Nonlinear Dynamics of Financial Development on Trade Balance

Transcription

1 International Journal of Economics and Finance; Vol. 8, No. 5; 2016 ISSN X EISSN Published by Canadian Center of Science and Education Nonlinear Dynamics of Financial Develoment on Trade Balance Muhammet Belen 1 1 Faculty of Economics and Business Administration, Karabuk University, Karabuk, Turkey Corresondence: Muhammet Belen, Faculty of Economics and Business Administration, Karabuk University, Karabuk, Turkey. Tel: mbelen@karabuk.edu.tr Received: March 4, 2016 Acceted: Aril 18, 2016 Online Published: Aril 25, 2016 doi: /ijef.v8n5286 URL: htt://dx.doi.org/ /ijef.v8n5286 Abstract The goal of this study is to investigate the nonlinear dynamics of financial develoment on trade balance in Turkey. For the eriod of 1987Q12015Q2, by emloying nonlinear autoregressive distributed lag (NARDL) aroach to cointegration, it is found that there is cointegration between these variables. This finding suggests that both increase and decrease in financial develoment have different imacts on trade balance. The relationshi between financial develoment and trade balance has also negative sign. This finding asserts that an imrovement in financial develoment will cause a significant deterioration in trade balance while oosite is also true but with less imact. The results of the study suort the emirical literature which has identified the relationshi between financial develoment and growth as being highly correlated. But it does not suort using financial develoment as a olicy tool to reduce trade balance. Keywords: financial develoment, trade balance, real exchange rate, asymmetric cointegration, nonlinear ARDL 1. Introduction Trade balance is the difference between a country s exorts and imorts. In a country, if exorts are greater than imorts then trade balance will be in surlus and if imorts are greater than exorts then trade balance will be in deficit. Trade balance is very imortant for an economy and it is argued that trade deficit is not good for the long economic growth (Muzammil & Ahad, 2015). In this study, the cointegration relationshi between financial develoment and trade balance has been investigated in Turkey, by emloying nonlinear autoregressive distributed lag (NARDL) aroach to cointegration imroved by Shin, Yu, and GreenwoodNimmo (2014) as an extension of the linear cointegration techniue roosed by Pesaran, Shin, and Smith (2001). According to Shin et al. (2014) the nonlinearity of many macroeconomic variables and rocesses has long been recognized and nonlinearity is endemic within the social sciences and that asymmetry is fundamental to the human condition (Keynes, 1936,. 314; Kahneman & Tversky, 1979; Shiller, 1993, 2005). Turkey in the 1960s and 1970s was the develoing country following an imort substitution strategy. This strategy was successful in manufacturing outut but disaointing in exorts. Between1970 and 1980, Turkey initiated a first wave of economic reforms to develo a more outwardoriented and marketbased economy. In its second wave reforms in the mid1990s, Turkey continued its trade liberalization efforts. And the reforms of the 2000s were on macroeconomic olicy frameworks and structural reforms. Turkey s reform efforts have resulted in a considerably more oen and globally integrated economy. The comrehensive and more balanced financial sector reforms have also resulted in a more robust financial system. Following the establishment of Custom Union with the Euroean Union in 1996, the exort market share of manufactured industrial roducts has increased substantially. For the last six years, manufacturing exorts have been about 93% of total exorts (see Table 1). On the other hand, as the Turkish economy has develoed beyond the traditional laborintensive industries, manufacturing has reuired more raw materials and sohisticated machinery and euiment, most of which continue to be imorted (Orhan & Nergiz, 2014). Because of this imort deendent structure of the Turkish economy, any growth in GDP causes to remain high current account deficits. The high and volatile current account deficit in Turkey is detrimental to financial stability and has been at the core of olicy discussions in recent years. 286

2 Table 1. Turkey s foreign trade statistics (USD million) Exorts Manufacturing Exorts by Balance of Volume of Proortion of Exorts Imorts ISIC, Rev.3 Foreign Trade Foreign Trade Covered by Exorts Years Value Change (%) Value % of Total Exorts Value Change (%) Value Value (%) , , , , , , , , , , , , , , , , , , , , , , , , Source: TurkStat, Foreign Trade Statistics. In theory, a country s trade balance deends on its own level of economic activity, the level of economic activity in the rest of the world and the exchange rate, all in real terms. Economic growth at home results in more imorts and deteriorates trade balance. On the other hand, economic growth in the rest of the world results in more exorts and imroves trade balance. By stimulating exorts and reducing imorts, dereciation at home currency is also exected to imrove the trade balance of a country (BahmaniOskooee & Fariditavana, 2015). In the emirical literature, financial develoment has also some imacts on trade balance. It is evidenced that countries with a relatively welldeveloed financial sector have a comarative advantage in industries and sectors that rely on external finance (Kletzer & Bardhan, 1987). The link between financial develoment and the structure of the trade balance was exlored by Beck (2002). In the study using a 30year anel for 65 countries, it is evidenced that financial develoment exerts a large causal imact on the level of both exorts and the trade balance of manufactured goods. Countries with a betterdeveloed financial system have a higher exort share and trade balance in manufactured goods. Samba and Yan (2009) investigates the relationshi between the level of financial develoment of a country and its comarative advantage in international trade, by using VAR models. To rovide long run relationshis between financial develoment and international trade in manufactured goods, for most of the countries from East Asia they suggest that international trade in manufactured goods enhances financial develoment. The emirical uestion of whether a country s level of manufacturing trade is affected by its financial sector develoment was also studied by Kiendrebeogo (2012) and he found that countries with betterdeveloed financial systems tend to secialize in industries that rely on external finance in roduction. He used ure crosssectional and anel secifications on a samle of 75 countries over the eriod and revealed that financial develoment exerts a ositive effect on manufacturing exorts strongly and robustly, even after controlling for the effect of banking crises. For Pakistan, by using time series data from 1972 to 2014, Muzammil and Ahad (2015) have investigated the imact of financial develoment on trade balance. In the study, it is found that financial develoment has ositive significant imact on trade balance in the long run but in the short run there is no significant relationshi between these variables. The study offers government to enhance financial develoment by managing lending interest rates to imrove trade balance. The relationshi between financial develoment and trade balance reuires further emirical work and the existing literature does not rovide any evidence regarding imact of financial develoment on trade balance in the case of Turkey. If financial develoment has imacts on trade balance, it would mean that financial sector develoment is also imortant for economic develoment and therefore it shows riorities for olicy makers. The rest of the aer is organized as follows: In section 2, the data and methodology which is used in the study are introduced. In section 3, the results of analysis are given. And section 5 concludes. 2. Data and Method The data used in this study has been mainly obtained from IMF data server. The imort and exort values have been retrieved form IMF Direction of Trade Statistics (DOTS) in US dollar terms. For having an index of financial develoment, the ratio of total rivate credits to Nominal Gross Domestic Product (GDP) has been used, all in national currency. The values of rivate credits have been calculated from the database of Central Bank of Turkey which contains rivate sector credits of commercial banks, articiation banks and develoment banks. Exchange rate has been defined as national currency er US dollar and consumer rice indexes of Turkey and 287

3 US have been used for calculating real exchange rate. Using uarterly data of Turkey, the study has been conducted for the eriod of 1987Q12015Q2. Instead of using exorts minus imorts, the ratio of exorts to imorts (X/M) has been used as a roxy for trade balance in the study. This modification of the variable has some advantages. Without modification the variable may have a negative sign and therefore logarithmic transformation cannot be used. By using this ratio the variable also becomes insensitive to nominal values of exorts and imorts (BahmaniOskooee, 1991). Desite the wellknown fact that macroeconomic variables ossess asymmetric and nonlinear features (Keynes, 1936; Kahneman & Tversky, 1979; Shiller, 1993, 2005), research on the relationshi between financial develoment and trade balance has been tested only within a linear framework so far. While there have been some studies about nonlinear relationshi between real exchange rate and trade balance (Karamelikli, 2016), nonlinear relationshi between financial develoment and trade balance has not been studied yet. Previous studies have also not included financial develoment as a key variable in their models. Following (Muzammil & Ahad, 2015), the model of this study has been set as below: LTB t = α 0 α 1 LRER t α 2 LFD t ε t (1) In the model, LTB t is defined as ln( X M ) t, and RER t is defined as E tp* t P t where E is the nominal effective exchange rate, and P and P* are the domestic and foreign rice levels resectively. LFD reresents the ratio of rivate credits to country s gross domestic roduct in logarithmic form. While CPI and LRER are used together in the model of (Muzammil, 2015) as indeendent variable, the CPI variable is omitted in this study to avoid multicollinearity roblem. In this study, an asymmetric cointegration aroach is used to estimate asymmetric effect of financial develoment on trade balance. The nonlinear autoregressive distributed lag (NARDL) model estimates shortand longrun nonlinearities via ositive and negative artial sum decomositions of exlanatory variables. The asymmetric longrun relationshi can be formulated as below: y t = β x t β x t u t (2) where x t is a k 1 vector of regressors decomosed as x t = x 0 x t x t, and x and x are artial sum rocesses of ositive and negative changes in x t. The error correction model (ECM) can be constructed as follows: ΔLTB t = β 0 j=1 β 1j ΔLTB tj j=0 β 2j ΔLRER tj m j=0 β 3j ΔLFD tj θε t1 e t (3) where reresents the first differences of the variables and the errorcorrection term ε is the OLS residuals series from the longrun cointegrating regression (Euation 1). Putting Euation (1) and Euation (3) together yields following ECM euation: ΔLTB t = ψ η 0 LTB t1 η 1 LRER t1 η 2 LFD t1 j=1 β 1j ΔLTB tj β 2j ΔLRER tj j=0 m j=0 β 3j ΔLFD tj e t (4) where ψ = β 0 θα 0, η 0 = θ, η 1 = θα 1, η 2 = θα 2. In addition, θ = η 0, α 1 = η 1, α θ 2 = η 2 θ are the long run coefficients of LTB, LRER and LFD variables, while β 1, β 2, β 3 are the short run coefficients of the variables. In order to determine the existence of asymmetric cointegration relationshi between financial develoment and trade balance, the aroach of Schorderet (2002, 2003) and Shin et al. (2014) is followed in this study. This aroach reuires LRER and LFD variables to be decomosed to ositive and negative shocks. Then LRER, LRER, LFD and LFD are the artial sums of ositive and negative changes in the LRER and LFD variables. More recisely: LRER t = t i=1 ΔLRER i = t i=1 max(δlrer i, 0) ; LRER t = t i=1 ΔLRERi = t i=1 min(δlrer i, 0) LFD t = t i=1 ΔLFD i = t i=1 max(δlfd i, 0) ; LFD t = t i=1 ΔLFDi = t i=1 min(δlfd i, 0) (5) 288

4 Considering Euation (5), the long run relationshi in Euation (1) can be redefined as follows: LTB t = α 0 α 1 LRER t α 1 LRERt α2 LFD t α 2 LFDt εt (6) Following Shin et al. (2014) it is straightforward to rewrite Euation (4) in asymmetric form as follows: ΔLTB t = ψ η 0 LTB t1 η 1 LRER t1 η 1LRERt1 η 2 LFD t1 η 2LFDt1 j=1 β 1j ΔLTB tj (β j=0 2j ΔLRER tj β 2j ΔLRER tj ) m (β j=0 3j ΔLFD tj β 3j ΔLFD tj ) e t (7) where ψ = β 0 θα 0, η 0 = θ, η 1 = θα 1, η 1 = θα1, η2 = θα 2, η 2 = θα2. And α 1 = η 1, α θ 1 = η 1, α θ 2 = η 2, α θ 2 = η 2 θ are asymmetric negative and ositive longrun coefficients of the real exchange rate and financial develoment resectively. In order to determine the longrun asymmetric cointegration, Shin et al. (2014) rooses the Pesaran et al. (2001) bounds test as in linear ARDL aroach. In Pesaran et al. (2001) bounds test, null hyothesis is defined as η 0 = η 1 = η 2 = 0 against alternative hyothesis η 0 0 or η 1 0 or η 2 0. And for the longrun asymmetry, null hyothesis is η 0 = η 1 = η 1 = η 2 = η 2 = 0. Then the calculated Fstatistic is comared with the tabulated F values by Pesaran et al. (2001). Null hyothesis of α 1 = α 1 and α 2 = α 2 is tested by the Wald Ftest for checking of ossible longrun symmetry. To test the existence of shortrun symmetry, Wald Ftest is used to test the null hyothesis of i=0 β 2i = i=0 β 2i and i=0 β 3i = i=0 β 3i. When the null hyothesis of symmetry is rejected, the alternative hyothesis that claims asymmetric relationshi could be acceted. Then asymmetric dynamic multiliers of LREER, LREER, LFD and LFD could be calculated resectively. The cumulative dynamic multilier effects of LREER, LREER, LFD and LFD on LTB can be figured out as follow: LTB ti LRER t m h h h = i=0 ; m h = LTB ti i=0 LTB ti LFD t LRER t w h h = i=0 ; w h h = LTB ti i=0 LFD t (8) Note that, as h then m h α 1, m h α 1, w h α 2, w h α 2, where α 1, α 1, α 2 and α 2 are the asymmetric longrun coefficients. The changes from initial euilibrium to the new oint that caused by negative or ositive shocks of LRER and LFD is catured by the dynamic multiliers (Shin et al., 2014). 3. Results Euation (7) can be called as Model I. In this model, all variables have asymmetric characteristics and it means that any increase or decrease in indeendent variables has distinct effects both in short and longrun. As a result, this model is defined as a full asymmetric model. The contents of Table 2 give evidences of the rejection of null hyothesis of no cointegration in the both models. This table also claims that longrun symmetry is rejected. The null hyothesis of symmetry in the shortrun for LFD cannot be rejected in Model I. Tests for basic assumtions of normality, heteroscedasticity and serial correlation are satisfying. By remodeling symmetric attern of LFD in shortrun euation no (9) could be derived. ΔLTB t = ψ η 0 LTB t1 η 1 LRER t1 η 1LRERt1 η 2 LFD t1 η 2LFDt1 j=1 β 1j ΔLTB tj (β j=0 2j ΔLRER tj β 2j ΔLRER tj ) m j=0 β 3j ΔLFD tj e t (9) Euation (9) is called as Model II and the estimation results for this Model are also resented in the Table 2. The null hyothesis of symmetric relationshi in the longrun is rejected for this model and it can be acceted as the best fitted model. The estimation results of full asymmetric model (Model I) and artial asymmetric model (Model II) can be seen in Table

5 Table 2. Emirical results of nonlinear relationshi Full Asymmetry (Model I) Partial Asymmetry (Model II) INTERSECT *** *** LTB(1) *** *** LRER_N(1) LRER_P(1) * LFD_N(1) ** LFD_P(1) ** *** D(LTB(1)) *** *** D(LTB(2)) ** ** D(LTB(3)) ** * D(LTB(4)) *** ** D(LRER_N) D(LRER_P) *** D(LRER_N(1)) D(LRER_P(1)) *** *** D(LFD_N) D(LFD_P) D(LFD_N(1)) ** D(LFD_P(1)) D(LFD_N(2)) D(LFD_P(2)) ** D(LFD_N(3)) D(LFD_P(3)) D(LFD_N(4)) D(LFD_P(4)) D(LFD) D(LFD(1)) * D(LFD(2)) *** Fstatistic *** *** R JarueBera Heteroskedasticity F Bounds Values 2.86 & & 4.01 F Bound WLR For LFD *** WSR For LFD WLR For LRER *** WSR For LRER *** ECM(1) *** Note. ***, ** and * denote statistical significance at the 1, 5 and 10 ercent levels resectively. The longrun coefficients of negative and ositive decomositions of LRER and LFD are calculated as 0,048, 0,227, 0,098 and 0,167 resectively. As seen in Table 2, LRER is statistically insignificant while LRER, LFD and LFD are significant at %10, %5 and %1 resectively. Therefore we can conclude that ositive decomosition of financial develoment is the most significant among other variables where negative decomosition of real exchange rate is not significant at any level. The results show that the negative changes in real exchange rate are statistically rejected while ositive changes in real exchange rate could be acceted only at %10 level of significance. The longrun multilier of negative shocks is not statistically accetable. While longrun multilier of ositive shocks on real exchange rate is statistically weak, it is estimated as having ositive sign. Exchange rate assthrough to domestic rices in Turkey can exlain these results. In the theory, there is a ositive relationshi between real exchange rate and inflation in the longrun (Karamelikli & Korkmaz, 2016). Therefore, changes in real exchange rate could be reflected on domestic rices. In the economic theory, it is exected that any changes in exchange rate cause rice advantages or disadvantages for tradable goods. So in the case of exchange rate assthrough to rices, it can be 290

6 exected that the rice advantages or disadvantages are to be eliminated in the longrun. Longrun multiliers of ositive and negative decomositions of financial develoment are significant at %1 and %5 level and the coefficients of both variables are negative. Then an increase in financial develoment will cause a decrease in trade balance while a negative change in financial develoment causes a decrease in trade balance less than ositive change. Desite of the same signs of both ositive and negative decomositions of financial develoment in the longrun, ositive changes have more imact on trade balance. 4. Conclusions In this study, by emloying asymmetric autoregressive distributed lag (NARDL) aroach to cointegration, the relationshi between financial develoment and trade balance has been investigated. The results reveal that there is cointegration between financial develoment and trade balance in Turkey. The existence of asymmetric relationshi between these variables suggests that any increase or decrease in financial develoment affects trade balance with different degrees. Therefore, future studies related to this subject should consider asymmetric feature of financial develoment. The study also finds that there is a negative relationshi between financial develoment and trade balance. This result suorts the literature which suggests that financial develoment fosters economic growth. In develoing countries, financial sector lays a critical role in facilitating economic growth. Financial sector affects domestic savings, caital accumulation, technological innovation and growth. Through economic growth and roviding access to financial services, financial develoment can also imact overty reduction. Therefore these countries attach great imortance to financial develoment. The findings of this study are convenient with imort deendency characteristics of Turkish economy where any growth in GDP causes high current account deficits. Trade liberalization olicies may also lead to faster growth of imorts than exorts in Turkey. For olicy recommendations, this study does offer not to use financial develoment as a olicy tool to decrease trade deficit. Because ositive imrovements in financial develoment have negative imacts on trade balance. And ositive shocks on financial develoment have twofold higher imact than negative shocks. Then olicy makers should be aware of the asymmetrical characteristics of financial develoment as a olicy tool considering trade balance. References BahmaniOskooee, M. (1991). Is there a longrun relation between the trade balance and the real effective exchange rate of LDCs? Economics Letters, 36(4), htt://dx.doi.org/ / (91)90206z BahmaniOskooee, M., & Fariditavana, H. (2015). Nonlinear ARDL aroach, asymmetric effects and the Jcurve. Journal of Economic Studies, 42(3), htt://dx.doi.org/ /jes Beck, T. (2002). Financial develoment and international trade. Journal of International Economics, 57(1), htt://dx.doi.org/ /s (01) Kahneman, D., & Tversky, A. (1979). Prosect Theory: An Analysis of Decision under Risk Daniel Kahneman; Amos Tversky. Econometrica, 47(2), htt://dx.doi.org/ /j x Karamelikli, H. (2016). Linear and Nonlinear Dynamics of the Turkish Trade Balance. International Journal of Economics and Finance, 8(2), htt://dx.doi.org/ /ijef.v8n270 Karamelikli, H., & Korkmaz, S. (2016). The Dynamics of Exchange Rate PassThrough to Domestic Prices in Turkey, htt://dx.doi.org/ /pressacademia.2016 Keynes, J. (1936). The general theory of emloyment, interest and money. London: Macmillan. Retrieved from htt://cas.umkc.edu/economics/eole/facultypages/kregel/courses/econ645/winter2011/generaltheory.df Kiendrebeogo, Y. (2012). The Effects of Financial Develoment on Trade Performance and the Role of Institutions. Economics Bulletin, 32(3), Kletzer, K., & Bardhan, P. (1987). Credit markets and atterns of international trade. Journal of Develoment Economics, 27(12), htt://dx.doi.org/ / (87)90006x Muzammil, M., & Ahad, M. (2015). Imact of Financial Develoment on Trade Balance: An ARDL Cointegration and Causality Aroach for Pakistan. Retrieved from htts://mra.ub.unimuenchen.de/id/erint/68829 Orhan, O. Z., & Nergiz, E. (2014). Turkey s Current Account Deficit Problem and Its Effects on the Euroean Union Accession. İstanbul Gelişim Üniversitesi Sosyal Bilimler Dergisi, 1(1), Retrieved from 291

7 htt://dergiark.ulakbim.gov.tr/igusbd/article/view/ Pesaran, M. H., Shin, Y., & Smith, R. J. (2001). Bounds testing aroaches to the analysis of level relationshis. Journal of Alied Econometrics, 16(3), htt://dx.doi.org/ /jae.616 Samba, M. C., & Yan, Y. (2009). Financial Develoment and International Trade in Manufactures: An Evaluation of the Relation in Some Selected Asian Countries. International Journal of Business and Management, 4(12), htt://dx.doi.org/ /ijbm.v4n1252 Schorderet, Y. (2002). A Nonlinear Generalization of Cointegration: A Note on Hidden Cointegration (No ). Genève. Retrieved from htt:// Schorderet, Y. (2003). Asymmetric cointegration (No ). Cahiers du déartement d économétrie, Faculté des sciences économiues et sociales, Université de Genève. Retrieved from htt:// Shiller, R. J. (1993). Macro Markets: Creating Institutions for Managing Society s Largest Economic Risks. Clarendon Press. Shiller, R. J. (2005). Irrational Exuberance (2nd ed.). Princeton University Press. Shin, Y., Yu, B., & Greenwoodnimmo, M. (2014). Modelling asymmetric cointegration and dynamic multiliers in a nonlinear ARDL framework. In R. C. Sickles, & W. C. Horrace (Eds.), Festschrift in Honor of Peter Schmidt Econometric Methods and Alications ( ). htt://dx.doi.org/ / Coyrights Coyright for this article is retained by the author(s), with first ublication rights granted to the journal. This is an oenaccess article distributed under the terms and conditions of the Creative Commons Attribution license (htt://creativecommons.org/licenses/by/3.0/). 292