@article { author = {Amrollahi Bioki, Elham and Hojabr Kiani, Kambiz and Memarnejad, Abbas and Abtahi, Seyed Yahya}, title = {State-dependent effects of liquidity growth on exchange rate volatility in Iran's economy}, journal = {The Journal of Economic Policy}, volume = {13}, number = {25}, pages = {241-280}, year = {2021}, publisher = {Yazd University}, issn = {2645-3967}, eissn = {2645-3975}, doi = {10.22034/epj.2021.15510.2145}, abstract = {Introduction: Monetary variables serve as the main conditions for economic stability. In the economy of Iran, liquidity has tremendously grown in recent years and, by affecting the exchange rate as an important macroeconomic variable, has led to its volatility. Therefore, identifying the relationship between liquidity and exchange rate volatility is of great importance from the perspective of economic policymakers in order to make decisions in macroplans. This study analyzes the dynamic response of nominal exchange rate volatility in different states of liquidity growth in the Iranian economy.Methodology: At first, using the statistics of the Central Bank of the Islamic Republic of Iran, the quarterly data were extracted of the nominal exchange rate, liquidity, export price index, and import price index for the period of 1990-2019.Then, the liquidity growth variable was calculated through the logarithmic difference of that variable. Also, with regard to the terms of trade from the ratio of export price index to import price index, the nominal exchange rate volatility was calculated by using MS-EGARCH (1,1), MS-TGARCH (1,1) and MS-GJRGARCH(1,1) with the norms of std, ged, sstd, and sged conditional distributions. The results indicated that the MS-EGARCH (1,1) model with the sstd conditional distribution and the lowest value of Akaike Information Criterion (AIC) is the optimal model for calculating nominal exchange rate volatility; Then, the stationary of liquidity growth, terms of trade and nominal exchange rate volatility were confirmed by using Lee and Strazicich test (2003) with two structural breaks. Using the linear vector autoregressive model, the optimal interval length of the model was found to be 4. According to test conducted by Hansen (1999), the number of regimes turned to be 2. As a result, the threshold vector autoregressive model (TVAR) with two regimes was used to investigate the effect of the liquidity growth and terms of trade on the nominal exchange rate volatility.Results and Discussion: Considering the nominal exchange rate volatility as a threshold variable with the value of 12.57, the results indicate that, in the low regime of the nominal exchange rate volatility, lagged liquidity growth does not have a significant effect on the nominal exchange rate volatility. However, by exceeding the threshold and being in the high regime of nominal exchange rate volatility, the lagged liquidity growth have positive and significant effect on nominal exchange rate volatility, because the growth of liquidity as an expansionary monetary policy leads to growth in demand for goods and services. Because the supply of goods and services is limited in the short time, this leads to inflation and the exchange rate volatility increases. Also, the terms of trade reduce the nominal exchange rate volatility in the high regime, which is consistent with the findings of Chipili (2012). In addition, in order to explain the dynamics of liquidity growth and nominal exchange rate volatility, the Markov Switching Vector Autoregressive (MSVAR) model was used. The results show that, in the equations of the nominal exchange rate volatility and liquidity growth, the autoregressive coefficients are significant in both regimes. The results of Granger causality test based on MSVAR equations indicate that, in the low regime of the nominal exchange rate volatility, liquidity growth is not the Granger cause of this volatility, while liquidity growth is the Granger cause in the high regime. The terms of trade are also the Granger cause of the nominal exchange rate volatility in the high regime. According to the results of the study, in the high regime, the growth of liquidity and terms of trade are effective in the volatility.Conclusion: If the liquidity is directed to production according to the quantity theory of money, the volume of production increases and part of the liquidity effect will be neutralized. Otherwise, speculators and traders in the market assets such as gold, currency and housing will increase the price of these assets as well as inflation and exchange rate volatility in the country; Also, if the government adjusts its budget deficit through supply-side policies instead of borrowing from the central bank, it will reduce the turbulence in the exchange market by reducing the amount of liquidity. The terms of trade can be improved by factors such as exporting goods to countries with low elasticity demand, maintaining the monopoly status of the country's exports, supporting of exporters, and decreasing share of oil in the country's exports, Therefore, if the liquidity growth is controlled and the terms of trade are improved, the nominal exchange rate volatility decreases, which policymakers consider as a strategic point.}, keywords = {Threshold Vector Autoregressive (TVAR) Model,Markov Switching Vector Autoregressive (MSVAR) Model,Markov Switching GARCH (MSGARCH) Model,Liquidity,Nominal exchange rate volatility}, title_fa = {اثرات وابسته به وضعیت رشد نقدینگی بر نوسانات نرخ ارز در اقتصاد ایران}, abstract_fa = {این مطالعه واکنش پویای نوسانات نرخ ارز اسمی را به وضعیت‌های مختلف رشد نقدینگی در اقتصاد ایران و طی دوره زمانی 1398:04-1369:03 مورد تحلیل قرار می‌دهد. بدین منظور نوسانات نرخ ارز اسمی با استفاده از مدل MS-EGARCH(1,1)[1] با توزیع شرطی sstd محاسبه شده است. نتایج مطالعه با بهره‌گیری از مدل خودرگرسیون برداری آستانه‌ای[2] (TVAR) و لحاظ نوسانات نرخ ارز اسمی به عنوان متغیر آستانه حاکی از آن است که در رژیم پایین نوسانات نرخ ارز اسمی، وقفه‌های رشد نقدینگی اثر معنی‌داری بر نوسانات نرخ ارز اسمی ندارند اما در رژیم بالای نوسانات نرخ ارز اسمی، وقفه‌های رشد نقدینگی اثر مثبت و معنی‌داری بر نوسانات نرخ ارز اسمی دارند؛ همچنین رابطه مبادله تجاری، نوسانات نرخ ارز اسمی را در رژیم بالای نوسانات نرخ ارز اسمی، کاهش می‌دهد. نتایج مطالعه با استفاده از مدل خودرگرسیون برداری با امکان تغییر رژیم مارکوف[3] (MSVAR) نشان می‌دهد که در معادلات نوسانات نرخ ارز اسمی و رشد نقدینگی، ضرایب خودرگرسیونی در هر دو رژیم معنی‌دار هستند. همچنین نتایج آزمون علیت گرنجر بر پایه معادلات MSVAR حاکی از آن است که در رژیم پایین نوسانات نرخ ارز اسمی، رشد نقدینگی علت گرنجری نوسانات نرخ ارز اسمی نمی‌باشد اما رشد نقدینگی علیت گرنجر نوسانات نرخ ارز اسمی در رژیم بالای نوسانات نرخ ارز اسمی خواهد بود. از طرف دیگر، رابطه مبادله تجاری نیز علیت گرنجر نوسانات نرخ ارز اسمی در رژیم بالای نوسانات نرخ ارز اسمی است. با توجه به نتایج مطالعه در رژیم بالای نوسانات نرخ ارز اسمی، رشد نقدینگی و رابطه مبادله تجاری بر نوسانات نرخ ارز اسمی مؤثر می‌باشند، لذا چنانچه رشد نقدینگی کنترل شود و رابطه مبادله تجاری بهبود یابد، باعث کاهش در نوسانات نرخ ارز اسمی می‌شود که می‌تواند به عنوان یک نکته راهبردی مورد توجه سیاست‌گذاران قرار گیرد.}, keywords_fa = {مدل خودرگرسیون برداری آستانه‌ای (TVAR),مدل خودرگرسیون برداری با امکان تغییر رژیم مارکوف (MSVAR),مدل مارکوف سوئیچینگ گارچ (MSGARCH),نقدینگی,نوسانات نرخ ارز اسمی}, url = {https://ep.yazd.ac.ir/article_2253.html}, eprint = {https://ep.yazd.ac.ir/article_2253_9f4463f2108d2580ea982465396468df.pdf} }