring the Level of International Capital Mobility for MENA Countries Muhittin KAPLAN Melikşah University, Faculty of Economics and Management, Department of Economics, Kayseri, Turkey. mkaplan@meliksah.edu.tr Hüseyin KALYONCU Melikşah University, Faculty of Economics and Management, Department of International Trade and Business, Kayseri, Turkey. hkalyoncu@meliksah.edu.tr ABSTRACT To achieve sustainable development, it is vitally important to sustain macroeconomic stability, which is closely related to the extent of capital mobility allowed by a country. This paper attempts to measure the level of international capital mobility empirically by estimating the Feldstein-Horioka coefficients employing the panel data for the MENA countries over the period 1963-2007. In empirical analysis, time series properties of the data are examined using recently developed techniques of panel unit root. Having obtained that variables of the model are stationary variables, we use the fixed effect panel model in the analysis of data.The results indicate that capital mobility has always been high in MENA countries but this is particularly obvious for the period 1980-2007, which corresponds to the liberalization period. For the subperiod of 1963-1980, the estimated coefficients are relatively higher, implying the presence of a relatively lower level of capital mobility. Keywords: Feldstein-Horioka puzzle; Capital mobility; Fixed Effect Panel Volume 1 Number 1 January 2011 25 �Muhittin KAPLAN & Hüseyin KALYONCU Introduction This paper examines the degree of capital mobility for MENA countries. It is well known that one of the important aspects of achieving sustainable development is to preserve macroeconomic stability, which is closely related to the extent of capital mobility. For this reason, measuring the level of capital mobility is an important task to achieve. While higher capital mobility was encountered as one of the reasons behind the recent worldwide financial crisis, the subject is also important for policy makers and firms for a number of reasons; (i) the effectiveness of macroeconomic policies is closely related to the degree of international capital mobility; (ii) higher international capital mobility helps firms to allocate resources efficiently and achieve risk diversification; (iii) higher international capital mobility may also increase volatility which may end up with financial crisis. For example, the global financial crisis began in the USA and spread to Europe and then to the whole world. Today we see that devastating effect of global financial crisis is more pronounced in developed countries than the developing countries. When we consider the MENA countries in this sense, although they are not composed of a homogeneous group, they seem that they are not as much affected as developed countries. It can be argued that this is because most of the MENA countries are oil-exporting countries and high oil prices following the invasion of Iraq led to an accumulation of significant amounts of dollars in these countries. Even though this is true, this cannot be the sole reason. To be sure, we first need to measure whether capital mobility is high in these countries. If capital mobility is high for this group of countries, we can accept that the capital they accumulated helped them to stabilize their economies during the crisis and that is why they are less affected by the global financial crisis. If capital mobility is low, then we say that these countries were exempt from the crisis because they were luckily not allowing free movement of capital. Review of the empirical literature shows that most of the studies on the measurement of the level of international capital mobility have focused on estimating the Feldstein and Horioka (1980, hereafter FH) model. FH model involves examining the relationship between savings and investment empirically. Intuitively, the FH model implies that the correlation between savings and investment will be one if capital movement is not allowed and otherwise it will be zero if there is perfect capital mobility. Since then, many studies have been carried out to estimate the relationship between savings and investment, producing an enormous literature on the subject. In this paper, we aim to estimate the FH coefficient for MENA countries using the fixed effect panel model. FH coefficients will be estimated for MENA countries over the period of 1963-2007 and sub-periods of 1963-1980, 1981-2007, 1981-1997 and 1998-2007. The small coefficients will be interpreted as increased capital mobility. Literature Review Given the importance of the subject for open economies, a number of different empirical 26 Journal of Economic and Social Studies �Measuring the Level of International Capital Mobility for MENA Countries methodologies were developed aiming to measure the extent of capital mobility. The FeldsteinHorioka model has found widespread use in the empirical literature because the model is simple as well as providing an intuitive explanation for the level of capital mobility. The model suggested and empirically estimated by Feldstein and Horioka (1980) is as follows: (I/Y) = cons + b (S/Y) (1) where I, S and Y represent domestic investment, domestic saving and gross domestic product respectively. The coefficients cons and b denote constant term and savings-retention coefficients and they are the coefficients that will be ultimately estimated. In equation (1), dependent variable, domestic investment and independent variable, domestic saving is given as shares of the gross domestic product. Using data over 1960-74, Feldstein and Horioka (1980) found that the savings-retention coefficient is very close to the one for 16 OECD countries, implying low capital mobility. Since then, an enormous literature has accumulated to test the Feldstein-Horioka puzzle and explain the puzzle. Apergis and Tsoumas (2009) provide a detailed survey of these studies of empirical literature. In general, the empirical literature on the subject provides mixed results for both developed and developing countries. Studies testing the puzzle for developing countries found out that the savingretention coefficient is small, indicating that the level of capital mobility is high in these countries (Payne and Kumazawa, 2006; Apergis and Tsoumas, 2009;Coakley et.al., 1999). On the contrary, some studies provide evidence that capital mobility is low in developing countries (Murthy, 2008; Ghosh and Ostry, 1995). While Wong (1990) argue that the high capital mobility observed in developing countries can be attributed to the size of the non-traded sector, Kasuga (2004) argue that small-sized and inefficient financial mechanisms in developing countries lead to high capital mobility. Ozmen (2005), Bahami-Oskooee and Chakrabarati(2005), and Sinha and Sihna (2004) find that the correlation between saving and investment is high in larger economies. Bangake and Eggoh (2010) mention the importance of the legal protection system provided for investors in relation to capital mobility. They tested the Feldstein-Horioka puzzle for 37 African countries using the panel cointegration technique and found that savings and investment are a non-stationary and cointegrating series. Their estimation results indicate that capital mobility is higher (0.34) in the countries with strong legal protection of investors than in countries with worse protection (0.85). Overall, the test of the Feldstein-Horioka puzzle for the developing countries, including Middle East countries, shows high capital mobility because the magnitude of foreign aid and the extent of the non-traded sector are high in these countries and they have weak financial markets and are relatively open economies (Apergis and Tsoumas, 2009). Econometric Methodology and the Data This paper attempts to investigate the relationship between investment rate and saving rate to measure the level of capital mobility for MENA countries. The data subject to empirical analysis is taken from IMF International Financial Statistics for 12 countries in the MENA region over the period 1963- Volume 1 Number 1 January 2011 27 �Muhittin KAPLAN & Hüseyin KALYONCU 2007. The data set is determined by the availability of the data. In other words, those countries that have unbroken series of data over the sample period are included in the data set. These countries are Algeria, Israel, Iran, Egypt, Jordan, Kuwait, Libya, Morocco, Saudi Arabia, Syria, Tunisia, and Turkey. The variables employed in the empirical work involve gross domestic investment and gross domestic savings as percentages of gross domestic product. As in any empirical study employing time series data, it is vital to determine the level of integration of series. For this reason, we first check the level of integration of investment rate and saving rate variables. The integration level of variables can be determined by the standard unit root tests such as the Augmented Dickey-Fuller (ADF) test. However, it is well-known that standard unit root tests which are test based on individual time series have low power against stationary alternatives. For this reason, the recently developed panel unit root tests were frequently employed in the investigation of the time series properties of data. Since panel data increases the power of the test by enhancing the time series dimension of the data by the cross section, the results will be more reliable. Some of the most popular panel unit root tests are as follows: the LLC (Levin, Lin and Chu, 2002), the IPS (Im, Pesaran and Shin, 2003), ADF - Fisher Chi-square (Maddala and Wu,1999), and PP - Fisher Chi-square (Choi, 2001). While the LLC test allows for heterogeneity of individual deterministic effects and a heterogeneous serial correlation structure, it assumes the presence of a homogeneous autoregressive root under the alternative. The latter is identified as a serious limitation for the LLC test. The LLC test procedure involves using pooled t-statistics of the estimator to evaluate the hypothesis of non-stationarity of each individual time series. The more recently developed IPS tests overcame the limitation of the LLC test by allowing for heterogeneity of the autoregressive root under the alternative. The IPS test is simple to calculate and allows for residual serial correlation and heterogeneity of dynamics across groups. However, simulations indicate that the IPS test is sensitive to a correct choice of lag orders in the underlying ADF regressions; the power of the t-bar test is more favorably affected by a rise in time dimension of the data than the cross-section units of the data; and the interpretation of the IPS test results are difficult because of the heterogeneous nature of the alternative hypothesis. Maddala and Wu’s (1999) and Choi’s (2001) tests were similar in the way that both suggested panel unit root tests performed using a Fisher statistic, but they were developed to overcome the shortcomings of the LLC and the IPS tests. Maddala and Wu’s (1999) and Choi’s (2001) tests solves the problems related to previously mentioned tests by providing the combination of probability values for a unit root tests applied to each group in the data set. With this in mind, we employed the LLC, the IPS, ADF-Fisher and PP-Fisher panel unit root tests in this paper. For the LLC and IPS test, the optimal lag length is determined according to Schwarz criteria. The time series properties of the variables involved determined our choice of the empirical methodology to use in the analysis of measuring the extent of capital mobility model. As shown below, since both independent and dependent variables of the empirical model are stationary variables, we did not test for cointegration and decided to estimate the model with fixed effect panel data model. The empirical findings are provided in the next section. 28 Journal of Economic and Social Studies �Measuring the Level of International Capital Mobility for MENA Countries Empirical Results In this section, the estimation results obtained from panel unit root tests and the equation (1) which shows the relationship between investment rate and saving rate will be provided. Table 1 and Table 2 provide panel unit root tests results for investment and saving variables respectively. In the first column, the LLC, the IPS, ADF-Fisher and PP-Fisher panel unit root tests are given. While the second column provides panel unit root test results with constant, results with constant and trend are given in the third column. It is worth mentioning that the optimal lag length for the tests were determined according to Schwarz criteria. Examination of the tables shows that the null hypothesis of non-stationarity is rejected at 1% level by all tests. Therefore, we conclude that the variables subject to empirical analysis of the paper are stationary at levels and hence there is no danger of regression results being spurious. Table 1. Panel Unit root test results for the Investment variable Levin, Lin and Chu Im, Pesaran and Shin W-stat ADF - Fisher Chi-square PP - Fisher Chi-square With constant Statistics Probability* -3.26816 0.0005 -4.65013 0.0000 70.1163 0.0000 63.3885 0.0000 With constant and trend Statistics Probability* -2.45043 0.0071 -3.54615 0.0002 57.4535 0.0001 43.4409 0.0089 Note: *Probabilities for Fisher tests are computed using an asymptotic Chi-square distribution. All other tests assume asymptotic normality. Table 2. Panel Unit root test results for the Saving variable With constant Statistic Probability* Levin, Lin & Chu t* -2.59357 0.0047 Im, Pesaran and Shin W-stat -3.20335 0.0007 ADF - Fisher Chi-square 52.5655 0.0007 PP - Fisher Chi-square 53.0406 0.0006 With constant and trend Statistic Probability* -3.06139 0.0011 -2.80969 0.0025 47.3121 0.0031 48.4657 0.0022 Note: *Probabilities for Fisher tests are computed using an asymptotic Chi-square distribution. All other tests assume asymptotic normality. Having established that both saving and investment variables are integrated I(0), fixed effect panel methodology is used in the estimation of the F-H model. Table 3 presents the empirical findings for the whole period and the sub-periods. Examination of the table shows that the independent variables of the model and tests related to the model are given in the first column. The following five columns provide estimation results for different time periods. Estimation results for the whole period 19632007 are given in the second column. The sub-periods are determined by the main changes which occurred in the world economy. For example, 1980 marks the beginning of the liberalization period for developing countries. The year of financial crisis, 1997, in Asia is also a turning point in terms of Volume 1 Number 1 January 2011 29 �Muhittin KAPLAN & Hüseyin KALYONCU financial system in the world. Theoretically, we expect that capital mobility has increased over these periods in line with financial liberalization policies. Examination of the Table 3 indicates that there is no statistically significant relationship between saving and investment rates implying the presence of very high capital mobility in the MENA region. The saving-retention coefficient (b) is almost zero for the period 1963-2007. Considering the subperiods, it is clearly obvious that after 1980 capital mobility is significantly high compared to the period of 1963-1980. While the saving-retention coefficient is 0.13 and statistically significant at the 5% level for 1963-1980, it is very close to zero and not statistically significant for 1981-2007 periods. This indicates that liberalization policies had an immense effect in the MENA region in terms of increased capital mobility. Another important observation about the saving-retention coefficient is that this coefficient remained about the same during the sub-periods of 1981-1997 and 1998-2007. Table 3. Fixed Effect Panel Estimation Results Period Constant Savings R-squared S.E. of regression F-statistic Wald Test (Chi-square) Fixed Effects Tests: Cross-section F-Test Period F-Test 1963-2007 0.234 (50.8)* 1963-1980 0.2001 (11.87)* 1981- 2007 0.243 (52.2)* 1981-1997 0.250 (47.1)* 1998-2007 0.240 (23.4)* 0.0051 (0.285) 0.130083 (1.94)** -0.02902 (-1.54) -0.040 (-1.78)*** -0.056 (-1.48) 0.533 0.0547 9.854* 0.081 0.666408 0.055731 12.813* 3.753** 0.561 0.046 9.567* 2.373 0.487 0.0508 5.935* 3.179*** 0.862 0.025 29.22* 2.185 28.514* 5.403* 17.802* 13.269* 25.929* 2.942* 10.789* 2.964* 52.831* 3.562* Note: Values in brackets are t-values. *,**,*** denote significance at the levels of 1%, 5% and 10% respectively. We also tested whether estimated savings-retention coefficients are different from zero using the Wald test. As seen from the table, except for the 1963-80 and 1981-1997 periods, the savings-retention coefficients are not different from zero statistically, implying the presence of perfect capital mobility. Finally, it is seen that both cross-section and period fixed effects contribute statistically significantly to the explanation of the dependent variable. In particular, significant period effects imply the existence of close connections among countries of the MENA region over time. Considering the economic policies followed by the countries in the MEAN region, these findings provided above seem to be suspicious. As we know, most of these economies are closed economies, their financial markets are not developed and they have undertaken liberalization policies very recently. In this sense, we can argue that the findings of this study are in agreement with the findings 30 Journal of Economic and Social Studies �Measuring the Level of International Capital Mobility for MENA Countries of the literature. As mentioned above, the level of capital mobility is found to be higher in relatively closed economies than open economies, in countries with inefficient financial markets than those with financially developed markets, etc. Since some of the MENA countries are resource-rich countries, they accumulate enormous savings; since they do not have financially developed markets, they need foreign capital and aid. Taking these together, we conclude that capital mobility is high in the MENA region. Conclusion In this study, we attempted to measure the degree of capital mobility in the MENA region. Time series properties of the data investigated using panel unit root tests indicated that both variables of interest are stationary. Therefore, we estimated the Feldstein-Horioka equation with fixed effect panel methodology. The findings of the study provided a number of important insights into the level of capital mobility in the region. First, characteristics of sub-periods are very different from each other in terms of the level of capital mobility. While the capital mobility is relatively low during the period of 1963-1980, it is pretty high during the period of 1981-2007. Secondly, the results imply the presence of perfect capital mobility in the period of 1981-2007. Thirdly, although the sub-periods of 1981-1997 and 1998-2007 are slightly different from each other, it seems that the perfect capital mobility assumption holds in these periods as well. References Apergis, N. &Tsoumas, C. (2009). A survey of the Feldstein-Horioka puzzle: What has been done and where we stand. Research in Economics, 63, 64-76.doi:10.1016/j.rie.2009.05.001 Bahami-Oskooee, M. &Chakrabarati, A. (2005). Openness, size and the saving-investment relationship. Economic Systems, 29, 289-293.doi:10.1016/j.ecosys.2005.06.001 Bangake, C. &Eggoh, J. (2010). International capital mobility in African countries: Do the legal origins matter?. Economics Bulletin, 30, 1-10. Choi, I. (2001). Unit root tests for panel data. Journal of International Money and Finance, 20, 249–272.doi:10.1016/S0261-5606(00)00048-6 Coakley, J. &Hasan, F. & Smith, R. (1999). Saving, investment and capital mobility in LDCs. Review of International Economics, 7, 632-640.doi:10.1111/1467-9396.00188 Feldstein, M. &Horioka, C. (1980). Domestic saving and international capital flows. Economic Journal, 90, 314-329.doi:10.2307/2231790 Ghosh, A. R. &Ostry, J. D. (1995). The current account in developing countries: A perspective from Volume 1 Number 1 January 2011 31 �the consumption smoothing approach. The World Bank Economic Review, 9, 305-333.doi:10.1093/ wber/9.2.305 Im, K. S., Pesaran, M. H., & Shin, Y. (2003). Testing for unit roots in heterogeneous panels. Journal of Econometrics, 115, 53–74.doi:10.1016/S0304-4076(03)00092-7 Kasuga, H. (2004). Saving-investment correlations in developing countries. Economics Letters, 83, 371-376.doi:10.1016/j.econlet.2003.11.017 Levin, A., Lin, C. F., & Chu, C. (2002). Unit root tests in panel data: Asymptotic and finite-sample properties. Journal of Econometrics, 108, 1–24.doi:10.1016/S0304-4076(01)00098-7 Maddala, G. S. & Wu, S. (1999). A Comparative Study of Unit Root Tests with Panel Data and A New Simple Test. Oxford Bulletin of Economics and Statistics, 61, 631–52.doi:10.1111/14680084.61.s1.13 Murthy, N. R. Vasudeva (2009). The Feldstein–Horioka puzzle in Latin American and Caribbean countries: a panel cointegration analysis. Journal of Economics and Finance, 33(2), 176-188. doi:10.1007/s12197-008-9051-5 Ozmen, E. (2005). Macroeconomic and institutional determinants of current account deficits. Applied Economics Letters, 12, 557-560.doi:10.1080/13504850500120714 Payne, J. &Kumazawa, R. (2006). Capital mobility and the Feldstein-Horioka puzzle: Reexamination of less-developed countries. The Manchester School, 74, 610-616.doi:10.1111/j.14679957.2006.00512.x Sinha, T. &Sihna, D. (2004). The mother of all puzzles would not go away. Economics Letters, 82, 259-267.doi:10.1016/j.econlet.2003.06.002 Wong, D. Y. (1990). What do saving-investment relationship tell us about capital mobility? Journal of International Money and Finance, 9, 60-74.doi:10.1016/0261-5606(90)90005-K � Dublin Core The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/. Extent The size or duration of the resource. 1053 Title A name given to the resource Measuring the Level of International Capital Mobility for MENA Countries Author Author KAPLAN, Muhittin KALYONCU, Hüseyin Abstract A summary of the resource. To achieve sustainable development, it is vitally important to sustain macroeconomic stability, which is closely related to the extent of capital mobility allowed by a country. This paper attempts to measure the level of international capital mobility empirically by estimating the Feldstein-Horioka coefficients employing the panel data for the MENA countries over the period 1963-2007. In empirical analysis, time series properties of the data are examined using recently developed techniques of panel unit root. Having obtained that variables of the model are stationary variables, we use the fixed effect panel model in the analysis of data.The results indicate that capital mobility has always been high in MENA countries but this is particularly obvious for the period 1980-2007, which corresponds to the liberalization period. For the subperiod of 1963-1980, the estimated coefficients are relatively higher, implying the presence of a relatively lower level of capital mobility. Publisher An entity responsible for making the resource available International Burch University Date A point or period of time associated with an event in the lifecycle of the resource 2011-01 Keywords Keywords. Article PeerReviewed Identifier An unambiguous reference to the resource within a given context ISSN 1986 – 8502 H Social Sciences (General),HB Economic Theory,HG Finance https://omeka.ibu.edu.ba/files/original/f43b49a2e7cfbe6f379294d80381be55.pdf 9edca5e109172d6859379d13400658cb PDF Text Text Journal of Economic and Social Studies Paradigm Shift for Sustainable Development: The Contribution of Islamic Economics Z. Hafsa ORHAN ASTRÖM International University of Sarajevo (IUS), Strategic Analysis and Risk Assessment (SARA) Program, Sarajevo, Bosnia and Herzegovina. hafsaorhan82@hotmail.com ABSTRACT Sustainable development is a common concept of the 21st century. However, the expected changes towards sustainable development are slow. We believe every change starts with changes in understanding of the subject matter. If sustainable development is the aim, it should start with changes in understanding. This paper aims to explain the necessary paradigm shift for sustainable development by the contribution of Islamic economics. While doing this, the reasons of paradigm shift, the content of such a paradigm shift, and the possible contributions of Islamic economics will be analyzed. Keywords: Sustainable Development; Paradigm Shift; Islamic Economics Volume 1 Number 1 January 2011 73 �Z. Hafsa ORHAN ASTRÖM Introduction The concept of ‘sustainable development’ started to appear in the 1970’s and became a frequent hot topic for discussions about world politics, economy and the environment. However, the fulfillment of the aim of sustainable development is yet to happen. There are different reasons for this unfulfillment, such as the lack of political will, the effect of financial crises, etc. But we believe the problem is at a deeper level, i.e., at the paradigm level. The aim of this paper is to discuss the possible contribution of Islamic economics to the necessary paradigm shift for the aim of sustainable development. In this sense, there are two hypotheses: H1: A paradigm shift is necessary for the aim of sustainable development H2: Islamic economics can contribute to the paradigm shift for the aim of sustainable development While trying to find out the validity of the above hypotheses, we will analyze the reasons for the necessity of a paradigm shift, the structure of a possible paradigm shift, and why and how Islamic economics can contribute to such a paradigm shift. The next section will give background information on sustainable development, the paradigm shift concept and Islamic economics. The third section will be about the relationship between sustainable development and Islamic economics. The last section will conclude the paper. Literature Review To be able to analyze the possible contributions of Islamic economics to the paradigm shift for sustainable development, some concepts should be clear first. The three basic concepts of this paper will be introduced. These concepts are sustainable development, paradigm shift and Islamic economics. The concept of sustainable development started to appear in the 1970s. The reason behind the emergence of such a concept was the growing awareness of the depletion of natural resources, and worsening environmental conditions on the one hand and increasing world population on the other hand, i.e., demand-supply discrepancy. Despite the common use of the concept, there is a lack of an explicit definition. One of the first attempts towards the definition of this concept was done by World Conservation Strategy (1980): “For development to be sustainable, it must take account of social and ecological factors, as well as economic ones; of the living and non-living resource base; and of the long-term as well as the short-term advantages and disadvantages of alternative action.” However, today the most commonly used definition is the one established by the World Commission on Environment and Development (1987): “Economic and social development that meets the needs of the current generation without undermining the ability of future generations to meet their own needs.” This definition can sound vague since it does not indicate what is meant by ‘meeting the needs of the current generation’ and ‘without undermining the future generations’ ability of meeting 74 Journal of Economic and Social Studies �Paradigm Shift for Sustainable Development: The Contribution of Islamic Economics their own needs’. At this point, Dalal-Clayton and Bass (2002) explain what they actually mean. According to them, meeting the needs of the present generation covers the economic needs, social, cultural and health needs plus the political needs. On the other hand, “without undermining the future generations’ ability to meet their own needs” refers to minimising usage, the sustainable use of renewable resources, and keeping within the absorptive capacity of local and global sinks for wastes. Efforts to maintain sustainable development include treaties and establishment of rules, laws and regulations. The actors taking part in these efforts are quite diverse, such as intergovernmental organizations, non-governmental organizations, nations, international organizations, the private sector and civil society. One of the earliest efforts in the global arena was the Stockholm Conference held in 1972. However, the first comprehensive attempt came with the Rio Declaration on Environment and Development (1992), issued as a result of the United Nations Rio Conference. The declaration covers 27 principles. The following Earth Summit was in South Africa. It ended with the Johannesburg Declaration (2002) where the commitment to sustainable development is mentioned once more and the challenges were indicated. Additionally, three sets of goals were established due to the three different time horizons: the short-term (for 2015) goals of the Millennium Declaration of the United Nations; the two-generation goals (for 2050) of the Sustainability Transition of the Board on Sustainable Development; and the long-term (beyond 2050) goals of the Great Transition of the Global Scenario Group. Another declaration was released by the International Law Association Committee on the legal aspects of sustainable development, called the New Delhi Declaration on the Principles of International Law Related to Sustainable Development (2002). The declaration includes 7 principles which are based on the previous declarations mentioned above. Some of these principles are the duty of states to ensure sustainable use of natural resources, the principle of equity and the eradication of poverty, and the principle of common but differentiated responsibilities. Another important document is the Kyoto Protocol (1997) which was adopted with a direct focus on global warming. Heretofore, around 200 countries have signed the protocol, which includes 28 articles. Efforts are not limited to the global summits. There are also efforts at the regional or state level. The fourth Asia-Pacific Forum for Environment and Development (2003) summarizes the efforts done in the region, e.g., a sub-regional cooperation in Southeast Asia, the ASEAN Regional Centre for Biodiversity Conservation established in the Philippines in 1999 to coordinate ASEAN initiatives on biodiversity conservation, biodiversity loss and degradation in the region; and the Regional Environment Programme (SPREP) established in 1982 by the governments of the South Pacific region to protect the regional environment. According to the report of the Minister of Public Works and Government Services (2001), the Government of Canada Action Plan 2000 on Climate Change outlines cost-effective measures that will take Canada one-third of the way to its Kyoto target. Despite all these efforts, there are different problems and challenges ahead for the aim of sustainable development. Some of the basic problems are lack of conformity among different actors who have their own agendas, properties and capabilities, financial inadequacies and lack of legal enforcement. However, there is another problem which we believe should attract basic concern, i.e. paradigmatic problems. Woods (2002) argues that sustainable development confers a contested paradigm since the term ‘sustainability’ is vague in itself and more importantly, it lacks an approach that tackles the Volume 1 Number 1 January 2011 75 �Z. Hafsa ORHAN ASTRÖM irreconcilability among economic, social and environmental dimensions. Henceforth, we argue that to be able to tackle such a problem, a paradigmatic shift is necessary. Paradigm and Paradigm Shift Literally, a paradigm is “a typical example or pattern of something; a model” (Hobson, 2004). The prominent figure known for the idea of paradigm shift is Kuhn. He identifies paradigm as “... what the member of scientific community, and they alone, share” (Kuhn, 1970). The idea of paradigm shift is a cyclic process which starts with an already settled paradigm. In time, anomalies and crises emerge as a natural process. As a response to these crises, scientific discoveries begin which result in scientific revolutions through paradigmatic shift. A typical paradigm shift cycle follows the scheme below: Normal science--> Model Drift--> Model Crisis--> Model Revolution--> Paradigm Change Islamic Economics Islamic economics has medieval roots, including an immense literature from Muslim scholars such as Al-Ghazali, Ibn-Khaldun, and Ibn-Qayyim. On the other hand, a better-organized contemporary Islamic economics paradigm flourished in the second half of the twentieth century, especially with the de-colonization of Muslim countries. Islamic economics is built upon the knowledge coming from the basic sources of the religion of Islam which are Quran and sunnah -sayings and living habits of the prophet Mohammed-, plus the accumulated knowledge of Islamic jurisprudence generated by consensus (ijma), analogy (qıyas) and independent interpretation (ijtihad). According to the well-known contemporary definition, economics is the science of allocating scarce resources due to unlimited wants. The difference in Islamic economics starts from this point, where the assumption is the lack of absolute scarcity (stock) even if there is relative scarcity (flow) in this world. According to a comprehensive definition (Ahmed, 2002) “Islamic economics is the science that studies the best possible use of all available economic resources, endowed by Allah, for the production of maximum possible output of Halal goods and services that are needed for the community now and in future and the just distribution of this output within the framework of shariah and its intents.” Rules and regulations in Islam follow the objectives of public welfare (maslahah) which are categorized as preservation of life, property, religion, reason and procreation. Asutay (2007) describes Islamic economics as a ‘system’ which owns its framework paradigm, value system, foundational axioms – such as doctrine of oneness (tawhid), justice and charity (adl wa’l-ihsan), self-development (tazkiyah), responsibility (fardh) - operational principles, specific methodology and functional institutions. Because of these peculiarities, Islamic economics is seen as an alternative paradigm (Presley & Sessions, 1994; Zaman, 2005). The primary properties of Islamic economics can be seen below: 76 Journal of Economic and Social Studies �Paradigm Shift for Sustainable Development: The Contribution of Islamic Economics 1. Islamic multi-faceted point of view connecting the different parts of life together, e.g., social, economic, political and religious issues 2. Dominance of a normative approach 3. Importance of social altruism 4. The approach of connecting financial sector and real sector 5. Acceptance of no absolute ownership, no absolute scarcity 6. Criticism of waste Discussion According to the Kuhn cycle, paradigm shifts occur when there are anomalies which need novel explanations. When the concept of sustainable development came into existence, there were anomalies which necessitated a paradigm shift. Such anomalies were increasing inequalities within and among the nations, increasing poverty, especially in developing countries, depletion of the ozone layer, global warming, depletion of some species of animals and plants, water and air pollution, etc. Sustainable development was an effort to change the way of thinking towards the planet. That is why the concept of development is preferred instead of growth, which is believed to reflect only a quantitative aspect of countries without taking into account some other qualitative items such as education, health and equality. However, the success of sustainable development as a new paradigm is yet to be clear. The basic problem arises due to the inconsistencies between its rhetoric and the axioms of the entrenched economic paradigm. In that regard, here comes the possible contribution of Islamic economics which is pronounced as an alternative economic system. Below, the possible contributions of Islamic economics to the paradigm shift of sustainable development will be analyzed according to different issues. The first possible contribution can come from the normative approach of Islamic economics. It is one of the criticisms of contemporary economic theory that it identifies itself as a value-free science. One of the reflections of such an identity can be seen in the determination of efficiency in society. As it is known, in the classical economic theory, economic efficiency is described by Pareto optimality, where it is impossible to make one better without making another one worse off. The problematic part in that approach is the lack of any statement on equity or social well-being. Similar criticisms are made for utilitarianism, the social utility theory of classic economics. According to Van Wyk (2001), the basic criticisms of utilitarianism are the consequentalism where “… the rightness of actions is judged entirely by the goodness of the consequent state of affairs” and welfarism in which “… the goodness of the state of affairs must be judged entirely by the goodness of the set of individual utilities in the respective of state affairs.” It was mentioned before that one of the foundational axioms of Islamic economics is adl wa’l-ihsan, which can be named as just balance or equilibrium where “a ‘maximal’ rate of economic growth must be maintained to satisfy the requirement of intergenerational equity. By the same token, from the many growth paths available, the choice will be restricted to those which satisfy the Islamic ethical constraints.” (Naqvi, 1997) In order to show the difference of the welfare structure in Islamic economics, the below model of Naqvi (ibid.) can be followed. First of all, welfare (W) is a function of u (the average individual utility function) which depends on k(t) (the per capital Volume 1 Number 1 January 2011 77 �Z. Hafsa ORHAN ASTRÖM consumption), C (the average commodity basket) which depends on k(t), x(t) (the per capita capital stock) and t (time). The welfare function can be seen below: (1) On the other hand, there are two constraints, motion and wealth: (2) (3) where, Q2 is the square of the difference of wealth holding in the society. The second constraint means that the tolerance for inequality is dependent on the level of per capita income. B is an upper limit in a constraint maximization problem. The next step is to maximize the first equation due to the constraints of 2 and 3. At the end, we get the equalities: below: (4) (5) According to Naqvi (ibid.), these equalities prove that “in order to maximize aggregate social welfare and the growth in the size of the commodity basket containing the wage goods in a growing economy, the inequality of wealth must be held at a minimum, while capital is priced efficiently.” Indeed, one of the problems with the paradigm of sustainable development is the Sisyphean effort of combining value free economic theory with non-value free social issues. Another possible contribution can be to adopt the altruistic point of view of Islamic economics instead of the self-interest driven, rational economic agent (homo-economicus) theory. In the economic dictionary, homo economicus or economic man is defined as “a person who makes rational decisions in order to achieve their most preferred outcome given the constraints upon choice” (Black, Hashimzade, & Myles, 2009). The most typical aspect of economic man is profit maximization ability. As a result of profit maximization, the equilibrium occurs where (Hasan, 1992): (6) (7) (8) when, (9) (10) 78 Journal of Economic and Social Studies �Paradigm Shift for Sustainable Development: The Contribution of Islamic Economics In these equations, a, b and c are positive constants and R refers to the profit. Meanwhile, equation 9 and 10 show the negative relationship between price (P) and output (Q) and the positive relationship between average cost (AC) and output (Q). It is quite a common criticism that the basic agent of the classic economics, homo-economicus, is a value-free agent. This was the basic reason why Kahneman and Tversky developed a new theory, called Prospect Theory later on, i.e., to include the ideas and values of the people into the economic decision-making system. On the other hand, there is another type of agent called homo-islamicus who “... is said to be both entrepreneurial and moral. He is an Islamic personality who defines his existence by combining private and public life and religious and economic activities through his Islamic ethical values and norms. He does not eschew economic activity and retreat to other-worldly asceticism because of his religion. Nor does he make concessions with regard to his religion and morals for his business activity. He is competitive, productive and innovative, rather than a rent-seeker and speculative. He thinks that being economically successful is a duty of any and every Muslim, as Islam condemns idleness, laziness and encourages hard work and resourcefulness” (Adas, 2006). According to the profit maximization results of a homo-islamicus agent (Hasan, 1992): (11) (12) (13) where, (14) The results of these two profit maximization calculations show that the equilibrium quantity and profit are higher for homo-islamicus agent, showing that the efficiency is higher, while the price is lower. The aforementioned properties of Islamic economics, normative and altruistic approaches, are especially the ones in which classic economic theory has problems with respect to becoming an alternative paradigm of sustainable development. Additional contributory properties of Islamic economics can be mentioned, such as the multi-faceted point of view where the coherence among the social, political, economic and environmental issues is not neglected, and the non-absolutist point of view. Non-absolutist point of view refers to the acceptance of not having absolute freedom and ownership. One of the problematic points of view of today’s generation is that they have the rights of limitless ownership without taking into account the responsibilities towards society and humanity. According to Islamic economy, what people earn is not immune from the rights of other people, e.g., according to the seventieth chapter, twenty third and fourth verses of the Holy Quran (Asad, 2006) “... And in whose wealth there is a right acknowledged for the beggar and the destitute...” It can be argued that there are limitations or constraints in terms of private ownership in other systems Volume 1 Number 1 January 2011 79 �Z. Hafsa ORHAN ASTRÖM besides Islam. However, the sources and contents of such restrictions are different for Islam. For instance; because of the faith in God and the hereafter, Muslim people are obliged to pay a tax called zakah to the needy people as a proof that people have rights on other people’s earnings. On the other hand, as an Islamic law, a private ownership can be limited if there is concern for social well-being. Islam is unique in its idea of ownership since it is seen by many, e.g. A. Kia and M. S. Chaudhry, as lying between capitalism, where private ownership is almost sacred, and socialism, where private ownership is not accepted. The last contributory part can come from the criticism of wasting in Islamic economics in the age of massive commercialization and consumption, e.g., according to the Holy Quran (7:31) “O Children of Adam! Wear your beautiful apparel at every time and place of prayer: eat and drink: But waste not by excess, for Allah loveth not the wasters.” Conclusion In this paper, we discussed why the sustainable development paradigm has not been turned into a paradigmatic shift of Kuhnian cycles. In this regard, we indicated why Islamic economics can contribute such a paradigmatic shift. In doing this, we elaborated on the possible contributory points. At this point, it can be argued that the practice of Islamic economics is not promising, e.g., the backwardness of Muslim communities in terms of economic, political, social and environmental issues. However, the success of applications depends on different aspects such as different comprehensions by different authorities, political power, historical dynamics, necessary institutions etc. Presently, the rejection of Islamic economic paradigm due to the negativeness in applications seems hasty. Another issue is the universal applicability of Islamic economics. First of all, it should be indicated that to get contributions from Islamic economics for the paradigm of sustainable development does not necessarily mean to be Muslim. As can be seen, the above-mentioned possible contributory properties include universally acceptable aspects. The peculiarity of Islamic economics is the combination of these aspects with economic theories. It is true that there are some other alternative points of view trying to put value-based analysis into current economic theory, such as cognitive economics or prospect theory, but none is as comprehensive as Islamic economics yet. Moreover, the criticism of Islamic economics against the current one is deeper than these alternatives. As a comprehensive alternative, socialism can be mentioned. However, its practicality is a big question mark. It can be asked at this point why Islam as an economic system is more viable in practice than socialism. My quick answer is because the former one is more compatible with human nature where private ownership, family and faith are not neglected. References Adas, E. B. (2006). The Making of Entrepreneurial Islam and the Islamic Spirit of Capitalism, Journal for Cultural Research, 10, 113-137. Ahmed, H. (2002). A Microeconomic Model of an Islamic Bank. Jeddah: Islamic Development Bank. 80 Journal of Economic and Social Studies �Paradigm Shift for Sustainable Development: The Contribution of Islamic Economics Asad, M. (2006). The Message of the Quran. Istanbul: Isaret Yay. Asutay, M. (2007). A Political Economy Approach to Islamic Economics: Systemic Understanding for an Alternative Economic System, Kyoto Bulletin of Islamic Area Studies, 1, 3-18. Black, J., & Hashimzade, N., & Myles, G. (2009). Oxford Dictionary of Economics. Oxford: Oxford University Press. Dalal-Clayton, B., & Bass, S. (2002). Sustainable Development Strategies: A Resource Book. London: Earthscan Ltd. Hasan, Z. (1992). Profit Maximisation: Secular versus Islamic. In S. Tahir et al (Eds.), Readings in Microeconomics: An Islamic Perspective (pp. 239-255). Malaysia: Longman. Hobson, A. (2004). The Oxford Dictionary of Difficult Words. Oxford: Oxford University Press. International Union for Conservation of Nature and Natural Resources & United Nations Environment Programme & World Wildlife Fund (1980). World Conservation Strategy. IUCNUNEP-WWF. United Nations Documents, Johannesburg Declaration on Sustainable Development. Retrieved May 20, 2010, from http://www.un-documents.net/jburgdec.htm Kuhn, T. S. (1970). The Structure of Scientific Revolutions. Chicago: The University of Chicago Press. Kyoto Protocol to the United Nations Framework Convention on Climate Change (1997). Retrieved on June 2010 from http://unfccc.int/resource/docs/convkp/kpeng.html Government of Canada (2000). Action Plan 2000 on Climate Change (ISBN: 0-662-29444-0). Retrieved on June 2010 from file:///tmp/M22-135-2000E.pdf Naqvi, S. N. H. (1997). The Dimensions of an Islamic Economic Model, Islamic Economic Studies, 4, 1-24. International Law and Agreements (ILA) New Delhi Declaration of Principles of International Law Relating to Sustainable Development (2002, April). International Environmental Agreements: Politics, Law and Economics, 2, 211-216. Presley, J. R., & Sessions, J. G. (1994). Islamic Economics: The Emergence of a New Paradigm, The Economic Journal, 104, 584-596. United Nations Environment Programme (UNEP). Rio Declaration on Environment and Development. Volume 1 Number 1 January 2011 81 �Retrieved on May 2010 from http://www.unep.org/Documents.multilingual/Default.asp?Documen tID=78&ArticleID=1163 Van Wyk, M. W. (2001). Equal Opportunity and Liberal Equality. (Unpublished doctoral dissertation). The University of Johannesburg, Johannesburg. Woods, D. (2002). Sustainable Development A Contested Paradigm: Economics Forum of the Foundation for Water Research. Birmingham, UK. Zaman, A. (2005). Towards a New Paradigm for Economics, 18, 49-59. � Dublin Core The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/. Extent The size or duration of the resource. 1059 Title A name given to the resource Paradigm Shift for Sustainable Development: The Contribution of Islamic Economics Author Author ORHAN ASTRÖM, Z. Hafsa Abstract A summary of the resource. Sustainable development is a common concept of the 21st century. However, the expected changes towards sustainable development are slow. We believe every change starts with changes in understanding of the subject matter. If sustainable development is the aim, it should start with changes in understanding. This paper aims to explain the necessary paradigm shift for sustainable development by the contribution of Islamic economics. While doing this, the reasons of paradigm shift, the content of such a paradigm shift, and the possible contributions of Islamic economics will be analyzed. Publisher An entity responsible for making the resource available International Burch University Date A point or period of time associated with an event in the lifecycle of the resource 2011-01 Keywords Keywords. Article PeerReviewed H Social Sciences (General),HB Economic Theory,HG Finance https://omeka.ibu.edu.ba/files/original/51e3553dcfd1e2bc534c1725dd5fc425.pdf a52b502341316ca56b9704c2f0628bf3 PDF Text Text Journal of Economic and Social Studies CONFERENCE NOTES AND REPORTS Symposium Notes on Ethics and Social Responsibility, 14th and 15th of April 2011 ISCTE – Lisbon University Institute Maja SAVEVSKA Université Libre de Bruxelles Institut d’Études Européennes Maja.Savevska@ulb.ac.be This much awaited symposium brought together practitioners and scholars involved in some aspect of the broad concept of social responsibility. Although not necessarily limited to, the symposium mainly focused on the individual and corporate responsibility in different organizational settings. The symposium was roughly divided in panel and poster sessions covering a wide range of topics, such as: corporate misconduct, individual ethics in organizations, ethical decision making, business ethics, ethics and organizational performance etc. The symposium was opened and closed by two key note speakers, Prof. Daniel Arenas Vives and Prof. Jan Jonker respectively. Both professors delivered inspiring reflections on the need for a sustainable business practices. Overwhelming number of researchers coming from Portugal, UK, Turkey, Brasil, USA, Bosnia and Herzegovina, Italy, Belgium, Denmark, Netherlands, Spain, presented their findings at the symposium. We witnessed a lot of collaborative efforts among researchers with the aim of providing a large scale analysis of across sectoral and industry specific socially responsible initiatives. The reappearing theme that deserves further explanation was Corporate Social Responsibility (CSR herein), understood in its widest terms. Originating from the business and management literature in the 1970s, the concept of CSR has recently gained impetus thanks to globalization. Although we lack a scientific consensus on the exact Volume 1 Number 2 July 2011 155 �Maja SAVEVSKA definition of the concept of CSR, we can agree at least on three essential constituents that define the membership boundary of the CSR concept: (i) its voluntary nature, (ii) its multi-stakeholder participation, (iii) its objective to balance between profit orientation and social concerns. The symposium styled itself as a forum where both practitioners and researchers explore empirical and theoretical aspects of the CSR regime. The numerous research projects presented at the symposium were significantly skewed in favor of eclectic case studies as opposed to theoretical exploration of the state of art or meta-CSR empirical analysis. Notwithstanding the variations, once could discern three major research streams at this symposium: one focused on consumer attitudes, another on corporate mis/conduct and employee’s perception of it, and the third on the impact of the CSR strategies, be it internal or external. For the purpose of this report, I would have to single out presentations that captured my attention and would illustrate the diversity of case studies. Within the first research stream Selin Türkel and Burcu Öksüz from Izmir University of Economics and Ana Patrícia Duarte and Carla Mouro from ISCTE presented innovative ways of measuring consumer attitudes towards CSR in Turkey and Portugal respectively. In their ongoing study, Türkel and Öksüz constructed experiments where they control for different communication platforms in order to explore their effects on the consumer attitudes. While the completed survey of Duarte and Mouro demonstrated that the external CSR initiatives have significantly greater effect on consumer behavior than the internal initiatives. The second stream that emerged at the symposium was focusing on the corporate conduct and the perception of the employees. The joint effort of Ana Patrícia Duarte and José Conçalves des Naves explored the effects of employees’ perception of the company’s socially responsible behavior on the employees’ commitment. Susana Leal, Arménio Rego and Arnaldo Coelho gave special contribution in that they provided a psychological twist to the research by looking at the way the employees’ perceptions mediated through the psychological capital influence the Organizational Citizenship Behavior (OSB).1 In similar lines, Raquel Matos and Eduardo Simões explored the effects of the ethical climate on the OCB. It is worthwhile mentioning the research of Chiara Mio and Alvise Favotto from University of Venice who looked at the differences in perceptions along the hierarchy line of a company. The third stream of analysis was exploring the effects of the CSR initiatives. Particularly interesting was the study of Maria Vieira de Melo and Diego César de Vasconcelos 1 The employees conduct, apart from their specific job description that influences the operation within the company. 156 Journal of Economic and Social Studies �CONFERENCE NOTES AND REPORTS / Symposium Notes on Ethics and Social Responsibility who explored the effects of the CSR practices in the construction industry on the industrialization and the environment in Brazil. Valuable conclusion was also reached by Ana Patrícia Duarte and Sónia Conçalves who controlled for the internal and the external CSR strategies. In turn, they detected a positive impact of the internal CSR strategies on the employees’ conduct. It was especially rewarding seeing how the concept of CSR has left its European hearthland and gained sufficient impetus in other regions. Mohammet Sait Dinc and Teoman Duman study on the employees’ perception of the marketing strategies in Bosnia and Herzegovina was a case in point from the Balkans. Pelin Baytekin and Deniz Maden analysis of the nexus of CSR and education where they investigated Turkcell’s CSR initiative was yet another promising example. It has been rewarding to learn that many companies have realized the need to reassert their competitive position by improving their Corporate Citizenship role within the communities. One presentation that deserves mention in this report was that of Robson Sø Rocha who challenged some of the tenets of Varieties of Capitalism scholarship. In his study, Rocha was exploring the survival strategies of liberal market economic actors that operate within a coordinated market economy setting. Namely, his case study of an equity firm that bought the Danish TDC represents a valuable contribution to the neo-institutional theories. To the great pleasure of those who put more emphasis on theory, myself included, the symposium offered a theoretical panel where Rosa Slegers and me provided Aristotelian and constructivist twist to the debate. Slegers put forward the idea of a virtue-based approach to business ethics by using insights from evolutionary biology. Myself, I tried to provide a critical reading of the existing theoretical interpretations of CRS and argued for a constructivist understanding of the CSR potential to foster good corporate behavior. This report, by singling out papers, cannot do justice to the tremendous effort put by all the scholars whose attendance was highly appreciated. The vast number of undertaken studies is indicative of the importance that both the academic and the practitioner communities place in the concept of social responsibility. Although we lack a meta-CSR analysis and the jury is still out there to judge the impact of the CSR regime, the symposium brought us a step further in our understanding of the CSR regime. One can only hope that the critical theorists would be proven wrong in their assessment of CSR as an oxymoronic concept that cannot replace the good old regulations’ approach. Volume 1 Number 2 July 2011 157 � Dublin Core The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/. Extent The size or duration of the resource. 1076 Title A name given to the resource CONFERENCE NOTES AND REPORTS Symposium Notes on Ethics and Social Responsibility, 14th and 15th of April 2011 ISCTE – Lisbon University Institute Author Author SAVEVSKA, Maja Abstract A summary of the resource. This much awaited symposium brought together practitioners and scholars involved in some aspect of the broad concept of social responsibility. Although not necessarily limited to, the symposium mainly focused on the individual and corporate responsibility in different organizational settings. The symposium was roughly divided in panel and poster sessions covering a wide range of topics, such as: corporate misconduct, individual ethics in organizations, ethical decision making, business ethics, ethics and organizational performance etc. Publisher An entity responsible for making the resource available International Burch University Date A point or period of time associated with an event in the lifecycle of the resource 2011-01 Keywords Keywords. Article PeerReviewed H Social Sciences (General),HB Economic Theory,HG Finance https://omeka.ibu.edu.ba/files/original/e01ac07689cc617a778c7be0b6563d6b.pdf 716cee5a42f2edd8aecb2e193fb0a560 PDF Text Text Journal of Economic and Social Studies The Day-of-the-Week Effect in the Saudi Stock Exchange: A Non-Linear Garch Analysis Talat ULUSSEVER Department of Finance and Economics King Fahd University of Petroleum and Minerals, Dhahran, Saudi Arabia, talat@kfupm.edu.sa Ibrahim GURAN YUMUSAK Department of Economics Kocaeli University, Izmit, Turkey, iyumusak@kocaeli.edu.tr Muhsin KAR Department of Economics Cukurova University, Adana, Turkey mkar@cu.edu.tr ABSTRACT It is a well-known fact that the day-of-the-week effect in stock markets is one of the most prominent puzzling seasonal anomalies in finance and has been increasingly attracting attention from researchers and practitioners, as well as academics. This paper scrutinizes the day-of-theweek effect in the emerging equity market of Saudi Arabia, TADAWUL. By using a non-linear GARCH model and covering the data from January 2001 to December 2009, the findings of the study reveal that the returns on the five trading days follow different process. This confirms that mean daily returns are significantly different from each other and validates the day-of-theweek effect in TADAWUL. Keywords: Day of the week effect; GARCH; Saudi stock exchange Volume 1 Number 1 January 2011 9 �Talat ULUSSEVER & Ibrahim GURAN YUMUSAK & Muhsin KAR Introduction Financial markets have witnessed the presence of calendar anomalies, which have been documented extensively for the last two decades. The most prominent ones are definitely the January Effect and the Day of the Week Effect. The day of the week effect in stock markets has been attracting attention from researchers and practitioners, as well as academics and thus has been investigated extensively in different markets. Cross (1973), French (1980), Keim and Stambaugh (1984) Rogalski (1984), Aggarwal and Rivoli (1989) studied the week effect in different stock markets and revealed that the distribution of stock returns varies according to the day of the week. For example, they generally found that the average return on Monday is significantly less than the average return over the other days of the week. The day of the week regularity is not limited to a few equity markets. It is well documented that the day of the week regularity is present in other international equity markets (Jaffe and Westerfield 1985; Solnik and Bousquet 1990; Barone 1990, among others) and other financial markets including the futures market, Treasury bill market, and bond market (Gibbons and Hess, 1981; Cornell 1985; Dyl and Maberly 1986). Although the majority of the studies has centered on the seasonal pattern in mean return, many recent empirical studies have also tried to investigate the time series behavior of stock prices in terms of volatility by using variations of the generalized autoregressive conditional heteroscedasticity (GARCH) models (French, Schwert, and Stambaugh 1987; Akgiray 1989; Baillie and DeGennaro 1990; Hamao, Masulis, and Ng 1990; Nelson 1991). French and Roll (1986) proposed that the variances for the days following an exchange holiday should be larger than other days. Harvey and Huang (1991) observed higher volatility in the interest rates and foreign exchange futures markets during the first trading hours on Thursdays and Fridays. Needless to say, it is important to know whether there are variations in volatility of stock returns by day-of-the-week patterns and whether there is a connection between high (low) return and a corresponding high (low) return for a given day. Obviously, having such knowledge may allow investors to adjust their portfolios by taking into account day of the week variations in volatility. For instance, Engle (1993) argued that investors who dislike risk may adjust their portfolios by reducing their investments in those assets whose volatility is expected to increase. Finding definite patterns in volatility may be helpful in many ways, including but not limited to the use of predicted volatility patterns in hedging and speculative purposes and use of predicted volatility in valuation of certain assets, specifically stock index options. The-day-of-the week effect is regularity in the stock market that usually takes the form of considerably negative mean returns on the first day of the trading week and peculiarly high mean returns on the last day of the trading week. Settlement procedures, bid-ask spread biases, dividend patterns, negative information release, thin trading, measurement errors, specialists behavior, and the concentration of certain investment decisions at the weekend have been considered as partially main factors of the day of the week effect phenomenon by several studies like Cross (1973), French (1980), Gibbons and Hess (1981), Lakonishok and Levi (1982), Kein and Stanbaugh (1984), Rogalski (1984), Jaffe and 10 Journal of Economic and Social Studies �The Day-of-the-Week Effect in the Saudi Stock Exchange: A Non-Linear Garch Analysis Westerfield (1985), Smirlock and Starks (1986), Penman (1987), Damodaran (1989), Al-Loughani and Chappell (2001) and Tonchev and Kim (2004). The purpose of this study is to investigate the presence of day-of-the-week effect in emerging stock market of Saudi Arabia, TADAWUL. To the best of our knowledge, there is no previous study that has tested the presence of the day-of-the-week effect in TADAWUL. The paper contributes to the literature by documenting the presence of the day-of-the-week effect patterns by using non-linear GARCH analysis in TADAWUL, which has not been investigated by any earlier studies. Literature Review There is a huge literature on day-of-the-week effect for the stock returns. Among studies investigating the day-of-the-week anomaly for the U.S. market, Cross (1973) studied the returns on the S&P 500 Index over the period of 1953 and 1970. His findings showed that the mean return on Friday is higher than the mean return on Monday. French (1980), who also studied the S&P 500 index for the period from 1953 to 1977, revealed similar results. Gibbons and Hess (1981) found negative Monday returns for 30 stocks of Dow Jones Industrial Index. Keim and Stambaugh (1984) examined the weekend effect by using longer periods for diverse portfolios. Their results confirmed the findings of previous studies. There are many studies that try to explain the Monday effect. We can cite, among them but not limited to, calendar time hypothesis (French 1980), the delay between trading and settlements in stocks (Gibbons and Hess 1981; Lakonishok and Levi 1982), and measurement errors (Gibbons and Hess 1981; Keim and Stambaugh 1984). These studies mainly measure Monday return between the closing price on Friday and the closing price on Monday. Rogalski (1984) tried to respond to the question of whether prices fall between Friday close and Monday opening or during the day on Monday. He incorporated daily returns into trading and non-trading day returns and discovered that all of the average negative returns from Friday close to Monday close take place during the non-trading hours. Average trading day returns (open to close) are alike for all days. Other U.S. markets are not exceptions to day-of-the-week patterns. The Treasury bill market, the futures market, and the bond market present a similar pattern to that of the equity market (Cornell 1985; Dyl and Maberly 1986). Several studies showed that other stock markets around the world have also witnessed the day-of-the-week effect. Among them, Jaffe and Westerfield (1985) scrutinized the weekend effect in four developed markets, namely Australia, Canada, Japan and the U.K. Their results indicated the presence of the weekend effect in all countries studied. In contrast to earlier studies of the U.S. market, surprisingly, the lowest mean returns for both Japanese and Australian stock markets were found to be on Tuesday. Solnik and Bousquet (1990) investigated day-of-weekeffect for Paris stock exchange, and revealed a strong and persistent negative return on Tuesday, which is in line with studies on Australia and Japan. Barone (1990) exposed similar results for the Italian stock market, with the biggest decline in stock prices taking place in the first two days of the week and more pronounced on Tuesday. Furthermore, Agrawal and Tandon (1994), Alexakis and Xanthakis (1995), and Balaban (1995) also showed that the distribution of stock returns varies by day-of-the-week for various countries. Overall, the day-of-the-week effect in stock returns is a Volume 1 Number 1 January 2011 11 �Talat ULUSSEVER & Ibrahim GURAN YUMUSAK & Muhsin KAR common phenomenon and has been documented in different countries and different stock markets. Some empirical studies examined the time series behavior of stock prices in terms of volatility by using variations of GARCH models (French, Schwert, and Stambaugh 1987; Akgiray 1989; Baillie and DeGennaro 1990; Hamao, Masulis, and Ng 1990; Nelson 1991; Campbell and Hentschel 1992; Glosten, Jagannathan, and Runkle 1993). French, Schwert and Stambaugh (1987) studied the relationship between stock prices and volatility and confirmed that unexpected stock market returns are negatively correlated with unexpected changes in volatility. Campbell and Hentschel (1992) revealed similar findings. They showed that an increase in stock market volatility increases required stock returns, and thus decreases stock prices. Nelson (1991) and Glosten, Jagannathan, and Runkle (1993), in contrast, found that positive unanticipated returns brought about reduction in conditional volatility, while negative unanticipated returns caused upward movements in conditional volatility. Baillie and DeGennaro (1990) reported no evidence of a relationship between mean returns on a portfolio of stocks and the variance or standard deviation of those returns. These findings were also confirmed by Chan, Karolyi and Stulz (1992), who reported a significant foreign influence on the time-varying risk premium for U.S. stocks but no significant relation between the conditional expected excess return on the S&P 500 and its conditional variance. Moreover, Corhay and Rad (1994) and Theodossiou and Lee (1995) examined the behavior of stock market volatility and its relationship to expected returns for major European stock markets. Both studies displayed the presence of significant conditional heteroscedasticity in stock price behavior found no relationship between stock market volatility and expected returns. The Saudi Stock Exchange The Saudi stock exchange, known as the TADAWUL, is the largest not only in the Gulf Community Council (GCC) countries, but also in the entire Arab World. By December 2009, its market capitalization was around $313 billion. The next largest is the Kuwait stock exchange, which had a market cap of $94 billion. As a percentage of GDP, the TADAWUL’s market cap was around 67% of 2008 GDP and around 82% of 2009 GDP. It is technologically advanced, and introduced the world’s first fully-electronic market in the 1990s, comprising trading, clearing, settlement and depository (The Saudi Stock Market: Structural Issues, Recent Performance and Outlook, December, 2009, SAMBA.) The main index, the TADAWUL All Share Index (TASI) reached its peak on 25th of February 2006, when it closed at 20,635. It was severely affected by the 2008 global crisis, like all the stock markets all over the world, and saw below 4000. It is currently trading around 6300. 12 Journal of Economic and Social Studies �The Day-of-the-Week Effect in the Saudi Stock Exchange: A Non-Linear Garch Analysis Figure 1. TADAWUL All Share Index for the last 5 years Source: Gulfbase.com Figure 2. TADAWUL All Share Index for the Last 3 Months Source: Gulfbase.com The Saudi Arabia Monetary Agency (SAMA) was responsible for supervising the market from 1984 until 2003. In July 2003, authority was handed over to the newly formed Capital Market Authority (CMA). The CMA is now the sole regulator and supervisor of Saudi Arabia’s capital markets, and issues the necessary rules and regulations to protect investors and ensure fairness and efficiency in the market. For many years, the TADAWUL was open only to Saudi nationals. In December 2007, as part of the move to establish a GCC common market, the TADAWUL was opened to GCC nationals, though their participation remains limited as they have tended to focus on their domestic markets. Until 2008, non-Arab foreigners who were not resident in the Kingdom could only participate through a few mutual funds. However, in August 2008 the CMA approved new rules that allowed non-Arab foreigners to participate in share trading through swap arrangements with local CMA-approved and Volume 1 Number 1 January 2011 13 �Talat ULUSSEVER & Ibrahim GURAN YUMUSAK & Muhsin KAR licensed intermediaries. The Saudi stock market currently lists 138 companies, divided into fifteen sectors. Financials and Basic Materials sectors are the dominant sectors, together accounting for around 70% of market capitalization. The biggest two companies by market share are Al RAJHI Bank and SABIC, a petrochemical producer, both of which command around 11% of the market. Some of the smaller sectors have larger numbers of companies: for example, the Consumer Goods sector contains 16 companies, despite accounting for just 4% of the market’s value. GCC and other Arab citizens accounted for 3% of buys, while foreign residents in the Kingdom registered just 0.2%. Foreign residents outside the Kingdom placed 1.2% of buy orders with a small number of transactions. Between 2003 and its peak in February 2006, the index gained a staggering 700%, with market capitalization soaring to $800 billion - around two-and-a-half times nominal GDP. At its peak, the TADAWUL was the world’s tenth largest stock market by value, despite having only 78 listed stocks, many with a limited free float. In July 2009, the US Dow Jones Index became the first international index provider to offer indices on the TADAWUL. Dow is now providing four Saudi indices based on real time data and prices from the Kingdom. Standard & Poor’s and Bloomberg have also reached similar agreements to provide indices. The run-up in the stock market during the middle part of the decade saw the TASI soar well above global equity benchmarks as speculators ignored fundamentals and gambled that prices would keep on rising. The subsequent crash saw the TASI lag behind global benchmarks for over a year. Since the beginning of 2008, the index has basically realigned itself with the direction of global equity markets. This realignment did not prevent another serious period of turbulence in 2008. Surging global equity markets and oil prices in the first part of the year prompted a spike in activity on the TADAWUL. However, this was followed by an abrupt collapse in the second half as the global financial system seized up. Although not as severe as the correction in 2006, the TASI still shed 49% between June and December, ending the year at 4800. Market capitalization fell to $244 billion. The biggest loser by sector was petrochemicals, which lost 63% of its value during the course of the year, with investors concerned about a global supply glut and an apparent shortage of gas feedstock in Saudi Arabia. The TASI continued to track emerging equity markets very closely in the first quarter of 2009. Performance was subdued as the global economic recession hardened and oil prices also tracked lower. In the second quarter, global economic conditions began to improve, with the first signs that financial markets had stabilized and the real economy was nearing, or at, its trough. Oil prices also began to move upwards again. 14 Journal of Economic and Social Studies �The Day-of-the-Week Effect in the Saudi Stock Exchange: A Non-Linear Garch Analysis Although the TASI initially tracked the benchmark higher, its recovery stalled in May 2009 as concerns about debt problems in the Saudi corporate sector began to emerge. The scale of these problems is almost impossible to quantify given a lack of publicly available data. Nevertheless, this opacity itself unsettled investors; the TASI remained subdued, adding just 19% during the second quarter. Data and Methodology The data we used is daily return data that covers January 2001 to December 2009, except the official religious holidays. The Saudi Stock Exchange operates from Saturday to Wednesday, while Thursday and Friday are the official weekend in which there is no transaction. The returns are oneday logarithmic returns. If the following day is a non-trading day, then the return is calculated using the closing price indices of the latest trading day and that day. The earlier studies of the day-of-the-week effect can be divided into four categories based on the methodology employed. The first category employs the methodology by calculating returns means and variances for each day of the trading week, or estimating the coefficients of the equation (1) below and using standard t and F test or ANOVA to check the significance and equality of mean returns, without paying attention to the time series properties of the sample data (Santesmases, 1986; Solnik and Bousquet, 1990; Athanassakos and Robinson, 1994; and Balaban, 1995). The second category of studies calculates mean daily returns or estimates the coefficients of equation (1). They, on the other hand, carry out hypothesis testing using t-statistics and χ2, calculated by using heteroscedasticity-consistent standard errors, proposed by White (1980). This approach does not inspect the distributional properties of the data used (Chang, 1993; Peiro, 1994; Abraham and Ikenberry, 1994). However, it should be mentioned that Chang, 1993) performed a more thorough investigation of the time series properties of the sample data using the Jarque-Bera test of normality and Breusch-Pagan-Godfrey test for heteroscedasticity and found that regression residuals are nonnormal, heteroscedastic and auto-correlated. Therefore, they employ tests that adjust regression errors for departures from conventional assumptions. The third category tests the normality of returns via the Kolmogorov-Smirnov D Statistic. If the returns are found to be normally distributed, the t and F-tests or ANOVA are employed. Otherwise, non-parametric tests are used to test for the existence of the day-of-the-week effect (Board and Sutcliffe, 1998; Wong, 1992). The fourth category begins with reporting descriptive statistics of the distributional properties of the return series. These statistics show that the series are highly leptokurtic relative to the normal distribution. Then, this outcome is used as a justification for the use of a GARCH (generalized autoregressive conditional heteroscedasticity) model to examine the presence of the day-of-the-week effect (Najand and Yung, 1994; Alexakis and Xanthakis, 1995). Volume 1 Number 1 January 2011 15 �Talat ULUSSEVER & Ibrahim GURAN YUMUSAK & Muhsin KAR In this study, we extend the works of the fourth category by explicitly testing for independently and identically distributed (IID) in the empirical residuals. We first utilize a standard method to test for daily seasonality in stock market returns by estimating the following regression (the basic model): Rt = β1 + β2D2 + β3D3 + β4D4 + β5D5 + Ut (1) where Rt is the rate of return on day t, β1, β2, β3, β4, β5 are parameters, D2, D3, D4, and D5 are binary dummy variables for Sunday, Monday, Tuesday, and Wednesday (i.e. D2 = 1 if t is Sunday, 0 otherwise) and Ut is a stochastic error term. To be able to confirm the existence of the day-of-theweek effect, at least two coefficients must be statistically significant and unequal. Standard t and F statistics are used to test these hypotheses. Obviously, the values of these test statistics are insignificant if the conventional assumptions about OLS error terms are violated. Daily stock returns are likely to violate these assumptions (Chang, 1993). The estimate of β1 is the sample mean return for Saturdays, while the estimates of the remaining coefficients are equal to the difference between the sample mean of the corresponding day and the sample mean for Saturday. Under the null hypothesis of no the-day-of-the-week effect, β2 = β3 = β4 =β5 = 0 and residual should be IID random variables. This approach is equivalent to regressing the returns on five daily dummies, with no constant term, and testing for the equality of all parameters. We will examine the IID assumption through the application of the Brock, Dechert and Scheinkman (BDS) test proposed by Brock (1987). BDS statistics gives a statistical test of IID within a time series, and is based upon the correlation dimension (Grassberger and Procaccia, 1983). Brock (1987) shows that for a time series which is IID, the BDS statistic is asymptotically N (0, 1). Let where Cm(ε) represents the fraction of all m-tuples in the series which are “close” to (within ε of) each other and σm(ε) is an estimate of the standard deviation. Wm (ε) is the BDS statistic and provides a formal test of the IID assumption. If the null hypothesis of IID can be rejected at this stage, then the implication is that the residuals contain some hidden, possibly non-linear, structure. We will illustrate that this is indeed the case, and it is due to the time varying volatility of stock returns data. To check this possibility, we will employ a GARCH model (Bollerslev, 1987) to the returns series. The model to be employed is of the form: 16 Journal of Economic and Social Studies �The Day-of-the-Week Effect in the Saudi Stock Exchange: A Non-Linear Garch Analysis We then carry out the BDS tests on the normalized residuals from the GARCH model to check for any remaining unexplained structure. We further carry our analysis by checking for the existence of relationships between groups of parameters of the GARCH model. For that purpose, Wald tests of coefficient restrictions are employed. a relationship between the parameters two variables makes it employed.The The existence existence ofofa relationship between the parameters of two of variables makes it possible employed. employed. employed. employed. employed. The The TheThe existence existence existence The existence existence ofof ofaone aof relationship a relationship of relationship avariable relationship a relationship between between between between between the the theparameters the parameters parameters the parameters parameters of of oftwo two of two of variables two variables variables two variables variables makes makes makes makes makes it itit it it possible to express in terms of the other, thus simplifying the model and increasing to express one variable in terms ofthethe the other, thus simplifying the model and increasing the degree possible possible possible possible possible toto to express express express to to express express one one one variable one variable variable one variable variable in in in terms terms terms in in terms of terms of of of the other, of the other, other, the other, thus other, thus thus simplifying thus simplifying simplifying thus simplifying simplifying the the the model the model model the model and model and and increasing and increasing increasing and increasing increasing the degree of freedom. offreedom. freedom. the the the degree the degree degree the degree of degree of of freedom. freedom. of of freedom. freedom. In the final step, the GARCH model is re-estimated with the accepted coefficient restrictions Instep, the final step, the GARCH model isre-estimated re-estimated the accepted coefficient restrictions being InIn Inthe the In thefinal In the final final the final step, final step, step, the the step, theGARCH the GARCH GARCH the GARCH GARCH model model model model is model is isre-estimated re-estimated re-estimated issubject is re-estimated with with with the the with theaccepted the accepted accepted the accepted accepted coefficient coefficient coefficient coefficient coefficient restrictions restrictions restrictions restrictions restrictions being imposed. Once again, we thewith normalized residuals from the restricted GARCH being being being being imposed. being imposed. imposed. imposed. imposed. Once Once Once Once again, Once again, again, again, we again, we wesubject we subject subject we subject subject the the the normalized the normalized normalized the normalized normalized residuals residuals residuals residuals residuals from from from the from the the restricted the restricted restricted the restricted restricted GARCH GARCH GARCH GARCH GARCH imposed. Once again, we subject the normalized residuals from the restricted GARCH model the to model to the BDS testing. If these residuals turn out to befrom IID, then this final model is to used model model model model to model to to the the to the BDS to the BDS BDS the BDS testing. BDS testing. testing. testing. Ifequations IfIf these these these Ifresiduals If these residuals these residuals residuals residuals turn turn turn out turn out turn toout to to be out be to be IID, to be IID, IID, be IID, then then then IID, then this this this then final this final final this final model model final model model is model isis used used used is is used toto used toseparate to tothese BDS testing. Iftesting. these turn out toout be IID, then this final model is used to derive derive separate forresiduals each day of the trading week. If the specifications of derive derive derive derive derive separate separate separate separate separate equations equations equations equations equations forfor for each for each each foreach day each day day offollows of of day the of the the of trading the trading trading the trading week. week. week. week. If week. IfIfthe the If the specifications Ifthe specifications specifications thespecifications specifications ofare of ofthese of these these ofthese these equations are not identical, it day that the five daily are drawn from different equations for each day of the trading week. Iftrading the specifications ofreturns these equations not identical, equations equations equations equations equations are are are not are not not are identical, not identical, identical, not identical, identical, it it it follows follows follows it it follows follows that that that that the the the that five the five five the five daily daily daily five daily returns daily returns returns returns returns are are are drawn are drawn drawn are drawn drawn from from from from different from different different different different distributions, and hence a day-of-the-week effect does indeed exist. it follows that the five daily returns are drawn from different distributions, and hence a day-of-thedistributions, distributions, distributions, distributions, distributions, and and and hence and hence hence and hence ahence day-of-the-week aa day-of-the-week day-of-the-week a day-of-the-week a day-of-the-week effect effect effect effect does effect does does does indeed indeed does indeed indeed exist. indeed exist. exist. exist. exist. week effect does indeed exist. Empirical Results Empirical Empirical Empirical Empirical Empirical Empirical Results Results Results Results ResultsResults Equation 6 shows the results of estimating the basic model. Equation Equation Equation Equation Equation 6 6shows 6 shows shows 6 shows 6 the shows the the results results results the results of results of of estimating estimating estimating of of estimating estimating the the basic the basic basic the basic model. basic model. model. model. model. Equation 6the shows the results of the estimating the basic model. (6) Rt = 0.003265 – 0.003167β2 – 0.003021β3 - 0.002556β4 – 00.2602β5 + Ut 0.003265 0.003265 =t 0.003265 = 0.003265 – –0.003167 – 0.003167 0.003167 – 0.003167 –(5.98) 0.003167 β2ββ–22–0.003021 –β0.003021 0.003021 β–20.003021 – 0.003021 β3ββ-330.002556 -β - 0.002556 β- 3(-3.97) 0.002556 - 0.002556 β4ββ–44 –00.2602 β–400.2602 00.2602 β–400.2602 – 00.2602 β5ββ+ β+ U5U β+t5t U +t(-3.91) (6) U(6) R RtRR=tt == Rt0.003265 2(-3.89) 30.002556 55 + tU t(6)(6)(6) (-3.37) 2 (5.98) (5.98) (5.98) (5.98) (-3.89) (-3.89) (-3.89) (-3.89) (-3.97) (-3.97) (-3.97) (-3.37) (-3.37) (-3.37) (-3.91) (-3.91) (-3.91) (-3.89) (-3.97) (-3.97) (-3.37) (-3.37) (-3.91) (-3.91) (R(5.98) = 0.091) 2 22 2 2 (R(R (R =(R =0.091) =(R 0.091) 0.091) = 0.091) = 0.091) As it is clearly seen from the results, all t-statistics of the estimated parameters are greater than Asclearly itseen is seen clearly seen from the results, all t-statistics of the estimated parameters are greater than the ittheis seen from the all t-statistics of the estimated parameters are greater than AsAs As itAs itis it As isis clearly it clearly clearly is clearly seen seen from from from from the the the the results, results, results, all all all t-statistics all t-statistics t-statistics t-statistics ofof of the the of the estimated the estimated estimated estimated parameters parameters parameters parameters are greater are greater greater greater than than than than critical value atresults, the 5%results, significance level. This confirms that all are ofare the differences between critical value at the 5% significance level. This confirms that all of the differences between the mean the critical at the significance level. This confirms that all of the differences between the the the critical the critical critical critical value value value value atvalue at at the the the at5% the 5% 5% significance 5% significance significance significance level. level. level. level. This This This This confirms confirms confirms confirms that that that all that all all of all ofthe of the differences the differences differences differences between between between between the mean returns of Saturday and each other trading day are significantly different from zero. returns of Saturday and each other trading day are significantly different from zero. Therefore, the the mean returns of Saturday and each other trading day are significantly different from zero. the the the mean the mean mean mean returns returns returns returns ofof of Saturday Saturday of Saturday Saturday and and andeach and each each each other other other other trading trading trading trading day day day are day are are significantly are significantly significantly significantly different different different different from from from from zero. zero. zero. zero. Therefore, the results are supportive of the day-of-the-week effect. Therefore, the results are supportive of the day-of-the-week effect. Therefore, Therefore, Therefore, Therefore, the the the results the results results results are are supportive are supportive supportive supportive ofof of the the the ofday-of-the-week the day-of-the-week day-of-the-week day-of-the-week effect. effect. effect. results areare supportive of the day-of-the-week effect.effect. Table (1) reports the results of applying the BDS test to the residuals of the basic model. The Table (1) reports the results of applying the BDS test to the residuals of the basic model. The at Table Table Table Table (1) (1) (1) reports (1) reports reports reports the the the results the results results results ofof of applying applying of applying applying the the the BDS the BDS BDS BDS test test test to test to to the the to the the residuals residuals residuals ofhypothesis of of the the of the basic the basic basic basic model. model. model. The The The The calculated test statistics are quite high, indicating that the null of themodel. IID iscalculated rejected Table (1) reports the results of applying the BDS test toresiduals the residuals of the basic model. The calculated test statistics are quite high, indicating that the null hypothesis of the IID is the rejected at the calculated calculated calculated calculated test test test statistics test statistics statistics statistics are are are quite are quite quite quite high, high, high, high, indicating indicating indicating indicating that that that the that the the null the null null hypothesis null hypothesis hypothesis hypothesis of of of the the of the IID the IID isIID isis rejected rejected rejected is explained rejected atat at atlevel. the 5% level. This finding suggests that variations in daily cannot be by test statistics are quite high, indicating that the null hypothesis ofreturns the IID isIID rejected at 5% the 5% level. This finding suggests that variations in daily returns cannot be explained by the the the the 5% the 5% 5% level. 5% level. level. level. This This This This finding finding finding finding suggests suggests suggests suggests that that that variations that variations variations variations in in in daily daily in daily daily returns returns returns returns cannot cannot cannot cannot be be be explained be explained explained explained by by by the by the the the basic linear model. This finding suggests that variations in daily returns cannot be explained by the basic linear model. basic linear model. basic basic basic basic linear linear linear linear model. model. model. model. Table 1. BDS tests on the basic model residuals Table 1. BDS tests on the basic model residuals BDS tests on the basic model residuals Table Table Table Table 1.Table 1. 1. BDS BDS BDS 1. 1. BDS tests tests tests tests onon on the the on the basic the basic basic basic model model model model residuals residuals residuals residuals m=4 m=5 m=6 m=7 m=8 ε mm m = =4= m44m = 4=7.876 4 mm m = =5= m55m = 5=8.879 5 mm m = =6= m66m = 6=10.002 6 mm m = =7= m77m = 7=11.378 7 mm m = =8= m88m = 8=14.056 8 ε εε ε ε 0.042 0.042 0.042 0.042 0.042 0.042 7.876 7.876 7.876 7.876 7.876 8.879 8.879 8.879 8.879 8.879 10.002 10.002 10.002 10.002 10.002 11.378 11.378 11.378 11.378 11.378 14.056 14.056 14.056 14.056 14.056 0.084 8.148 9.067 10.117 11.109 12.067 0.084 0.084 0.084 0.084 0.084 8.148 8.148 8.148 8.148 8.148 9.067 9.067 9.067 9.067 9.067 10.117 10.117 10.117 10.117 10.117 11.109 11.109 11.109 11.109 11.109 12.067 12.067 12.067 12.067 12.067 0.168 9.657 10.112 10.302 10.598 10.675 0.168 0.168 0.168 0.168 0.168 9.657 9.657 9.657 9.657 9.657 10.112 10.112 10.112 10.112 10.112 10.302 10.302 10.302 10.302 10.302 10.598 10.598 10.598 10.598 10.598 10.675 10.675 10.675 10.675 10.675 The results of the BDS test suggest that we should fit a GARCH model. Table (2) reports the final The results of the BDS test suggest wegeneral shouldtofit a GARCH model. Table show (2) reports results of estimating a GARCH modelthat using specific modeling. The results that thethe The The The results The results results The results results ofof of the the of the BDS of the BDS BDS the BDS test BDS test test suggest test suggest suggest test suggest suggest that that that we that we we that should we should should we should should fitfit fit a afit GARCH a GARCH GARCH fit ageneral GARCH a GARCH model. model. model. model. model. Table Table Table Table (2) Table (2) (2) reports (2) reports reports (2) reports reports the the thethethe final results of estimating GARCH model using specific modeling. The results show GARCH model provides aa better explanation than the basic to model. final final final final results results final results results of results of of estimating estimating of estimating of estimating estimating a aGARCH amodel GARCH GARCH a GARCH a GARCH model model model model using using using using general using general general general general toto to specific specific specific tothan to specific specific modeling. modeling. modeling. modeling. modeling. The The The results The results results The results show results show show show show that the GARCH provides amodel better explanation the basic model. that that that the that the the that GARCH the GARCH GARCH the GARCH GARCH model model model model provides model provides provides provides provides a better aa better better a better aexplanation better explanation explanation explanation explanation than than than than the the the than basic the basic basic the basic model. basic model. model. model. model. The results of performing the BDS tests on the standard residuals of the GARCH model are The The Theresults The results results The results results ofof ofperforming performing of performing of performing performing the the the the BDS BDS the BDS tests BDS tests tests tests onon tests on the on the the on standard the standard standard the standard standard residuals residuals residuals residuals residuals of of ofthe the of theGARCH of the GARCH GARCH the GARCH GARCH model model model model are model are areareare given in Table (3). ItBDS is absolutely clear that these residuals are indeed IID. given given given given ingiven in in Table Table Table in in Table (3). Table (3). (3). It(3). Itis It(3). isis absolutely Itabsolutely absolutely is It absolutely is absolutely clear clear clear clear that that clear that these that these these that these residuals these residuals residuals residuals residuals are are are indeed are indeed indeed are indeed IID. indeed IID. IID. IID. IID. Volume 1 Number 1 January 2011 17 �Talat ULUSSEVER & Ibrahim GURAN YUMUSAK & Muhsin KAR The results of performing the BDS tests on the standard residuals of the GARCH model are given in Table (3). It is absolutely clear that these residuals are indeed IID. Table 2. Maximum Likelihood Estimates of the GARCH (1,1) model Table 3. BDS tests on the GARCH (1,1) model residuals Table (4) reports the results of applying various Wald tests of restrictions on the parameters of the GARCH model. These results suggest that variable terms in the original GARCH model should be replaced by a new set of dummy variables, namely D1, D34, and D24, such that, D1 = 1 if day t is a Saturday and 0 otherwise, D34 = 1 if day t is a Monday or Tuesday and 0 otherwise, and D24 = 1 if day t is a Sunday or a Tuesday and 0 otherwise. Table (5) shows the estimates of the GARCH model with new dummy variables. The change in the model specification slightly increases the explanatory power of the model. The final model explains about 8% of the variation in daily returns. The BDS test statistics were calculated for the residuals of this final model and the results are reported in Table (6). Again, the null hypothesis of IID cannot be rejected. This result indicates that the final GARCH model can adequately describe the daily return process of the TADAWUL stock price index. Table 4. Wald tests for coefficient restrictions Null Hypothesis ß1 + ß2 = 0 ß1 + ß3 = 0 ß1 + ß4 = 0 18 χ2 2.255 0.067 0.038 P 0.133 0.802 0.853 Journal of Economic and Social Studies �The Day-of-the-Week Effect in the Saudi Stock Exchange: A Non-Linear Garch Analysis ß1 + ß5 = 0 ß6 + ß9 = 0 ß6 + ß10 = 0 ß6 + ß11 = 0 ß6 + ß12 = 0 ß1 = ß2 0.034 4.128 0.339 4.597 0.588 0.043 0.867 0.042* 0.0576 0.031* 0.0436 0.0834 * Significant at the 5% level Table 5. Maximum Likelihood Estimation of the GARCH (1,1) model (the coefficient restriction imposed) Table 6. BDS tests on the restricted GARCH (1,1) model residual In Table (5), the GARCH coefficient α3 is highly significant. This implies that a significant part of the current volatility of TADAWUL stock index returns can be explained by past volatility, and that the past volatility tends to persist over time. The parameter estimates of the final GARCH model can be used to construct the equations from 7 to 11 for five days of the trading week. The five returns equations clearly reveal that the mean daily returns are significantly different from each other. Consequently, based on the results of Table (5), we can confirm the presence of day-ofthe-week effect on daily stock returns in the Saudi Stock Exchange. Volume 1 Number 1 January 2011 19 �Talat ULUSSEVER & Ibrahim GURAN YUMUSAK & Muhsin KAR Conclusion The presence of the day of the week effect in stock market returns has been one of the hotly debated issues in the finance literature. Settlement procedures, bid-ask spread biases, dividend patterns, negative information release, thin trading, measurement errors, specialists’ behavior, and the concentration of certain investment decisions have been considered as main factors behind the day of the week effect phenomenon in the empirical studies. In this study, covering the daily stock return data from January 2001 to December 2009 and employing a non-linear GARCH model, we intended to test the presence of the day-of-the-week effect in the Saudi Stock Exchange (TADAWUL), which is a recently modernized stock market and offers a unique opportunity to test for seasonal anomalies. It should be noted that trading takes place from Saturday to Wednesday in TADAWUL as opposed to the more traditional Monday through Friday trading. The empirical results of the study confirm that all of the differences between the mean returns of Saturday and each other trading day are significantly different from zero, which are supportive of the day-of-the-week effect (Equation 6). Furthermore, the findings (Equations 7-11) reveal that the returns on the five trading days follow different processes, which obviously confirms the presence of day-of-the-week effect in daily stock returns in TADAWUL. This implies that there is room for investors to adjust their portfolios by taking into account day of the week variations in volatility in the Saudi Stock Exchange. References Abraham, A. and Ikenberry, D.L. (1994). The Individual Investor and the Weekend Effect. Journal of Financial and Quantitative Analysis, 29, 263-277. Agrawal, A. and Tandon, K. (1994). Anomalies or Illusions? Evidence from Stock Markets in Eighteen Countries. Journal of International Money and Finance, 13, 83-106. Aggarwal, R. and Rivoli, P. (1989). Seasonal and Day-of-the Week Effects in Four Emerging Stock Markets. Financial Review, 24, 541-550. Akgiray, V. (1989). Conditional Heteroscedasticity in Time Series of Stock Returns: Evidence and Forecast. Journal of Business, 62, 55-80. Alexakis, P. and Xanthakis, M. (1995). Day of the Week Effect on the Greek Stock Market. Applied Financial Economics, 5, 43-50. Al-Loughani, N. and Chappell, D., (2001). Modelling the day-of-the-week effect on the Kuwait Stock Exchange: A GARCH representation. Applied Financial Economics 11 (4), 353–359. 20 Journal of Economic and Social Studies �The Day-of-the-Week Effect in the Saudi Stock Exchange: A Non-Linear Garch Analysis Athanassakos, G. and Robinson, M. (1004). The Day of the Week Anomaly: The Toronto Stock Exchange Experience. Journal of Business and Accounting, 21, 883-856 Baillie, R. T. and DeGennaro, R.P. (1990). Stock Returns and Volatility. Journal of Financial and Quantitative Analysis, 25, 203-214. Balaban, E. (1995). Day of the Week Effects: New Evidence from an Emerging Market. Applied Economics Letters, 2, 139-143. Board, J.E.G. and Sutcliffe, C.M.S. (1988). The Weekend Effect in UK Stock Market Return. Journal of Business Finance and Accounting, 15, 199-213 Barone, E. (1990). The Italian Stock Market: Efficiency and Calendar Anomalies. Journal of Banking and Finance, 14, 483-510. Bollerslev, T. (1987). Generalized Autoregressive Conditional Heteroscedasticity. Journal of Econometrics, 31, 307-327. Brock, W.A., Dechert, W.D. and Scheinkman, J.A. (1987). A Test of Independence Based on the Correlation Dimension, Manuscrip., Social System Research Unit. University of Wisconsin. Campbell, J. Y. and Hentschel, L. (1992). No News is Good News: An Asymmetric Model of Changing Volatility in Stock Returns. Journal of Financial Economics, 31, 281-318. Chan, K. C., Karolyi, G.A. and Stulz, R.M. (1992). Global Financial Markets and the Risk Premium on U.S. Equity. Journal of Financial Economics, 32, 137-167. Chang, E.C., Pinegar, J.M., and Ravichandran, T. (1993). International Evidence on the Robustness of the Day-of-the-Week Effect. Journal of Financial and Quantitative Analysis, 23, 497-513 Corhay, A. and Rad, A. T. (1994). Statistical Properties of Daily Returns: Evidence from European Stock Markets. Journal of Business Finance and Accounting, 21, 271-282. Cornell, B. (1985). The Weekly Patterns in Stock Returns Cash versus Futures: A Note. Journal of Finance, 40, 583-88. Cross, F. (1973). The Behavior of Stock Prices on Fridays and Mondays. Financial Analysts Journal, 29, 67-69. Damodaran, A. (1989). The Weekend Effect in Information Releases: A Study Earnings and Volume 1 Number 1 January 2011 21 �Talat ULUSSEVER & Ibrahim GURAN YUMUSAK & Muhsin KAR Dividend Announcements. Review of Financial Studies, 2, 607-623 Dyl, E. and Maberly, E. (1986). The Daily Distribution of Changes in the Price of Stock Futures. Journal of Futures Markets, 6, 513-521. Engle, R. (1993). Statistical Models for Financial Volatility. Financial Analysts Journal, 49, 72-78. French, K. R. (1980). Stock Returns and The Weekend Effect. Journal of Financial Economics, 8, 55-69. French, K. R. and Roll, R. (1986.) Stock Return Variances: The Arrival of Information of the Reaction of Traders. Journal of Financial Economics, 17, 5-26. French, K. R., Schwert, G.W. and Stambaugh, R.F. (1987). Expected Stock Returns and Volatility. Journal of Financial Economics, 19, 3-29. Gibbons, M., and Hess, P. (1981). Day of the Week Effects and Asset Returns. Journal of Business, 54, 579-596. Glosten, L. R., Jagannathan, R.and Runkle, D.E. (1993). On the Relation between the Expected Value and the Volatility of the Nominal Excess Returns on Stocks. Journal of Finance, 48, 17791801. Grassberger, P. and Procaccia, I. (1983). Measuring the Strangeness of Strange Attractors. Physica, 90, 189-208 Hamao, Y., Masulis, R. and Ng, V. (1990). Correlations in Price Changes and Volatility across International Stock Markets. Review of Financial Studies, 3, 281-308. Harvey, C. and Huang, R. (1991). Volatility in Foreign Exchange Futures Markets. Review of Financial Studies, 4, 543-570. Jaffe, J. and Westerfield, R. (1985). The Weekend Effect in Common Stock Returns: The International Evidence. Journal of Finance, 40, 433-454. Keim, D. B. and Stambaugh, F. (1984). A Further Investigation of Weekend Effects in Stock Returns. Journal of Finance, 39, 819-840. Lakonishok, J. and Levi, M (1982). Weekend Effects on Stock Returns: A Note. Journal of Finance, 37, 883-889. Najad, M. and Yung, K. (1994). Conditional Heteroskedasticity and the Weekend Effect in S&P 500 22 Journal of Economic and Social Studies �The Day-of-the-Week Effect in the Saudi Stock Exchange: A Non-Linear Garch Analysis Index Futures. Journal of Business Finance and Accounting, 21, 603-612 Nelson, D. B. (1991). Conditional Heteroskedasticity in Asset Returns: A New Approach. Econometrica, 59, 347-370. Pagan, A. 1984. Econometrics Issues in the Analysis of Regressions with Generated Regressors. International Economic Review 25: 221-47. Peiro, A. (1994). Daily Seasonality in Stock Returns: Further International Evidence. Economics Letter, 45, 227-232 Penman, S. H. (1987). The Distribution of Earnings News Over Time and Seasonality in Aggregate Stock Returns. Journal of Financial Economics, 18, 199-228. Rogalski, R. J. (1984). New Findings Regarding Day-of-the-Week Returns over Trading and NonTrading Periods: A Note. Journal of Finance, 35, 1603-1614. Santesmases, M. (1986). An Investigation of the Spanish Stock Market Seasonalities. Journal of Business Finance and Accounting, 13, 267-276 Samirlock, M. and Starks, L. (1986). Day-of-the-Week and Interday Effects in Paris Bourse. Journal of Financial Economics, 17, 197-210 Solnik, B. and Bousquet, L. (1990). Day-of-the-Week Effect on the Paris Bourse. Journal of Banking and Finance, 14, 461-468. Theodossiou, P., and Lee, U. (1995). Relationship Between Volatility and Expected Returns Across International Stock Markets. Journal of Business Finance and Accounting, 22, 289-300. Tonchev, D. and Kim, T.-H., (2004). Calendar effects in Eastern European financial markets: Evidence from the Czech Republic, Slovakia and Slovenia. Applied Financial Economics 14 (14), 1035–1043. White, H. (1980). A Heteroskedasticity-Consistent Covariance Matrix Estimator and a Direct Test for Heteroskedasticity. Econometrica, 48, 817-838 Wong, K.A., Hui, T.K. and Chan, C.Y. (1992). Day-of-the-Week Effect: Evidence from Developing Stock Markets. Applied Financial Economics, 2, 49-56. Volume 1 Number 1 January 2011 23 � Dublin Core The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/. Extent The size or duration of the resource. 1051 Title A name given to the resource The Day-of-the-Week Effect in the Saudi Stock Exchange: A Non-Linear Garch Analysis Author Author ULUSSEVER , Talat GURAN YUMUSAK, Ibrahim KAR, Muhsin Abstract A summary of the resource. It is a well-known fact that the day-of-the-week effect in stock markets is one of the most prominent puzzling seasonal anomalies in finance and has been increasingly attracting attention from researchers and practitioners, as well as academics. This paper scrutinizes the day-of-theweek effect in the emerging equity market of Saudi Arabia, TADAWUL. By using a non-linear GARCH model and covering the data from January 2001 to December 2009, the findings of the study reveal that the returns on the five trading days follow different process. This confirms that mean daily returns are ignificantly different from each other and validates the day-of-the-week effect in TADAWUL. Publisher An entity responsible for making the resource available International Burch University Date A point or period of time associated with an event in the lifecycle of the resource 2011-01 Keywords Keywords. Article PeerReviewed Identifier An unambiguous reference to the resource within a given context ISSN 1986 – 8502 H Social Sciences (General),HB Economic Theory,HG Finance https://omeka.ibu.edu.ba/files/original/25edaf0150c73848f7c8fd0bda0468a0.pdf 573d769a7e84c6ac2f0c4bef5d9cdd63 PDF Text Text Journal of Economic and Social Studies Econometric Analysis of Import and Inflation Relationship in Turkey between 1995 and 2010 Volkan Ulke Faculty of Economics International Burch University, Sarajevo, BiH volkanulke@yahoo.com Uğur ERGÜN Faculty of Economics International Burch University, Sarajevo, BiH uergun@ibu.edu.ba Abstr ct In economics, the relation between import volume and inflation rate has been discussed several times for different countries. This study investigates the relationship between inflation and import volume by using monthly time series data for the Turkish economy over the period 1995-2010. The study applies a number of econometric techniques: Augmented Dickey-Fuller unit root test, univariate cointegration test, error correction model, and Granger causality test. The results of this dissertation show that there is long term and short term co-integration relation between inflation and import volume. Indeed, there is one-way Granger-causality from import to inflation. Key words: Import, Inflation, Cointegration, Granger Causality Jel odes: E52, F43 Volume 1 Number 2 July 2011 69 �Volkan U & Uğur ERGÜN Introduction There are a few different reasons that can account for the inflation in goods and services. Three major types of inflation, as part of what Robert J. Gordon (1988) call the “triangle model”; demand-pull inflation, cost-push inflation and built-in inflation. Demand-pull inflation refers to the idea that the economy actual demands more goods and services than available. This shortage of supply which enables sellers to raise prices until equilibrium is put in place between supply and demand. The cost-push theory, also known as “supply shock inflation,” suggests that shortages or shocks to the available supply of a certain good or product will cause a ripple effect through the economy by raising prices through the supply chain from the producer to the consumer. According to demand-pull and cost-push theory we accept a relation between import volume and inflation. Some of the studies which focus on the relationship between inflation and import are as follows; Bayraktutan and Arslan (2003) studied on the relationship among wholesale price index, foreign exchange rate, and import volume an annual data of the 1980-2000 periods. Their study shows that there is a direct and interaction among wholesale price index, foreign exchange rate, and import volume. Research by Cheng and Laura (1997) shows determinants of inflation in Turkey over the period 1970 to 1995. Bahmani-Oskooee and Domaç (2003), central banks can eliminate inflation by interfering with monetary aggregates, particularly, the monetary base. However, it is noted that the supported correlation between money and prices is not an indicator of the direction of causality. In Bahmani-Oskooee and Domaç (2003) the external shocks followed by exchange rate depreciations, changes in public sector prices, and inflationary inertia are all found to be factors influencing inflation in Turkey. According to Domaç (2004), increases in inflation expectations can be followed by exchange rate depreciation since the monetary authority buys foreign currency to keep purchasing power stable. Domaç (2004) found that monetary variables such as money or real exchange rate direct the inflationary process of Turkey. Their findings for Turkey by stating: “The empirical findings show that inflationary pressures in Turkey have their origin in the following factors: (i) the presence of external shocks which engender sharp exchange rate depreciations; (ii) changes in public sector prices; and (iii) inflationary inertia”. Saatçioğlu and Korap (2005) investigated the potential causes of chronic-high inflationary environment in Turkish economy for the period 1988-2004 using monthly observations. Research by Albert, Maurice and Barrie (2005) supports the relationship between 70 Journal of Economic and Social Studies �Econometric Analysis of Import and Inflation Relationship in Turkey between 1995 and 2010 domestic market pressure and inflation depends on openness to international trade. The possibility that international trade is responsible for the apparent breakdown of the relationship between excess demand and inflation is suggested in the analyses of Phillips curve relationships studied by Gordon (1998) and Rich and Rissmiller (2000). Gylfason (1998) studied on export and population, per capita income, agriculture, primary exports, and inflation by statistical methods in cross-sectional data from the World Bank covering 160 countries. He showed two important hypotheses for these countries macroeconomic factors.First, concerning exports, inflation is inversely correlated with real exchange rates as long as nominal exchange rates do not adjust instantaneously to prices, even if high inflation may impede exports and growth through other channels as well. Second, economic growth has been linked to a host of variables in recent work, two of which are quite robust: initial income, reflecting catch-up and convergence, and investment.The ratio of exports to Gross domestic product (GDP) inversely related to population. This paper analyzes inflation and import relationship in Turkey between 1995 and 2010 using bivariate cointegration model assumptions. Consumer price index (CPI) is used as indicator of inflation. In this study, we investigate the relationship between inflation and import volume by using monthly time series data for the Turkish economy over the period 1995-2010. In the study, existence of a co-integration relationship and causality between import and inflation is tested. Methodology and Data Before starting the analysis,time series are transformed to eliminate spurious problem. In the first step of the co-integration analysis, Augmented Dickey Fuller test is used as stationary test. If the series are nonstationary in levels but stationary in first difference, co-integration test can be applied. In the second step, error correction model (ECM) is performed to investigate dynamic relationship between inflation and import. In the last step, granger causality test is applied to clarify the existence and direction of the causality between variables. The data used in this study belong to the period of 1995-2010. Import volume and CPI data was organized by using the Central Bank of the Republic of Turkey (CBRT) databases.CPI is used as indicator of inflation. Monthly CPI and import data table can be seen appendix A.1. Volume 1 Number 2 July 2011 71 �Volkan U & Uğur ERGÜN Table 1 Descriptive Statistics Table M Mean 894.88 368.48 Median 425 244.03 St. Deviation 237.30 231.85 kewness 0.418 0.819 urtosis 2.373 2.442 Jarque- era 8.733 23.988 Stationarity Tests The main element of econometric studies with time series is to test whether series are stationary or not. Stationary process is a type of stochastic process that has got a great deal of attention and close examination by time series analysts. Generally, a stochastic process is said to be stationary if its mean and variance are constant over time and the value of covariance between the two time periods depends only on the distance or gap or lag between the two time periods and not the actual time at which the covariance is computed (Gujarati, 2003:797). Despite of the fact that our interest is stationary time series, we often face with nonstationry time series. In econometric practice, using of nonstationary time series can cause serious problems. A number of empirical works have been shown that, in general the statistical properties of regression analysis using nonstationary time series are doubtful. Models, generated by time series including stochastic or deterministic trend, can give spurious regression results (Utkulu, 1993). Co-integration Test As mentioned before, using nonstationary time series in econometric analyses may cause serious problems. The time series, which include stochastic or deterministic trend, can give spurious regression results. Hence test statistics can be invalid. Most of the macroeconomic time series include trend. Some researchers suggest to difference time series until transforming them to stationary series. It was proved that this method can cause losing some of long-run information which is of interest to economists (Utkulu, 1997:39). This problem of econometric studies can be solved by the co-integration concept presented by Engle and Granger (1987). With the help of co-integration analysis, nonstationary variables can be included to the regression without causing spurious results. Also this analysis provides efficiency in testing, estimating and modeling of long-run relationships among time series variables. 72 Journal of Economic and Social Studies �Econometric Analysis of Import and Inflation Relationship in Turkey between 1995 and 2010 Engle-Granger Two-Step Co-integration Method A method of estimating a long-run equation was presented by Engle and Granger (1987) and this method has been widely applied by researchers. One of the main advantages of this method is that the long-run equilibrium relationship can be modeled by directly involving the levels of the variables. In the first step all dynamics are ignored and the long-run equation is estimated: Yt= βXt+ ut (1) In order for Yt and Xt to be cointegrated, the estimated residuals from equation (1) must be stationary. In this case the co-integration regression is said to be sufficient. As the variables are nonstationary, we can face the spurious regression problem. Therefore, R2 and DW must be carefully inspected. If all indicators are satisfactory, we can proceed to the next step. The second step includes estimating of a short-run model. In the short-run there may be disequilibrium. Hence, we can treat the error term as the “equilibrium error” (Gujarati, 2003:824). And we can use this error term to tie the short-run behavior of GDP to its long-run value. The error correction mechanism (ECM) first used by Sargan (1984) and later popularized by Engle and Granger corrects for disequilibrium. An important theorem, known as Granger Representation Theorem, states that if two variables Y and X are co-integrated, then the relationship between the two can be expressed as Error Correction Mechanism (ECM) (Engle and Granger, 1987:255). Simply, we can write ECM for equation (1) as follows: ∆Yt= α0 + α1∆Xt + α2ut-1+ εt (2) where ∆ denotes the first difference, εt is an error term, ut-1 is the lagged value of the error term from co-integration regression (1). According to the Granger Representation Theorem α2 is expected to be negative and statistically significant. The absolute value of α2 shows how quickly the equilibrium is restored. Also α2 should take a value between -1 and 0, otherwise the process is explosive (Ghatak, Milner and Utkulu, 1997). Volume 1 Number 2 July 2011 73 �Volkan U & Uğur ERGÜN Error Correction Model (Hendry’s General-to-Specific Approach) Above the simple form of ECM is showed, but for obtaining the best error correction model for our analysis, in this study Hendry’s (1995) general-to-specific approach will be used. General-to-specific modeling is formulation of a fairly unrestricted dynamic model, in this manner called general, which is afterwards transformed, tested and reduced in size by performing a number of tests for restrictions. The general model is usually depicted as autoregressive distributed lag form (ADL). The ADL form means that a dependent variable, Yt, is described as a function of its own lagged values, and the current and lagged values of independent variables. In the literature Lr (lag operator) is used for notation of ADL model. Lr is defined for variable Xt as: LrXt= Xt-r (3) Let’s consider a simple first order autoregressive model: Yt = αYt-1 + εt (4) We can rewrite this using lag operator as: (1 - αL)Yt= εt (5) Also consider a finite distributed lag model: Yt = β0Xt + β1Xt-1 + β2Xt-2 +...+ βnXt-n + εt (6) Using lag operator, equation (6) becomes: Yt = b(L)Xt + εt (7) If we add lagged values of dependent variable (Yt) to distributed lag model (6), the result will be ADL model, and is denoted as: Yt = α0Yt-1 + α2Yt-2 +…+ αkYt-k + β0Xt + β1Xt-1 + β2Xt-2 +...+ βnXt-n + εt (8) In more succinct notation, using polynomial lag operator, it can be denoted as: a(L)Yt = b(L)Xt + εt 74 (9) Journal of Economic and Social Studies �Econometric Analysis of Import and Inflation Relationship in Turkey between 1995 and 2010 In our investigation, for estimating short-run dynamics we will apply simple form of(8): (10) Generally, Hendry’s general-to-specific model consists of four steps : 1. General model is established. This model must include variables of the theoretical model and bound the dynamic of process in possible minimum. 2. After reparameterisation of the model, more orthogonal and more explainable parameters, from the long-run equilibrium’s point of view, are obtained. 3. By simplifying the model, a short-run model with consistent data set is obtained. 4. Coefficients, error terms and power of the estimation are tested. In economic theories, generally, no information about the adaptation process from short-run to the long-run are presented. Consequently, short-run dynamics of the models are determined according to variables of the time series. Granger Causality In econometrics, the notion of causality changes its philosophical matter and is more explicit. In empirical econometrics, researchers want to know whether an increase in one economic series results increases in another economic series or decreases; to identify the direction of relationship among series. The most widely econometrical definition of causality has been introduced by Granger (1969). In literature it is called as Granger definition of causality and can be formulated simply as follows: If present value of Y can be predicted by using past values of X, then X is a Granger cause of Y; and causality from X to Y is denoted as X → Y. The basic aims of investigation of causality relationship between X and Y can be arranged as (Işiğiçok, 1994:90): - Prediction of future periods by using current values of X and Y; - Whether Y can be predicted by its past values or by past values of X; - Identifying exogenity and endogenity of variables; Volume 1 Number 2 July 2011 75 �Volkan U - & Uğur ERGÜN Finding direction of causality; To find out after how many periods the change in one variable affects another variable; To determine the structural changes in parameters. The Granger causality test was originally suggested by Granger (1969) and modification was suggested by Sargent (1976). The Granger test assumes that information related to the prediction of the variables, Y and X, is included only in the time series data on these variables. The test involves estimation of following regressions: (11) (12) Regressions (11) presumes that current value of Y is related with the past values of X; and (12) postulates that current value of X is related with the past values of Y. The first step of the Granger causality test is establishing of hypotheses: H0: ∑αi = 0: X does not Granger cause Y H1: ∑αi ≠ 0: X Granger causes Y For testing null hypothesis, we apply F test: (13) where RSSR is restricted residual sum of squares, obtained running regression with including all lagged Y, but without including X; RSSUR is unrestricted residual sum of squares, obtained by running regression including lagged X; m is number of restrictions; k is number of parameters in the unrestricted regression; n is number of observations. The final step is the comparison of the computed F value with the critical F value. If computed F value exceeds the critical F value at the significance level (%1, %5, 76 Journal of Economic and Social Studies �Econometric Analysis of Import and Inflation Relationship in Turkey between 1995 and 2010 %10) then we reject H0. Rejection of the null hypothesis indicates causality relationship between variables. Since the Granger causality tests are very sensitive to the lag length selection, Akaike information criterion (AIC) will be used in this study (Kasman and Emirhan, 2007). For choosing the lag length, we will start with one lag and increase them by AIC. The lag of the model with the least AIC value will be our model’s lag length. There are four possible cases that can appear when testing causality between Xand Y: i. X → Y: Unidirectional causality from X to Y. It occurs when the estimated coefficients of the lagged X in (11) are statistically different from zero (∑αi≠0); and coefficients of the lagged Y in (12) are not statistically differentfrom zero (∑δj= 0). ii. Y → X: Unidirectional causality from Y to X. The estimated coefficients of the lagged X in (12) are not statistically different from zero (∑αi= 0); and coefficients of the lagged Y in (12) are statistically different from zero (∑δj0). iii. X ↔ Y: Bilateral causality. The coefficients of X and Y are statistically different from zero. iv. Independence. The coefficients of X and Y are not statistically significant. Before the development of the error correction model, the standard Granger test had been using for testing causality between two variables. According to Granger, if there is co-integration between two variables, then the advantages of standard Granger causality test are not valid (Oskooee and Alse, 1993). Therefore, if there is co-integration between variables, then error correction term, obtained from longrun equation, is included to standard Granger test. Otherwise, standard Granger test is implied without including error correction term (Giles D., Giles J. and McCann, 1993:201). So, causality relationship is tested using error correction model. The Granger error correction model can be formulated as follows: (14) (15) Volume 1 Number 2 July 2011 77 �Volkan ULKE & Uğur ERGÜN In these equations ECt-1 and EC’t-1 are stationary error terms, obtained from equations (14) and (15) respectively; and are called error correction terms. ∆ indicates the first difference. In Granger error correction model, we test whether estimated coefficients of lagged values of all variables are significant or not by using F test (Oskooee, Mohtadi and Shabsigh, 1991). Let’s consider equation (15). For saying Y Granger causes X, not only all λ2i must be statistically significant, but also δ2 must be significant. For functioning of the mechanism also the coefficient of error correction term must be negative and the same time has to be between 0 and -1 (Ghatak, Milner and Utkulu, 1997:217). Empirical Results As a preliminary stage to co-integration analysis, the stationarity of each variable was tested using graphical analysis and unit root tests. First of all, the graphs of the variables (CPI, import volume) are presented in Figure below. Figure1.1 Variation of import versus inflation 78 Journal of Economic and Social Studies �Econometric Analysis of Import and Inflation Relationship in Turkey between 1995 and 2010 Figure above implies that the variables have been fluctuating and increasing togetherover the sample period. That is, showing an upward trend, intimating perhaps that the mean of all variables have been altering. This implies that the series of the variables are not stationary. On the other hand if the first differences of the variables are taken, it looks like purified from trend. Therefore the first differences of two variables seem stationary. However, these outcomes must be supported by the unit root test results which are presented in Table 1. Table 2 Unit Root Test Result Levels D rob. D rob. 894.88 368.48 425 244.03 With intercept Inflation mport First Difference 237.30 231.85 0.418 0.819 Note: t-values are reported in the table.* denote rejection of null hypothesis at 1%, 5% and 10% respectively. Critical values are based on MacKinnon (1991); 1%, 5%, 10% is -3.45775, -2.87349 and -2.57322. According to the unit root test, we cannot reject H0, and all variables are nonstationary in levels (I(0)). After taking the first differences for variables, we reject the null hypothesis at 1% significance level. Test results show that time series are stationary from the first order (I(1)). After showing that all variables are integrated of order one, we can proceed to the cointegration test. By using cointegration analysis, we will test whether there is a long-run relationship between inflationand import. Table 3 Cointegration Test Results Co-integration eq: ons. erm Yt= βXt+ut -0.006447 (-0.000799) Coefficient -0.367997 (-6.515491) R2 0.18342 DW 2.04625 D -6.515491* Note: The numbers in parenthesis are t-statistics. *denote rejection of null hypothesis at 1%, 5% and 10%. Critical values are based on MacKinnon (1991); 1%, 5%, 10% is -3.45775, -2.87349 and -2.57322. It can be seen from Table 3 that coefficient of regression has negative sign and is statistically significant. In other words, increases in independent variable decreases Volume 1 Number 2 July 2011 79 �Volkan U & Uğur ERGÜN dependent variable. Thus, our results are collateral with the demand-pull and costpush theory. An increase in import volume will decrease inflation. Stationarity of the error term, which is obtained from cointegration equations, shows a long-run relation between inflation and imports. Yt= -0.367997Xt– 0.006447 (16) According to equation (16), 1 unit increase in independent variable will decrease dependent variable by 0.367997. As seen in Table 2, the value of R2 is low. However, it could be higher, if more independent variables (like, exchange rate, unemployment, export, oil price etc.) are added to the equation (16). According to Granger Representation Theorem, if there is co-integration between variables, error correction mechanism must work. Consequently error correction mechanism will be examined in the next step. To examine whether a long-run equilibrium relationship between inflation and independent variables exists, co-integration tests are employed. It is found that inflation and imports are co-integrated; which means that a long run or equilibrium relationship exists between them. In the short-run relationship there may be disequilibrium. Therefore, one can treat the error term as the “equilibrium error” (Gujarati, 2003: 824). And we can use this error term to tie the short-run behavior of inflation to its long-run value. The short-run dynamics will be examined by employing an error-correction model. In the next step, insignificant parameters were dropped and remaining parameters can show significant effects of used parameters to inflation. Our error correction model is employed for determining short-run dynamics. Table 4 Error Correction Model Test Result o-int. eq: inflation(-1) oint q1 1 mport(-1) -0.82753 -0.1103 [-7.50256] c R2 dj. R2 -91.3408 0.17278 0.150179 Note: The numbers in parenthesis are the t-statistics. The critical values at 10% and 5% are 1.29 and 1.66respectively (1-tail). Two diagnostic tests (R2, Adjusted R2) were presented in the tables. The results, reported in Table 4, show that the adjusted R2 is not high, which implies that the model used in this study is not affected from problem of autocorrelation. The coefficients of the error correction terms, estimated for both models, are statistically significant and have correct (negative) signs, confirming the evidence for 80 Journal of Economic and Social Studies �Econometric Analysis of Import and Inflation Relationship in Turkey between 1995 and 2010 co-integration of the variables in the long-run model established earlier. These coefficients indicate what proportion of the discrepancy between the actual and longrun or equilibrium value of inflation is eliminated or corrected each month (Kasman A. and Kasman S., 2005). Coefficient of the error term, estimated for the first model, is - 91.3408. Finally, and perhaps most importantly, it can be concluded that there is a dynamic relationship between inflation and import. The evidence from our error-correction models and from long-run models shows that both long-run and short-run dynamics are significant. Therefore, our findings support validness of an equilibrium relationship between the dependent and independent variables in each co-integration equations. If there is a co-integration vector between inflation and import, there must be causality among variables at least in one direction (Granger, 1986). Hence, Granger causality test is used to examine the nature of this relationship. Granger (1986) and Engle and Granger (1987) supply a test of causality, which takes into account the information, provided by the co-integrated properties of variables. Table 5 Granger Causality Test result ull Hypothesis mport does not Gr. cause inflationdoes not Gr. cause Import bs 190 1.48607 F-Statistic 9.12461 (-6.515491) robability 0.00017 0.22895 Table 5 reports results of the causality analysis of inflation and import. It can be seen that there is unidirectional causation between inflation and import. Table 5 indicates that since F-statistic value of import is significantly big, therefore import does Granger causes on inflation. Simply it shows the inflation does not affect import. As a result there is unidirectional causation between inflation and import from import to inflation. Conclusion In this study, we investigate the relationship between inflation and import volume by using monthly time series data for the Turkish economy over the period 1995 to 2010. In the study, existence of a co-integration and dynamic relationship and Volume 1 Number 2 July 2011 81 �Volkan U & Uğur ERGÜN causality between import and inflation is tested by performing econometric methods such as co-integration, error correction model and Granger causality. Our test results indicate that; (a) long-run and dynamic relationships are found between inflation and import, (b) there is unidirectional causality from import to inflation. Also this result supports the theoretical approach. Our results imply that policy makers who are responsible for optimum inflation rate for sustainable development can use import to reach the planned inflation rate target through changing imposed tax rates on import. Future studies may focus on the relationship between inflation and import including the other related factors and changes in degree of relationship over financial crises and time. References Albert S. Dexter, Maurice D. Levi, & Barrie R. Nault (2005). International Trade and the Connection between Excess Demand and Inflation, Review of International Economics, 13(4), 699–708, 2005 Bayraktutan Y. & Aslan I. (2003). Türkiye’de döviz kuru, ithalat ve enflasyon, Afyon Kocatepe Üniversitesi, i.i.B.F. Dergisi ( 102 C..V, S.2, 2003) Barro Robert & Grilli Vittorio (1994). European Macroeconomics, Ch. 8, p. 139, Fig. 8.1. Macmillan Bahmani-Oskooee, M.; Domaç, I. (2003). “On the Link between Dollarisation and Inflation: Evidence from Turkey.” Comparative Economic Studies, September, Vol. 45, iss. 3, pp. 306-28. Cheng H. L. & Laura P. (1997). An Econometric Analysis of the Determinants of Inflation in Turkey, International Monetary Fund European I Department ,December 1997 Dickey, D.A & Fuller, W.A. (1979). Distribution of the Estimators of Autoregressive Time Series with a Unit Root, Journal of the American Statistical Association, 74:427-431. Dickey, D.A. & Fuller, W.A. (1981). Likelihood Ratio Statistics for Autoregressive Time Series with Unit Root, Econometrica, 49:1057-1072. 82 Journal of Economic and Social Studies �Domaç, I. (2004). “Explaining and Forecasting Inflation in Turkey.” The World Bank, Policy Research Working Paper Series, No. 3287. Engle, R.F. & Granger, C. W. J. (1987). Cointegration and Error Correction: Representation, Estimation and Testing, Econometrica, 55: 251-276. Ghatak, S., Milner, C. and Utkulu, U. (1997). Exports, Export Composition and Growth: Cointegration and Causality Evidence for Malaysia, Applied Economics, 29(2):213-223. Giles, D.E.A., Giles, J.A. and McCann, E. (1993). Causality, Unit Roots and ExportLed Growth: The New Zealand Experience, Journal of International Trade and Economic Development, 1(1):195-218. Gordon, R. J. (1998). Foundations of the Goldilocks Economy: Supply Shocks and the Time-varying NAIR, Brookings Papers on Economic Activity 2 (1998):297– 346. Granger, C.W.J. (1986). Developments in the Study of Cointegrated Economic Variables, Oxford Bulletin of Economics and Statistics, 48(3):213-228. Granger, C.W.J. (1969). Investigating Causal Relations by Econometric Models: cross Spectral Methods, Econometrica, 37(3):424-438. Gujarati, D.N. (2003). Basic Econometrics, West Point: US Military Academy. Gylfason,Thorvaldur (1998). Exports, Inflation, and Growth, University of Iceland; SNS— Center for Business and Policy Studies, Stockholm, Sweden; and CEPR. Hendry, D.F. (1995). Dynamic Econometrics, Oxford University Press: Oxford. Kasman, A. and Kasman, S. (2005). Exchange Rate Uncertainty in Turkey and ItsImpact on Export Volume, METU Studies in Development, 32(2):41-58. Kasman, A. & Emirhan, P.N. (2007). Export-Led or Import-Led Economic Growth Evidence from Cointegration Tests with and without Regime Shifts, Yapı Kredi Economic Review, 18(1):3-19. Lars E.O. & Svensson (2003). Escaping from a Liquidity Trap and Deflation: The Foolproof Way and Others, Journal of Economic Perspectives, Volume 17, Issue 4 Fall 2003, Mackinnon, J.G. (1991). Critical Values of Cointegration Tests, Long Run Economic Relationship: Readings in Cointegration, New York: Oxford University Press. Oskooee Bahmani, M., & Alse, J. (1993). Export Growth and Economic Growth: �An Application of Cointegration and Error-Correction Modelling, The Journal of Developing Areas, 27(4):535-542. Pagan, A.R. (1987). Three Econometric Methodologies: a Critical Appraisal, Journal of Economic Surveys, 1:3-24. Oskooee - Bahmani, M., Mohtadi, H. and Shabsigh, G. (1991). Export, Growth and Causality in LDCs, Journal of Development Economics, 36(2):405-415. Phillips, P.C.B. and Perron, P. (1988). Testing for Unit Roots in Time Series Regression, Biometrika, 75:335-346. Richard H. Clarida (1991). Co-Integration, Aggregate Consumption, And The Demand For Imports; A Structural Econometric Investigation, Working Paper No. 3812 National Bureau Of Economic Research, August 1991 Rich, Robert W. & Donald Rissmiller (2000). Understanding the Recent Behavior of U.S. Inflation Current Issues in Economics and Finance, Federal Reserve Bank of New York 6, 8 (2000):1–6. Saatçioğlu C. & Korap H. L. (2005). An Empirical Analysis of Turkish Inflation (1988-2004): some non-monetarist estimations, Erciyes Üniversitesi İktisadi ve İdari Bilimler Fakültesi Dergisi, Sayı 26, Ocak – Haziran 2005, ss. Sargan, J.D. (1984). Wages and Prices in the United Kingdom: A Study in Econometric Methodology, in K.F. Wallis and D.F. Hendry, eds., Quantitative Economics and Econometric Analysis, Basil Blackwell, Oxford, U.K. Sargent, T.J. (1976). A Classical Macroeconomic Model for the United States, Journal of Political Economy, 84:207-238. Utkulu, U. (1993).Cointegration Analysis: An Introduction Survey with Applications to Turkey, 1. Ulusal Ekonometri ve İstatistik Sempozyumu Bildirileri, İzmir: Ege Universitesi Basımevi, 11-12 November 1993, 303-323. Utkulu, U. (1997). How to Estimate Long-run Relationships in Econometrics: An Overview of Recent Developments, Dokuz Eylül Üniversitesi İİBF Dergisi, 12(2):39-48. �Appendix A A.1 Monthly CPI and Import Data year/month CPI/Import 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 jan feb mar apr may jun 100 119,3 165,85 106,25 212,5 340 100 108,85 117,97 142,33 123,22 145,27 81,928 151,11 121,43 101,49 151,11 272 150,42 136,4 178,47 174,7 185,96 163,17 115,25 119,3 125,93 103,03 144,68 234,48 172,31 147,83 182,7 167,06 205,95 184,67 94,444 154,55 158,14 144,68 194,29 283,33 148,09 184,78 208,08 173,27 199,07 198,72 141,67 212,5 165,85 138,78 234,48 206,06 106,17 132,95 145,21 158,99 162,57 171,02 138,78 183,78 234,48 295,65 309,09 971,43 153,98 187,47 198,56 214,18 224,03 236,73 272 377,78 111,48 66,019 133,33 219,35 194,15 171,4 148,25 144,87 169,7 157,31 128,3 377,78 566,67 323,81 1133,3 1133,3 163,66 144,87 187,82 200,88 205,26 187,72 261,54 295,65 219,35 323,81 425 425,34 211,03 199,61 274,45 248,49 263,79 273,1 445,26 447,56 451,85 454,1 456,06 455,46 301,84 292,76 403,03 378,24 381,03 403,78 486,39 486,47 487,75 491,23 495,73 496,24 344,27 396,92 486,21 457,56 467,87 474,35 524,96 526,11 527,55 534,61 544,63 546,46 388,42 467,13 553,38 552,53 605,32 594,43 577,09 579,55 584,86 591,92 594,89 593,45 505,07 542,81 631,07 616,05 712,18 680,27 624,25 632,32 638,39 649,1 658,78 656,4 779,1 764,22 801,68 853,06 920,61 928,74 683,55 681,22 688,73 688,86 693,28 694,04 442,57 432,72 501,74 482,59 518,22 596,09 739,5 750,21 754,58 759,09 756,37 752,12 557,48 561,78 716,31 712,55 702,21 726,25 �year/month jul aug sep oct nov dec 234,48 137,16 174,36 85 87,179 123,64 194,29 152,21 146,22 153,19 170,43 205,93 323,81 141,67 111,48 104,62 130,77 200 186,61 167,92 158,3 171,87 188,05 218,45 107,94 109,68 93,151 81,928 103,03 133,33 197,12 198,31 207,79 208,57 207,59 235,62 200 170 101,49 111,48 158,14 206,06 199,92 177,84 174,52 173,46 167,45 184,55 178,95 161,9 113,33 107,94 161,9 115,25 172,53 151,56 173,75 169,71 183,5 211,45 309,09 309,09 219,35 219,35 183,78 272 223,05 232,66 221,43 239,58 255,71 211,6 283,33 234,48 115,25 111,48 161,9 212,5 163,87 167 163,32 160,43 169,73 164,07 485,71 309,09 194,29 206,06 234,48 425 219,35 210,71 215,05 230,32 236,63 256,06 424,53 425,21 430,95 434,94 440,46 442,33 298,86 284,91 295,91 313,55 250,11 392,65 457,63 461,11 465,49 475,94 482,18 483,71 416,22 375,92 404,68 385,66 408,57 499,45 493,4 497,6 502,7 511,71 518,89 521,05 457,59 489,83 494,3 481,13 461,26 557,08 551,09 548,67 555,72 562,77 570,04 571,35 558,36 585,38 579,51 534,05 614,97 622,18 589,11 589,24 595,31 606,11 617,92 619,27 725,48 700,09 689,48 745,15 793,09 768,67 660,19 658,57 661,55 678,75 684,4 681,6 980,28 918 852,82 712,53 575,72 543,84 695,79 693,71 696,42 713,2 722,25 726,08 613,05 610,9 595,35 609,06 601,67 716,2 748,51 751,52 760,74 774,68 774,89 772,55 766,69 735,95 745,87 824,77 817,07 980,27 CPI/Import 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 Journal of Economic and Social Studies � Dublin Core The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/. Extent The size or duration of the resource. 1068 Title A name given to the resource Econometric Analysis of Import and Inflation Relationship in Turkey between 1995 and 2010 Author Author Ulke, Volkan ERGÜN, Uğur Abstract A summary of the resource. In economics, the relation between import volume and inflation rate has been discussed several times for different countries. This study investigates the relationship between inflation and import volume by using monthly time series data for the Turkish economy over the period 1995-2010. The study applies a number of econometric techniques: Augmented Dickey-Fuller unit root test, univariate cointegration test, error correction model, and Granger causality test. The results of this dissertation show that there is long term and short term co-integration relation between inflation and import volume. Indeed, there is one-way Granger-causality from import to inflation. Publisher An entity responsible for making the resource available International Burch University Date A point or period of time associated with an event in the lifecycle of the resource 2011-01 Keywords Keywords. Article PeerReviewed H Social Sciences (General),HB Economic Theory,HG Finance https://omeka.ibu.edu.ba/files/original/222a92feb92bbdb73295eccff7ab3968.pdf 92f662319f5f42d4e534f8ea4f4a1765 PDF Text Text Journal of Economic and Social Studies Financial Determinants of Investment for Turkey Okyay UÇAN Çukurova University, Department of Economics, Adana, Turkey. oucan@cu.edu.tr Özlem ÖZTÜRK Çukurova University, Department of Economics, Adana, Turkey. oozturk@cu.edu.tr ABSTRACT One of the fundamental aims of economic policies is to increase capital accumulation in terms of investment that is necessary to maintain a desirable and sustainable growth rate in the developing countries. The majority of empirical studies show that per capita GDP growth, foreign trade, capital flows, external debt, public sector borrowing requirements, inflation and interest rate are the main determinants of investment rate. Recently, there is an increasing emphasis on the role of the financial sector in this process, since a financial system, in essence, mobilizes saving to investment. In particular, it can be argued that a well-functioning and developed financial system may efficiently mobilize available resources for investment. Therefore, the aim of this study is to investigate whether financial development has contributed to an increase in investment in Turkey. To reach an empirical and firm conclusion, an investment function, including the traditional potential determinants along with financial development, is estimated by utilizing the developments in the time series econometrics in terms of unit root tests that allow structural breaks and co-integration for the period 1970-2009 in Turkey. Keywords: VAR; Unit root; Co-integration; Investment; Financial Development; Turkey. Volume 1 Number 1 January 2011 83 �Okyay UÇAN & Özlem ÖZTÜRK Introduction Analysis of data from a large sample of countries has consistently shown that the rate of accumulation of physical capital is an important determinant of economic growth (Levine and Renelt, 1992). Until 1980, Turkey pursued an industrialization strategy based on state-led import-substitution. Under this strategy the economy enjoyed rapid economic growth up to 1976. As a consequence of the rapid expansion of public demand and an investment boom, imports were, however, growing much more rapidly than exports, and the economy became increasingly dependent on foreign borrowing. From 1977 to 1980, growth collapsed virtually to zero, inflation accelerated, and foreign debt continued to increase. Controls on nominal interest rates caused extreme financial depression in this period. Administratively imposed ceilings on deposit and lending rates, credit rationing, and excessive reliance of corporations on credit rather than equity finance and other direct security issues were common characteristics of the pre-1980 financial regime in Turkey (Atiyas and Ersel, 1995). Investments in the Turkish economy in the period after 1970 have increased consistently. However, during economic crises such as 1980, 1998 and 2001, numerical increase in investment has been observed, but relatively the rate of growth has decreased. In the last 30-year period only in the previous year of 2009 has a decrease in investment occurred. This decline in investment is a matter of concern, given the close connection between the level of investment and the rate of economic growth as documented in recent studies (Ben-David, 1998; Chari, Kehoe & McGrattan, 1997; Barro, 1991; Khan & Reinhart, 1990; Kormendi & Meguire, 1985). It is therefore worthwhile to investigate the factors that determine the level of domestic investment in these countries. This paper investigates the role of financial factors in determining domestic investment and private investment in Turkey. The premise of this study is that financial development facilitates the channeling of resources from savers to the highest-return investment activities, increases the quantity of funds available for investment, and thus mitigates the liquidity constraints faced by entrepreneurs. Thus a large and liquid financial system reduces the overall costs and risks of investment, which stimulates capital accumulation. The analysis is based on a reduced-form investment model that relates a country’s domestic investment to the level of financial development while controlling for other nonfinancial factors. Following a standard practice in time series analysis, the investment equation is specified as a dynamic serial correlation model (see Hsiao, 1986; Anderson & Hsiao, 1982, 1981). To test the effects of financial development on investment, four indicators are used alternatively: credit to the private sector, total liquid liabilities of financial intermediaries, credit provided by banks, and a composite index combining these three indicators. Theoretical Approaches Despite the remarkable attention devoted to investment behavior, the literature has not yet produced a full-fledged model of investment applicable to the context of developing countries. Conventional models such as the flexible accelerator proved quite successful in explaining aggregate investment in industrial countries. The main underlying assumptions of these models, however (such as the assumption of perfect capital markets, absence of liquidity constraints, and abstraction from the role of government), are highly questionable in the context of developing economies. Research in the 84 Journal of Economic and Social Studies �Financial Determinants of Investment for Turkey past decades has shifted attention toward the role of financial factors in explaining investment over time and across countries. Studies that emphasize the role of financial determinants for investment in developing countries have revived the original ideas of Schumpeter (1932) about the importance of the financial system in promoting technological progress. These studies also embed the Keynesian view that the ``state of credit’’ is an important determinant of investment (Keynes, 1937, 1973). Gurley and Shaw (1955) provided vital impetus to the Schumpeterian and Keynesian insights by tying economic growth directly to financial development. Gurley and Shaw suggested that “economic development is retarded if only self-finance and direct finance are accessible, if financial intermediaries do not evolve” (Gurley & Shaw, 1955, pp. 518-519). One key difference between developed and underdeveloped countries, as Gurley and Shaw argued, is the level of organization and sophistication of financial intermediaries, especially because of their role in facilitating the flow of loanable funds between savers and investors. McKinnon (1973) and Shaw (1973) offered a theoretical and empirical foundation for the relationship between monetary factors and investment. These authors advanced the hypothesis that investment in developing countries is positively associated with the accumulation of real money balances. The McKinnon-Shaw hypothesis is based on the assumption that limited access to credit in developing countries forces investors to accumulate enough real balances before they can initiate investment projects. This view establishes a positive relationship between real interest rate and investment. Higher interest rates on deposits attract more real balances, which allow them to finance more investment. This contradicts the neoclassical view that higher interest rates increase the user cost of capital and thus reduce investment. Evidence for this neoclassical interest rate (relative price) effect is mixed at best. Recent studies go beyond the McKinnon- Shaw tradition and relate investment to financial development in general by emphasizing the special services that financial intermediaries provide to investors. The financial system is the key to matching financial resources to investors’ needs both through short-term credit expansion and, through its maturity transformation function, by channeling saving into long-term credit markets. Financial markets play an important role in allocating investment capital to high return activities (Greenwood & Smith, 1997). Financial intermediaries have a special function in alleviating information problems, reducing liquidity risk, reducing monitoring costs, and channeling credit to certain classes of borrowers that cannot access nonintermediated forms of credit (Levine, 1997). This analysis implies that low investment in developing countries may be due to low financial intermediation characterized by a limited range of financial instruments, limited long-term lending, inefficient lending practices (for example, politically motivated lending), direct credit control, and crowding out of private investment by public borrowing for consumption purposes. The emphasis on the role of finance for investment constitutes a major improvement on the traditional view that domestic investment is primarily determined by domestic saving. This traditional view holds that the level of saving determines the interest rate and thus the cost of investment, which in turn influences the demand for new capital. Indeed, a number of studies have documented a close connection between low investment rates in developing economies and low domestic saving (Bayoumi, 1990; Dooley, Frankel & Mathieson, 1987; Feldstein & Horioka, 1980). These studies find that countries with low saving rates also have low investment rates. The positive relationship between domestic saving and domestic investment is often viewed as evidence of imperfect international capital flows Volume 1 Number 1 January 2011 85 �Okyay UÇAN & Özlem ÖZTÜRK and various country-specific institutional and noninstitutional rigidities (see Feldstein & Horioka, 1980). However, this approach, that assumes that saving directly causes investment, has important limitations. First, this view is an equilibrium (static) approach. Second, this view only considers the real side of the saving behavior and regards saving as a residue of income after consumption. As some authors argue, it is more appropriate to consider saving as a financial phenomenon. Under this view, saving is regarded as a mechanism of supply of funds (directly and indirectly) to the capital markets that channel the funds into the investment process. In that sense, the financial sector benefits from positive externalities from the real sector through the volume of saving. This approach implies an important role of the financial system in the determination of the level of investment. Empirical studies have shown that a number of nonfinancial factors also affect domestic investment in developing countries. This paper pays particular attention to three categories of factors: factors hypothesized by conventional investment theory (output growth and interest rate); factors related to government policy (government consumption, government borrowing, and inflation); and openeconomy factors (trade flows, foreign debt, and black market activities). Neoclassical investment theory suggests that the growth rate of real output is positively related to investment because it indicates changes in aggregate demand for output that investors seek to meet. Empirical evidence is consistent with this accelerator effect and shows that high output growth is associated with high investment rates (Fielding, 1997, 1993; Greene & Villanueva, 1991; Wai & Wong, 1982). Empirical tests have been less successful in establishing a robust negative relationship between the interest rate and investment. Neoclassical theory suggests that high interest rates raise the cost of capital, which reduces the investment rate. Government policies affect domestic investment through various channels, too. The first is that government consumption spending may crowd out domestic investment by raising interest rates, by reducing the pool of funds in the markets, and by increasing distortionary taxation on investment activities. It is also possible, however, for government spending to “crowd in” domestic investment through the accelerator channel. The net effect is theoretically unpredictable; it can only be determined empirically. Government borrowing from the domestic financial system is another factor that can reduce investment. Fischer (1993, p. 487) argues that the inflation rate serves as an indicator of the overall ability of the government to manage the economy. Since there are no good arguments for very high rates of inflation, a government that is producing high inflation is a government that has lost control. Evidently, there is little incentive to invest in a country where the government has lost control over the macroeconomic environment. Trade flows, external debt, and black market activities also affect the rate of investment in sub-Saharan African economies. Empirical evidence shows that among the many measures of openness, the flow of trade (imports and exports) appears to have the most consistent relationship with investment. Levine and Renelt (1992) find that a positive relationship between investment and trade holds whether trade flows are measured by imports, exports or total trade (imports plus exports). Studies on developing countries in general in particular find a negative relationship between external debt and domestic investment (Jenkins, 1998; Greene & Villanueva, 1991). High debt can depress investment in various ways. First, high debt implies that a higher proportion of domestic output is used to meet debt obligations. This phenomenon, referred to as “debt overhang” (Krugman, 1988), creates 86 Journal of Economic and Social Studies �Financial Determinants of Investment for Turkey a disincentive effect that discourages domestic investment. Second, high debt obligations adversely affect the country’s position in international credit markets and can even cause credit rationing. The credit rationing effect can be important. Credit rationing thus amplifies the debt overhang effect in reducing domestic investment. Third, high levels of external debt depress investment by making the macroeconomic environment more uncertain. Chronic trade deficits combined with ill-advised monetary and exchange policies have created a shortage of foreign Exchange. This and other factors have caused black market activities to flourish. High black market premia tend to induce domestic investors to substitute foreign currency hoarding for investment in physical capital. In addition, black market premia may simply be a symptom of overvaluation of national currencies. This by itself can depress investment by reducing foreign demand for domestic products. Black market premia can also be a sign of pervasive price distortions in the economy that adversely affect investment. Love and Zicchino (2006) applied vector auto regression (VAR) to firm-level panel data from 36 countries to study the dynamic relationship between firms’ financial conditions and investment by using orthogonalized impulse-response functions in order to separate the ‘fundamental factors’ from the ‘financial factors.’ They found that the impact of financial factors on investment, which indicates the severity of financing constraints, is significantly larger in countries with less developed financial systems. Their finding emphasizes the role of financial development in improving capital allocation and growth and shows that the availability of internal funds is more important in explaining investment in countries with less developed financial systems. Furthermore, the impact of a positive shock to cash flow on investment was significantly higher in countries with a ‘low’ level of financial development than in countries with a ‘high’ level of financial development and they found that positive shock to marginal productivity has less impact on investment of firms in countries with low levels of financial development. Saumitra N. Bhaduri (2005) investigated the impact of financial liberalization on the investment patterns in a developing economy, India. The empirical findings revealed mixed evidence in favor of the hypothesis that the liberalization effort has succeeded in relaxing financial constraint faced by the Indian firms. The sample set consisted of a composite and heterogeneous mix of 362 firms whose annual accounts were reported without any gap for the financial years 1989–1990 to 1994–1995. The small and young firms in the sample experienced a significant increase in financial constraint in the post-liberalization period. Koo and Maeng (2005) found that firstly, financial liberalization significantly reduced the financial constraints confronted by firms. Secondly, the effect of financial liberalization on financial constraints was stronger for small and non-chaebol firms than large and chaebol firms. This suggested that various liberalization policies implemented in financial markets helped firms to get wider access to external finance. This paper investigated whether financial liberalization affected firms’ investments in Korea. They tested for the hypothesis that financial liberalization had an impact on firms’ investment behavior. Scaperlanda and Laurence (1969) studied, by employing the least-squares multiple regression technique, the empirical data from the 1952-66 periods related to U.S. direct investment in the European Economic Community (E.E.C.). While the primary orientation of this study was to evaluate statistically the determinants of direct foreign investment, the findings also have a bearing on the effect of the E.E.C. on the patterns of U.S. direct investment. No statistical evidence was found in Volume 1 Number 1 January 2011 87 �Okyay UÇAN & Özlem ÖZTÜRK support of the tariff-discrimination hypothesis and while the relationship between the size of market variable and U.S. direct investment appeared to have been somewhat affected by the establishment of the E.E.C., the stability of the market-size elasticity between the pre- and post-E.E.C. periods indicated that the E.E.C. has had little impact on the sensitivity of I to changes in Y. Günçavdı and Bleaney (2005) tested for shifts in aggregate private investment functions for Turkey as a consequence of financial liberalization in the early 1980s. Results for a neoclassical model in error correction form suggested that the short-run dynamics of investment were altered by financial liberalization, with reduced sensitivity to the availability of credit, but with no evidence of increased sensitivity to the cost of capital. Estimation of an Euler equation model indicated that credit constraints remained binding after liberalization. They interpreted this as evidence of significant structural change to private investment functions after financial liberalization, but with credit constraints continuing to operate. They attempted to estimate the impact of financial liberalization on the dynamics of private investment in Turkey by using an error correction version of the standard neoclassical model and also by an Euler equation approach based on the first-order conditions for dynamic profit maximization. In both cases, the results were limited by the use of aggregate data but it was important to test to what extent aggregate investment equations were liable to structural change under financial liberalization. In the case of the error-correction model, they were able to find statistically significant evidence of structural change. Credit variables became much less important after liberalization, as expected, although cost variables did not become more important. The Euler equation model, on the other hand, displayed no significant structural break associated with financial liberalization. The results for the Euler equation model suggested that credit constraints were binding both before and after liberalization, which would explain why this model does not display structural instability, since the specification depends only on whether the constraint is binding (Güncavdı, Bleaney, 2005, 445;452). Ndikumana (2006) investigated the effects of financial development on domestic investment in a sample of 30 sub-Saharan African countries. It was based on a dynamic serial-correlation investment model including various indicators of financial development, controlling for country-specific fixed effects and nonfinancial factors of investment. The results indicated a positive relationship between domestic investment (total investment and private investment) and various indicators of financial development. Higher financial development led to higher future levels of investment, implying a potent long-run effect of financial development on domestic investment. The findings implied that financial development could stimulate economic growth through capital accumulation. Ang (2009) suggested that significant directed credit programs favoring certain priority sectors tended to discourage private capital formation in both countries. Interest rate controls appeared to have a positive impact on private investment, with the effect being more pronounced in Malaysia. While high reserve and liquidity requirements exerted a negative influence on private investment in India, the effect was found to be positive in Malaysia. The empirical evidence showed a significant steady-state relationship between private investment and its determinants. The results suggested that financial repressionist policies, in the form of significant directed credit controls, appear to have retarded private investment in both India and Malaysia. However, contrary to the financial liberalization thesis, interest rate restraints appeared to be an effective device in stimulating private investment in both countries. While high reserve and liquidity requirements tended to have an 88 Journal of Economic and Social Studies �Financial Determinants of Investment for Turkey undesirable effect on private investment in India, they were found to be favorable in Malaysia. Malaysia had lower corruption and better law and order compared to India. In sum, their results tended to support the proposition that some form of financial restraint might have stimulated private investment. Jongwanich and Kohpaiboon (2008) examined patterns and determinants of private investment in an attempt to understand why levels of private investment in South East Asia have not yet fully recovered, using Thailand as a case study. The private investment equation was estimated during the period 1960–2005. They found that it was capital fund shortages rather than existing spare capacity that hindered short-run investment recovery. In the long run, policy emphasis should have been on promoting a conducive investment climate. The key finding was that private investment in Thailand had borne the brunt of aggregate demand contraction since the outbreak of the Asian financial crisis in 1997. Among the short-run investments, credit shortage was the most important constraint on investment recovery following the crisis. In the long run, private investment was mostly determined by business opportunities and investment costs. Spiegel (2000) indicated that financial development positively influenced both rates of investment and total factor productivity growth. In summary, they found that different aspects of financial development had positive influences on total factor productivity growth and rates of factor accumulation. While the ratio of private-sector liabilities to income had a fairly robust impact on total factor productivity growth, it entered with its predicted sign only in the accumulation of physical capital, and even then the performance of that variable was not robust to the inclusion of country fixed effects. The indicator of liquid liabilities as a share of income performs similarly, although it was not as robust to the inclusion of fixed effects as the private liabilities ratio in the growth equations. Morck et al. (1990) tried to question how the stock market affected investment. For their analysis, they investigated the effect of stock returns and the growth in fundamental variables on investment growth in order to see how important the stock market was after controlling for fundamentals. The firm-level regressions showed that movements in relative share prices were associated with fairly large and statistically significant investment changes when fundamentals were held constant, but the incremental R2 from relative stock returns was fairly small. The cross-sectional variability of investment was sufficiently large that relative stock returns could account for only a small part of it. They argued that the explanatory power of relative stock returns for investment was unlikely to be evidence that the stock market provided new information to managers, since managers probably learned little from the market about their own firms’ idiosyncratic prospects. They also provided evidence that the relation between relative stock returns and investment was not driven by the costs of external financing. The explanatory power of relative stock returns for investment might be evidence of the market exerting pressure on managers, although it also seemed likely that the market was picking up the effect of imperfectly measured fundamentals. Gordon (1964) presented the results of empirical work undertaken to test the theory’s ability to explain the annual investment of the firm. 23 large chemical corporations were selected and data was obtained to provide the value of each variable for the years 1954 through 1960, thereby providing 161 observations on each variable. This study’s conclusions might be summarized as follows: First, a large proportion of the scale deflated variation in investment in the sample was explained by the security flow variable, the investment consistent with maintaining a satisfactory level of security made Volume 1 Number 1 January 2011 89 �Okyay UÇAN & Özlem ÖZTÜRK possible by the period’s operations. Second, the correlation with the profitability variable was quite small. While this might be due to lack of influence on the level of investment, it was quite possible that the cause was the limitations of the change in sales as a measure of rate of return on investment. Methodology Let us first define d to be the degree of integration of a time series At. More precisely, if At achieves stationary after being differenced d times, it is said to be integrated of order d, denoted by At ~ I(d). Thus, an I(1) variable is a variable that achieves stationarity after being differenced one. A necessary condition for testing for a long-run relationship between two or more variables is that these variables are I(1), i. e., stationary in first differences. We, therefore, first test for a unit root using the conventional unit root tests such as ADF (see Dickey and Fuller, 1981; Said and Dickey, 1984) and KPSS (Kwiatkowski-Phillips-Schmidt-Shin, 1992) tests. Besides, in order to capture any possible structural shift over the estimation period, we also employ Perron (1997). This test allow for a single break at an unknown time under the alternative hypothesis of trend-stationarity. While it is true that the Perron (1997) test, by virtue of accounting for one structural break, is an advance over standard ADF and KPSS tests, it is argued that the Perron test may lose power when confronted with two or more breaks (see Lee and Strazicich, 2003). To address this problem, there is a procedure devised by LS (2003) which proposes a model that tests endogenously for two or more structural breaks. If the examined series for the given country are deemed to be I(1), the next step is to establish whether they are cointegrated or not. To this end, we relied on the Johansen & Juselius (1990) cointegration test. If the series are not cointegrated we follow with the Vector Autoregressive (VAR) model. The VAR model, recently, is one of the most used time series analysis. The VAR model is flexible, easy to estimate, and it usually gives a good fit to macroeconomic data. However, the possibility of combining long-run and short-run information in the data by exploiting the cointegration property is probably the most important reason why the VAR model continues to receive the interest of both econometricians and applied economists. Since, there is no constraint in the VAR approach and it gives dynamic relations between the variables, it is used more for the time series (Keating, 1990:453-454). One of the advantages of the VAR approach is that variables are endogenously determined. Secondly, the VAR approach allows the variables depending more than white noise terms or lag of variables. Thus VAR model becomes more flexible than AR models and since the VAR approach includes most properties of data, it offers a strong structure. Moreover, every equation may be solved by OLS (Ordinary Least Squares) because there is no simultaneous terms right side of the equation. As Sims (1980) claims, huge scaled structural models fail to forecast out of sample estimation (Brooks, 2008:291; Greene, 1993:553). Simply, as y1t and y2t are the variables, the VAR model can be defined as follows (Brooks, 2008:290): (1) 90 Journal of Economic and Social Studies �Financial Determinants of Investment for Turkey Current values of variables depend on the k lag of the variables and error terms. Here it is white noise error term and while i=1,2 E (uit) =0 ve E (u1tu2t)=0. The assumption of uncorrelated error terms with their lagged values makes no constriction to the VAR model. The reason for this is that by increasing the lag length of the variables autocorrelation problem is achieved (Güloğlu ve Özgen, 2004:96). Lag length is determined by Akaiki Information Criteria (AIC), Schwarz Information Criteria (SIC), Hannan-Quinn Information Criteria (HQ) and Final Prediction Error (FPE). All variables have to be stationary in the VAR approach. But, Sims (1980) and Sims, Stock and Watson (1990) claim that whether variables have unit root or not, differencing procedure needs not to be applied. When differencing procedure is applied, long-run relationship between the variables can be lost (Enders, 2004:270). On the other hand, Fuller (1976) asserts that differencing a series doesn’t show an asymptotic efficiency (Günçavdı et al., 2000: 160). Thus, if the series are not cointegrated, VAR approach is used with differencing I(1) variables. In this paper we use both Sims and Fullers approach, and then we write the best one that explains the economic theory. To interpret the VAR results impulse-response function, variance decomposition and granger causality are used. Data and Formation of Variables All data are gathered from International Financial Statistics online services reported by the International Monetary Fund (IMF). This publication has annual data for Turkey from 1970 to 2009. The variables used in this paper are: Total gross domestic investment as a percentage of gross domestic products (GDP) : It Private domestic investment as a percentage of GDP: PIt Real per capita gross domestic product : PGDPt Growth rate of GDP deflator (Inflation) : Inft Discount rate (real interest rate): r Financial development indicators (see Ndikumana, 2000; Levine, 1997; Demirgüc-Kunt and Maksimovic, 1996; Lynch, 1996 for a discussion of measuring the items of financial development). By following Ndikumana (2000), Total credit to the private sector as a percentage of GDP (FD1) The ratio of broad money to GDP is used as a measure of size of the financial sector (FD2) The relative importance of banks in the supply of credit is measured by total domestic credit provided by the banking sector as a percentage of GDP (FD3) Claims on government as a percentage of GDP (FD4) A composite index of financial development (FDindex) The formula for the FDindex that is developed by Demirgüc-Kunt and Levine (1996) is adapted to our paper as the following: (1) Volume 1 Number 1 January 2011 91 �Okyay UÇAN & Özlem ÖZTÜRK where FDi is an indicator of financial development, is the sample mean of indicator i. Since we have four financial development indicators, we use 4 for the maximum number of total symbol. In this paper, we create two main models dealing with It and PIt. In these models besides real interest rate (r), real per capita gross domestic product (PGDPt), inflation (inf); financial development indicators are changed to each other to see their individual effect on It and PIt, i.e.: Empirical Results Unit root test results To start with the ADF and KPSS tests were applied both on levels and on the first differences of It, PIt, PGDPt, Inft, r, FD1, FD2, FD3, FD4 and FDindex. The results are summarized in Table 1. The tests on the level series (Table 1) unequivocally indicate that all variable are non-stationary for the ADF but KPSS test indicates that PIt, FD1, FD2, FD4 and FDindex are stationary. In first differences, except Inft and FD2 with KPSS test, all variables are stationary. Table 1. ADF and KPSS Unit-root Test Results ADF KPSS First First Level Level Difference Difference Variables tψ tψ tψ tψ -2,028 -6,517(*) 0,157 0,053 It -1,410 -4,592(*) 0,109 0,065 PIt -4,274 -7,519(*) 0,768 0,118 PGDPt -0,021 -5,697(*) 0,175 0,112 r -0,729 -6,141(*) 0,179 0,170 Inft FD1 -2,287 -4,907(*) 0,100 0,133 FD2 -3,337 -8,246(*) 0,135 0,500 FD3 -1,154 -4,800(*) 0,148 0,076 FD4 -2,608 -6,854(*) 0,083 0,053 FDindex -1,699 -5,546(*) 0,120 0,056 %1 -4,211 -4,211 0,216 0,216 Critical %5 -3,529 -3,529 0,146 0,146 Values %10 -3,196 -3,196 0,119 0,119 Note: All regression variables are in logarithm. Asterisks *, **, ***, show significance of 1%, 5% and 10% levels respectively. The critical values are obtained from MacKinnon (1991) for the ADF test and from Kwiatkowski et al. (1992) for the KPSS test. ADF test examines the null hypothesis of a unit root against the stationary alternative. KPSS tests the stationarity null hypothesis against the alternative hypothesis of a unit root. 92 Journal of Economic and Social Studies �Financial Determinants of Investment for Turkey I.e., Since ADF and KPSS tests are in conflict, as we emphasize in the methodology part, there may be structural breaks. So we apply Perron and LS unit root tests to see more accurate results. Perron (1997) and Lee-Strazicich (2003) tests indicate that regime shifts in all the variables. Table 2. Perron and LS Unit-root Test Results Perron(1997) First Difference TB tψ tψ -4,83 2002 -7,54* Level Level Variables It TB 1997 PIt 1991 -5,52** 2002 -5,23 2000D PGDPt 1996 -5,81* 1997 -9,03* 1983DT Inft 1992 -3,57 1980 -6,74* 1999DT TB 1997DT LS(2003) First Difference tψ TB tψ -3,97 2005DT -7,78* 1996DT -6,52* -3,69 2002DT -4,58 1996DT -8,42* 1979DT -7,91* -4,53 1982DT 1981DT -4,04 1978DT -7,32* 2004DT 1998D 1994DT FD1 2006 -5,10 1996 -7,06* -5,91 -5,90* 2005DT 1994D FD2 2007 -4,57 2008 -4,16 1987DT -5,17 1979DT -9,10* 1986DT -4,74 1990D -6,00* FD3 1987 -4,48 2000 -5,20 1992DT 1989D 1987DT -11,39* -4,61 FD4 1989 -4,42 1989 -10,65* 1992DT 1991DT FDindex 1987 -4,52 1989 -6,34* -----4,04 1985D -6,33* %1 -6,32 -6,32 -6,28 -6,28 Critical %5 -5,59 -5,59 -5,62 -5,62 Values %10 -5,29 -5,29 -5,24 -5,24 Note: All regression variables are in logarithm. Asterisks *, **, ***, show significance of 1%, 5% and 10% levels respectively. TB shows the structural break time. D is dummy variable and DT is trend dummy. The critical values are obtained from Perron (1997) for the Perron test and from Lee and Strazicich (2003, 2004) for the LS test. Perron and LS tests examine the null hypothesis of a unit root against the stationary alternative. r 1986 -2,46 1979 -7,58* The results of Perron and LS unit root tests presented in Table 2 provide further evidence of the existence of a unit root when breaks are allowed. The estimated results of unit root tests indicate that all variables are I(1) as the null hypothesis of non-stationary can not be rejected at conventional significance levels. Co-integration test results Since it is agreed that all variables are I(1), we start to investigate whether there is a long run relation between the variables in Model A and B. Table 3 shows the co-integration relation of (total gross domestic investment as a percentage of gross domestic products) It with financial development indicators and a list of control variables consisting of inflation, growth rate of real per capita GDP and real interest rate. Volume 1 Number 1 January 2011 93 �Okyay UÇAN & Özlem ÖZTÜRK Table 3. Johansen Co integration Results Where r=Number of Co-integrating Vectors for Model A   max Variables in Cointegrating Vector Model with FD1 Model with FD2 Model with FD3 Model with FD4 Model with FDindex Alternative 1 lag %95 Critical Value 1 lag %95 Critical Value r 1 r2 r 1 r2 r 1 r2 r 1 r2 r 1 r2 69,98 42,31 69,81 47,85 27,67 24,47 33,87 27,58 62,19 40,19 69,81 47,85 21,99 17,79 33,87 27,58 64,01 39,46 69,81 47,85 24,55 17,84 33,87 27,58 59,99 30,87 69,81 47,85 29,12 12,12 33,87 27,58 60,82 36,99 69,81 47,85 23,83 15,06 33,87 27,58 Trace Null r0 r 1 r0 r 1 r0 r 1 r0 r 1 r0 r 1 Table 4 shows the co-integration relation of (private domestic investment as a percentage of GDP) PIt with financial development indicators and a list of control variables consisting of inflation, growth rate of real per capita GDP and real interest rate. Table 4. Johansen Co-integration Results Where r=Number of Co-integrating Vectors for Model B   max Variables in Cointegrating Vector Model with FD1 Model with FD2 Model with FD3 Model with FD4 Model with FDindex Alternative 1 lag %95 Critical Value 1 lag %95 Critical Value r 1 r2 r 1 r2 r 1 r2 r 1 r2 r 1 r2 65,84 39,63 69,81 47,85 26,21 21,30 33,87 27,58 65,15 39,98 69,81 47,85 25,17 18,68 33,87 27,58 66,41 42,87 69,81 47,85 23,53 21,69 33,87 27,58 57,10 33,53 69,81 47,85 23,56 16,10 33,87 27,58 62,99 41,01 69,81 47,85 21,98 20,26 33,87 27,58 Trace Null r0 r 1 r0 r 1 r0 r 1 r0 r 1 r0 r 1 Thus, since series are not co-integrated in both Table 3 and Table 4, the VAR approach is used with differencing all I(1) variables to make them stationary. 94 Journal of Economic and Social Studies �Financial Determinants of Investment for Turkey Variance Decomposition Results Model A Here, for all models, total gross domestic investment is affected by itself along all 10 periods time. Especially, Model A-FD1 analysis shows that relative to the other models at the end of 10th period effect of total credit to private sector (FD1) is evidently determined (See Appendix-i (Model A)). Model B At the end of the 10th period private domestic investment, for all models, is affected from inflation and real interest rate variables as a percentage of 30% (See Appendix-i (Model B)). Especially, in the model with FD3 (domestic credit provided by the banking sector), relative to other financial development items, at the end of 10th period domestic credit’s effect is higher on private domestic investment (See Appendix-i (Model B-FD3)). Impulse Response Analysis Impulse response functions describe how the economy reacts over time to exogenous impulses, which economists usually call ‘shocks’, and are often modeled in the context of a vector auto-regression. Impulse response functions describe the reaction of endogenous macroeconomic variables at the time of the shock and over subsequent points in time. Impulse Response Analysis results are given in the appendix-ii by graphs. We investigate the dynamic responses of variables against one unit standard error shocks to the other variables. While dotted lines show the significance bounds, straight line gives the estimates. Model A Against the effect of an exogenous shock or innovation in FD1, FD3, and FD4 there exists a positive effect on the total investment in the short term and then at the end of 10th period total investment tends to its mean value. Against the effect of an exogenous shock in FD2 there exists a negative effect on the total domestic investment. As a total effect of financial development items we can analyze the effect of a shock in FDindex. As a result of this a positive effect also occurs on total domestic investment (It) in the short term but It again tends to its mean at the end of the 10th period. On the other hand, as an impact to the effect of an exogenous shock in control variables (inflation, real interest rate and real per capita GDP), a negative effect occurs on the total domestic investment in the short term. Model B Against the effect of an exogenous shock or innovation in FD1,FD2, and FD4 there exists a positive effect on the private investment in the short term and then at the end of the 10th period private investment tends to its mean value. As an impact to the effect of an exogenous shock in FD3 there Volume 1 Number 1 January 2011 95 �Okyay UÇAN & Özlem ÖZTÜRK exists a negative effect on the private domestic investment. As a total effect of financial development items we can analyze the effect of a shock in FDindex. As a result of this a positive effect also occurs on private domestic investment (It) in the short term but It again tends to its mean at the end of 10th period. On the other hand, against to the effect of an exogenous shock in inflation and real interest rate, there exists a negative effect on the private domestic investment. But, after an effect of an exogenous shock in real per capita GDP, there exists a positive effect on private investment. Granger Causality Results The term causality suggests a cause and effect relationship between two sets of variables. Table 5 shows the Granger Causality results of Model A with FD1. Model A Table 5. Model A with FD1 Null Hypothesis: Obs F-Statistic Probability DLNPGDP does not Granger Cause DLNI DLNI does not Granger Cause DLNPGDP 37 1.00114 1.84336 0.37868 0.17470 DR does not Granger Cause DLNI DLNI does not Granger Cause DR 37 0.74486 0.18083 0.48285 0.83542 DLNFD1 does not Granger Cause DLNI DLNI does not Granger Cause DLNFD1 37 4.48833 0.60694 0.01913 0.55117 DLNINF does not Granger Cause DLNI DLNI does not Granger Cause DLNINF 37 0.78625 1.96594 0.46415 0.15657 In the Table 5 it is seen that total credit to the private sector (FD1) may Granger-cause total gross domestic investment. Since no other Granger cause is found, other tables are given in the appendixiii. Model B Here, inflation and real interest rate may Granger-cause private domestic investment for all Granger causality results for model B. On the other hand, there is not any significant Granger cause between the other variables (See appendix-iii (Model B)) 96 Journal of Economic and Social Studies �Financial Determinants of Investment for Turkey Conclusion The study is based on Turkish data from 1970 to 2009. The results mainly indicate a positive relationship between total domestic investment and all four indicators of financial development as we create a composite index of financial development items. The results are qualitatively similar for total domestic investment and private investment, with stronger effects of financial factors on private investment than on total domestic investment. The findings also suggest that high financial development is a predictor of future levels of domestic investment. Higher financial development in the 1980s is associated with higher investment levels in the 1990s and 2000s. The results also confirm stylized facts for other determinants of investment. Inflation and real interest rate negatively affect total domestic investment. Domestic investment, however, is negatively affected by real per capita GDP growth (accelerator effect). On the other hand, private investment is positively affected by real per capita GDP growth. References Anderson, T. W., & Hsiao, C. (1981). Estimation of dynamic models with error components. Journal of the American Statistical Association, 76 (375), 598- 606. Anderson, T. W., & Hsiao, C. (1982). Formulation and estimation of dynamic models using panel data. Journal of Econometrics, 18, 83-114. Ang, J., B., (2009). Private Investment and Financial Sector Policies in India and Malaysia. World Development, 37(7), 1261–1273, Atiyas, I. Ersel, I. (1995). “The Impact of Financial Reform: The Turkish Experience,” in G. Caprio, I. Atiyas and J. Hanson (eds.), Financial Reform: Theory and Evidence, (New York: Cambridge University Press). Barro, R.J. (1991). Economic growth in a cross section of countries. Quarterly Journal of Economics, 106 (2), 407-443. Bayoumi, T. (1990) Saving-investment correlations, Immobile capital government policy or endogenous behavior? IMF Staff Papers, 37 (2), 360-387. Ben-David, D. (1998). Convergence clubs and subsistence economies. Journal of Development Economics, 55, 155-171. Benhabib, J., & Spiegel, M. (2000). The Role of Financial Development in Growth and Investment. Journal of Economic Growth, 5, 341–360 Bhaduri, S., N. (2005). Investment, financial constraints and financial liberalization: Some stylized Volume 1 Number 1 January 2011 97 �Okyay UÇAN & Özlem ÖZTÜRK facts from a developing economy, India. Journal of Asian Economics, 16, 704–718 Brooks, C. (2008). Introductory Econometrics for Finance, Second Edition, Cambridge University Press. Chari, V. V., Kehoe, P. J., & McGrattan, E. R. (1997). The poverty of nations: a quantitative investigation. Federal Reserve Bank of Minneapolis, Research Department Staff Report 204. Dooley, M., Frankel, J. & Mathieson, D. J. (1987). International capital mobility: what do the saving investment correlations tell us? IMF Staff Papers, 34, 503-530. Enders, W. (2004). Applied Econometric Time Series, New York:Wiley. Feldstein, M., & Horioka, C. (1980). Domestic saving and international capital flows. The Economic Journal, 90, 314-329. Fielding, D. (1997). Adjustment, trade policy and investment slumps: evidence from Africa. Journal of Development Economics, 52, 121-137. Fielding, D. (1993). Determinants of investment in Kenya and C’ote d’Ivoire. Journal of African Economies, 2 (3), 299-328. Fischer, S. (1993). The role of macroeconomic factors in growth. Journal of Monetary Economics, 32, 485- 512. Fuller, W.A. (1976). Introduction to Statistical Time Series, New York:Wiley. Gordon, J., M. (1964). Security and Investment: Theory and Evidence. The Journal of Finance, 19(4), 607-618 Greene, J., & Villanueva, D. (1991). Private investment in developing countries. IMF Staff Papers, 38 (1), 33-58. Greenwood, J., & Smith, B. D. (1997). Financial markets in development and the development of financial markets. Journal of Economic Dynamics and Control, 21, 145-181. Güloğlu B., & Özgen, F. B. (2004). Türkiye’de İç Borçların İktisadi Etkilerinin VAR Tekniğiyle Analizi, ODTÜ Gelişme Dergisi, 31(Haziran). Guncavdı Ö., Bleaney M., & McKay A. (1998). Financial liberalisation and private investment: evidence from Turkey. Journal of Development Economics, 57, 443–455 Günçavdı, Ö., Levent H. ve Ülegin B. (2000). Yüksek ve Değişken Enflasyon Tahmininde Alternatif 98 Journal of Economic and Social Studies �Financial Determinants of Investment for Turkey Modellerin Karşılaştırılması: Türkiye Örneği, ODTÜ Gelişme Dergisi, 149-171. Gurley, J. G., & Shaw, E. S. (1955). Financial aspects of economic development. American Economic Review, 45 (4), 515-538. Hsiao, C. (1986). Analysis of panel data. Cambridge, New York: University Press. International Finance Corporation (1998). Trends in private investment in developing countries. Statistics for 1970-96, Discussion paper 34. Jenkins, C. (1998). Determinants of private investment in Zimbabwe. Journal of African Economies, 7 (1), 34-61. Jongwanich, J. (2008). Private Investment: Trends and Determinants in Thailand. World Development, 36(10), 1709–1724 Keating, J. W. (1990). Identifying VAR Models Under Rational Expectations. Journal of Monetary Economics, 25. Keynes, M. J. (1937). The General Theory of Employment, Interest, and Money. New York, Harcourt Brace Jovanovich, reprinted, 1953. Khan, M. S., & Reinhart, C. M. (1990). Private investment and economic growth in developing countries. World Development, 18 (1), 19-27. Koo, J., & Maeng, K. (2005). The effect of financial liberalization on firms’ investments in Korea, Journal of Asian Economics, 16, 281–297 Kormendi, R. C., & Meguire, P. G. (1985). Macroeconomic determinants of growth: cross-country evidence. Journal of Monetary Economics, 16, 141-163. Krugman, P. (1988). Financing vs. forging a debt overhang, Journal of Development Economics, 29, 253-268. Levine, R. (1997). Financial development and economic growth: views and agenda. Journal of Economic Literature, 35, 688-726. Levine, R., & Renelt, D. (1992). A sensitivity analysis of cross-country growth regressions, American Economic Review, 82 (4), 942-963. Love, I., & Zicchino L. (2006). Financial development and dynamic investment behavior: Evidence from panel VAR. The Quarterly Review of Economics and Finance, 46, 190–210 McKinnon, R. (1973). Money and capital in economic development, Washington DC, Brookings Volume 1 Number 1 January 2011 99 �Okyay UÇAN & Özlem ÖZTÜRK Institution. Ndikumana, L. (2000). Financial Determinants of Domestic Investment in Sub-Saharan Africa: Evidence from Panel Data. World Development, 28(2), 381-400 Scaperlanda, A., & Mauer L., J. (1969). The Determinants of U.S. Direct Investment in the E.E.C. The American Economic Review, 59(4), 558-568 Schiantarelli, F. (1996). Financial constraints and investment methodological issues and international evidence. Oxford Review of Economic Policy, 12(2), 70-89 Schumpeter, J. A. (1932). The theory of economic development, Cambridge: Harvard University Press. Shaw, E. S., (1973). Financial deepening in economic development, New York: Oxford University Press. Wai, T. U., & Wong, C., H. (1982). Determinants of private investment in developing countries. Journal of Development Studies, 19 (1), 19-36. 100 Journal of Economic and Social Studies �Financial Determinants of Investment for Turkey Appendix-i MODEL A Variance Decomposition Tables Variance Decomposition of DLNI:Model A-FD1 Period S.E. DLNI DLNFD1 1 2 3 4 5 6 7 8 9 10 0.108189 0.114674 0.116760 0.117296 0.117427 0.117456 0.117465 0.117468 0.117468 0.117468 100.0000 91.31028 88.10730 87.56561 87.37640 87.34178 87.33265 87.32892 87.32812 87.32794 0.000000 4.141264 4.600590 4.802083 4.845550 4.843689 4.849448 4.850649 4.850598 4.850696 Variance Decomposition of DLNI:Model A-FD2 Period S.E. DLNI DLNINF 1 2 3 4 5 6 7 8 9 10 0.108458 0.115583 0.117364 0.117540 0.117549 0.117552 0.117553 0.117553 0.117553 0.117553 100.0000 90.72336 88.51779 88.43920 88.42730 88.42511 88.42507 88.42506 88.42505 88.42505 0.000000 2.729410 3.847332 3.842494 3.852711 3.854248 3.854273 3.854279 3.854284 3.854284 Variance Decomposition of DLNI: Model A-FD3 Period S.E. DLNI DLNINF 1 2 3 4 5 0.108065 0.115639 0.116991 0.117374 0.117446 100.0000 89.56568 87.60536 87.10785 87.01222 Volume 1 Number 1 January 2011 0.000000 3.279691 4.342173 4.333262 4.388973 DLNINF DLNPGDP DR 0.000000 0.614225 3.056185 3.037059 3.127242 3.160769 3.161032 3.162232 3.163013 3.163040 0.000000 0.982688 0.962059 1.091725 1.091581 1.094995 1.096584 1.096610 1.096644 1.096680 0.000000 2.951546 3.273865 3.503523 3.559228 3.558767 3.560289 3.561588 3.561628 3.561647 DLNPGDP DR DLNFD2 0.000000 3.081509 3.520241 3.530372 3.529824 3.529811 3.529792 3.529793 3.529793 3.529793 0.000000 3.400433 3.662834 3.721340 3.723527 3.723440 3.723418 3.723430 3.723431 3.723431 0.000000 0.065293 0.451802 0.466594 0.466635 0.467393 0.467443 0.467443 0.467444 0.467444 DLNPGDP DR DLNFD3 0.000000 3.311463 3.892787 3.910996 3.910662 0.000000 3.312430 3.398263 3.517622 3.516091 0.000000 0.530732 0.761415 1.130270 1.172049 101 �Okyay UÇAN & Özlem ÖZTÜRK 6 7 8 9 10 0.117453 0.117453 0.117453 0.117453 0.117453 87.00204 87.00186 87.00182 87.00181 87.00181 4.391217 4.391203 4.391203 4.391203 4.391203 3.912851 3.912857 3.912859 3.912859 3.912859 Variance Decomposition of DLNI: Model A-FD4 Period S.E. DLNI DLNINF DLNPGDP 1 2 3 4 5 6 7 8 9 10 0.108420 0.115961 0.117277 0.117396 0.117410 0.117412 0.117412 0.117412 0.117412 0.117412 100.0000 89.84121 88.01535 87.85444 87.83481 87.83338 87.83337 87.83335 87.83335 87.83335 0.000000 3.438107 4.623120 4.616059 4.628621 4.629506 4.629501 4.629511 4.629513 4.629513 0.000000 3.219958 3.656558 3.655720 3.658869 3.659176 3.659171 3.659171 3.659171 3.659171 Variance Decomposition of DLNI: Model A -FDindex Period S.E. DLNI DLNINF DLNPGDP 1 2 3 4 5 6 7 8 9 10 0.107382 0.115634 0.116954 0.117296 0.117336 0.117341 0.117341 0.117341 0.117341 0.117341 100.0000 88.22844 86.28256 85.89115 85.84443 85.83787 85.83779 85.83778 85.83777 85.83777 0.000000 3.415930 4.611005 4.591966 4.617251 4.616976 4.616990 4.616989 4.616989 4.616989 0.000000 3.567548 4.194833 4.224296 4.221599 4.225583 4.225597 4.225599 4.225602 4.225602 3.520461 3.520542 3.520550 3.520550 3.520550 1.173432 1.173535 1.173564 1.173576 1.173580 DR DLNFD4 0.000000 3.360777 3.555080 3.704141 3.708052 3.708235 3.708251 3.708257 3.708257 3.708258 0.000000 0.139953 0.149895 0.169644 0.169644 0.169705 0.169706 0.169706 0.169706 0.169706 DR DLNFDINDEX 0.000000 3.338606 3.307889 3.413893 3.425451 3.428218 3.428222 3.428231 3.428232 3.428232 0.000000 1.449480 1.603709 1.878697 1.891273 1.891353 1.891399 1.891403 1.891406 1.891407 DLNPGDP DR MODEL B Variance Decomposition Tables Variance Decomposition of DLNPI: Model B-FD1 Period S.E. DLNPI DLNFD1 DLNINF 102 Journal of Economic and Social Studies �Financial Determinants of Investment for Turkey 1 2 3 4 5 6 7 8 9 10 0.111704 0.131884 0.137132 0.138684 0.139195 0.139354 0.139404 0.139424 0.139430 0.139432 100.0000 72.90510 68.00580 67.20303 66.76867 66.64278 66.62735 66.61298 66.60742 66.60673 0.000000 0.361952 0.997606 1.899030 2.102646 2.097979 2.116462 2.124081 2.123991 2.124521 0.000000 11.97497 16.18442 15.85363 15.95689 16.05452 16.04905 16.05042 16.05463 16.05470 Variance Decomposition of DLNPI: Model B-FD2 Period S.E. DLNPI DLNINF DLNPGDP 1 0.111371 100.0000 0.000000 0.000000 2 0.134236 70.59143 12.47682 0.038478 3 0.137751 67.04588 15.11256 0.149176 4 0.139147 66.73769 14.82891 0.782973 5 0.139750 66.77630 14.85455 0.790853 6 0.139950 66.64188 14.96127 0.790099 7 0.139992 66.60480 14.98001 0.791385 8 0.140010 66.60431 14.97613 0.794327 9 0.140019 66.60358 14.97733 0.794818 10 0.140022 66.60177 14.97888 0.794796 Variance Decomposition of DLNPI: Model B-FD3 Period S.E. DLNPI DLNINF DLNPGDP 1 2 3 4 5 6 7 8 9 10 0.113117 0.132713 0.137096 0.139309 0.139943 0.140037 0.140066 0.140079 0.140082 0.140083 100.0000 73.91439 69.90781 68.45069 67.95800 67.87165 67.86658 67.85956 67.85689 67.85654 0.000000 8.857569 9.850386 9.871808 10.25433 10.34304 10.33949 10.34367 10.34636 10.34638 0.000000 0.283118 0.284117 0.564892 0.565699 0.572337 0.577291 0.578136 0.578120 0.578234 Variance Decomposition of DLNPI: Model B-FD4 Period S.E. DLNPI DLNINF DLNPGDP Volume 1 Number 1 January 2011 0.000000 0.144374 0.608559 1.005528 1.002418 1.019089 1.030832 1.031342 1.031667 1.032140 0.000000 14.61360 14.20362 14.03878 14.16937 14.18564 14.17631 14.18118 14.18229 14.18191 DR DLNFD2 0.000000 15.53199 15.75905 15.52096 15.38871 15.42208 15.43548 15.43307 15.43141 15.43176 0.000000 1.361280 1.933336 2.129465 2.189593 2.184673 2.188322 2.192163 2.192866 2.192788 DR DLNFD3 0.000000 16.86352 17.15470 16.61792 16.48073 16.47663 16.47040 16.46838 16.46852 16.46845 0.000000 0.081402 2.802990 4.494690 4.741239 4.736345 4.746241 4.750247 4.750106 4.750396 DR DLNFD4 103 �Okyay UÇAN & Özlem ÖZTÜRK 1 2 3 4 5 6 7 8 9 10 0.111575 0.135002 0.139053 0.139725 0.139963 0.140040 0.140055 0.140058 0.140060 0.140060 100.0000 69.47911 65.95623 65.97560 65.89605 65.82322 65.81694 65.81719 65.81617 65.81589 0.000000 12.06639 15.14679 15.04366 15.07943 15.12762 15.13298 15.13232 15.13292 15.13313 0.000000 0.130478 0.157194 0.358112 0.370389 0.370459 0.371933 0.372470 0.372480 0.372491 Variance Decomposition of DLNPI: Model B-FDindex Period S.E. DLNPI DLNINF DLNPGDP 1 2 3 4 5 6 7 8 9 10 104 0.112972 0.133776 0.137682 0.139539 0.140144 0.140245 0.140275 0.140291 0.140296 0.140296 100.0000 72.90162 69.28895 68.46558 68.19073 68.09424 68.08200 68.07689 68.07403 68.07341 0.000000 10.08486 12.04079 11.86915 12.10474 12.19810 12.19796 12.19878 12.20180 12.20225 0.000000 0.190314 0.179788 0.484045 0.483038 0.488077 0.491548 0.492549 0.492521 0.492605 0.000000 16.39682 16.91693 16.75737 16.77625 16.80167 16.80126 16.80084 16.80122 16.80129 0.000000 1.927210 1.822856 1.865255 1.877879 1.877024 1.876882 1.877178 1.877210 1.877201 DR 0.000000 16.60307 16.82740 16.41971 16.27832 16.27891 16.27659 16.27285 16.27223 16.27227 0.000000 0.220144 1.663077 2.761518 2.943173 2.940671 2.951904 2.958937 2.959425 2.959468 Journal of Economic and Social Studies �Financial Determinants of Investment for Turkey Appendix-ii MODEL A Impulse Response Analysis Volume 1 Number 1 January 2011 105 �Okyay UÇAN & Özlem ÖZTÜRK MODEL B Impulse Response Analysis1 1 In this analysis “invy” , “pinvy” , “pery” are used instead of “I”, “PI”, PGDP” respectively. 106 Journal of Economic and Social Studies �Financial Determinants of Investment for Turkey Appendix-iii MODEL A Granger Causality Results Model A – FD2 Null Hypothesis: Obs F-Statistic Probability DLNINF does not Granger Cause DLNI DLNI does not Granger Cause DLNINF 38 0.84739 4.02848 0.36359 0.05251 DLNFD2 does not Granger Cause DLNI 38 0.66461 0.42045 1.04756 0.31309 DLNI does not Granger Cause DLNFD2 DR does not Granger Cause DLNI DLNI does not Granger Cause DR 38 1.48965 0.10038 0.23043 0.75326 DLNPGDP does not Granger Cause DLNI 38 1.44872 0.23681 0.43703 0.51289 Obs F-Statistic Probability DLNINF does not Granger Cause DLNI DLNI does not Granger Cause DLNINF 38 0.84739 4.02848 0.36359 0.05251 DR does not Granger Cause DLNI DLNI does not Granger Cause DR 38 1.48965 0.10038 0.23043 0.75326 DLNPGDP does not Granger Cause DLNI DLNI does not Granger Cause DLNPGDP 38 1.44872 0.43703 0.23681 0.51289 DLNFD3 does not Granger Cause DLNI DLNI does not Granger Cause DLNFD3 38 1.41139 0.35330 0.24282 0.55608 Obs F-Statistic Probability 38 0.84739 0.36359 DLNI does not Granger Cause DLNPGDP Model A – FD3 Null Hypothesis: Model A – FD4 Null Hypothesis: DLNINF does not Granger Cause DLNI Volume 1 Number 1 January 2011 107 �Okyay UÇAN & Özlem ÖZTÜRK DLNI does not Granger Cause DLNINF 4.02848 0.05251 DR does not Granger Cause DLNI DLNI does not Granger Cause DR 38 1.48965 0.10038 0.23043 0.75326 DLNPGDP does not Granger Cause DLNI 38 1.44872 0.23681 0.43703 0.51289 38 0.10555 0.59435 0.74720 0.44591 Obs F-Statistic Probability DLNINF does not Granger Cause DLNI DLNI does not Granger Cause DLNINF 38 0.84739 4.02848 0.36359 0.05251 DR does not Granger Cause DLNI DLNI does not Granger Cause DR 38 1.48965 0.10038 0.23043 0.75326 DLNPGDP does not Granger Cause DLNI DLNI does not Granger Cause DLNPGDP 38 1.44872 0.43703 0.23681 0.51289 DLNFDINDEX does not Granger Cause DLNI DLNI does not Granger Cause DLNFDINDEX 38 2.10028 0.51455 0.15617 0.47793 Obs F-Statistic Probability DLNFD1 does not Granger Cause DLNPI DLNPI does not Granger Cause DLNFD1 38 0.04863 0.52332 0.82674 0.47424 DLNINF does not Granger Cause DLNPI DLNPI does not Granger Cause DLNINF 38 4.31446 3.04715 0.04519 0.08965 DLNI does not Granger Cause DLNPGDP DLNFD4 does not Granger Cause DLNI DLNI does not Granger Cause DLNFD4 Model A – FDindex Null Hypothesis: MODEL B Granger Causality Results Model B– FD1 Null Hypothesis: 108 Journal of Economic and Social Studies �Financial Determinants of Investment for Turkey DLNPGDP does not Granger Cause DLNPI DLNPI does not Granger Cause DLNPGDP 38 0.64549 2.09254 0.42715 0.15692 DR does not Granger Cause DLNPI DLNPI does not Granger Cause DR 38 13.4414 0.63531 0.00081 0.43079 Obs F-Statistic Probability 38 4.31446 0.04519 3.04715 0.08965 Model B – FD2 Null Hypothesis: DLNINF does not Granger Cause DLNPI DLNPI does not Granger Cause DLNINF DLNPGDP does not Granger Cause DLNPI DLNPI does not Granger Cause DLNPGDP 38 0.64549 2.09254 0.42715 0.15692 DR does not Granger Cause DLNPI 38 13.4414 0.00081 0.63531 0.43079 38 0.99414 0.00598 0.32558 0.93880 Obs F-Statistic Probability DLNINF does not Granger Cause DLNPI DLNPI does not Granger Cause DLNINF 38 4.31446 3.04715 0.04519 0.08965 DLNPGDP does not Granger Cause DLNPI DLNPI does not Granger Cause DLNPGDP 38 0.64549 2.09254 0.42715 0.15692 DR does not Granger Cause DLNPI DLNPI does not Granger Cause DR 38 13.4414 0.63531 0.00081 0.43079 DLNFD3 does not Granger Cause DLNPI DLNPI does not Granger Cause DLNFD3 38 0.00060 1.32403 0.98052 0.25767 Obs F-Statistic Probability DLNPI does not Granger Cause DR DLNFD2 does not Granger Cause DLNPI DLNPI does not Granger Cause DLNFD2 Model B – FD3 Null Hypothesis: Model B– FD4 Null Hypothesis: Volume 1 Number 1 January 2011 109 �DLNINF does not Granger Cause DLNPI 38 DLNPI does not Granger Cause DLNINF 4.31446 0.04519 3.04715 0.08965 DLNPGDP does not Granger Cause DLNPI DLNPI does not Granger Cause DLNPGDP 38 0.64549 2.09254 0.42715 0.15692 DR does not Granger Cause DLNPI DLNPI does not Granger Cause DR 38 13.4414 0.63531 0.00081 0.43079 DLNFD4 does not Granger Cause DLNPI DLNPI does not Granger Cause DLNFD4 38 0.04707 0.03570 0.82950 0.85122 Obs F-Statistic Probability DLNINF does not Granger Cause DLNPI DLNPI does not Granger Cause DLNINF 38 4.31446 3.04715 0.04519 0.08965 DLNPGDP does not Granger Cause DLNPI DLNPI does not Granger Cause DLNPGDP 38 0.64549 2.09254 0.42715 0.15692 DR does not Granger Cause DLNPI DLNPI does not Granger Cause DR 38 13.4414 0.63531 0.00081 0.43079 DLNFDINDEX does not Granger Cause DLNPI DLNPI does not Granger Cause DLNFDINDEX 38 0.04343 0.88874 0.83612 0.35228 Model B– FDindex Null Hypothesis: � Dublin Core The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/. Extent The size or duration of the resource. 1060 Title A name given to the resource Financial Determinants of Investment for Turkey Author Author UÇAN, Okyay ÖZTÜRK, Özlem Abstract A summary of the resource. One of the fundamental aims of economic policies is to increase capital accumulation in terms of investment that is necessary to maintain a desirable and sustainable growth rate in the developing countries. The majority of empirical studies show that per capita GDP growth, foreign trade, capital flows, external debt, public sector borrowing requirements, inflation and interest rate are the main determinants of investment rate. Recently, there is an increasing emphasis on the role of the financial sector in this process, since a financial system, in essence, mobilizes saving to investment. In particular, it can be argued that a well-functioning and developed financial system may efficiently mobilize available resources for investment. Therefore, the aim of this study is to investigate whether financial development has contributed to an increase in investment in Turkey. To reach an empirical and firm conclusion, an investment function, including the traditional potential determinants along with financial development, is estimated by utilizing the developments in the time series econometrics in terms of unit root tests that allow structural breaks and co-integration for the period 1970-2009 in Turkey. Publisher An entity responsible for making the resource available International Burch University Date A point or period of time associated with an event in the lifecycle of the resource 2011-01 Keywords Keywords. Article PeerReviewed H Social Sciences (General),HB Economic Theory,HG Finance https://omeka.ibu.edu.ba/files/original/503075330bb975c14277699da52a7758.pdf 35a7a277d265e3a81cffbc101571959b PDF Text Text A comparison of ANFIS and ARIMA techniques in the forecasting of electric energy consumption of Tokat province in Turkey A comparison of ANFIS and ARIMA techniques in the forecasting of electric energy consumption of Tokat province in Turkey Rüstü YAYAR Faculty of Economics and Administrative Science Gaziosmanpasa University, Tokat, Turkey rustu.yayar@gop.edu.tr Mahmut HEKIM Tokat Vocational School Gaziosmanpasa University, Tokat, Turkey mahmut.hekim@gop.edu.tr Veysel YILMAZ Suşehri Timur Karabal Vocational School Cumhuriyet University, Sivas, Turkey veyselyilmaz@cumhuriyet.edu.tr Fehim BAKIRCI Faculty of Economics and Administrative Science Atatürk University, Erzurum, Turkey fehim.bakirci@atauni.edu.tr Abstr ct In this study, the electric energy demand of Tokat province was estimated by means of ANFIS and ARIMA techniques. Seven different forecasting experiments were implemented for the subscriber groups and the consumption of electric energy which is the dependent variable. The electric energy demand of the province for the first six months of the year 2011 was estimated by means of ANFIS and ARIMA techniques. The obtained results were compared and interpreted in order to illustrate the forecasting success of these techniques. We showed that the ANFIS is more appropriate than the ARIMA in point of the forecasting of electric consumption. Keywords: Electric energy consumption, Forecasting, ANFIS, ARIMA, and Tokat. Jel odes: C22, C45, Q47 Volume 1 Number 2 July 2011 87 �Rüstü YAYAR & Mahmut HEKIM & Veysel YILMAZ & Fehim BAKIRCI Introduction Nowadays, energy is an important source of life. By entering into the life of mankind in 1880s, the electric energy gradually became an indispensable part of modern life and industry (Şekerci Öztura, 2007). In order to generate the electric energy, which is a secondary energy source, the support of primary energy sources is needed. Electric energy is widely used in many fields, and a large part of energy resources for the benefit of the people are converted into electric energy (Demir, 1968). The electric energy demand has been continuously increasing in parallel with the growing population, urbanization, industrialization, technologic deployment, and enhancement of welfare. The spread of the usage fields of electric energy which is one of the most important parts of all types of economic activities increases the electric energy demand. Also, since the distribution network provided a great deal of the electric energy of even the smallest residential areas, the share of electric energy in the total energy consumption increased (Kılıç, 2006). The province of Tokat is geographically located between 39-51, 40-55 North latitudes and 35-27, 37-39 East longitudes, in the inner side of middle Black Sea part of Black Sea Region. Covering the 1.3% of mainland Turkey, elevation from the sea level of the province is 623m and surface area of it is 9.958km² (Governorship of Tokat, 2006). There are 12 districts of Tokat province including with the central district¹. Historical periods of the province consist of Hattie, Hittite, Phrygian, Med, Persian, Alexander the Great, Roman, Byzantine, Arabian, Danishmend, Anatolian Seljuk, Mongolian, Ilkhanid, Ottoman Governments and Emperors (Provincial Department of Environment And Foresty of The Governorship of Tokat, 2007:1). The province of Tokat became a province with the proclamation of the republic in 1923 (Turkish Statistical Institute, 2010:10). By 2010 the population of the province is 550.703 (Turkish Statistical Institute, 2010). The province has a wide potential of both agriculture and other sectors (such as tourism). In the economic structure of the industry, agriculture, livestock sector plays an important role. Particularly in the food industry, rock and land-based industries, forest products industry and in recent years, textile weaving and garment sector is the backbone of the economy in Tokat (Provincial Department of Environment And Foresty of The Governorship of Tokat, 2007:130). 88 Journal of Economic and Social Studies �A comparison of ANFIS and ARIMA techniques in the forecasting of electric energy consumption of Tokat province in Turkey Electric energy was firstly provided in the city center of Tokat in 1935 by the Hydroelectric power plant which was built on Aksu for the city lightening. This power plant consists of 2 tribunes with 175 horsepower (HP), and each tribune operates by 230 m³ water passing through the water channel. The power plant provided electric energy of 135 kWh for 1580 subscribers. However, the electricity production was insufficient for the city. Because of Almus Hydroelectric power plant built in 1966, Tokat had continuously the electric energy, and still it provides the electric energy of the city. Besides, transmission lines were renewed and electric energy of every part of the city was provided by using 18 transformers (Governorship of Tokat, 2006). The installed power plant in Tokat consists of Almus, Köklüce, Ataköy Hydroelectric power plants within Tokat. Almus Hydroelectric power plant became operational in 1966 and Number of Unit – Power is 3X9 MW and its installed power is 27 MWAnnual production of the plant is 100 GWh. Köklüce entered service in 1998 and Number of Unit – Power is 2X46 MW and its installed Power is 90 MW. Annual production of the power plant is 588 GWh (Electricity Generation Corporation, 2011). The construction of Ataköy Hydroelectric power plant dam was completed in 1977. Having 5.5 MW Power, its annual production is 8 GWh (VIII. Regional Directorate of State Hydraulic Works, 2011). In this study, electric energy demand are estimated by Adaptive Neuro-Fuzzy Inference System (ANFIS) and Autoregressive Integrated Moving Average (ARIMA) techniques by using the electric energy consumption data of Tokat province in the time of period between January 2002 and December 2010. Matlab 7.04 package program is used for the ANFIS model and Minitab 14.0 package program is used for the ARIMA model. Aim This study aims at contributing to the planning of supply by estimating electric demand in the future. The demand for electric energy was forecasted in Tokat province by means of the ANFIS and the ARIMA models. The generation planning of electric energy is very important because it cannot be stored, and therefore must be consumed shortly after its generation. If these kind local studies were generalized into all country, it helps the supply planning and provides a more effective usage of resources. Therefore we estimated the electric energy demand by ANFIS and ARIMA models for Tokat province. Volume 1 Number 2 July 2011 89 �Rüstü YAYAR & Mahmut HEKIM & Veysel YILMAZ & Fehim BAKIRCI Literature Energy consumption and demand are among the most debated topics, and many studies have been made available in the literature. It is seen that the studies especially focused on the causality between the electricity consumption and economic growth and controversial results has been achieved. There are scarcely any studies on electric energy consumption and demand through ANFIS and ARIMA models. These models are often used in engineering studies. There are many studies in various fields by using the fuzzy logic method. In one of them, Tufan and Hamarat (2003) analyzed the “Aggregation of The Financial Ratios of publicly-traded companies Through Fuzzy Logic Method”. The first detailed study examining the demand for electricity with the help of econometric models is the work of Houthakker (1951). The study includes the econometric analysis of the household electricity demand through cross-section data from the period of 1937-1938 for 42 residential center in England. Another study was conducted by Fisher and Kaysen. Fisher and Kaysen (1962) using time-series and cross sectional data, examined the demand for electricity with the help of multiple regression and analysis of covariance. Electric demand was taken up in four components, including electricity demand, household electricity demand, industrial electricity demand and the short and long period determinant of them. Accordingly, short run is in question if the stock of electric appliances is fixed and long run is in question if the stock is variable. In the case of industry, short run is in question if there is an assumption that technology is invariable and long run is in question if there is an assumption that technology is variable (Tak, 2002). Al-Garni and Javeed Nizami (1995) developed a model of artificial neural networks for electric energy consumption with the data from seven years such as temperature, moisture, solar radiation and population. After the comparison of the model of artificial neural networks with the regression model, it was revealed the model of artificial neural networks was a better forecasting model. Al-Garni and Abdel-Aal (1997) estimated the Electric Energy consumption for five years on the east of Saudi Arabia for the consumption on the sixth year, developing monthly ARIMA models by using the univariate and Box-Jenkins time-series analysis. When ARIMA models are compared to the abductive network machinelearning models, it is seen that ARIMA models needs less data and coefficient and give better results as well. 90 Journal of Economic and Social Studies �A comparison of ANFIS and ARIMA techniques in the forecasting of electric energy consumption of Tokat province in Turkey Brown and Koomey (2002) examined the increase in electric demand in their work named “Electric Use in California: Past Trends and Present Usage Patterns. They took sector (settlement, commercial, industrial, agricultural and other) of electric consumption from the previous period as data. They brought forward that there had been a great increase in the electric demand in 1990, compared to 1980 and that it stems from the increase in buildings and the tendencies in the building sector (Enduse Forecasting and Market Assessment, 2011). It can be seen that the studies on the modeling of the energy consumption and demand, which is also crucial for the economy of Turkey, accelerated in 2000s. There are not many studies, done widely on the provincial (regional) electricity consumption and demand. State institutions and organizations of which subject of activity is electricity became obliged to state the electricity consumption of the provinces on their annual report on the basis of subscriber groups after “Legislation On The Annual Reports Drawn Up By The Public Administrations” were published by the Ministry of Finance on the Official Gazette dated 17.03.2006 and became effective on 01.01.2006 (Official Gazette, 2006). These institutions conduct studies also on the monthly and yearly electric data which must be sent to Turkish Statistical Institute (TÜİK) and Ministry of Energy and Natural Resources (ETKB) by these institutions even though these data changes in the ensuing years. In the work of Terzi (1998) which analyzes the relationship between the electric consumption and economical growth for the period of 1950-1991, the relationship between the electricity consumption and of the commerce house, industry and household and economical growth; the long period relationship between the variables were determined through the Engle-Granger cointegration method and the short run dynamics were analyzed through the debugging tool. It was determined through this econometric method that the income and price elasticity were in fact inelastic. A meaningful and two-way relationship between electricity consumption and economical growth came along in the business and industry sector. Sarı and Soytaş (2004) employed the technique of generalized forecast error variance decomposition and came to the conclusion that the electricity demand and variance in national income growth are as important as employment (Lise & Van Monfort, 2005). Çebi and Kutay (2004) used artificial neural networks while estimating the long run electric energy consumption and compared the results to the Box-Jenkins models and regression technique. The results revealed that using artificial neural networks was a good forecasting method for electric energy consumption. Volume 1 Number 2 July 2011 91 �Rüstü YAYAR & Mahmut HEKIM & Veysel YILMAZ & Fehim BAKIRCI In addition, we must specify the MAED (Model for Analysis of Energy Demand) model study of Ministry of Energy and Natural Resources, the most important study that was conducted in our country, which reveals the medium and long run general energy demand and electric energy demand in this demand. Research Method In this study, monthly electricity consumption data of Tokat province was used. This data covers the period between January 2002 and December 2010. The subscriber groups consist of private houses, industry, business firms, government agencies and other subscribers. The total electricity used in agricultural irrigation, fresh water, work-sites, temporary activities, state-owned enterprises, municipalities, internal activities, prefectures, sanctuaries, and local government lightening systems was also included in these groups. Seven different experiments were implemented for investigating the success of the ANFIS and the ARIMA models in the forecasting of the total electric energy consumption of all subscriber groups. Time Series and Box-Jenkins Forecasting Model Time Series A time series is simply a sequence of numbers collected in regular intervals (as day, month and year) over a period of time (Dikmen, 2009: 227). Time series monitors the motion of a variable in a time sequence. For example, monthly unemployment ratio, monthly increase ratio of money supply, annual inflation ratio, monthly electric use, etc. Time series can be also used as a resource of knowledge acquisition and a method for forecasting the future. While in the evaluation process, analysis is important in degrading the trend, growth trend, seasonality, cyclical, and irregular fluctuations (Bozkurt, 2007). The selection of method used for forecasting the future values depends on the estimation of purpose, type and elements of time series, amount of data, and the length of the estimation period (Asilkan & Irmak, 2009). 92 Journal of Economic and Social Studies �A comparison of ANFIS and ARIMA techniques in the forecasting of electric energy consumption of Tokat province in Turkey Box-Jenkins Forecasting Model Box-Jenkins method is the most widely used model for stationary time series modeling. For the implementation of Box- Jenkins method, the time series must be stationary (with constant mean, variance and autocorrelation). If the series is not stationary, it should be made stationary by taking the difference of a few times (Gujarati, 2009). Box-Jenkins method is based on the principle that each time series is a function of past values and may only be explained by means of them. Some assumptions cannot be applied based on the econometric models, but there is not any restrictive assumption for Box-Jenkins method (Bircan & Karagöz, 2003). In this method; • In contrast to the regression models that explain yt with a k number of explanatory variables of x1, x2, x3,…, xk, • The dependent variable Yt can be explained by its own past or lagged values ​​and stochastic error terms. The most important stage of the Box-Jenkins method is the selection of the appropriate ARMA (p, q) model by examining the autocorrelation and partial autocorrelation coefficients. Experience of the researcher is very important because of this phase is not able to determine mechanically. If the time series is not stationary, artificial autocorrelations will prevent the model to determine. Non-stationary time series is transformed stationary time series by logarithmic transform or taking differences. Autocorrelation and partial autocorrelation coefficients of the distribution can be examined with the help of graphs (autocorrelation function (ACF) and partial autocorrelation function (PACF). When the autocorrelation coefficients are seen to be approaching zero exponentially, AR model must be applied; while the partial autocorrelation coefficients are realized to be approaching the same level mentioned above, then MA model must be used; if both of these approach zero exponentially; ARMA model must be applied in this situation. In ARMA model, the degree of AR is determined by the number of partial autocorrelation coefficients (p), while the degree of MA is determined by the number of autocorrelation coefficients (q) (Önder & Hasgül, 2009: 65–66). Volume 1 Number 2 July 2011 93 �(PACF). When the autocorrelation coefficients are seen to betoofapproaching zerozero exponentially, Autocorrelation andand partial autocorrelation coefficients the distribution cancan be be examined (PACF). When the autocorrelation coefficients are seen be approaching exponentially, Autocorrelation partial autocorrelation coefficients of the distribution examined l autocorrelation function Autocorrelation and partial autocorrelation coefficients of the distribution can be examine AR model must be applied; while the partial autocorrelation coefficients are realized to be with the help of graphs (autocorrelation function (ACF) and partial autocorrelation function AR model must be applied; while the partial autocorrelation coefficients are realized to be with the help of graphs (autocorrelation function (ACF) and partial autocorrelation function ching zero exponentially, with the help of partial graphsautocorrelation (autocorrelationcoefficients function (ACF) and partial autocorrelation functio Autocorrelation and ofbemust the distribution can be approaching theto same level mentioned above, thenthen MA model be used; if both of examined these (PACF). When the autocorrelation coefficients are seen to approaching zero approaching the same level mentioned above, MA model must be used; ifexponentially, both of these (PACF). When the autocorrelation coefficients are seen to be approaching zero exponentially, icients are realized be (PACF). When the autocorrelation coefficients are seen to be approaching zero exponentiall withmodel the help of graphs (autocorrelation function (ACF) autocorrelation function approach zero exponentially; ARMA model must be applied inand this situation. AR must be applied; while the partial autocorrelation coefficients areare realized to to be be approach zero exponentially; ARMA model must be applied inpartial this situation. model must be applied; while the partial autocorrelation coefficients realized be used; ifAR both ofmodel these AR must be applied; while the partial autocorrelation coefficients are realized to b (PACF). When the autocorrelation coefficients are seen to be approaching zero exponentially, Rüstü YAYAR & Mahmut HEKIM & Veysel YILMAZ & Fehim BAKIRCI approaching the same level mentioned above, then MA model must be used; if both of these approaching thethe same level mentioned above, then MA model must be be used; if both of of these situation. AR approaching same level mentioned above, then MA model must used; if both model must bethe applied; while partial autocorrelation coefficients areautocorrelation realized to bethe In ARMA model, the degree of ARMA AR is the determined by number ofthis partial autocorrelation approach zero exponentially; model must bethe applied in in this situation. In approach ARMA model, degree ofARMA AR is model determined by the number ofsituation. partial zero exponentially; must be applied approach zero exponentially; ARMA model must be applied in this situation. approaching the same mentioned above, then MAbymodel must beofused; both of these coefficients (p), (p), while thelevel degree of MA is determined the number autocorrelation coefficients while the degree of MA is determined by the number of if autocorrelation of partial autocorrelation approach zero exponentially; ARMA model must be applied in this situation. (q)model, (Önder &the Hasgül, 2009: 65–66). ARMA the degree of AR is is determined byby thethe number of of partial autocorrelation coefficients (q) (Önder &degree Hasgül, 2009: 65–66). In ARMA model, of AR determined number partial autocorrelation umbercoefficients ofIn autocorrelation In ARMA model, the degree of ARthese is determined by the number of partial autocorrelatio ARMA models consists of four models, are AR, MA, ARMA and ARIMA. These coefficients (p), while the degree of MA is determined by the number of autocorrelation coefficients (p), while the degree of MA is determined by the number of autocorrelation coefficients (p), while the of (Demirel MA is determined by the numberautocorrelation of autocorrelatio In ARMA model, of AR isthese determined by the number of partial models will(Önder bethe explained in thedegree following: & et al.,MA, 2010). ARMA models consists ofdegree models, are are AR, MA, ARMA and ARIMA. These coefficients (q) &four Hasgül, 2009: 65–66). ARMA models of four models, these AR, ARMA and ARIMA. These coefficients (q)consists (Önder & Hasgül, 2009: 65–66). coefficients (q) (Önder & Hasgül, 2009: 65–66). coefficients (p), while the degree of MA is determined by the number of autocorrelation models will be explained in the following: (Demirel & et al., 2010). models will be explained in the following: (Demirel & et al., 2010). MA and coefficients ARIMA. These (q) (Önder &ofHasgül, 2009: 65–66). ARMA models consists four models, these areare AR, MA, ARMA andand ARIMA. These ARMA models consists of four models, these AR, MA, ARMA ARIMA. These ARMA models consists of four models, these are AR, MA, ARMA and ARIMA. The AR (p) Model AR models (p)models Model bebe explained in in thethe following: (Demirel && et et al.,al., 2010). AR (p) will Model will explained following: (Demirel 2010). models will consists be explained in the following: et al.,ARMA 2010). and ARIMA. These ARMA models of four models, these(Demirel are AR,&MA, In AR (p) model, Y value is the linear function of stochastic error term and weighted models will be explained (Demirel al., 2010). t in In AR (p)(p) model, Yt value is the the linear function of & stochastic errorerror term and and weighted AR Model In AR (p) model, Yt value is following: the linear function ofetstochastic term weighted AR Model ARaggragates (p) Model of past values in p period of the series. AR (p) model is shown as follows: aggragates of past values in p period of the series. AR (p) model is shown as follows: aggragates of past values in p period of the series. AR (p) model is shown as follows: error termAR and (p)weighted Model In AR of ofstochastic error term andand weighted Y Φt 1 Ymodel,  Φ1 2YYΦ tt ... Φ linear δplinear  at (1) Y(p) Yvalue ...ispYisthe Φ δlinear function atfunction (1) In (p) stochastic error term weighted t AR tΦ 11model, t t22Y t  ppthe hown as follows: 2 value In AR (p)Yt model, Yvalue isYt the function of stochastic error term and weighte t aggragates of past values in p period of the series. AR (p) model is shown as follows: aggragates of past values in p period of the series. AR (p) model is shown as follows: (1) (1) aggragates of past values is in the p period of the series.ofAR (p) modelerror is shown follows: In AR model, Yt Yvalue stochastic term asand weighted Y(p) t Φ Yt1Y1 t Φ  ... ...ΦYpΦ Ytplinear δfunction at (1) 1Φ Φ  Y δ  at t Y 2Y t2Y 2Y YY ptt− Y Y Φ Y Y Y Y 1t  t  2 p Φ Y Φ Φ p t − p t  p p Y  Φ  Φ Y  ...  Φ Y  δ  at (1)the Φ21follows: Φ2 ,...,is(1) tt−−22 ,..., 2 2,..., pthe aggragates ofabove values series. model is shown1 , as 1in tpast 1model, t11, t t  2,..., t  pthe is values of(p) past observation, In the In the above model, istpof the values ofAR past observation, ,..., In the above model, , tt−−tp11t2,period ispthe the values past observation, ,,..., is Φ Ytp isthe Φ Φ2of Yt 2for  ...  values Φobservation, δδis atδthe (1) the coefficients the observation, is value the constant and Φ1 coefficients 1Yvalues t 1  p Yt  of p past observation, and and atvalue isatthe error term. is constant the constant value is the error the values of ation, coefficients , Φ2 ,..., for isfor the YtpYtp Φ pΦterm. Yt Y1past Y1t past Φ Φ Y 2 Φ Φ 1 2 t  t  2 at is the error term. In In thethe above model, , Y,t 1 ,..., is the values of past observation, , ,..., ispΦthethe 1Φ 2Φ Y Y above model, ,..., is the values of past observation, , ,..., t  p theterm. above model, , t 2 ,..., is the values of past observation, 1 , 2 ,..., isp is th e and at is the In error δ is δ Y Φ MA(q) Model coefficients forfor thethe values of past observation, the constant value and at is the error term. Y Y MA(q) Model Φ Φ coefficients values of past observation, is the constant value and at is the error term. t  p p 1 , t  2 ,..., 1 , at2 ,..., In the above model, is the valuesδofis past observation, is theterm coefficients for the tvalues of past observation, the constant value and is the error δ is the constant value and at is the error term. Model coefficients for the values past observation, In MA (q)MA(q) model, Yt value isofthe linear function of average pastpast errorerror terms in qinperiod MA(q) Model In MA (q) model, Y value is the linear function of average terms q period MA(q) Model MA(q) Model t backwards. MA (q ) model is shown as follows: backwards. MA (q ) model is shown as follows: error terms in In qModel period MA (q) model, Yt value is the linear function of average past error terms in q MA(q) In MA (q) linear of average past error terms in in q (2) period YtMA period t (q) model, at  1aYtt- 1Y1value  a(q) t -... is... linear at-function Y at atMA - 22model athe  In model, value is average past error terms q period tis q qq afunction tY - 1 2 2the tas - qfunction backwards. shown follows:of of In MA (q) model, is the linear average past error terms in q(2) perio t value backwards. MA (q ) model is shown as follows: backwards. MA (q ) model is shown as follows: (2) backwards. MAYt(qvalue ) model is asfunction follows:of average past error terms in q period In MA model, theshown linear Y(q) t  a1 at1-aa1 t- 1 is ... (2)(2)   at a at 2 at2-a2ta a ......q atq-aqt - aq t Y  q (2) 2 a a a 1 ,21,,… t q Y  at   a a 2 , ,… t -1 2 ,……, t - qthe error backwards. MA (qt ,) model follows: t , , t -1t -1,is t tshown -t1-2 ,……, 2 as t - 2is t terms, -q In the above model, is theq error terms, ,is qthe is coefficients the coefficients ( In the above model,  q error Yterms   and  at  1average at - 1 average   2of at - 2the  ...   q at - q (2) isthe series. terms t and error … , of isofthe coefficients at -qaseries.  q q aist athe at t -1at -1at -2at -2 of the     1 2 t q , ,at, a,t -1 ,……, isa the error terms, ,1 ,… , , isthe coefficients In In thethe above model, 2 at -2 error terms, coefficients above model, - q the q the 2, … is the error terms, the above , , ,……, ,……, tis is the error terms, ,1 ,,… , isis is the coefficien In In the above model, model  ARMA (p,q) Model is the average of the series. of error terms and ARMA (p,q) Model a  is the average of the series. of of error terms and a a a    t q q the coefficients of error terms and m is the average of the series. series. error terms ,……, of the is the error terms, 1 , 2 , … , is the coefficients In the above model,andt , t -1is, thet -2average  ismost the average of thepreocess series. of error terms andthe ARMA model, the most stochastic preocess models, is the linear function of past ARMA (p,q) Model model, stochastic models, is the linear function of past ARMA (p,q) Model ARMA (p,q) Model observations and error terms. ARMA (p,q) model is generally shown as follows: observations and ARMA (p,q)error Modelterms. ARMA (p,q) model is generally shown as follows: linear function of past ARMA (p,q) Model ARMA model, the most preocess thethe function of(3) past Y  Φ Y  Φ YΦ ... Φstochastic - atδ -at θ 1 atθ-models, θ atθ- 22is θ qlinear atθ-linear Y  Φ Y stochastic ...  amodels, at -... ... the ARMA model, is t t  22 Y p YΦ t  p Yt δ 11  qq a t - q n as follows: ARMA t 1 t 1model, 1 t 1 2the 2 most p  preocess t -12models, 2 is thetmost stochastic preocess linearfunction functionof (3) ofpast pa observations and error terms. ARMA (p,q) model is generally shown as follows: ARMA(3) model, the most stochastic preocess models,is isgenerally the linearshown function past observations andand error terms. ARMA (p,q) model asofas follows:  θ q at -q ARMA observations error terms. ARMA (p,q) model is generally shown follows: the most the as linear of past (3)(3) Yt model, Yt1Y  ΦY2error Yt2Y2 tterms. 2Ystochastic ... ... Φ YtpYpt preocess (p,q) δ -δat  θmodels, aθt1generally θ aist2-shown θfollows: aθtq-qafunction observations ARMA model Yt Φ 1Φ Y Φ  ...pΦ -δat a2θt- 2a...  ... 1 tand 1is -1a t qΦ 1Y 1Y Y p  -1a2 θ tΦ - qa Φ p Φ p Y Y Φ Y  Φ  Φ   Φ Y  at  θ  ...  θ t  p Φ t  t  2 t  p observations and error terms. ARMA (p,q) model is generally shown as follows: 1 2 t  1 t  2 t 1 ,t 1 , ,..., 2 ,..., t  2 is the ppast t  p past 1 t -1values, 2 t -2 q 2t ,..., -q observation , 1 ,,..., is the In third equation, is the observation values, is the ( In third equation, Φ pYt  Φ1Yt 1  Φ2Yt  2  ...  Φ p Yt  p  δ - at  θ1 at -1 a θa2aat - 2a  (3) a ... at -θ2 q at -aq t -q(3) at -the 1 , Φ2 ,..., t , t -1 δ is δthe q Φ error es, Φcoefficient thepast t ,, t -1t -,2 ,……, Y constant, is forisfor past observation values, Yt Y1 ,t Y1t Y2 t,..., is the constant, ,……, is the error coefficient observation values, Φ Φ Y Φ t  p p Φ Φ 1 2 t  p p is observation values, , 1 the is InaInthird equation, 2 ,..., Y is Φ pisthethe Y Y2 t,..., thepast past , ,..., third equation, pthe 2 Y 1 ,Φ 2 ,..., at -2 istthe past observation values,F1,values, F2values, ,...,Fp is Φ third equation, ,..., is the pastobservation observation is th third equation,  q  q Yt-1,,t Y1t-2,, ..., t - q InIn  t-p   ,……, is the error 1 2 1 2 , , … , is the coefficients of error terms. terms and at -qat -Φ , past , for … past ,observation isYthe coefficients of error terms.a , aat ,aaat t……,a terms and Yvalues, Yt 1observation -1a,t -1at -2a,……, δ Φ Φ t values, p p qathe is the ercoefficient is the constant, t  2 t 2 δ 1 2 is the constant, is error coefficient for a a a , observation ,..., values, isvalues, the is past observation values, ,..., is the Incoefficient third equation, thethe constant, , t , ,t -1t-q, ,……, error forfor past observation t t-1, t-2 t - q the t -,2 ,……, δ is constant, isis the err coefficient past ror  terms and q1, qq2, … ,qq is the coefficients of error terms. a a  a a   1 2 q t q t t -1 t 2 δ ARIMA (p,d,q) Model , 1 ,,past … coefficients of error terms. terms andand 2 , observation ARIMA (p,d,q) Model , is, the isq the coefficients error terms. terms isof the constant, , ,……, is the error coefficient for values, 1 ,,… 2, … is the coefficients of error terms. , terms and 1 , 2 , … ,  q is the coefficients of error terms. terms and To ARIMA make a non-stationary timetime series stationary, one one or two times the the difference-making (p,d,q) Model To make a (p,d,q) non-stationary series stationary, or two times difference-making ARIMA Model ARIMA (p,d,q) Model process is carried out and the result are shown with d. The model that is applied to the process is carried out and the result are shown with d. The model that is applied to series the series es the difference-making ARIMA (p,d,q) Model To make a non-stationary time series stationary, one or two times the difference-making 94make Journal of or Economic and Socialthe Studies Toto make a non-stationary time series stationary, oneone twotwo times difference-making hat is applied the series To a non-stationary time series stationary, or times the difference-makin process is carried outout andand thethe result areare shown with d. d. The model that is applied to to thethe series process is carried result shown with The model that is applied series process is carried out and theseries resultstationary, are shown one with or d. two The model that difference-making is applied to the seri To make a non-stationary time times the process is carried out and the result are shown with d. The model that is applied to the series �A comparison of ANFIS and ARIMA techniques in the forecasting of electric energy consumption of Tokat province in Turkey ARIMA (p,d,q) Model To make a non-stationary time series stationary, one or two times the differencemaking process is carried out and the result are shown with d. The model that is applied to the series stationary by differencing is called as non-stationary linear stochastic model or integrated model shortly (Bircan & Karagöz, 2003). Figure 1. Box-Jenkins Procedure Box-Jenkins process operates as follows: (Dobre & Alexandru, 2008: 157). Plot Series Is it Stationary? Static? Identify Possible Model Difference “Integrate” Series Seriler Diagnostic OK? Make Forecast Adaptive Neuro-Fuzzy Inference System (ANFIS) Fuzzy Inference System (FIS) consists of three conceptual components: fuzzy rule base, data base and inference. In this system, the fuzzy rules and membership functions of input and output variables are determined by the user. The most important step is to set the membership degrees of input and output variables. FIS techniques aim at providing of significant inferences by using the linguistic rules (Ross, 2004). Fuzzy systems do not have the skill to learn things, so they heavily depend on expert opinion. Adaptive Neuro-Fuzzy Inference System (ANFIS), a hybrid model, was first developed by Jang in 1993 in order to overcome this problem. This system has combined the learning skill by artificial neural networks with inference skill of expert opinion based FIS models (Jang, 1993). It adjusts the membership functions of input and output variables and generates the rules related to input and output, automatically. ANFIS can produce all the rules by using the dataset and enables the researchers to interpret these rules. Therefore, it is the widely used model in the studies of classification and estimation. Volume 1 Number 2 July 2011 95 �Adaptive Neuro-Fuzzy Inference System (ANFIS), a hybrid model, was first developed by Jang in 1993 in order to overcome this problem. This system has combined the learning skill by artificial neural networks with inference skill of expert opinion based FIS models (Jang, 1993). It adjusts the membership functions of input and output variables and generates the rules related to input and output, automatically. can produce all the rules by using the Rüstü YAYAR & Mahmut HEKIM & Veysel YILMAZ & ANFIS Fehim BAKIRCI dataset and enables the researchers to interpret these rules. Therefore, it is the widely used model in the studies of classification and estimation. In an ANFIS modelmodel consisting of two output,the thesetset rules In an ANFIS consisting of twoinputs inputsand and one one output, of of rules is as is as follows (Jang, 1993): follows (Jang, 1993): Rule 1 : If x is A1 and y is B1 , then f1  p1 x  q1 y  r1 Rule 2 : If x is A2 and y is B 2 , then f 2  p 2 x  q 2 y  r2 where xxand the the inputs, Ai andABand are the fuzzythe sets, fi are sets, the outputs the within the Bi are fuzzy fi are within the outputs where andy are y are inputs, i i , q and r are the design fuzzy region specified by the fuzzy rule, and parameters p fuzzy region specified by the fuzzy rule, and parameters pi,i qii and rii are the design parameters thatduring are determined duringprocess. the training Thearchitecture ANFIS architecture that areparameters determined the training Theprocess. ANFIS to implement these to implement these two rules is shown in Figure 2, in which a circle indicates a whereas fixed two rules is shown in Figure 2, in which a circle indicates a fixed node, a square Figure 2. 2. Structure Structure of of an an ANFIS. ANFIS. Figure Figure 2. Structure of an ANFIS. node, whereas a square indicates an adaptive node (Jang, 1993). indicates an adaptive node (Jang, 1993). Figure 2. Structure of an ANFIS. In the first layer, each node produces membership grades to which they belong to each In the first layer, each node produces membership grades to which they belong to each of the In layer, each produces membership grades totooutputs which Inthe thefirst first layer, eachnode nodesets produces membership grades whichthey they belong toeach eachof ofthe the of the appropriate fuzzy using membership functions. The of thisbelong layer areto appropriate fuzzy sets using membership functions. The outputs of this layer are the fuzzy appropriate fuzzy sets using membership functions. The outputs of this layer are the fuzzy appropriate fuzzy sets using membership Thebyoutputs of this layer are the fuzzy the fuzzy membership grade of the inputs,functions. which are given membership grade of the inputs, which are given by membership grade of the inputs, which are given by membership grade of the inputs, which are given by 1 O (4) (4) OOii1i1   A AAii i(((xxx))) (4) (4) 1 11  B ( ) O y (5) (5) OOii i BBii i 222((yy)) (5) (5) (small, large, etc.) areare the the linguistic Where, and the y arecrisp the crisp inputs ith node, Ai andBBi (small, xx and yx are inputs to iito th node, A large, etc.) Where, crisp toto th AAii iand and large, etc.) Where, xand andyyare arethe the crispinputs inputs ithnode, node,membership andBBii i(small, (small, large, etc.)are are, the thelinguistic linguistic Where,linguistic and mB labels characterized by appropriate functions mA i i  A and  B , respectively. labels characterized by appropriate membership functions i and i , ,respectively.  A  B labels by appropriate membership functions  A and  B respectively. labelscharacterized characterized by appropriate membership functions ii ii respectively. In the the second layer,layer, every node incomingsignals signals sends the product product out. In the second every nodemultiplies multiplies the incoming and and sendssends the prodIn In the second second layer, layer, every every node node multiplies multiplies the the incoming incoming signals signals and and sends the the product out. out. Each node output represents the firing strength of a rule. uct out. Each node output represents the firing strength of a rule. Each node output represents the firing strength of a rule. Each node output represents the firing strength of a rule. 2 O (6) OOii2i2  w wwii i   AAAii (((xxx)))  BBBii (((yyy))) (6) (6) (6) i i In the third layer, the main objective is to calculate the ratio of each iith rule’s firing strength In Inthe thethird thirdlayer, layer,the themain mainobjective objectiveisistotocalculate calculatethe theratio ratioof ofeach each th ithrule’s rule’sfiring firingstrength strength wi is is taken taken as as the the normalized normalized firing firing to the the sum sum of of all all rules’ rules’ firing firing strength. strength. Consequently, Consequently, w to wi is taken as the normalized firing to the 96 sum of all rules’ firing strength. Consequently, Journal of iEconomic and Social Studies strength strength strength w Oi333  (7)  w  wwii i O (7) Oi i wwii i  w1  w (7) ww11ww222 �A comparison of ANFIS and ARIMA techniques in the forecasting of electric energy consumption of Tokat province in Turkey In the third layer, the main objective is to calculate the ratio of each ith rule’s firing strength to the sum of all rules’ firing strength. Consequently, wi is taken as the normalized firing strength Oi3 = wi = wi w1 + w2 (7) In the fourth layer, the nodes are adaptive nodes. The output of each node in this layer is simply the product of the normalized firing strength and a first order polynomial (for a first order Sugeno model). Thus, the outputs of this layer are given by Oi4 = wi f i = wi ( pi x + qi y + ri ) (8) where wi is the ith node’s output from the previous layer. Parameters pi, qi and ri are the coefficients of this linear combination and are also the parameter set in the consequent part of the Sugeno fuzzy model. In the fifth layer, there is only one single fixed node. This single node computes the overall output by summing all the incoming signals as follows f = ∑w i i fi ∑w (p x + q = ∑w i i i i i y + ri ) (9) i Accordingly, the defuzzification process transforms each rule’s fuzzy results into a crisp output in this layer. In this study, the ANFIS was trained by hybrid learning algorithm which is highly efficient in training the ANFIS. This learning algorithm adjusts all parameters {ai, bi, ci} and {pi, qi, ri} to construct the ANFIS output match the training data. When the premise parameters ai, bi and ci of the membership functions are fixed, the output of the ANFIS becomes as follows: f = w1 f1 + w2 f 2 = w1 ( p1 x + q1 y + r1 ) + w2 ( p 2 x + q 2 y + r2 ) (10) Where, p1 , q1 , r1 , p 2 , q 2 and r2 are the adjustable resulting parameters? The least squares method is widely used to easily identify the optimal values of these parameters (Jang, 1993). Volume 1 Number 2 July 2011 97 �Rüstü YAYAR & Mahmut HEKIM & Veysel YILMAZ & Fehim BAKIRCI ARIMA and ANFIS Applications We benefited from autocorrelation and partial autocorrelation functions of the related series in order to obtain ARIMA models. The appropriate models for applications were investigated based on the monthly electric data between years 2002–2010. After choosing the models, the monthly electric values pertaining to the period between the first half of 2010 and second half of 2011 were estimated by the models. Then, the forecasting values relating to the second half of 2010 were compared with the real values of the same period. Seven different models based on the subscriber groups were tested for the investigations. They are total electric use (Model 1), electric use in private houses (Model 2), electric use in industrial organizations (Model 3), electric use in business firms (Model 4), electric use in government agencies (Model 5), electric use in other subscriptions (Model 6), and electric use in business firms-government agencies-other subscriptions (Model 7). In the implemented experiments, we observed that there is no ARIMA model appropriate for Model 1 and 7. Experiment 1. Analysis of total energy use (Model 1) In the construction stage of an appropriate ARIMA model related to the consumption of total energy use, we determined that the change of energy consumption versus months is non-stationary. The difference of the monthly energy series was taken once in accordance with the autocorrelation and partial autocorrelation function graphs. When the difference was once taken, the series had got stationray character in model validation stage. However, the ARIMA techniques generated through tests are to pass the suitability test in order to be used for future forecastings. Thus, it is indicated that the autocorrelation coefficients of the forecastings generated through the ARIMA techniques show a trend of systematicity in the suitability test. Despite all those efforts put forward, no model was detected as appropriate for this purpose. When the ANFIS model was trained by the training dataset for 1000 iterations, it found three rules for this experiment. The results obtained by ANFIS model are shown in Figure 3, where the test result of the model is red line and the forecasting of the next six month period is green line as follows: 98 Journal of Economic and Social Studies �A comparison of ANFIS and ARIMA techniques in the forecasting of electric energy consumption of Tokat province in Turkey Figure 3. The ANFIS output for Model 1. As seen in Figure 3, there is a small decrease possibility, but the stationary case will be lasting in the next six months. The comparative results of ARIMA and ANFIS techniques are given in Table 1: Table 1. Forecasting of total electric use (Model 1) Month Jul.10 Real ARIMA Forecasting ANFIS MSE Forecasting 42,0229 50,9282 ug.10 51,4488 50,9275 ep.10 59,9116 50,9268 ct.10 43,4847 50,9261 ov.10 46,3052 50,9254 Dec.10 48,5912 Jan.11 o appropriate model has been found 50,9240 50,9233 Mar.11 50,9226 pr.11 50,9219 May.11 50,9212 Jun.11 50,9205 Number 2 July 2011 40,4117 50,9247 eb.11 Volume 1 MSE 99 �Rüstü YAYAR & Mahmut HEKIM & Veysel YILMAZ & Fehim BAKIRCI As seen in Table 1, the ARIMA could not achieve any forecasting result. However, when the ANFIS was tested by six-month testing dataset, its mean square error (MSE) ratio was 40.4117. The forecasting obtained by the ANFIS for the next six months was close to the obtained test results of the ANFIS, so it can be observed that there was approximately a stationary case. Experiment 2. Analysis of the electric use in private houses (Model 2) In the construction stage of an appropriate ARIMA model related to the consumption of electric energy use in private houses, we determined that the change of energy consumption versus months is non-stationary. The difference of the energy series is once taken in accordance with the autocorrelation and partial autocorrelation function graphs. When the difference taken, the series showed stationary haracter in the model validation stage. As a result of the tests, ARIMA(1,1,2) was chosen as the model. When the ANFIS model was trained by the training dataset for 4000 iterations, it found eight rules for this experiment. The results obtained by ANFIS model are shown in Figure 4, where the test result of the model is red line and the forecasting of the next six month period is green line as follows: Figure 4. The ANFIS output for Model 2. 100 Journal of Economic and Social Studies �A comparison of ANFIS and ARIMA techniques in the forecasting of electric energy consumption of Tokat province in Turkey As seen in Figure 4, there is a constant increase in the next six months in Model 2. The comparative results of ARIMA (1,1,2) and ANFIS techniques and the real values for the period between 2002-2010 are given in Table 2: Table 2. Forecasting of the electric use in private houses (Model 2) ARIMA(1,1,2) ANFIS Month Real Jul.10 18,9146 21,1999 22,0048 ug.10 22,9082 21,3377 22,4716 ep.10 32,8206 21,4178 ct.10 19,9593 21,5264 ov.10 20,9913 21,6209 23,8837 Dec.10 23,7073 21,7224 24,3547 Jan.11 21,8205 24,8256 eb.11 21,9202 25,2962 Mar.11 22,0191 25,7668 pr.11 22,1184 26,2372 May.11 22,2175 26,7077 Jun.11 22,3167 27,1781 Forecasting MSE 24,0842 Forecasting 22,9416 23,4125 MSE 21,3410 As seen in Table 2, the MSE ratio was 24.0842 for ARIMA (1,1,2) and 21.3410 for ANFIS in the result of forecasting implemented by the six-month testing data for Model 2. Thus, it can be said that the ANFIS provided a more successful forecasting than the ARIMA. Experiment 3. Analysis of the electric use in industry (Model 3) We experienced that the monthly energy change is originally unsuitable for the validation of a model in the set stage of an appropriate ARIMA model to be applied in electric use in industry. After the analysis on the autocorrelation and partial autocorrelation function graphs of the series employed in industrial electric use, the difference was taken twice. Then, the series was made suitable for the analysis. After a few testing, the model was selected as ARIMA (1,2,1). Volume 1 Number 2 July 2011 101 �Rüstü YAYAR & Mahmut HEKIM & Veysel YILMAZ & Fehim BAKIRCI When the ANFIS model was trained by the training dataset for 1500 iterations, it found two rules for this experiment. The results obtained by ANFIS model are shown in Figure 5, where the test result of the model is red line and the forecasting of the next six month period is green line as follows: Figure 5. The ANFIS output for Model As seen in Figure 5, the next six months is constantly stationary for Model 3. The comparative results of the real values in the period between 2002-2010 and ARIMA (1,2,1) and ANFIS techniques are given in Table 3. Table 3: Forecasting of the electric use in the industry (Model 3) ARIMA(1,2,1) Months Jul.10 Real Forecasting ANFIS MSE Forecasting 6,5457 11,1835 6,4739 ug.10 6,6035 8,3942 6,1838 ep.10 6,7541 9,1898 ct.10 7,0919 8,2417 ov.10 6,7017 8,1281 6,1088 Dec.10 5,9649 7,6015 6,0901 Jan.11 7,2656 6,0714 eb.11 6,8283 6,0527 Mar.11 6,4308 6,0340 pr.11 6,0049 6,0153 May.11 5,5836 5,9967 Jun.11 5,1508 5,9780 102 6,1138 6,1471 6,1275 MSE 0,3079 Journal of Economic and Social Studies �A comparison of ANFIS and ARIMA techniques in the forecasting of electric energy consumption of Tokat province in Turkey As seen in Table 3, the MSE ratio is 6.1138 for ARIMA (1,1,2) and 0.0379 for ANFIS in the result of forecasting made through the six-month testing data for Model 3. Thus, it can be said that ANFIS provided a more successful forecasting than the ARIMA. Experiment 4. Analysis of the electric use in business firms (Model 4) While a model related to the electric use in business firms was constructed, monthly energy change was analyzed. But, it was determined that the series was originally unsuitable to construct an appropriate model. After the analysis on the autocorrelation and partial autocorrelation function graphs, the series was observed to be appropriate for an analysis stage when the difference was once taken. After a few testing, ARIMA(1,1,1) was determined to be the most appropriate model for this experiment. When the ANFIS model was trained by the training dataset for 3600 iterations, it found four rules for this experiment. The results obtained by ANFIS model are shown in Figure 6, where the test result of the model is red line and the forecasting of the next six month period is green line as follows: Figure 6. The ANFIS output for Model 4. As seen in Figure 6, there was very quickly rise in the next six months for Model 4. The comparative results of the real values in the period between 2002-2010 and ARIMA(1,1,1) and ANFIS techniques are given in Table 4. Volume 1 Number 2 July 2011 103 �Rüstü YAYAR & Mahmut HEKIM & Veysel YILMAZ & Fehim BAKIRCI Table 4: Forecasting of electric use in business firms (Model 4) Months Jul.10 Real ARIMA(1,1,1) Forecasting ANFIS MSE Forecasting 5,1655 6,6010 6,9789 ug.10 9,5100 7,1028 7,1073 ep.10 9,4926 7,0126 ct.10 5,8424 7,1121 ov.10 6,5950 7,1508 7,5016 Dec.10 7,4564 7,2089 7,6370 Jan.11 7,2608 7,7745 eb.11 7,3147 7,9142 Mar.11 7,3680 8,0561 pr.11 7,4215 8,2001 May.11 7,4749 8,3462 Jun.11 7,5283 8,4943 2,6646 7,2370 7,3683 MSE 2,8888 As seen in Table 4, the MSE ratio is 2.6646 for ARIMA (1,1,1) and 2.888 for ANFIS as a result of the forecasting through the six month testing data for Model 4. Thus, it can be said that ARIMA is little more successful in the implemented experiment when compared to ANFIS. Experiment 5. Analysis of the monthly electric use in government agencies (Model 5) While constructing a model for the electric use in government agencies, the graphs of monthly energy change were used in hand as a basis. It was determined that the series was originally unsuitable for model validation. After the analysis on the autocorrelation and partial autocorrelation function graphs, the series was, then, found to be suitable for an analysis stage when the difference was once taken. After a few testing, the model was selected as ARIMA (2,1,1). When the ANFIS model was trained by the training dataset for 5000 iterations, it found four rules for this experiment. The results obtained by ANFIS model are shown in Figure 7, where the test result of the model is red line and the forecasting of the next six month period is green line as follows: 104 Journal of Economic and Social Studies �A comparison of ANFIS and ARIMA techniques in the forecasting of electric energy consumption of Tokat province in Turkey Figure 7. The ANFIS output for Model 5. As seen in Figure 7, there is a constant decrease in the next six months for Model 5. The comparative results of ARIMA (2,1,1) and ANFIS techniques and the real values in the period between 2002-2010 are given in Table 5. Table 5: Forecasting of the electric use in government agencies (Model 5) ARIMA(2,1,1) Month Jul.10 Real Forecasting MSE ANFIS Forecasting 4,2125 3,3762 3,5093 ug.10 3,9930 3,6692 3,5053 ep.10 2,6223 3,4032 ct.10 3,0580 3,4401 ov.10 3,6314 3,3647 3,4604 Dec.10 4,2085 3,3635 3,4337 Jan.11 3,3431 3,4011 eb.11 3,3429 3,3627 Mar.11 3,3415 3,3184 pr.11 3,3467 3,2686 May.11 3,3531 3,2131 Jun.11 3,3620 3,1522 Volume 1 Number 2 July 2011 0,3909 MSE 0,3841 3,4961 3,4812 105 �Rüstü YAYAR & Mahmut HEKIM & Veysel YILMAZ & Fehim BAKIRCI As seen in Table 5, the MSE ratio is 0.3909 for ARIMA(2,1,1) and 0.3841 for ANFIS in the result of the forecasting made through the six month testing data for Model 5. Thus, it can be said that ANFIS is little more successful in the forecasting when compared to ARIMA. Experiment 6. Analysis of electric consumption of the others (Model 6) Model testing named the others about the electric consumption of the subscriber groups for six months was done. It was determined that the series was not appropriate to construct a model in its original. Thus, the series became appropriate after the difference was once taken through the autocorrelation and partial autocorrelation function graphs. After several tests, the model were determined as ARIMA (1,1,1). When the ANFIS model was trained by the training dataset for 5000 iterations, it found three rules for this experiment. The results obtained by ANFIS model are shown in Figure 8, where the test result of the model is red line and the forecasting of the next six month period is green line as follows: Figure 8. The ANFIS output for Model 6. As seen in Figure 8, there will be a little increase in the next six months for Model 6. In Table 6, there are comparative results of the real values from the period of 2002-2010 and the techniques of ARIMA (1,1,1) and ANFIS in Model 6. 106 Journal of Economic and Social Studies �A comparison of ANFIS and ARIMA techniques in the forecasting of electric energy consumption of Tokat province in Turkey Table 6: Electric Consumption Forecasting of the Others (Model 6) Months ARIMA(1,1,1) Real Forecasting MSE ANFIS Forecasting em.10 7,1845 8,1967 7,4670 Ağu.10 8,4338 8,2828 7,5274 yl.10 8,2217 8,3353 ki.10 7,5329 8,3873 as.10 8,3855 8,4392 7,7085 ra.10 7,2539 8,4912 7,7689 ca.11 8,5431 7,8293 Şub.11 8,5951 7,8897 Mar.11 8,6470 7,9501 is.11 8,6990 8,0104 May.11 8,7509 8,0708 Haz.11 8,8029 8,1312 0,5540 7,5877 7,6481 MSE 0,3401 As seen in Table 6, MSE ratio is 0.3401 for ANFIS whereas it is 0.5540 for ARIMA (1,1,1) as a result of the forecasting which was done by using test data for six months for Model 6. Thus, ANFIS provided a more successful forecasting than ARIMA for Model 6. Experiment 7. The analysis of electric consumption of the Business firm, State office and others (Model 7) It was determined in the forecasting of modeling about the electric consumption of the business firm, state office and others that consumed energy is non-stationary. The series became stationary by taking the difference for once according to the autocorrelation and partial autocorrelation function graphs of the related series. The ARIMA techniques which are obtained through several experiments must pass the compliance test to be used as an forecasting model regarding the future. In the compliance test which was done for this reason, it was determined that the autocor- Volume 1 Number 2 July 2011 107 �Rüstü YAYAR & Mahmut HEKIM & Veysel YILMAZ & Fehim BAKIRCI relation coefficients of the forecasting errors of the forecastings obtained through the ARIMA techniques shows a systematic tendency. The appropriate model could not be determined. When the ANFIS model was trained by the training dataset for 5000 iterations, it found three rules for this experiment. The results obtained by ANFIS model are shown in Figure 8, where the test result of the model is red line and the forecasting of the next six month period is green line as follows: Figure 9. The ANFIS output for Model 7. As seen in Figure 9, there will be a linear increase in the next six months for Model 7. In Table 7, there are the results of ANFIS model for Model 7. 108 Journal of Economic and Social Studies �A comparison of ANFIS and ARIMA techniques in the forecasting of electric energy consumption of Tokat province in Turkey Table 7: Electric Consumption Forecasting of the Commerce Houses, State Offices and Others (Model 7) Months ARIMA Real Forecasting ANFIS MSE Forecasting em.10 16,562 17,5558 Ağu.10 21,937 17,6570 yl.10 20,336 17,7583 ki.10 16,433 17,8599 as.10 18,612 17,9616 ra.10 18,918 ca.11 o appropriate model has been found MSE 4,8571 18,0636 18,1660 Şub.11 18,2690 Mar.11 18,3730 is.11 18,4789 May.11 18,5878 Haz.11 18,7017 As seen in Table 7, there is not any forecasting through ARIMA for Model 7. MSE ratio became 4.8571 as a result of the forecasting through ANFIS which was done by using test data for six months. The estimated results of the next six months are similar to the results obtained through ANFIS but there has been an increase at the least. Conclusions We implemented different seven experiments for estimating and analyzing electric energy consumption in order to plan the production, transmission and distribution of electric energy and to determine the consequences of the events occurring in the electric market. These experiments focus on the electric energy consumption forecasting implemented by using ANFIS and ARIMA techniques including the analyses of total energy consumption, household electric consumption, industrial electric consumption, commerce house electric consumption, monthly electric con- Volume 1 Number 2 July 2011 109 �Rüstü YAYAR & Mahmut HEKIM & Veysel YILMAZ & Fehim BAKIRCI sumption in state offices, electric of the others and the electric consumption of the commerce houses, state offices and the others. This is especially guiding key for the investors planning the investments in the electric sector. There has been a preparatory work for the necessary precautions regarding the electric in Tokat, revealing the electric consumption structure of Tokat province. The electric demand structure of Tokat regarding the previous consumption of Tokat province and the estimated electric energy for the future has been revealed. Although the study which was conducted through ANFIS and ARIMA techniques in the field of electric energy consumption was conducted on regional basis, it can be a pilot study for the extensive national and international studies on the energy consumption. Electric energy demands for the tested periods and the first six months of the year 2011 were estimated by using the ARIMA and the ANFIS techniques. Every forecasting experiment related to the electric consumption performed by ANFIS and ARIMA showed that ANFIS is more successful estimator. The ARIMA was more successful in only an forecasting experiment. In addition to this, an appropriate ARIMA model could not have been found for two forecasting experiments. As a result, the ANFIS is more appropriate than the ARIMA in the forecasting studies regarding the electric consumption. References Al-Garni, A.Z. & Abdel-Aal R.E. (1997). Forecasting Monthly Electric Energy Consumption in Eastern Saudi Arabia Using Univariate Time Series Analysis, Energy, Volume 22, Issue 11, 1059–1069. Al-Garni, A.Z. & Javeed Nizami, S. S. A. K. (1995). Forecasting Electric Energy Consumption Using Neural Networks, Energy Policy, Volume 23, Issue 12, 1097–1104. Asilkan, Ö. & Irmak, S. (2009). Forecasting the Future Prices of the Second Hand Automobiles Through Neural Networks, Journal of Faculty of Economics And Administrative Sciences, Volume: 14, 2, 378. Bircan, H. & Karagöz, Y. (2003). A Practice on The Monthly Exchange Rate Forecasting Through the Box-Jenkins Models, Journal of Social Sciences of Kocaeli University, Issue:6/2, 49, 51. 110 Journal of Economic and Social Studies �Bozkurt, H. (2007). Analysis of Time Sequences, Ekin Publishing House, Bursa, 7–8. Brown R.E. & Koomey J.G. (2002). Electricity Use in California: Past Trends and Present Usage Patterns, Enduse Forecasting and Market Assessment, http://enduse.lbl.gov/info/LBNL-47992.pdf. Çebi, H. & Kutay, F. (2004). Forecasting of The Turkey’s Electrical Energy Consumption Until 2010 Through Artificial Neural Networks, J. Fac. Eng. Arch. Gazi Univ., Volume 19, No 3, 227-233, http://www.mmfdergi.gazi.edu.tr/2004_3/227234.pdf. Demir, A. (1968). A Study on World’s Economy, SBF Publications of Ankara University No.259 Ankara, 68. Demirel, Ö. & Kakilli, A. & Tektaş, M. (2010). Forecasting of Electrical Energy Load Through Anfis And Arima Models, Journal of Faculty of Engineering And Architecture Volume J. Fac. Eng. Arch. Gazi Uni. 25, No:3, 602. Dikmen, N. (2009). Basic Concepts and Applications of Econometrics, Nobel Publication Distribution, Ankara. Dobre, I. & Alexandru, A. A. (2008). Modelling Unemployment Rate Using BoxJenkins Procedure, Journal of Applied Quantitative Methods, 157. Electricity Generation Corporation (2011). www.euas.gov.tr. Governorship of Tokat (2006). Province Annual of 2006, Tokat, 39, 146. Gujarati, D.N.(2009). Translators Şenesen, G. G. & Şenesen, Ü., Basic Econometrics, Sixth Edition, Literature Publication, İstanbul, 378. Jang, J. S. R., (1993). ANFIS: Adaptive-Network Based Fuzzy Inference Systems, IEEE Transaction on Systems, Man, and Cybernetics, 23 (03): 665-685. Kılıç, N. (2006). A General Overview On Turkey’s Electrical Energy Consumption And Production, Izmir Chamber of Commerce, R&D Journal, July 2006, İzmir, 12. Lise, W. & Van Montfort, K. (2005). Energy Consumption And Gdp in Turkey: is There a Cointegration Relationship? June 29 – July 2, 2005, Istanbul, Turkey, 5. Official Gazette, (2006). Regulations on Activity Report by Public Administrations, 17.03.2006, Issue: 26111. Önder, E. & Hasgül, Ö. (2009). Use of Box-Jenkins Model on the Forecasting of Foreign Visitors, Time Sequence Analysis Through the Winters Method And Artificial Neural Networks Winters, Administration, Year: 20, Issue: 62, 62-83. �Provincial Department of Environment and Foresty of The Governorship of Tokat, (2007). Environment Status Report of 2007 of Tokat, Tokat, 1, 130. Ross, T. J. (2004). Fuzzy Logic with Engineering Applications, Second edition, John Wiley. Şekerci Öztura, H.(2007). Today and Tomorrow of Electrical Energy In Our Country, VI. Energy Symposium of Union of Chambers of Turkish Engineers And Architects - Global Energy Policies And The Reality of Turkey, Ankara, 268. Tak S. (2002). Methods of Electrical Energy Demand Estimation and An Econometric Practice for Turkey (1970–2005), Unpublished Postgraduate Thesis, Atatürk University, Erzurum, 47. Terzi, H. (1998). Relationship Between Electricity Consumption and Economic Growth in Turkey : A Sectoral Comparision, Journal of Business and Finance, Volume: 13, Issue:144, March 1998, 62-71. Tufan, E. & Hamarat, B. (2003). Clustering of Financial Ratios of the Quoted Companies through Fuzzy Logic Method, Journal of Naval Science and Engineering, Vol. 1, No:2, 123-140. Turkish Statistical Institute (TUIK) (2011). http://report.tuik.gov.tr/reports/ rwservlet?ad nksdb2=&ENVID=adnksdb2Env&report=ikametedilen_ il.RDF&p_kod=2&p_kil1=60&p_yil=2010&p_dil=1&desformat=html. Turkish Statistical Institute (TUIK) (2009). Regional Indicators, 2009 TR 83 Samsun, Tokat, Amasya, April 2010 Ankara, 10. VIII. Regional Directorate of State Hydraulic Works - Samsun, (2011). http:// www2.dsi.gov .tr/bolge/dsi7/tokat.htm#atakoy. � Dublin Core The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/. Extent The size or duration of the resource. 1070 Title A name given to the resource A comparison of ANFIS and ARIMA techniques in the forecasting of electric energy consumption of Tokat province in Turkey Author Author YAYAR, Rüstü HEKIM, Mahmut YILMAZ, Veysel BAKIRCI, Fehim Abstract A summary of the resource. In this study, the electric energy demand of Tokat province was estimated by means of ANFIS and ARIMA techniques. Seven different forecasting experiments were implemented for the subscriber groups and the consumption of electric energy which is the dependent variable. The electric energy demand of the province for the first six months of the year 2011 was estimated by means of ANFIS and ARIMA techniques. The obtained results were compared and interpreted in order to illustrate the forecasting success of these techniques. We showed that the ANFIS is more appropriate than the ARIMA in point of the forecasting of electric consumption. Publisher An entity responsible for making the resource available International Burch University Date A point or period of time associated with an event in the lifecycle of the resource 2011-01 Keywords Keywords. Article PeerReviewed H Social Sciences (General),HB Economic Theory,HG Finance https://omeka.ibu.edu.ba/files/original/6b1eaf536440a11a8f10e43b3f84a8eb.pdf e0d73488a26f9a393063a8f4ce297516 PDF Text Text Journal of Economic and Social Studies Construction of Multi Dimensional Performance Measurement Model in Business Organizations: An Empirical Study Feyyaz YILDIZ Faculty of Business and Administrative Science, Afyon Kocatepe University, Turkey. feyyaz.yildiz@gmail.com Mustafa HOTAMIŞLI Faculty of Business and Administrative Science, Afyon Kocatepe University, Turkey. hotamisli@aku.edu.tr Ali ELEREN Faculty of Business and Administrative Science, Afyon Kocatepe University, Turkey. eleren@aku.edu.tr ABSTRACT The studies of performance measurement in firms have been conducted for a long period of time. However, the performance models and methods used in previous studies were limited. The purpose of this study is to test a performance based model that uses a modified approach in firms’ performance measurement. The new performance model used in this study is based on expectations in terms of performance measurement and evaluation of the firms with multiple dimensions. Different from the conventional gap models, the method used in this study is “Performance Measurement Method Based on Gap Percentages” developed by Eleren (2009). This method allows the researcher to use quantitative and qualitative data together. The model was tested with data collected from 42 firms engaged in business activities in marble industry in the Turkish province of Afyonkarahisar. Keywords: Performance Measurement, Multi-Dimensional Performance Evaluation Model, Gap Percentages Analysis, Marble Sector, Afyonkarahisar. Volume 1 Number 1 January 2011 33 �Feyyaz YILDIZ & Mustafa HOTAMIŞLI & Ali ELEREN Introduction Under the conditions of competitions which changed and became even more difficult with the globalization, the importance of performance management for the firms (enterprises) has become even more important. Performance management is taken into consideration within the management information systems concept-wise and consists of functions such as measurement of performance and development. Following the performance in general from the individuals and units to the general bears importance in terms of power and sustainability under the conditions of competition. Thus, the managers of the enterprises (firms) give a greater importance to performance management systems today. Sometimes the strategy and goals developed by the business organizations in the course of time may be in conformity to all the activities of the organization and obtaining a performance model based on exceeding the goals previously will become an obligation. Performance measurement and evaluation for firms was executed with simple and limited methods with single measure only until recently and it was based on partial measurements. However today, with the developments in the performance management systems as well as the use of improved statistical and mathematical methods, many models and methods have been developed in measurement and assessment of performance with multiple criterions and multiple dimensions. If the goal in performance measurement and assessment models is the evaluation of the performance of the enterprise in question, the goals to be determined and the criterions of evaluation should have gaps based on the structure of the sector. Thus, the sector based precision must be taken into consideration during the formation of these models. As the models are being prepared, other than the models prepared in the way to address to all sectors in the way to address them constantly, the importance of designing the models based on the sectors exclusively have gained importance due to reasons such as sector gaps, changes in conditions. Furthermore, changes based on time oblige the models to become more flexible so as to use it in the subsequent time interval. Many methods have been used in performance measurement. One of them is Gap methods which is used for measurement of performance even if not frequently. This method which we meet in performance measurement based on the quality of service (Servqual or Serperf) is based on the principle of comparison of expected (targeted) results and the realized (factual) results. Literature Review There are many studies related to performance, performance management and measurement of performance in literature. The concept “Performance management system” was first used by Beer and Ruh (1976). Thereafter, Bell created a foundation for development of the system to a further point with his studies in (1978) and (1987). The studies in this field started to increase in number since 1990s.There are many definitions available in the literature on performance, performance management and performance models. In their study named Auditing Productivity in the firms, Baş and Artar (1991) explained performance as; “the quantitative and qualitative explanation of intended goals that is related with an individual, a group or an enterprise engages and performs, in other 34 Journal of Economic and Social Studies �Construction of Multi Dimensional Performance Measurement Model in Business Organizations:.... words it is a quantitative and qualitative explanation of what they achieved and performed related with their tasks”. According to Akal (1992), performance is “the concept which determines what was obtained as the result of a purposeful and planned activity in general context.” According to Macey (2001), performance management is an extensive process to make a firm reach its goals with performance management and functional strategies. Barutçugil (2002) defined performance management as “the management process which undertakes to perform collection of information for the current and future position of the organization, to compare the same and to commence and continue the required and new activities to provide constant development of the performance so as to direct the business organizations to the objectives”. Harrington (1996) defined it as “the series of operations which determine at which rate the organizations can reach to the previously determined objectives”. According to Tekeli (2003) the performance measurement is, “the information obtained by the comparison or association of the factors which affect the success of a firm”. In more technical terms, the performance measurement is “the process of regular and systematic data collection, analysis and reporting to be used by a firm to follow up the resources it uses, the results it obtained with the produced goods and services”. You may find briefly the primary models in performance literature and the performance dimensions used in these models as listed in the following Table 1. Table 1. Multi Dimensional Performance Evaluation Models (Ağca, 2009, p.56). Innovation Learning and Development Employees √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ Vision/ Strategy √ √ √ √ √ Laitinen 2002 √ Integrated performance measurement model for SMEs √ EFQM 1991-1999 √ European Quality Foundation Perfection Model √ √ Neely et al. 2002 √ √ Performance Prism Model √ √ Chen-nel et al. 2000 √ √ Organizational Performance Evaluation Model √ Atkin-son et al. 1997 √ Responsibility Based Performance Evaluation Model √ Bititci et al. 1997 Market Product/Quality of Processes Product /Speed of Process Efficiency/ Productivity Flexibility Integrated Performance Measurement Model Fitzgerald et al. 1991 Results Determinants Model √ √ Kaplan and Norton 1992 Lynch and Cross 1991 Performance Pyramid √ √ Balanced Scorecard Indicator Keegan et al. 1989 Financial Customer PERFORMANCE DIMENSIONS Performance. Measurement Matrix MULTI DIMENSIONAL PERFORMANCE EVALUATION MODELS √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ Volume 1 Number 1 January 2011 √ √ √ √ √ √ √ √ √ √ 35 �Feyyaz YILDIZ & Mustafa HOTAMIŞLI & Ali ELEREN Competition Social Responsibility and External Environment √ √ √ √ √ √ √ √ √ √ √ √ Other than the models used for performance measurement, there are also methods of measuring. The information relating to these methods are given in the following Table 2 in brief. As the table is analyzed, it can be seen that the simulation and statistical methods are predominantly preferred. However, it can be observed that there is significant increase in the use of Decision Making Methods with Multiple Criterions. The reason for preferring these methods are other than the fact that they are methods which are easily applicable, it can work with quantitative and qualitative data and it allows a model consisting of different dimensions and variables to be transformed into a single performance variable. The most frequently used method among the Decision Making Methods with Multiple Criterions is the Analytical Hierarchy Process and TOPSIS method. At the same time, the approaches of these methods taken into consideration with fuzzy logic are preferred. Table 2. Examples from the methods used in the measurement of the performance (Akyüz, 2006, p.26. ; Eleren,.2009, p.1304). AUTHORS Jagadees and Babu (1994) Chenhal (1996) Tong and Chen (1998) Berry and Cooper (1999) Caporaletti et al. (1999) Lo and Pushpakumara (1999) Martin et al. (1999) Suwignjo et al. (2000) Bititci et al. (2001) MacCarthy and Wasuri (2001) Selen and Asheyeri (2001) Chan et al. (2002) Corbett and Pan (2002) Yurdakul (2002) Chan et al. (2003) Sarkis (2003) Chen and Chen (2004) Triantis and Otis (2004) Agus (2005) Ali and Wadhwa (2005) Meer et al. (2005) Silandria (2005) Pearn and Wu (2006) Sandrock et al. (2006) Eleren and Özgür(2006) Eleren (2007) Eleren and Soba (2009) METHODS USED IN THE MEASUREMENT OF THE PERFORMANCE SPC √ PE √ FA MCDM DEA SEM LP NLP FUZ √ REG SIM √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ SPC: Statistical Process Control, PE: Process Efficiency, NLP: Non Linear Programming, DEA: Data Enandlope Analysis, SEM: Structural Equation Model, LP: Linear Programming, FA: Factor Analysis, FUZ: Fuzzy Logic, REG: Regression – SIM: Simulation, MCDM: Multi Criteria Decision Making, (egg; AHP, and TOPSIS.Model.) 36 Journal of Economic and Social Studies �Construction of Multi Dimensional Performance Measurement Model in Business Organizations:.... Methodology In forming a performance model, determination of performance dimensions and variables, and weighing of variables are required. This research was conducted through a survey among the senior managers of the firms which continue their business activities in the marble industry as registered to the Chamber of Industry and Trade in Province of Afyonkarahisar in TurkeyThe purpose of this study is to develop a multidimensional performance measurement model and to determine the dimensions of this model, variables within each dimension and weighing of each variable for a sector. The sample of the study is composed of the 42 firms engaged in business activities in marble industry in the Province of Afyonkarahisar, registered to the Chamber of Industry and Commerce of Afyonkarahisar and the information relating to these firms for the year 2009.Primary data were used in the study. In order for the researcher to reach its goals, the original data he needs, the data he has collected with the use of relevant devices are named as the primary data (Altunışık et al., 2005). At this point, in order to reach the primary data, face to face interview among conventional survey methods was used. In the selection of the sampling, the method used was sampling method which is not random and based on probability. In preparing the scale used to obtain the data, the scale used by Eleren and Soba (2009) was considered as the basic scale. However, although it originally consists of six dimensions, the dimensions at this stage were limited to four dimensions as employee satisfaction, finance, production and marketing functions. Two staged scale was used consisting of questions with the purpose of collecting data for each variable the questions relating to weighing the dimensions and the variables relating to such dimensions. In the survey section prepared to collect information, there are questions relating to each dimension. The target relating to the relevant variables in the questions and the results which were realized have been asked to be evaluated. The questions relating to the first of the dimensions were answered by the employees and the others were answered by the senior management. Empirical Results Firms participated in this study; 100% of them are classified as SME (according to the criteria of workforce, turnover and capital). 18% of the firms consist of single person enterprises, 68% consist of limited liability companies and others consist of joint stock corporations. 86% of the firms are family businesses and family members are assigned in management positions. 62% of the senior managers of the firms consist of persons with bachelor’s degree or higher proves that they attach importance to education although they are SMEs or family businesses. 92% of the workers consist of men and their average age is 29 and this qualifies as young work force. Despite this, their average work experience is 14 years which proves that they started business at a very young age. Volume 1 Number 1 January 2011 37 �Feyyaz YILDIZ & Mustafa HOTAMIŞLI & Ali ELEREN Model is formed in three stages which are listed as follows; • • Determining and weighting dimensions of the model, Determining and weighting variables related with the dimensions of the model, In the first stage, the senior managers were asked to evaluate the dimensions between 1-5 and as the result of these evaluations, the average points were proportioned to the total points based on the significance levels. They are as follows; • • • • Employee expectations and satisfaction (W=0,189), Production Management (W=0,274), Marketing Management (W=0,261), Financial Management (W=0,276) In the second stage, the senior managers were asked to evaluate the variables relating to all dimensions between 1-5 and as the result of this evaluation; the significances of the variables within the dimension was calculated. The results of this evaluation is as follows: 1. In terms of Employee expectations and satisfaction, the purpose was to determine the employees’ level of satisfaction from the enterprise and the management. The evaluation questions were asked to only 145 of 489 employees who work in 42 enterprises. Table 3. Employee expectations and satisfaction factor and its variables 1.EMPLOYEE EXPECTATIONS AND SATISFACTION I am satisfied with the salary and wage against what I perform as my job. The working/living quality provided is satisfactory. Peace and safety has priority in terms of work satisfaction. Everyone has fair share of speaking in management. I believe that the distribution of wages and bonuses is fair and just. We believe that the work load is suitable. I believe that we have sufficient work safety We work in team spirit. It is satisfying that the theoretical and applied trainings are provided. All workers have adopted the culture of the enterprise. N : 489 / n: 145 SIGNIFANCE LEVEL AVERAGE W/w 2.99 0.189 4.75 0.118 4.42 0.110 4.31 0.107 4.27 0.106 4.20 0.104 4.11 0.102 3.93 0.098 3.59 0.089 3.49 0.087 3.22 0.080 2. The evaluation of the variables relating to production management function was 38 Journal of Economic and Social Studies �Construction of Multi Dimensional Performance Measurement Model in Business Organizations:.... conducted by the business owners/senior management. Most of the variables consist of quantitative data. Table 4. Product Management Factor and its Variables SIGNIFANCE LEVEL 2. PRODUCT MANAGEMENT Diversity of Products (*) Age of production technology (*) Rate of capacity usage (*) Rate of Wastage % (*)(-) Number of patents developed (*) Number of patents owned (*) Level of professionalism in production (1-5) Vocational training studies ( hour / year) (*) Number of projects performed during last five years (*) Number of work accidents and sicknesses incurred during last five years (*)(-) N : 124 / n: 42 Note: [(*) Quantitative Data ; ( - ) Negative Directional]. AVERAGE 4.33 4.54 4.48 4.31 4.22 4.04 3.93 3.91 3.41 3.36 3.25 W/w 0.274 0.115 0.114 0.109 -0.107 0.102 0.100 0.099 0.086 0.085 -0.082 3. The evaluation of the variables relating to marketing management function was conducted by the business owners/senior management. Most of the variables consist of quantitative data. Table 5. Marketing Management Factor and its variables SIGNIFANCE LEVEL 3. MARKETING MANAGEMENT Rate of increase in annual sales (*) Rate of decrease in customer complaints (1-5) Ratio of exports in all sales (*) Number of trade mark registered products (*) Level of professionalism in marketing management (1-5) Total number of products (*) Number of Web based / e-trade sales % (*) Level of cooperation with Professional logistics companies (1-5) Training of sales personnel ( … hour / year) Rate of marketing costs in total costs % (*)(-) N : 124 / n: 42 Note: [(*) Quantitative Data; ( - ) Negative Dimensional]. AVERAGE W/w 4.12 4.11 4.05 3.92 3.83 3.79 3.61 3.52 3.37 3.34 3.28 0.261 0.116 0.114 0.107 0.104 0.103 0.101 0.092 0.091 0.087 -0.085 4. The evaluation of the variables relating to financial management function was conducted by the business owners/senior management. Most of the variables consist of quantitative data Volume 1 Number 1 January 2011 39 �Feyyaz YILDIZ & Mustafa HOTAMIŞLI & Ali ELEREN Table 6. Financial Management Factor and Variables 4. The evaluation of the variables relating to financial management function was SIGNIFANCE LEVEL conducted by the business owners/senior management. Most of the variables AVERAGE W/w consist of quantitative data. 4. FINANCIAL MANAGEMENT Table 6. Financial Management Factor and Variables Level of professionalism in financial management (1-5) Periodical conduct of Financial planning, analysis and audits (1-5) Management Accounting application (1-5) 4. FINANCIAL MANAGEMENT Equity Capital / Total Assets Level of professionalism in financial management (1-5) Periodical Rate conductofofAccounts Financial planning, analysis and audits (1-5) Turnover Receivables Management Accounting application (1-5) Liquidity (Current Ratio) Equity Capital / Total Assets Turnover of Accounts Receivables Net ProfitRate / Equity Capital Liquidity (Current Ratio) Net TotalCapital Assets NetProfit Profit / /Equity Net Profit / Total Assets Stock Turnover Stock Turnover Net Capital Rate of turnover NetWorking Working Capital Rate of turnover 124 / /n:n: 42 42 N N: :124 Note:Note: [(*)[(*) Quantitative Data]. Quantitative Data]. 4.36 0.276 4.66 0.114 0.112 W/w0.109 0.276 0.100 0.114 0.1120.100 0.109 0.096 0.100 0.1000.095 0.096 0.0950.094 0.094 0.091 0.091 0.0900.090 SIGNIFANCE LEVEL 4.62 AVERAGE 4.55 4.36 4.34 4.66 4.62 4.15 4.55 4.09 4.34 4.15 4.01 4.09 3.93 4.01 3.93 3.87 3.87 3.71 3.71 The factors and variables and their weights to be used in forming the performance model and The factors and variables and their weights to be used in forming the performance model their weights were determined in the previous section. At this point, the model below was and their weights were determined in the previous section. At this point, the model below generated with the use of the data mentioned here. was generated with the use of the data mentioned here. Performance Function is denoted by by f(x), Wi , , Performance Function is denoted f(x),factor factor(dimension) (dimension)weights weightsare are denoted denoted by by W i variables of the gap percentage are denoted by x and the weights of the variables are denoted ij variables of the gap percentage are denoted by xij and the weights of the variables are by wi; denoted by wi; f(x) = W1*F1 + W2*F2 + W3*F3 + W4*F4 (1) = W1*(w11*x11 + w12*x12 + …) + W2*(w21*x21 + w22*x22+ ….) + …. (2) Discussion Discussion The data have been prepared in an M.S. Excel file with all factors and the related variables. have been prepared in an participated M.S. Excel file factors and related variables. As The all data 42 enterprises which have in with the all research werethetransferred to the As all 42 enterprises which haveincreased participated in the research were transferred to the worksheet, since the size of the file excessively, 10 enterprises have been selected among the enterprises define thefile small and mid-scaled so ashave to represent them worksheet, since thetosize of the increased excessively,enterprises 10 enterprises been selected andamong performance model wastoapplied the enterprises define on thethese smallenterprises. and mid-scaled enterprises so as to represent them and performance model was applied on these enterprises. Calculation of Gaps and Gap The data to all dimensions and variables Calculation of Gaps and Percentages: Gap Percentages: Therelating data relating to all dimensions and for each enterprise were entered in M.S. Excel worksheet. The data entered consist of binary variables for each enterprise were entered in M.S. Excel worksheet. The data entered consist data system. These are the realized and expected performance values. These values are of binary the realized and expected and performance values. These values classified intodata foursystem. groupsThese beingare quantitative and qualitative positive and negative are classified into four groups being quantitative and qualitative and positive and negative dimensional. Likert questions consist of qualitative values between 1 and 5. Moreover, dimensional. Likert questions consist of qualitative values between 1 and 5. Moreover, quantitative data such as rate of capacity usage or liquidity consist of rations or numbers quantitative datavariables. such as rate of capacitythe usage or liquidity consist rations which or numbers which express these Furthermore, variables such as rate ofofwastage is not which express these are variables. the variables as rate coefficients of wastage which is desired to be increased definedFurthermore, as negative directional andsuch the weight have the sign (-). As the differences are calculated, the formulations mentioned below will be used: value (3) Journal of expected) Economic and Social Studies 40 gap = (Performance value realized) – (Performance �Construction of Multi Dimensional Performance Measurement Model in Business Organizations:.... not desired to be increased are defined as negative directional and the weight coefficients have the sign (-). As the differences are calculated, the formulations mentioned below will be used: gap = (Performance Value Realized) – (Performance Value Expected) (3) The result being zero means that the expected prediction was not provided hence low performance. If the result is zero, it means that full performance was maintained and if it is over zero, it means that it was exceeded. Performance gap percentages are other indications of the gap and since the rate defined for performance calculation is between -1/+1 , it allows that the data will be standardized before they were used in performance model. gap percentage=(Performance Value Realized–Performance Value Expected) / (Performance Value Expected) (4) After formation of the performance function f(xi), by using all ratios, groups and weights, performance points can be determined. The points are calculated as the result of the operations below respectively. For each enterprise involved in performance evaluation individually; • The gaps and the gap percentages between the performance values expected and realized for each observation will be calculated for all performance dimensions and variables. • The weighted gap percentages will be calculated by multiplying the gap percentages with the weights of the variables. • The weighted gap percentages of the variables at all dimensions will be calculated and the dimension scores will be found. • The weighted dimension scores will be calculated by multiplying the score dimensions with their own dimensional weights. • At the last stage, the weighted score for each dimension will be summed and the total scores of the enterprises will have been obtained. As the scores were ranked in order of amplitude, the performance order of the enterprises will have been formed. If the score is negative, it is interpreted that the enterprise failed to reach its goals in terms of all dimensions. If it is zero, it means that it fully reached its target and if it is a positive number than it will be interpreted that it has exceeded its targets and became more successful. • Theoretically, it is assumed that total points vary between -1,00 and +1,00. Moreover, since the performance scores based on dimensions were found by summing them, it should be taken into consideration that the numbers of variables should be different in all dimensions. For instance, in this study, each dimension was defined with 10 variables (questions). If different number of variables were present in dimensions, it needs to be balanced after summing the dimension scores taking the Volume 1 Number 1 January 2011 41 �Feyyaz YILDIZ & Mustafa HOTAMIŞLI & Ali ELEREN number of the variables in consideration comparatively. Table 7. The Dimensions of the Enterprises and the Performance Points and Ranking in Total PERFORMANCE DIMENSIONS 1. EMPLOYEE EXPECTATIONS AND SATISFACTION I am satisfied with the salary and wage against what I perform as my job. The working/living quality provided is satisfactory. Peace and safety has priority in terms of work satisfaction. Everyone has fair share of speaking in management. I believe that the distribution of wages and bonuses is fair and just. We believe that the work load is suitable. I believe that we have sufficient work safety We work in team spirit. It is satisfying that the theoretical and applied trainings are provided. All workers have adopted the culture of the enterprise. 2. PRODUCT MANAGEMENT Diversity of Products (*) Age of production technology (*) Rate of capacity usage (*) Rate of Wastage % (*)(-) Number of patents developed (*) Number of patents owned (*) Level of professionalism in production (1-5) Vocational training studies ( hour / year) (*) Number of projects performed during last five years (*) Number of work accidents and sicknesses incurred during last five years (*)(-) 42 Journal of Economic and Social Studies �Construction of Multi Dimensional Performance Measurement Model in Business Organizations:.... ENTERPRICES A01 A02 A03 A04 A05 A06 A07 A08 A09 A10 AVR - 0,004 - 0,010 - 0,023 - 0,021 - 0,015 - 0,010 - 0,015 0,002 - 0,013 0,004 - 0,011 0,004 - 0,003 - 0,003 - 0,003 - 0,001 - 0,002 - 0,004 - 0,002 - 0,003 - 0,002 -0,002 - 0,002 - 0,001 - 0,003 - 0,002 0,004 - 0,003 0,004 0,002 - 0,001 0,003 0,000 - 0,002 - 0,003 - 0,001 - 0,002 - 0,003 0,003 - 0,001 0,002 0,001 - 0,001 -0,001 - 0,001 - 0,003 - 0,003 - 0,003 - 0,003 - 0,002 - 0,003 - 0,000 - 0,003 0,002 -0,002 0,001 0,003 - 0,003 - 0,002 - 0,003 - 0,003 - 0,002 0,003 - 0,002 - 0,001 -0,001 - 0,002 - 0,002 - 0,003 - 0,001 - 0,003 - 0,001 - 0,002 - 0,002 - 0,001 0,004 -0,001 - 0,002 - 0,001 - 0,002 - 0,002 - 0,001 0,003 - 0,002 - 0,001 - 0,002 - 0,000 -0,001 - 0,001 - 0,001 - 0,001 - 0,001 - 0,002 - 0,001 - 0,002 - 0,001 0,002 - 0,001 -0,001 0,003 0,003 - 0,002 - 0,003 - 0,001 - 0,002 - 0,002 0,002 - 0,003 - 0,002 -0,001 - 0,001 - 0,002 - 0,001 - 0,002 - 0,003 - 0,001 - 0,002 - 0,001 - 0,002 0,002 -0,001 0,002 - 0,006 - 0,011 - 0,012 - 0,001 0,001 - ,005 - 0,009 0,002 - 0,001 -0,004 - 0,000 - 0,003 - 0,000 - 0,001 0,001 - 0,000 0,002 0,002 0,004 - 0,003 0,000 0,002 - 0,001 - 0,001 - 0,002 - 0,002 0,001 0,003 - 0,001 - 0,003 - 0,001 -0,001 0,001 - 0,001 - 0,001 0,003 0,002 0,001 0,001 0,001 0,000 0,002 0,001 0,019 0,018 0,018 0,019 0,017 0,016 0,019 0,017 0,019 0,016 0,018 0,000 0,000 - 0,006 - 0,005 0,000 - 0,006 - 0,006 - 0,004 0,000 0,000 -0,003 0,000 0,000 0,000 - 0,006 0,000 - 0,004 - 0,005 0,000 0,000 0,000 -0,002 - 0,004 - 0,003 - 0,002 - 0,002 - 0,003 - 0,002 - 0,003 - 0,004 - 0,003 0,003 -0,002 - 0,002 - 0,002 - 0,003 - 0,003 - 0,002 - 0,003 - 0,002 - 0,002 - 0,002 - 0,003 -0,002 - 0,005 - 0,005 - 0,006 - 0,007 - 0,006 - 0,006 - 0,006 - 0,006 - 0,005 - 0,006 -0,006 0,003 - 0,008 - 0,008 - 0,007 0,002 0,004 - 0,007 - 0,008 0,002 - 0,007 -0,003 Volume 1 Number 1 January 2011 43 �Feyyaz YILDIZ & Mustafa HOTAMIŞLI & Ali ELEREN 3. MARKETING MANAGEMENT Rate of increase in annual sales (*) Rate of decrease in customer complaints (1-5) Ratio of exports in all sales (*) Number of trade mark registered products (*) Level of professionalism in marketing management (1-5) Total number of products (*) Number of Web based / e-trade sales % (*) Level of cooperation with Professional logistics companies (1-5) Training of sales personnel ( … hour / year) Rate of marketing costs in total costs % (*)(-) 4. FINANCIAL MANAGEMENT Level of professionalism in financial management (1-5) Periodical conduct of Financial planning, analysis and audits (1-5) Management Accounting application (1-5) Equity Capital / Total Assets Turnover Rate of Accounts Receivables Liquidity (Current Ratio) Net Profit / Equity Capital Net Profit / Total Assets Stock Turnover Net Working Capital Rate of turnover PERFORMANCE SCORE RAN. NUM. 44 Journal of Economic and Social Studies �Construction of Multi Dimensional Performance Measurement Model in Business Organizations:.... - 0,015 0,012 - 0,017 - 0,018 0,003 - 0,011 - 0,019 - 0,013 - 0,018 - 0,003 -0,010 - 0,004 0,003 - 0,004 - 0,005 - 0,004 - 0,001 - 0,002 - 0,005 - 0,005 - 0,004 -0,003 - 0,003 - 0,001 0,001 - 0,002 - 0,000 - 0,002 - 0,002 0,003 - 0,002 0,000 -0,001 - 0,005 - 0,004 - 0,003 - 0,005 0,006 - 0,006 - 0,005 - 0,006 - 0,005 - 0,005 -0,004 - 0,006 0,006 - 0,005 - 0,006 - 0,004 - 0,005 - 0,006 0,004 - 0,003 - 0,004 -0,003 - 0,003 - 0,002 - 0,004 - 0,003 - 0,002 - 0,002 - 0,003 - 0,003 - 0,002 - 0,004 -0,003 - 0,003 - 0,002 - 0,002 - 0,003 - 0,002 - 0,002 - 0,003 - 0,003 - 0,002 - 0,003 -0,003 - 0,006 - 0,008 - 0,007 - 0,006 - 0,005 - 0,004 - 0,004 - 0,007 - 0,006 0,007 -0,005 - 0,002 - 0,003 - 0,002 - 0,002 - 0,003 - 0,004 - 0,003 - 0,002 - 0,002 - 0,003 -0,002 0,002 0,000 0,002 - 0,000 0,000 - 0,002 - 0,003 0,000 - 0,000 - 0,005 -0,001 0,015 0,022 0,007 0,015 0,017 0,017 0,011 0,005 0,009 0,018 0,014 - 0,009 0,010 - 0,019 - 0,010 - 0,009 0,003 - 0,004 - 0,009 - 0,010 0,000 -0,006 - 0,003 0,003 - 0,003 - 0,003 0,004 - 0,002 - 0,004 - 0,003 - 0,003 - 0,003 -0,002 - 0,003 - 0,002 - 0,002 - 0,003 - 0,003 - 0,002 - 0,003 - 0,002 - 0,003 0,003 -0,002 - 0,004 - 0,003 - 0,003 - 0,003 - 0,004 0,002 - 0,003 0,004 - 0,003 - 0,003 -0,002 - 0,003 - 0,002 - 0,003 - 0,002 - 0,003 0,002 0,002 - 0,003 - 0,002 - 0,002 -0,002 0,004 0,001 0,002 0,003 0,004 0,000 0,002 0,001 0,002 0,002 0,002 - 0,000 0,001 - 0,003 - 0,001 0,000 - 0,001 - 0,001 - 0,000 - 0,002 0,001 -0,001 - 0,002 0,004 - 0,002 - 0,000 - 0,002 - 0,004 - 0,001 - 0,004 - 0,002 0,003 -0,001 0,001 0,005 - 0,002 0,001 - 0,006 0,002 0,001 0,000 - 0,000 0,001 0,000 0,004 0,003 0,000 0,002 0,002 0,004 0,003 - 0,002 0,003 0,002 0,002 - 0,002 0,002 - 0,003 - 0,004 - 0,001 0,000 - 0,001 - 0,000 0,000 - 0,001 -0,001 - 0,068 - 0,061 - 0,012 - 0,017 - 0,043 - 0,024 - 0,030 0,000 - 0,026 10 9 3 5 8 6 7 2 - 0,015 4 0,008 1 Volume 1 Number 1 January 2011 45 �Feyyaz YILDIZ & Mustafa HOTAMIŞLI & Ali ELEREN Total Performance Points and Ranking of The Firms: As the transactions mentioned in the previous section were followed up, the performance scores of the firms were calculated based on all dimensions and variables and it was shown above. Based on this; • All the firms at the model stage were not included in the sample application. Taking the matrix dimensions of the work sheet it was limited to 10 enterprises. The names of the enterprises were not mentioned since permission hasn’t been obtained. However they were denominated by numbers from 1 to 10. • The study is directly applicable to quantitative and qualitative data. • The Model is designed on an exclusive basis to the sectors taking the characteristics and the priorities of the sector in consideration. It also has the nature to be redesigned for each sector. Once, the model has been designed, the data based on each year can be used and be evaluated in comparative evaluations. • As the results were analyzed, it can be seen that the enterprise no. A02 takes the lead. It can be seen that especially the points which were obtained from finance and marketing dimensions were effective. Conclusion Many studies have been done on performance measurement and evaluation in the literature so far. Many models have been developed during these studies and different methods have been tried. It is of essential importance for the business organizations to determine their positions and their future goals precisely under the conditions of competition which became harder as well as following it up constantly. Due to this reason, it is inevitable that similar studies will continue on performance. The difference of the study in terms of the model and the method is based on re-evaluation of the differences based on gaps formed according to the differences between the previously used gap model and the results performed. The method applied allows separate scoring for all dimensions and by monitoring the scores, it is allowed to interpret how the scores have been formed and to analyze the quantitative - qualitative data together. The precision of the model varies based on the accurateness of data, participation of the significant rate of firms or enterprises in the sector, and the level of awareness of the participants as to the necessity of such a study. The Model is applicable in terms of individual evaluation of the enterprises (within its own course) and collective performance evaluations following its design for the sectors with dimensions, variables and weights. A02, A10 and A05 firms take the first three of ten enterprises denominated at the application stage of the study in code numbers. A02 firm which takes the first place has gained an advantage in terms of marketing and finance as it was evaluated in terms of dimensions. In terms of dimensions, A10 and A08 firms take the lead based on the in employee satisfaction,; A01 and A09 firms take the lead based on production, A02 and A05 firms take the lead based on marketing management and A02 and A06 firms take the lead based on financial management. 46 Journal of Economic and Social Studies �Construction of Multi Dimensional Performance Measurement Model in Business Organizations:.... References Agus, A. (2005). The Structural Linkages between TQM, Product Quality Performance, and Business Performance: Preliminary Empirical Study in Electronics Companies. Singapore Management Review, 27 (1), 87-105. Ağca, V. (2009). Türk İmalat İşletmelerinde Çok Boyutlu Performans Değerleme (PD) Modellerine Dayalı Performans Göstergelerinin Kullanılabilirliği, (The Usefulness of Performance Measures Based On Multi-Dimensional Performance Measurements Models in Turkish Manufacturing Firms), Dumlupınar University, Social Science Institute Journal, 23, 51-66. Altunışık, R., Coşkun, R., Bayraktaroğlu, S. & Yıldırım, E., (2005) Sosyal Bilimlerde Araştırma Yöntemleri: SPSS Uygulamalı (4th Ed.). Sakarya. Akal, Z. (1992). İşletmelerde Performans Ölçüm ve Denetimi. Ankara: MPM Yayınları. Akyüz, Gökhan (2006). Proses İmalatında Performans Ölçme Ve İyileştirmeye Yönelik Bir Modelleme Yaklaşımı: Bir Uygulama. (Basılmamış Doktora Tezi).Akdeniz Üniversitesi Sosyal Bilimler Enstitüsü, Antalya. Ali, M. & Wadhwa, S. (2005). Performance Analysis of Partial Flexible Manufacturing Systems. Global Journal of Flexible Systems Management, 6 (1),9-19. Andersen, B. (1999). Business Process Improvement. American Society for Quality: USA. Andersen, B. & Fagerhaug, T. (2001). Performance Measurement Explained,Designing and Implementing Your State-of-the-Art System, American Society for Quality: USA. Atkinson, A.A., Waterhouse, J.H. & Wells, R.B. (1997). A Stakeholder Approach to Strategic Performance Measurement. Sloan Management Review, 38 (3), 25-38. Atkinson, A. A. (1997). New Direction in Management Accounting Research. Journal of Management Accounting Research, 9, 285-311. Barutçugil, İ.(2002). Performans Yönetimi, Kariyer Yayınları: İstanbul. Baş, M. & Artar, A. (1991). İsletmelerde Verimlilik Denetimi. Ankara: MPM Yayınları No: 434. Beer, M. & Ruh, R. A. (1957). Employee Growth Through Performance Management. Harvard Business Review, July-August, 59-66. Beer, M., Ruh, R., Dawson, J. A., McCaa, B. B. & Kavanagh, M. J. (1978). A Performance Volume 1 Number 1 January 2011 47 �Feyyaz YILDIZ & Mustafa HOTAMIŞLI & Ali ELEREN Management System: Research, Design, Introduction and Evaluation. Personel Psychology, 31, 505535. Berry, W. L. & Cooper, M. C. (1999). Manufacturing Flexibility: Methods for Measuring the Impactof Product Variety on Performance in Process Industries. Journal of Operations Management, 17, 63-178. Bititci, U. S., Suwignjo, P. & Carrie, A. S. (2001). Strategy Management Through Quantitative Modelling of Performance Measurement Systems. International Journal of Production Economics, 69, 15-22. Chan, D. C. K., Yung, K. L., Ip, A. W. H. (2002). An Application of Fuzzy Sets to Process Performance Evaluation. Integrated Manufacturing Systems, 13 (4), 237-246. Chenhall, R. H. (1996). Strategies of Manufacturing Flexibility, Manufacturing Performance Measures and Organizational Performance: An Empirical Investigation. Integrated Manufacturing Systems, 7 (5), 25-32. Caporaletti, L. E., Dula, J. H. & Womer, N. K. (1999). Performance Evaluation Based on Multiple Attributes with Nonparametric Frontiers. Omega, 27, 637-645. Chen, J. P. & Chen, K. S. (2004). Quality And Reliability Corner Comparison of Two Process Capabilities by Using Indices Cpm: An Application to a Color STN Display. International Journal of Quality & Reliability Management, 21(1), 90-101. Chennell, A., Dransfield, S., Field, J., Fisher, N., Saunders, I. & Shaw, D. (2000). In Proceedings of the Performance Measurement –Past, Present and Future Conference 2000: OPM: A System for Organisational Performance Measurement. Cambridge. Corbett, C.J. & Pan, J. N., (2002). Evaluating Environmental Performance Using Statistical Process Control Techniques. European Journal of Operational Research, 139, 68-83. European Foundation For Quality Management (1998). EFQM Mükemmellik Modeli 2000 (EFQM ve KalDer Yayınları). İstanbul: Author. http://www.efqm.org Eleren, Ali. (2007). İMKB’ye Kayıtlı Çimento İşletmelerinin Finansal Tablolarının Bulanık Mantık Yaklaşımı İle Değerlendirilmesi. AKÜ İİBF Dergisi, 9 (1), 141-154. Eleren, A. & Özgür, E. (2006). Proceedins from 4. Ulusal Araştırmalar Sempozyumu’06: Gıda Sektöründe Faaliyet Gösteren İMKB’ye Kayıtlı İşletmelerde Analitik Hiyerarşi Süreci Yöntemi İle Performans Ölçümü. İstanbul: Turkey. 48 Journal of Economic and Social Studies �Construction of Multi Dimensional Performance Measurement Model in Business Organizations:.... Eleren, A. & Özgür, E. (2006). Türkiye’de Yabancı Sermayeli Mevduat Bankalarının Veri Zarflama Yöntemi İle Etkinlik Analizlerinin Yapılması. AKÜ İİBF Dergisi, 8 (2), 53-76. Eleren, A. & Soba, M. (2009). Proceedins from International Davraz Conress’09: The First International Davraz Congress on Social and Economic Issues Shaping the World’s Future: New Global Dialogue. Isparta: Turkey. Eleren, A. (2009). Proceeding from 6. KOBİ’ler ve Verimlilik Kongresi’09: The Sixth KOBİ’ler ve Verimlilik Kongresi. İstanbul: Turkey. Fitzgerald, L., Johnson, R., Brignall, S., Silvestro, R. & Voss, C. (1991). Performance Measurement in Service Businesses. London: CIMA. Frolick, M. N. & Ariyachandra, T. R. (2006). Business Performance Management: One Truth. Information Systems Management, 23(1), 41-48. Grady, M. W. (1991). Performance Measurement: Implementing Strategy. Management Accounting, 72 (June), 137-148. Heide, D. P. (1993). A Comprehensive Performance Management Model. (A Dissertation of Presented to the Graduate Faculty of the School of Human Behavior). United States International University Press. San Diego. Johnson, P. & Bell, C. (1987). Focused Vision for Focused Performance. Training and DevelopmentJournal, 41 (12), 56-59. Retrived from http://ehis.ebscohost.com/ehost/pdfviewer/pdfvi ewer?vid=4&hid=5&sid=fe3ddbe0-8ce1-4f9f-914a-d9ec39bab838%40sessionmgr11 Kaplan, R. ve Norton, D. (1992). The Balanced Scorecard: The Measures That Drive Performance. Harvard Business Review, Jan–Feb, 71-79. Retrived from http://hbr.org/2005/07/the-balancedscorecard/ar/1 Laitinen, E. K. (2002). A Dynamic Performance Measurement System: Evidence from Small Finnish Technology Companies. Scandinavian Journal of Management, 18, 65-99. Lo, E. K. & Pushpakumara, C. (1999). Performance and Partnership in Global Manufacturing Modelling Frameworks and Techniques. International Journal of Production Economics, 60 (1), 261269. Lynch, R. & Cross, K. (1991). Measure UP! Yardsticks for Continuous Improvement Oxford: Blackwell Publishing. Macey, S. (2001) An Integrated Model for Performance Management Based on ISO9000 and Business Volume 1 Number 1 January 2011 49 �Feyyaz YILDIZ & Mustafa HOTAMIŞLI & Ali ELEREN Excellence Models, Submitted in Partial Fulfilment of the Requirements for the Degree of Master of Applied Science. Industrial Engineering at Dalhous Univesity, Canada. MacCarthy, B. L. & Wasusri, T. (2001). Statistical Process Control for Monitoring Scheduling Performance-Addressing the Problem of Correlated Data. Journal of the Operational Research Society, 52 (7), 810-820. Martin, E. B., Morris, A. J., Kiparissides, C. (1999). Manufacturing Performance Enhancement Through Multivariate Statistical Process Control. Annual Reviews in Control, 23, 35-44. doi:10.1016/ S1367-5788(99)90055 Medori, D. & Steple, D. (2000). A Framework for Auditing and Enhancing Performance Measurement Systems. International Journal of Operations & Production Management, 20(5), 520-533. Retrived from http://www.emeraldinsight.com/journals.htm?issn=0144-3577&volume=20&issue=5 McGregor, D. (1957). An Uneasy Look at Performance Appraisal. Harvard Business Review, 35 (3), 89-94. Meyer, H. H., Kay, E. & French, J. R. P. (1965). Split Roles in Performance Appraisal. Harvard Business Review, 43(1), 123-129. Pearn, W. L. & Wu, C. W. (2006). Production Quality and Yield Assurance for Processes with Multiple Independent Characteristics. European Journal of Operational Research, 173, 637-647. Sandrock, C., Vaal, P. & Weightman, D. (2006). Performance Comparison of Controllers Acting on A Batch Pulp Digester Using Monte Carlo Modelling. Control Engineering Practice, 14, 949-958. Sarkis, J. (2003). Quantitative Models for Performance Measurement Systems-Alternate Considerations. International Journal of Production Economics, 86, 81-90. Selen, W. J. & Ashayeri, J. (2001). Manufacturing Cell Performance Improvement: A Simulation Study. Robotics and Computer Integrated Manufacturing, 17, 169-176. Silveria, G. J. C. (2005). Market Priorities, Manufacturing Configuration and Business Performance: An Emprical Analysis of the Order-Winners Framework. Journal Of Operations Management, 23, 662-675. Smith, PC. & Goddard, M. (2002). Performance Management and Operational Reseach: A Marriage Made in Heaven?. Journal of the Operational Research Society, 53, 247-255. Suwignjo, P., Bititci, U. S. & Carrie, A. S. (2000). Quantitative Models for Performance Measurement Systems. International Journal of Production Economics, 64, 231-241. 50 Journal of Economic and Social Studies �Construction of Multi Dimensional Performance Measurement Model in Business Organizations:.... Tong, L. I. & Chen, J. P. (1998). Lower Confidence Limits of Process Capability Indices For NonNormal Process Distributions. International Journal of Quality & Reliability Management, 15(8/9), 907-919. Triantis, K. & Otis, P. (2004). Dominance-Based Measurement of Productive and Environmental Performance for Manufacturing. European Journal of Operational Research, 154, 447-464. Yurdakul, M. (2002). Measuring A Manufacturing System’s Performance Using Saaty’s System with Feedback Approach. Integrated Manufacturing Systems, 13(1), 25-34. Volume 1 Number 1 January 2011 51 � Dublin Core The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/. Extent The size or duration of the resource. 1057 Title A name given to the resource Construction of Multi Dimensional Performance Measurement Model in Business Organizations: An Empirical Study Author Author YILDIZ, Feyyaz HOTAMIŞLI, Mustafa ELEREN, Ali Abstract A summary of the resource. The studies of performance measurement in firms have been conducted for a long period of time. However, the performance models and methods used in previous studies were limited. The purpose of this study is to test a performance based model that uses a modified approach in firms’ performance measurement. The new performance model used in this study is based on expectations in terms of performance measurement and evaluation of the firms with multiple dimensions. Different from the conventional gap models, the method used in this study is “Performance Measurement Method Based on Gap Percentages” developed by Eleren (2009). This method allows the researcher to use quantitative and qualitative data together. The model was tested with data collected from 42 firms engaged in business activities in marble industry in the Turkish province of Afyonkarahisar. Publisher An entity responsible for making the resource available International Burch University Date A point or period of time associated with an event in the lifecycle of the resource 2011-01 Keywords Keywords. Article PeerReviewed H Social Sciences (General),HB Economic Theory,HG Finance https://omeka.ibu.edu.ba/files/original/c581935d1c13eaaefe397f2a9dc106e1.pdf e2f418a747b5f89461e6dbff978f5c52 PDF Text Text Journal of Economic and Social Studies RFID Technology in Business Systems and Supply Chain Management Mehmet Erkan YÜKSEL Istanbul University, Department of Computer Engineering, Istanbul, Turkey, eyuksel@istanbul.edu.tr Asım Sinan YÜKSEL Indiana University, School of Informatics and Computing, Bloomington, Indiana, USA asyuksel@umail.iu.edu ABSTRACT In today’s fast-changing competition environment, companies and organizations need to renew their services and products, and change and replace their business processes with new ones continuously to benefit more from time and resources. Therefore, data capturing, gathering and management technologies are always needed by companies and organizations to support their decision-making and plans, and develop their strategies. One of the technologies that could help companies to handle data is RFID (Radio Frequency Identification). Many organizations are slow in warming up to the idea of using RFID to conduct more effective and efficient business processes, data mining applications, and cost savings. In this study, RFID technology and its system structure are proposed. The paper introduces a middleware for business models including RFID technology. Information about the advantages of RFID over today’s data gathering and Auto-ID (Automatic Identification) technologies is given. The impacts of RFID technology on business systems, especially supply chain management, are presented. Keywords: Radio Frequency Identification; Data Gathering; Item Tracking/monitoring; Process Management; Supply Chain Management Volume 1 Number 1 January 2011 53 �Mehmet Erkan YÜKSEL & Asım Sinan YÜKSEL Introduction Technological innovations and their consequences are becoming indispensable parts of our daily lives. RFID, as one of these innovations, is a system that provides easy, secure and quick data entry, storage and transmission. It is used in many places such as shops, stores, hospitals, pharmaceuticals companies, logistic services etc. where real time data should be used (Brown, 2007, Miles et al., 2008). At its core, RFID is an enabling technology that has the potential to help retailers provide the right product at the right place at the right time, thus maximizing sales and profits. It builds a basis for coding, storage and transmission systems. It improves the data management capabilities and resolves the problems caused by lack of information. It provides contactless and wireless technology to identify objects that are uniquely manufactured, shipped and sold, such as container, pallet, case and item, thus it provides the building blocks for increased visibility throughout the supply chain. It is important in improving efficiency and visibility, cutting costs, delivering better asset utilization, producing higher quality goods, reducing shrinkage and counterfeiting, and increasing sales by reducing out-of-stocks (Angeles, 2005, Brazeal, 2009). It helps in gathering data and improves the security of information about the objects. RFID has vast applications as it is relevant to any organization engaged in the production, movement or sale of goods. This technology includes retailers, distributors, logistics service providers, manufacturers and their suppliers, hospitals, pharmaceuticals companies, and the entire supply chain applications. RFID is an emerging technology consisting of data gathering, distribution, and management systems that has the ability to identify or scan information with increased speed and accuracy (Ahson & Ilyas, 2008). Although implementing RFID technology is a complicated process, the right planning and development of an RFID strategy can offer important advantages to business systems for efficient and successful supply chain management. While RFID technology has received a fair amount of attention in media recently, many are still unfamiliar with RFID and the benefits it can offer. In the face of the need for clear, extensive information about RFID and its benefits, this paper presents the opportunities offered by the technology for any organization involved in the production, management, or sale of goods. RFID Systems and System Components RFID is a wireless automatic identification (Auto-ID) and data capturing technology that gives the opportunity to monitor objects by using a tag that carries information. In RFID systems, there are different software and hardware requirements for data gathering and management. One of the most important components of RFID systems is tag. A tag can be identified as a microchip that has an electronic circuit and antenna on it. For the purpose of tracking the movement of goods, tags can be placed anywhere, such as containers, pallets, materials handling equipment, cases or even on individual products. Tags can be classified as passive (no battery), active (with battery) or semi-passive according to their power supply (Khan et al., 2009, Klaus, 2010). While active tags use an energy source that is integrated to a tag physically, passive tags obtain this energy from the readers in the communication field. Today, semi-passive tags that have some properties of both active and passive 54 Journal of Economic and Social Studies �RFID Technology in Business Systems and Supply Chain Management tags can be also used. The other component of RFID is reader which connects the tags to external world. Although readers can be classified as portable and mobile (Klaus, 2010), all of them consist of same components. In every reader, there are some parts that read tags, gather data and handle communication. While the reader antenna receives/sends the radio waves, it builds the signal and decrypts the signal which is sent from tags. There are eight main components for building RFID systems in a supply chain management: 4. Controller 7. Software System (ERP/MRP) 2. Antenna 5. Interrogator 8. Communication Infrastructre 3. Reader 6. Sensors 9. Annuciators/Actuators 1. RFID Tag RFID Tag An RFID tag consists of a microchip where the information about the object is stored, an antenna connected to the chip, on-board electronics, and a protecting film layer that covers these components. It is used as an electronic data carrier, and different information can be written and read in its environment. The microchip in the RFID tag can store information from 64 bit to 8 MB (Klaus, 2010), which means that the tag can carry some important information such as shipping history, order number, customer information, company/staff information and serial number (Ahson & Ilyas, 2008). There are several kinds of tags in different forms and sizes. A common way of categorizing tags is by their power source. This is also one of the main determining factors for the cost and longevity of a tag. There are three types of RFID tags: passive, active, and semipassive. A passive tag does not contain a battery; it obtains all of its energy from the reader by using different transmission methods. It uses the signal received from the reader to power the IC, and changes the signal level to transmit information back to the reader. Passive tags are the most common ones in cost-sensitive applications because they have no battery, no transmitter, and they are also very cheap. An active tag is a full-featured radio device with its own transmitting capability independent of the reader (Shepard, 2005, Roberts, 2006). It uses an on-board battery to power on-board electronics, microprocessor, memory, and external sensors for communication. Tags that use battery power for some functions but still allow the reader to power communications have been termed “active” as well. They are not only capable of supplying power for themselves but also they are able to initiate communications with other tags of their own kind without the aid of a reader. These tags are called two-way tags, battery-assisted passive tags or semi-passive tags (Shepard, 2005, Xiao et al., 2007). They use a battery to power the on-board electronics and microchip, and the passiveness of semi-passive tags depends on required signal levels between the tag and the reader. Volume 1 Number 1 January 2011 55 �Mehmet Erkan YÜKSEL & Asım Sinan YÜKSEL Table 5. Marketing Management Factor and its variables Although having an onboard battery makes RFID chips more expensive, semi-passive and active tags have several advantages over passive tags. In semi-passive systems, the reading range may be longer because the passive communications can use all of the power provided by the reader for data transmission rather than sharing some of the power with the chip. Active tags have an extremely long reading range and perform some functions in the absence of a reader (e.g., using battery power for environmental sensors). This capability can be very useful for tags to identify items such as perishable goods. These varieties of RFID tags can dynamically store data. Active RFID tags have large read/ write data storage, almost 128 kilobytes, and sophisticated data search/access capabilities. In a passive RFID, the data storage is less than 128 bytes with no search capabilities or data manipulation features (Klaus, 2010, Glover & Bhatt, 2006). RFID Reader An RFID reader is a specialized radio transmitter and receiver. It generates signals at the carrier frequency and modulates these carrier signals to convey information to the tags (Klaus, 2010). It must selectively receive and amplify responses from the tags, and convert the signal from the carrier frequency down to much lower frequencies. It is designed for fast and easy system integration without losing performance, functionality and security. Fig. 2 shows the components of an RFID reader. An RFID reader consists of a real time processor, operating system, virtual portable memory, and a transmitter/receiver unit in one small self-contained module that is easily installed in the ceiling or in any other convenient location. The reader is usually classified into two types (Klaus, 2010, Shepard, 2005). The portable reader is set to a definite place. It is the reader type that RF tags go through and by which they make communication. The mobile RFID reader includes a wireless interface, precisely Bluetooth (IEEE 802.15.1), ZigBee (IEEE 802.15.4) or Wi-Fi (IEEE 802.11b/g/n). This device uses 56 Journal of Economic and Social Studies �RFID Technology in Business Systems and Supply Chain Management short or long range radio links. Therefore, it can identify, read/write, remotely control and monitor RFID tags over wireless communication. It contains some software tools for communication with other mobile RFID readers, PDAs, laptops, etc. The mobile RFID reader facilitates the identification of the tags that are in dangerous fields where the reading process is difficult (Roberts, 2006, Xiao et al., 2007, Glover & Bhatt, 2006). Figure 2: The Components of a Reader RFID Controller An RFID controller is a machine such as computer, server, workstation etc. on which database or application softwares work. It also can be a network system which includes computers, servers and workstations connected to each other. The RFID Controller is the brain of an RFID system. It controls RFID middlewares, applications and database sytems (Brown, 2007, Shepard, 2005, Fosso Wamba et al., 2007). It is also used for connecting multi-queriers in a network and processing information centrally. The controller uses information that is collected by readers. It has some properties such as monitoring RFID sytem, rerouting data about items if necessary, remote controlling and managing devices, validating and authorizing identities, creating and managing accounts, stock analysis for products, coordination with ERP/MRP systems, informing suppliers when new product stock is required, etc. (Angeles, 2005, Brazeal, 2009, Glover & Bhatt, 2006). RFID Antenna It is the hardware that provides communication among readers and tags. In many situations, the use of an antenna is very important because tag reading ranges are very small. Although the antenna has a very simple structure according to its concept, it must receive the best signals in low power and adapt to special conditions. Antenna must be designed in different sizes, shapes and frequency intervals according to the properties and distances of the environment where the application will be implemented in (Klaus, 2010, Hossain & Karmakar, 2006). Antennas have two kinds of broadcasting: planarly or circularly. An antenna that broadcasts planarly Volume 1 Number 1 January 2011 57 �Mehmet Erkan YÜKSEL & Asım Sinan YÜKSEL concentrates on a unique axial for maximum income in the longest reading distance. An antenna that broadcasts circularly distributes the UHF energy generated by the reader to a longer distance equally. Therefore, with circular diffusion, it is possible that the antenna reads all surrounding tags. An antenna can be designed as to several factors, such as the following: - Reading distance - Reader/controller - Particular product types - Antenna polarization - Known orientation - Environmental changes - Arbitrary orientation - Speed of the tagged objects - Specific operating conditions RFID Interrogator An RFID interrogator is essentially a small computer. It has three basic parts: an antenna, an RF electronic module that is responsible for reading RFID tags, and a microcontroller module that is responsible for communication with controllers or readers. The interrogator acts as a bridge between the tag and the controller (or reader). It has just a few critical functions (Brown, 2007, Angeles, 2005, Ahson & Ilyas, 2008):  Reading data contents of an RFID tag.  Writing data to the tag, if required.  Relaying data from/to the controller.  Powering the tag, if required (for passive systems).  Implementing anti-collision measures to ensure simultaneous RW communication with many tags.  Authenticating tags to prevent fraud or unauthorized access to the RFID system.  Data encryption to protect the integrity of RFID data. Operating Frequencies in RFID Systems Operating frequency is the electromagnetic frequency which the tag uses to communicate or obtain power. The electromagnetic spectrum in the range in which RFID typically operates is usually broken up into low frequency (LF), high frequency (HF), ultra-high frequency (UHF), and microwave. Because of the fact that RFID systems broadcast electromagnetic waves, they are regulated as radio devices (Glover & Bhatt, 2006). 58 Journal of Economic and Social Studies �RFID Technology in Business Systems and Supply Chain Management Figure 3: Radio Frequency Spectrum Low Frequency (LF) Frequencies between 30 KHz and 300 KHz are considered as low. RFID systems commonly operate at the frequency range from 125 KHz to 134 KHz. LF RFID systems generally use passive tags, have low data-transfer rates from tag to reader, and are especially good if the operating environment contains metals, liquids, dirt, snow, or mud (a very important characteristic of LF systems). In LF RFID systems, active LF tags are also available from vendors (Lahiri, 2005). High Frequency (HF) High frequency RFID systems operate between 3 MHz and 30 MHz ranges. 13.56 MHz is the typical operating frequency used for HF RFID systems. A HF RFID system that uses passive tags, has a slow data-transfer rate from tag to reader, and offers fair performance in the presence of metals and liquids. The other frequency range is very high frequency (VHF) and lies between 30 and 300 MHz. Although there are applications for VHF systems such as FM radio/TV broadcasts, land mobile stations, marine and air traffic control communications, air navigation systems, etc., none of the current RFID systems operate in this range. Ultra High Frequency (UHF) UHF RFID systems operate between 300 MHz to 1 GHz ranges, and they can use both active and passive tags. These systems have fast data-transfer rates between tags and readers, but perform poorly in the presence of metals and liquids. Microwave Microwave RFID sytems that operate from 1 GHz upto 5.8 GHz. 2.4 GHz frequency are called Industry, Scientific, and Medical (ISM) band and accepted worldwide (Lahiri, 2005). Microwave RFID systems can be used for active, semi-active and passive tags. They have the fastest data-transfer rates between tags and readers. Because of the fact that antenna length is inversely proportional to the frequency, the antenna of a passive tag operating in the microwave range has the smallest length Volume 1 Number 1 January 2011 59 �Mehmet Erkan YÜKSEL & Asım Sinan YÜKSEL (which results in a small tag size because the tag microchip can also be made very small). RFID Applications RFID systems can be divided into two major groups, as mobile and immobile applications. Immobile RFID systems include RFID readers, antennas (usually 2 or 4 antennas for each reader), hosts, servers, middleware and external units such as light and sensors. These systems are also called RFID gates. In these systems, readers serve as gates, receive information from tagged objects and send these information to servers or controllers. Mobile systems use wireless communication to gather data and monitor objects. They are similar to fixed systems due to RFID system structure. They provide advantages such as data gathering and managing, reading/writing ranges and communication technologies. Reading/writing data from/to RFID tags is done by radio frequencies. Passive tags, which are widely used, are activated by the energy that is generated by RFID readers, and send their information to readers. RFID readers receive information and transfer this information to controllers, servers, database systems or RRP/ERP/MRP systems in supply chains. Fig. 4 shows an RFID system model in a supply chain management, and provides information about how RFID systems work. Figure 4: An RFID System Model 60 Journal of Economic and Social Studies �RFID Technology in Business Systems and Supply Chain Management RFID Data Model Used in a Supply Chain RFID data can be classified into two categories: the event data and the master data. The event data keeps real-time (or dynamic) information which is about RFID tagged objects such as containers, pallets, materials handling equipment, cases, automobiles, textiles, animals, and etc. The master data provides conditional information and verification about the event data. Fig. 5 shows an example of an RFID data model that contains 96 bit EPC code in a supply chain management, and provides information about how RFID data is structured. Figure 5: An RFID Data Model Event Data Event data is related to a definite time, and it provides the communication about RFID tagged objects during supply chain processes. It is created whenever some sort of transaction occurs. It is captured in distributed data repositories, and only the relevant event data has to be sent to the monitoring for further processing (Miles et al., 2008, Angeles, 2005, Brazeal, 2009). The processing and matching of automatically generated monitoring instructions with the event data gathered from distributed data repositories has to be performed by an appropriate event processing engine (Veronneau & Roy, 2009, Kwak et al., 2010). Event data creates information which is about investigating the existence of items somewhere at some time. It stores the identity, location and time information. Event data can be illustrated like this: “EPC X is observed at location L at 4:15 p.m. 28 June 2010”. Event data is currently used for tracking and tracing applications to monitor items associated with transportation processes and transported goods (Ferrer et al., 2010). The combination of new technologies provides the potential to use RFID based on event data for the automatic and near real time monitoring of processes in supply chain networks to detect anomalies according to specified objectives. By using RFID widely, applications may need more information and more sensor observations (Werner & Schill, 2009). Master Data Master data, also called reference data, describes an item and its general properties. It includes useful data about customers, products, employees, materials, suppliers, manufacturers etc. in a supply chain. It contains information such as source verification, product definition referenced by EPC (Electronic Product Code), manufacturer information, details about the object which event data is caught from, and storage information. It can define transactional processes and operations (Glover & Bhatt, 2006, Volume 1 Number 1 January 2011 61 �Mehmet Erkan YÜKSEL & Asım Sinan YÜKSEL Kwak et al., 2010, Tajima, 2007). Master data is a key information for quality-assurance, persisting demands, business operations and data-mining applications. It provides processes for collecting, aggregating, consolidating, matching, and distributing such data throughout a company to ensure control and consistency in the ongoing maintenance and application of this information (Chuang & Shaw, 2005). Master data is used in data management systems to define characteristics of an item that are used within other data centric processes. It is stored in different data systems across a company, and it may or may not be referenced centrally. Therefore, the possibility exists for duplicate and/or inaccurate master data. Usually, master data can’t grow as fast as event data. Data Size The effect of RFID on data gathering and management systems in a supply chain is related to which data will be collected by using RFID, how often it will be collected, and what will be done with RFID data. Due to the necessity of RFID system and network infrastructure, RFID data volume can overload the storage fields and supply chain network. Therefore, data size can be changed as a function that is dependent on the number of processes executed by the RFID system. RFID Middleware for Business Operations and Supply Chain Models RFID middleware is a software that bridges RFID system and enterprise IT applications. It helps data gathering and management for any RFID deployment in a supply chain. It consists of a set of services that allow multiple processes running on one or more RFID system to interact. RFID middleware assists with the filtering, aggregation, and routing of RFID data. It has built-in business rules that monitor the data stream and direct data to appropriate enterprise systems. It is used to manage data flow between the RFID networks and the IT systems within an organization (Sarac et al., 2010, Chuang & Shaw, 2005, Asif & Mandviwalla, 2005). Fig. 6 shows a sofware architecture in a supply chain management. RFID middleware can be integrated to the ERP/MRP system of a company. This integration helps RFID services to write the correct data to the desired places in time. The integration provides varying degrees of management and monitoring capabilities, service-oriented architecture integration capabilities, and built-in adapters to various ERP packages (Glover & Bhatt, 2006). The company which hosts the RFID service provides this integration by working with its trading partners which use current systems (Saygin et al., 2007, Gaukler & Seifert, 2007). RFID middleware provides connectivity with RFID devices while encapsulating the applications from the device interface and interconnections. It lowers the volume of information that applications need to process by filtering and grouping raw RFID observations captured by readers and sensors. It provides an application-level interface for querying RFID observations and managing RFID devices such as readers, controllers, interrogators and servers (Chuang & Shaw, 2007, Chen et al. 2008). 62 Journal of Economic and Social Studies �RFID Technology in Business Systems and Supply Chain Management Figure 6: An RFID Software Architecture For A Business Model (Glover & Bhatt, 2006). RFID middleware has four main functions for data management in business applications: 1. Data Gathering: Middleware is responsible for the extraction, aggregation, smoothing, and filtering of data from multiple RFID readers throughout an RFID network. It serves as a buffer between the volumes of raw data that are collected by RFID readers and the relatively small amount of data that is required by enterprise IT systems in the decision-making process. Without this middleware buffer, enterprise IT systems could quickly become overwhelmed by the flow of data (Hunt et al., 2007). 2. Data Routing and Management: Middleware facilitates the integration of RFID networks with enterprise systems. It determines which data must go where. It does this by routing data to appropriate enterprise systems within an organization (Saygin, 2007). 3. Process Management: RFID middleware is an application of knowledge, skills, tools, techniques and systems to define, visualize, measure, control, report and improve supply chain processes with the goal of meeting customer requirements profitably. It is the ensemble Volume 1 Number 1 January 2011 63 �Mehmet Erkan YÜKSEL & Asım Sinan YÜKSEL of activities of planning and monitoring the performance of a business process (Gaukler & Seifert, 2007, Bagchi et al., 2007). Middleware can be used to trigger events based on business rules. For example, an order is created on a company’s website, and a pallet lies at a dock door in a distant warehouse as though waiting for its marching orders. The enterprise IT system that is responsible for initiating this shipment would pass the purchase order to the middleware system. Middleware locates the specific dock door where the pallet is stored, and writes the delivery information on the pallet’s tag. Other events and processes such as unauthorized shipment and unexpected inventory, low stock, and stock cut can be managed by middleware. 4. Device Management: Middleware must contain technologies, protocols and standards used to allow the remote management of RFID devices, and involve updates of firmwares or other middlewares over the air. For example, the employer or the end-user can use middleware via a web portal to update the firmware, install middleware applications and fix bugs, as wireless and contactless. Thus, large numbers of RFID devices can be managed with middleware, and the end-user is freed from the technical service requirement to refresh or update the RFID system. For supply chain management and business applications, RFID device management means better control, update, safety, and management as well as increased efficiency, decreased possibility for device downtime. Middleware is used to monitor and coordinate devices such as readers, hosts, controllers, servers. Large organizations might have hundreds or thousands of different types and brands of readers to spread across their networks. Networking and monitoring RFID sytem devices, and keeping track of these devices` health and status, would be a major job and most efficiently done at the middleware level. Remote management of an RFID network could also be made possible through middleware (Sarma, 2004, Prabakar et al., 2006, Minli et al. 2008). Discussion The main feature of RFID technology is its ability to identify, locate, track, and monitor objects without a clear line of sight between the tag and the reader. Addressing all of the functional capabilities of RFID systems ultimately defines the RFID applications to be developed in a variety of industry, commerce, and service where data need to be gathered. The effectiveness of an RFID application in addressing desired functionality depends on several important factors: • Contactless: An RFID tag can be read without any physical contact between the tag and the reader. • Writable data: The data of a read-write (RW) RFID tag can be rewritten several times. • Absence of line of sight: A line of sight is generally not required for an RFID reader to read a tag. • Variety of reading ranges: An RFID tag can have a reading range from as small as a few inches to as large as more than 100 feet. • Wide data-capacity range: A tag can store from a few bytes of data to virtually any amount of data. 64 Journal of Economic and Social Studies �RFID Technology in Business Systems and Supply Chain Management • Support for multiple tag readings: It is possible to use an RFID reader to automatically read several RFID tags in its reading zone within a short period of time. • Durable: RFID tags and readers can easily operate under difficult conditions. • Perform smart tasks: In addition to the tasks of carrying and transmitting of data, an RFID system can be designed to perform some other tasks (e.g., acculturation to environmental conditions, operating at high or low temperature and pressure). • Extreme reading accuracy: Thanks to extreme reading accuracy advantage, RFID is an accurate and secure technology for data gathering and management. The advantages listed above are generic for any type of RFID sytems. Some additional factors are needed to be considered for applications such as data gathering, item monitoring, automatic identification, and access control mechanisms. These may include privacy and security concerns, data mining, and the integration of the RFID with other technologies such as biometric systems, Global Positioning Systems (GPS) and wireless communication technologies. In the near-feature, commercial applications of RFID technology in supply chain management will continue to develop and grow. Therefore, RFID industry must focus on applications that increase the volume of usage, lower the costs and develop effective business models. Companies gain competitive advantage over other companies by offering consumers greater values. These values can be provided by means of lower prices, investment or better services. New business strategies should be developed and new technologies should be adopted in order to compete with rival companies in industry. For the companies who will use RFID systems, a question arises: “Can RFID deliver a competitive advantage?” Firms who are the earlier adopters of RFID technology will gain a competitive advantage through business innovation and differentiation. Companies with supply chains have already begun to explore how to apply RFID technology with the goal of improving supply chain management and collaboration. Leader retailers around the world and national organizations have begun to suggest and insist that manufacturers and suppliers should attach RFID tags to products before shipment. A good example of a company who is an earlier adopter of RFID with the help of IBM is American Power Conversion. According to Rich Morrissey, APC’s director of eBusiness, APC has a better position with customers and RFID helped the company to achieve a competitive advantage through business innovation, differentiation and an established technological leadership position. In addition to these, the company protected its revenue streams, won customer loyalty, speeded up its decisionmaking and gained greater supply chain visibility. Another example is Wal-Mart, which is one of the largest American public organizations and runs a chain of large discount department stores and a chain of membership required warehouse stores. The company extended its existing edge and saw RFID as a way to reduce the cost of handling goods. Therefore, a firm who adopts the RFID technology saves billions of dollars and reduces the cost of its supply chain operations and can translate these savings into increased profit margins at the Volume 1 Number 1 January 2011 65 �Mehmet Erkan YÜKSEL & Asım Sinan YÜKSEL point of sale. Additionally, a firm can also cut prices that helps to gain low-cost advantage against its competitors. During supply chain processes, several data such as “which products use which parts” or “who assembles the product in what time” can be stored in database management systems. Therefore, products in which related parts are used can be automatically determined and applied to necessary operations. RFID systems give an opportunity to create a database with data on customers’ needs. By the use of RFID, information in database systems is updated and correct information that is necessary for reporting and analysis is obtained on time. RFID has many potential activities such as billing and delivering products, physical stock and identity account tracking. Information/data is easily accessed and updated dynamically in real-time, stocks are tracked and controlled, and storehouse and selling control can be done. Products that are taken from stock or that have remained in stock can be monitored with their costs and efficient stock management can be done. By using RFID data (e.g, EPC), correct information is obtained from the production line and used in different stages of the supply chain without human intervention. The products are directed to a definite route automatically and defective products are also prevented in the production line. Some critical processes during business operations such as stock management, item tracking/monitoring, transportation, delivery, device management, and software update are planned automatically. In this way, lost time and manpower can be decreased. There are several advantages of RFID applications in supply chain management. RFID technology provides collaborative business commerce solutions, enables more efficient and effective buying, selling, and cash management. Companies can control costs, increase sales, minimize risks, and enhance cash flow by using RFID systems. As a result of the real time item monitoring, RFID provides effective logistic management, and effective purchase and supplier/procurement management. In addition to these, RFID applications decrease repeating jobs and faults caused by using automation instead of manpower, decrease manpower costs and prevent problems by obtaining detailed information. An RFID tag has a unique identification code and data is protected with cryptographic algorithms. Advanced encryption methods prevent unauthorized access to the information in the RFID chip. The tag can be locked and becomes unusable if necessary. The security of RFID improves the delivery and control of goods, decreases theft and faults, increases anti-forgery, eliminates wrong data entry and prevents complications between similar products or the products that have similar codes. In this way, RFID data processing accuracy and sensitivity enable the proper verification of object data (Ahson & Ilyas, 2008, Hunt et al., 2007). RFID data contains a unique code (e.g., EPC) that provides the unique identification of each item. RFID contributes to supply chains and business operations in different ways through its advanced unique identification and real-time communication properties. Through the unique code of each tag and the easiness of scanning, RFID improves the accuracy, the speed of processes, the traceability and the visibility of products throughout supply chains. It also reduces handling and distribution 66 Journal of Economic and Social Studies �RFID Technology in Business Systems and Supply Chain Management costs and increases sales by reducing stock-outs. RFID ameliorates the efficiency of current supply chains, but it also supports the reorganization of supply chains to drastically enhance their overall performances. RFID is also used to minimize human faults and increase process speed in a scope like selling, storing, production, etc. In RFID systems, processes are executed faster than manual systems because information is transferred with electronic methods. With the speed increasing in the data entrance, work efficiency also increases and employees are shifted to more suitable locations. Another advantage of using RFID, is its being economic. With correct data entrance and increased speed of data entrance, the number of employees is reduced, leading the system to become more economic. An RFID application reduces the stationer and stores costs by eliminating data entry forms. Because it is not necessary to use complicated equipment – stationer for RFID tag, data recording/storing process becomes simpler and has less cost than other automatic identification (autoid) technologies. RFID is more durable, and can be applied more simply than other autoid when the factors like damage, frazzling, being torn during its usage are taken into account. RFID includes some technological features such as security, identification and authorization for commercial applications. Due to identification and authorization functions, RFID increases visibility, and develops effectiveness of communication among objects by matching entities automatically. For instance, RFID identification and authorization features in medical applications enable the identification, positioning and tracing of patients and related medical equipment consistently at the right time. In this way, double-checking in medical treatment services, workload and resources in advanced processes are decreased. RFID, integrated into wireless network systems, brings mobility to any tagged object by increasing the capacity of data communication. In this way, efficient use and management of goods is improved. For instance, the use of RFID with wireless network system enables the enumeration of medical equipment in a hospital from a fixed location. This dramatically decreases the search time. Moreover, by the use of RFID tags for patients, medical equipment, and patient registers in a hospital, data about entities can be traced fast. RFID integration enhances tracing of goods use, and provides current and the best sourcing use by decreasing loss or misplacement of goods. In this way, it improves the efficiency of material inventory. Due to RFID and wireless sensor network environment built with EPC global, aims and limits about location, storage space and time can be changed; RFID enables services existing in innovative ways or new services to be created. Standardized RFID network software architecture, which uses RFID system to enable the constant, temporaneous, direct identification, positioning and tracing of objects such as pallet, container, goods, medicines, and patients is used to transform data collected by RFID into information. This information is processed as useful information that can be used with back-end system in a company’s decision-making period (Smith, 2005). For instance, through the effective use of RFID and EPCglobal network in medical institutions (especially data processes with Volume 1 Number 1 January 2011 67 �Mehmet Erkan YÜKSEL & Asım Sinan YÜKSEL intensive workload such as search, data entry, comparison and verification), not only is medical error dramatically reduced, but treatment quality of the medical institution is also improved effectively, which enhances operational performance and capacity of hospital. Conclusion RFID is the general name of the systems that use radio waves in automatic object identification. It transmits the identification information of objects as a numerical serial number dynamically. These wireless systems give the opportunity to read without contact and being visibility. This property provides simplicity in difficult environments, as compared to conventional technologies such as barcode. Today, RFID technology with its different applications provides many advantages in the industry. Invention of new technologies decreases the operational costs of firms and companies and increases the efficiency and profitability. By using RFID technology, the changes in the working process can be analyzed and planned. The RFID system can set with the most suitable tag design and be started to be managed. Large scale applications with integrated databases such as inventory tracking, production band automation, stock management, staff and data tracking can be developed. RFID is a technology that is not limited to only product identification and tracking, but also it has large application fields for supply chain management and applications. It is very suitable for the companies and firms who need dynamic systems to control their entities, data and information regularly. The people who want to invest in RFID must investigate the RFID gains properly. It can be said that discovery duration about RFID is current, and to apply a system based RFID is both a science and an art. RFID improves the effectiveness of supply chain and provides the opportunity to collect various data about customer behaviours. It presents enterprise innovations that result in high performance on behalf of company value and applications that increase business operations. It has the structural features that activate business processes planning and management. Thanks to tracking of any object tagged with RFID and the intelligent management of data about the object, value-added operations of companies improve. RFID significantly alters companies’ capacity of obtaining real-time data about the locations and characteristics of tagged objects. When various processes related to business operations are used with RFID, companies can observe the location, history and changing situations of tagged objects more easily. When RFID is used with a company`s general communication infrastructure, it provides wide-ranging location and knowledge of goods to the company. RFID systems can be assessed both strategically and operationally. Strategical systems aim at integration of RFID technology and company processes, development of better business models by companies, customer satisfaction and development of new business opportunities. Besides, it leads an effective supply-chain system in order to coordinate companies’ members from overseas. Operational systems focus on effectiveness and flexibility in process redesign. RFID has an effective 68 Journal of Economic and Social Studies �RFID Technology in Business Systems and Supply Chain Management and competitive role in all stages of supply-chain management such as business value, organizational business strategies, cost, quality, service and speed, etc. Thanks to intelligent tracking of objects and automation of business processes, RFID decreases the costs of data collection, increases the effectiveness in business processes, and creates competitiveness among companies by interactive working with BPR, MRP, ERP, and SCM applications. Adoption of RFID by enterprises results in great changes for both business processes and company employees. RFID aims at activated restructuring studies, optimization of business processes, and providing effective business integration. In the near future, the adoption of sensor-based Radio Frequency Identification (RFID) technology will allow the creation of the real-time, sensor-connected manufacturing plant. By adding RFID tags to every product, tool, resource and item of materials handling equipment, manufacturers will be able to get better demand signals from customers and the market. References Brown, D. E., (2007). RFID Implementation, McGraw Hill Companies, NY, USA, ISBN: 978-007-226324-4. Miles, S. B., Sarma, S. E., and Williams, J. R., (2008). RFID Technology and Applications, Cambridge University Press, NY, USA, ISBN: 978-0-521-88093-0. Angeles, R., (2005). RFID Technologies: Supply-Chain Applications and Implementation Issues, Information Systems Management, Vol. 22, pp. 51-65. Brazeal, M., (2009). RFID: Improving the Customer Experience, Paramount Market Publishing, Inc., NY, USA, ISBN: 978-0-9801745-3-3. Ahson, S. A., & Ilyas, M., (2008). RFID Handbook, Applications, Technology, Security, and Privacy, CRC Press, FL, USA, ISBN: 978-1-4200-5499-6. Khan, M. A., Sharma M., and Prabhu R, B., (2009). A Survey of RFID Tags, International Journal of Recent Trends in Engineering, Vol. 1, No. 4, pp. 68-71. Klaus, F., (2010). RFID Handbook: Fundamentals and Applications in Contactless Smart Cards, Radio Frequency Identification and Near-Field Communication, Third Edition, John Wiley & Sons Ltd., West Sussex, UK, ISBN:978-0-470-69506-7. Shepard, S., (2005). RFID: Radio Frequency Identification, McGraw-Hill Companies, NY, USA, ISBN:0-07-144299-5. Roberts, C.M., (2006). Radio frequency identification (RFID), computers&security, Vol. 25, pp. 18–26. Volume 1 Number 1 January 2011 69 �Mehmet Erkan YÜKSEL & Asım Sinan YÜKSEL Xiao, Y., Yu, S., Wu, K., Ni, Q., Janecek, C., and Nordstad, J., (2007), Radio frequency identification: technologies, applications, and research issues, Wireless Communications And Mobile Computing, Vol. 7, No. 4, pp. 457–472. Glover, B., & Bhatt, H., (2006). RFID Essentials, O’Reilly Publishing, Sebastopol, CA, USA, ISBN: 978-0-596-00944-1. Fosso Wamba, S., Lefebvre, L. A., and Lefebvre, E., (2007). Integrating RFID Technology and EPC Network into a B2B Retail Supply Chain: a Step toward Intelligent Business Processes. Journal of Technology Management & Innovation, Vol. 2, No. 2, pp. 114-124. Hossain, S. S., & Karmakar, N., (2006). An Overview On RFID Frequency Regulations And Antennas, 4th International Conference on Electrical and Computer Engineering (ICECE), Dhaka, Bangladesh, pp. 424-427. Lahiri, S., (2005). RFID Sourcebook, Pearson Education, IBM Press, Indiana, USA, ISBN: 978-013-185137-5. Veronneau, S., & Roy, J., (2009). RFID benefits, costs, and possibilities: The economical analysis of RFID deployment in a cruise corporation global service supply chain, International Journal of Production Economics, Elseiver, Vol. 122, No. 2, pp. 692-702. Kwak, C., Cho, Y., Ko, J. M., and Kim, C. O., (2011). Adaptive Product Tracking in RFID-Enabled Large-Scale Supply Chain, Expert Systems with Applications, Elseiver, Vol. 38, No. 3. Ferrer, G., Dew, N., and Apte, U., (2010). When is RFID right for your service?, International Journal of Production Economics, Elsevier, Vol. 124, No. 2, pp. 414-425. Werner, K., & Schill, A., (2009), Automatic Monitoring of Logistics Processes Using Distributed RFID based Event Data, International Workshop on RFID Technology (IWRT), Milan, Italy, pp. 101-108. Tajima, M., (2007). Strategic value of RFID in supply chain management, Journal of Purchasing and Supply Management, Elseiver, Vol. 13, No. 4, pp. 261-273. Chuang, M.-L., & Shaw, W. H., (2005). How RFID Will Impact Supply Chain Networks, IEEE International Engineering Management Conference (IEMC 2005), Newfoundland, Canada, pp. 231235. Sarac, A., Absi, N., and Dauzere-Peres, S., (2010). A literature review on the impact of RFID technologies on supply chain management, International Journal of Production Economics, Elsevier, Vol. 128, No. 1, pp. 77- 95. 70 Journal of Economic and Social Studies �RFID Technology in Business Systems and Supply Chain Management Asif, Z., & Mandviwalla, M., (2005). Integrating the supply chain with RFID: a technical and business analysis, Communications of the Association for Information Systems, Vol. 15, No. 24, pp. 393–427. Saygin, C., Sarangapani, J., and Grasman, S. E., (2007). A Systems Approach to Viable RFID Implementation in the Supply Chain, Trends in Supply Chain Design and Management Technologies and Methodologies, Springer series in advanced manufacturing, Part 1, pp. 3-27. Gaukler, G. M., & Seifert, R. W., (2007). Applications of RFID in Supply Chains, Trends in Supply Chain Design and Management Technologies and Methodologies, Springer series in advanced manufacturing, Part 1, pp. 29-48. Bagchi, U., Guiffrida, A., O’Neill, L., and Hayya, J., (2007). The Effect of RFID On Inventory Management and Control, Trends in Supply Chain Design and Management Technologies and Methodologies, Springer series in advanced manufacturing, Part 1, pp. 71-92. Chuang, M.-L., & Shaw, W. H., (2007). RFID: Integration Stages in Supply Chain Management, IEEE Engineering Management Review, Vol. 35, No. 2, pp. 80-87. Chen, N.-K., Chen, J.-L., Chang, T.-H., and Lu, H.-F., (2008). Reliable ALE middleware for RFID network applications, International Journal Of Network Management, John Wiley & Sons, Vol. 19, No. 3, pp. 203-216 Hunt, V. D., Puglia, A., and Puglia, M., (2007). RFID: A Guide to Radio Frequency Identification, John Wiley & Sons Inc., New Jersey, USA, ISBN: 978-0-470-10764-5. Saygin, C., (2007). Adaptive inventory management using RFID data. The International Journal of Advanced Manufacturing Technology, Springer, Vol. 32, No. 9-10, pp. 1045-1051. Sarma, S., (2004). Integrating RFID, Queue, ACMDL, NY, USA, Vol. 2, No. 7, pp. 50-57. Prabakar, V., Kumar, B.V., and Subrahmanya, S.V., (2006). Management of RFID-centric business networks using Web Services, Advanced International Conference on Telecommunications and International Conference on Internet and Web Applications and Services (AICT-ICIW’06), Guadeloupe, French Caribbean, pp. 133. Minli, W., Gang, W., and Dajun, H. (2008). Device Management in RFID Public Service Platform, 3rd International Conference on Convergence and Hybrid Information Technology, Busan, Korea, Vol. 01, pp. 1133-1136. Smith, A. D. (2005). Exploring Radio Frequency Identification Technology and Its Impact on Business Systems, Information Management & Computer Security, Vol. 13, No. 1, pp. 16-28. Volume 1 Number 1 January 2011 71 � Dublin Core The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/. Extent The size or duration of the resource. 1058 Title A name given to the resource RFID Technology in Business Systems and Supply Chain Management Author Author YÜKSEL, Mehmet Erkan YÜKSEL, Asım Sinan Abstract A summary of the resource. In today’s fast-changing competition environment, companies and organizations need to renew their services and products, and change and replace their business processes with new ones continuously to benefit more from time and resources. Therefore, data capturing, gathering and management technologies are always needed by companies and organizations to support their decision-making and plans, and develop their strategies. One of the technologies that could help companies to handle data is RFID (Radio Frequency Identification). Many organizations are slow in warming up to the idea of using RFID to conduct more effective and efficient business processes, data mining applications, and cost savings. In this study, RFID technology and its system structure are proposed. The paper introduces a middleware for business models including RFID technology. Information about the advantages of RFID over today’s data gathering and Auto-ID (Automatic Identification) technologies is given. The impacts of RFID technology on business systems, especially supply chain management, are presented. Publisher An entity responsible for making the resource available International Burch University Date A point or period of time associated with an event in the lifecycle of the resource 2011-01 Keywords Keywords. Article PeerReviewed H Social Sciences (General),HB Economic Theory,HG Finance