بررسي اثرات متقارن و نامتقارن سياست مالي و توسعه تجارت بر توسعه مالي در ايران با استفاده از مدل غيرخطي NARDL

Evaluation of the Asymmetric Impacts of Fiscal Policy and Trade Development on Financial Development in Iran Using the NARDL Nonlinear Model

در مطالعه حاضر به بررسي اثرات متقارن و نامتقارن سياست مالي و توسعه تجارت بر توسعه مالي در ايران طي دورۀ زماني 2017-1973 پرداخته شده است. براي بررسي اثرات متقارن از روش جوهانسن- جوسيلوس استفاده شده و همچنين براي بررسي اثرات نامتقارن از رهيافت خودرگرسيو با وقفه‌هاي توزيعي غيرخطي (NARDL) استفاده شد. نتايج روش متقارن نشان مي­دهد كه در بلندمدت، افزايش مخارج دولت و تورم اثر منفي و معني­دار بر توسعه مالي دارند. همچنين نتايج بيانگر اين است كه توسعه تجارت اثر مثبت و معني‌داري بر توسعه مالي دارد. نتايج مربوط به الگوي ECM نيز نشان مي­دهد كه در هر دوره 0/078 از عدم تعادل يا خطاي كوتاه­مدت به سمت تعادل بلندمدت تعديل مي­شود. همچنين، نتايج روش (NARDL) نشان مي‌دهد كه شوك مثبت مخارج دولت اثر منفي و شوك منفي مخارج دولت اثر مثبت و معني‌دار بر توسعه مالي دارد. همچنين، شوك مثبت توسعه تجارت اثر مثبت و شوك منفي توسعه تجارت اثر منفي بر توسعه مالي دارد. علاوه بر اين، نتايج نشان مي ­دهد كه تورم اثر منفي و معني‌دار بر توسعه مالي دارد. درنهايت، نتايج آزمون والد هم نشان مي­دهد كه اثر شوك‌هاي مخارج دولت و توسعه تجارت هم در كوتاه‌مدت و هم در بلندمدت نامتقارن هستند.

1- INTRODUCTION Financial institutions play a crucial role in capital accumulation and financial development. These institutions by mobilization of savings; facilitation of trade, diversification and pooling of risks; monitoring of firms and exerting corporate governance; and facilitating exchange play an important role in achieving economic growth. In many empirical studies, the relationship between economic growth and financial development has been considered, and most researchers have identified the accumulation of physical and human capital and total productivity growth as the main channels for the impact of financial development on economic growth (Levine, 1997; Duramany-Lakkoh, 2020). 2- THEORETICAL FRAMEWORK Empirical studies show that the government spending has traditionally been considered a counterproductive policy tool for stimulating credit. Standard Keynesian and neoclassical theories express that an increase in government spending raises interest rates, thereby lowering private-sector investment. But there is the number of evidence to support the notion that government spending makes it stronger credit markets (Murphy & Walsh, 2018). In contrast, growing evidence from the United States and other advanced economies suggests that government spending could lower long-term interest rates (Miranda-Pinto et al, 2019). These studies indicate a gap in economists' understanding of the relationship between financial stimulus and credit markets. Also, Nanforosh & Dizaji (2016) show that government spending has a negative and significant impact on financial development, as well as trade, financial globalization and the quality index of legal institutions have a positive impact on financial development. 3- METHODOLOGY Considering symmetrical and asymmetrical effects of fiscal policy and trade development on financial development in Iran Using annual data over the period 1973-2017. That, the Johansen-Jocilus method was used to investigate the symmetric effects and the auto-regressive distributed lag (NARDL) model was used to investigate the asymmetric effects. In order to determine asymmetric pass-through of openness and government spending to financial development, we follow the approach of Shin et al. (2014). This approach requires the decomposition of the variable of interest. In this case, we decompose the LTARDE and LGOV variables into positive and negative sub-variables. The partial sums of positive and negative changes in openness are given by and , also, partial sums of positive and negative changes in government spending and . where LCREDIT is defined as financial development, LTRADE is degree of openness, LGOV is government spending and LCP is inflation. 4- RESULTS & DISCUSSION In time-series analyzes, before considering the model's estimation, it is necessary to test the statics variables of the research. ADF test results show that (LGOV+) and LTRADE variables is integrated at order zero I(0) and other variables are not stationary at the level. Also, PP test results show that (LGOV+) variables is integrated at order zero I(0) and other variables are not stationary at the level (table 1). Therefore, their first-order difference should be used in the Johansen-Jocilus method and NARDL model. The results indicate that co-integration is present. This result is supported by the fact that F-static is higher than the upper bound critical value at 1% critical value. Hence, the null hypothesis of no co-integration can be rejected. The results of the symmetric method show that in the long run, the increase in Government Spending and inflation have a significant negative effect on financial development. The results show that trade development has a positive and significant effect. Also, the results of the VECM model indicated that in each period, 0.078 of the imbalance or short-run error is adjusted towards the long-run equilibrium. Also, the results of the NARDL method show that the positive shock of government spending has a negative effect and the negative shock of government spending has a positive and significant effect on financial development. The positive trade shock has a positive effect and negative trade shock has a negative effect on financial development. Also, the results show that inflation has a significant negative effect on financial development. Finally, the results of the Wald test show that the effects of government spending shocks and trade are asymmetric in both the short and long run. The step toward achieving the research objectives is to examine the stability of the long-run parameter of the NARDL model by using the Cumulative Sum (CUSUM) and Cumulative Sum of Square (CUSUMSQ) tests following Pesaran et al., (1997). If the plots of these tests statistics stay within the critical bound of 5% level of significance, the null hypothesis of all coefficients of the regression are stable and cannot be rejected. Therefore, it implies that the coefficients in the error-correction model are stable. As observed in Figure (3), the plots of CUSUM and CUSUMSQ statistics stay within the critical 5% bound for the period. 5- CONCLUSIONS & SUGGESTIONS The present study evaluated the symmetrical and asymmetrical effects of fiscal policy and trade development on financial development in Iran Using annual data over the period 1973-2017. For this purpose, the Johansen-Jocilus method and (NARDL) model was used to estimate. The results show that in the long run, the increase in government spending (fiscal policy) and inflation have a significant negative effect on financial development. Also, the results show that trade development has a positive and significant effect on financial development.