DOI QR코드

DOI QR Code

An Analysis of the Exchange Rate Regime of Nepal: Determinants and Inter-Dynamic Relationship with Macroeconomic Fundamentals

  • DAHAL, Suresh Kumar (Nepal Rastra Bank, Department of Humanities and Social Sciences, Sri Sathya Sai Institute of Higher Learning) ;
  • RAJU, G. Raghavender (Department of Humanities and Social Sciences, Sri Sathya Sai Institute of Higher Learning)
  • 투고 : 2022.04.10
  • 심사 : 2022.07.05
  • 발행 : 2022.07.30

초록

The exchange rate is an important macroeconomic variable that influences internal and external balances. Nepal follows a dual exchange rate such that the Nepali rupee (NPR) is pegged with the Indian rupee (INR) but floats with the United States dollar (USD) and all other currencies. There have been very few studies on the exchange rate of Nepal, of which the majority focus on the bivariate relationship between exchange rate and another variable. However, this paper analyses the multivariate relationship between the USD-NPR exchange rate and major macroeconomic variables. Determinants of Nepal's exchange rate have been derived with multiple regression using the ordinary least square (OLS) approach. Since the explanatory variables could not significantly capture the movement of the dependent variable, a long-run relationship between Nepal and India's exchange rate has been analyzed using Engle-Granger cointegration to establish a relationship as suggested by a graphical representation. This explains that Nepal's exchange rate long run is determined by India's exchange rate than its own fundamentals. In addition, the macro-linkages of Nepal's macroeconomic variables have been analyzed using Standard Vector Autoregressive models followed by impulse response analysis which is useful for policy decisions. Some policy implications indicating the sustainability of Nepal's pegged regime have been drawn based on the empirical analysis.

키워드

1. Introduction

The exchange rate, the price of one currency against the other, is an important macroeconomic variable that facilitates the international trade of goods and services and the transfer of capital. While the history of international trade goes before the medieval period, the wide facilitation of the exchange rate has come into existence only after the adoption of the Gold standard1 and subsequently with the Bretton woods system2. However, in recent times, each nation adopts either fixed, free-floating, or a regime somewhere in between and maintains a bilateral exchange rate with major trading partners. The US dollar (USD) is considered an international reserve currency by most economies which also serves as a common currency for cross-exchange rates.

The exchange rate also indicates the external competitiveness of an economy. The undercurrent of this competitiveness is an economic performance that is reflected by other macroeconomic variables such as productivity, price level, interest rate differentials, etc. Thus, the exchange rate must be in tandem with other macroeconomic variables, at least in the long run, to avoid economic disequilibria, which can eventually result in financial and/or currency crises3. In a free-floating exchange regime, the exchange rate is adjusted continuously due to the market forces, and thus, arguably, it is believed to be in equilibria with other economic variables. However, in the case of a fixed exchange regime, it has to be revised, deliberately in case of significant change in economic performance over time as against the counterpart. The regime that follows is fixed with one currency but floating with other needs to assess both the arrangement scrupulously. Nepal follows a dual exchange rate system such that the Nepali rupee (NPR) is currently pegged with Indian Rupee (INR) but floats with the US dollar (USD) and other convertible currencies. Nepal is a landlocked country in South Asia, sharing its border with India from the East, West, and South, and China from the North. It is a small economy with a $33.65 billion nominal GDP4 and has an open porous border of about 1770 kilometers with India, resulting in high trade and cultural integration. This geopolitical background has directly impacted its exchange rate policy and reforms. The exchange rate between NPR and INR was stable from 1877 to 1932 with a fixed exchange regime. However, it went through a phase of instability from 1932–1960, when NPR replaced the former ‘Mohar’ currency in Nepal. Both countries went through a series of major political and economic transitions. (Maskay, 2007). The period from 1956 to pre-liberalization till 1993 witnessed the expansion of NPR, the end of the dual currency system, stability of the INR-NPR exchange rate, and the evolution of foreign exchange reserve management.

The exchange rate management before 1956 was in relation to INR alone, as foreign currency reserves were managed by the Reserve Bank of India (RBI). The buying and selling of other convertible foreign currencies began only after the inception of the Nepal Rastra Bank in 1956. While the INR rate was fixed by the Government, exchange rates against other convertible currencies were adopted from the rate published by the RBI. However, with a motive to enhance international trade, fixed exchange rates of the USD, Pound Sterling, and Swiss Franc against NPR were determined from May 14, 1960. Subsequently, the practice of dual exchange rate of other convertible currencies was introduced on March 31, 1978, but with a fixed USD rate. Eventually, USD started floating daily only from February 12, 1993, after the full convertibility of the current account (NRB, 1981, 2018). Since then, Nepal Rastra Bank and commercial banks have been publishing and quoting their exchange rates independently daily. The exchange rate of USD-INR-and USD-NPR and INR-NPR from 1994 to 2020 is depicted in the graphs below (Figures 1 and 2).

Figure 1: NPR and INR Exchange Rate Source: Nepal Rastra Bank

Figure 2: INR-NPR Exchange Rate (NPR Per INR)

As per the graph above (Figures 1 and 2), it can be observed that USD-NPR and USD-INR exchange rates have been moving upwards, nearly synchronized. Further, the INR-NPR rate has remained the same since February 12, 1993. In a dual exchange rate regime, the constant peg rate is possible only when the macroeconomic performance of Nepal is nearly similar to that of India. However, on the contrary, the economic landscape of these two countries has changed visibly. The graph below presents major exchange rate-determining variables as per existing theory, namely, inflation rate based on consumer prices, trade deficit, and GDP growth rate at current prices (see Figures 3, 4, and 5).

Figure 3: Inflation Rate (Percent) Based on the Consumer Price Index

Figure 4: Trade Deficit (Percentage of GDP)

Figure 5: GDP Growth Rate (Percent) Source: The World Bank

The graphical representation above depict that, the economic performance of India has been better than Nepal nearly consistently over the last three decades. For instance, the economic growth of India in the last 30 years has been nearly 6 percent on an average while Nepal’s average growth for the same period was only around 4.4 percent due to decade-long political turmoil and prolonged transition in the drafting of the new constitution. While the inflation rate of Nepal, on an average, was marginally high than that of India, the trade deficit has been a matter of serious concern to Nepal. Further, Nepal’s trade deficit is ever-widening that reaching a whopping 33.68 percent in 2019. Remittance inflow has been the major source for Nepal to finance its trade deficit.

The exchange rate is determined as per the performance of major macroeconomic fundamentals5. However, under a fixed regime, the disequilibrium is likely, especially when the regime is backed by the restricted capital flow. It is because a currency is pegged with unlimited convertibility such that the rate is kept constant by increasing/restricting the supply of pegged currency. However, in the case of Nepal, the pegged rate has not been revised since 1993 despite widened economic gap with India seems quizzical. The rate has been found overvalued which in turn has resulted in an export competitiveness trap (Paudel & Burke, 2015; Pandey, 2014).

The dual exchange rate system of Nepal requires a twofold analysis. Firstly, study the efficacy of conventional peg with INR, its alignment, sustainability, and reserves management. Secondly, analysis of the flexible exchange rate system with USD and other convertible currencies. While both are intertwined, the latter seems rationale to assess if the exchange rate reflects economic fundamentals. This is relevant, as there is no existing research work that analyses the USD-NPR exchange rate. Thus, this paper focuses on the second aspect with an empirical application as follows:

.  To identify major macroeconomic determinants of the  USD-NPR exchange rate.

.  To analyze the long-run relationship between USD- INR and USD-NPR exchange rate.

.  To estimate inter-dynamics among major economic  variables and draw some policy implications.

The second section of this paper contains a review of relevant literature, followed by methodology and empirical analysis in the third section. The fourth section concludes with some policy suggestions.

2. Literature Review

The exchange rate plays a role of a catalyst for macroeconomic stability. In an integrated world, it is a matter of concern for all. While for policymakers, the exchange rate is important to determine output and inflation in the economy, it is important for fund managers to plan their asset allocation (Bhanumurthy, 2006). Nations have adopted different variations of alternative regimes based on their economic objective. Though a fixed exchange rate encourages foreign trade, and investment and restrains inflation, it tends to have a long-term misalignment in the real exchange rate, requires high foreign exchange reserves, and results in inflexibility of economic policy. Similarly, though a flexible exchange rate helps to secure efficient allocation of resources and provides independence to monetary policy, it results in exchange rate volatility and possible speculative attack. This can have an adverse influence on foreign trade and investment in the short run and turn affect economic efficiency and growth in the medium and long run (Gylfason, 2000).

Time horizon is another important facet to determine factors that influence the exchange rate. There are microstructure approaches that mainly focus on the short term, called the order and flow approach. Under this, perspectives of spot-rate dynamics, bid-ask spread, and portfolio shift models try to explain the cursory movement of the exchange rate. The exchange rate dealers in India are of the view that two factors predominant affecting India’s exchange rate are speculation and central bank intervention (Bhanumurthy, 2006). Similarly, in Vietnam exchange rate intervention plays an important role to curb exchange rate volatility; however, it has a little influence on exchange rate movement (Ding & Wang, 2022) However, this may not hold in all the circumstances because all players in the market do not adapt to the same model as their expectation may vary across time period (Frankel & Froot, 1995). Thus, though the real exchange rate in the short run might get deviated due to sharp fluctuations and price rigidity, in the long run, it is determined by international trade and adjustment in the current account (Kaltenbrunner, 2015). Some literature argues that macro fundamental theories have no role in determining short-run exchange rate movement (Evans & Lyons, 1999).

However, the rationale for the fluctuation of the exchange rate in the short run and long run are analyzed differently. The long-run exchange rate determination adopts a macro- approach, known as the asset price-based approach or the stock approach (Chinn, 2012). Under this, most theories state that the exchange rate, in the long run, reflects macroeconomic fundamentals of an economy such as GDP growth, international trade, currency convertibility, inflation, interest rate, bond prices, cumulative trade, and economic policies (Frenkel, 1976; Dornbusch, 1976; Mundell, 1963; Fleming, 1962; Chinn, 2012).

There have been numerous studies on the determinants of the exchange rate in the long run. Studies before the 1970s were studied with the fixed price assumption, where the price level between two countries was assumed static. Mundell- Fleming (MF) model known as the IS-LM-BP model, an extension of the Keynesian IS-LM model, states that in a small economy with perfect capital mobility and static price level where the domestic interest rate is determined by world interest rate, exchange rate adjustment occurs through current and capital flows (Fleming, 1962; Mundel, 1963).

The collapse of the Bretton woods system in 1971 led to the adoption of a floating exchange regime by many industrial countries and in this shift, evolved monetary models. The augmentation of these models as compared to fixed price assumption models was that they included the effect of money demand and purchasing power parity such that inflation influenced the macroeconomic adjustments. It states that the exchange rate depends on the relative price of two currencies and the price, in turn, is dependent on the demand and supply of the stocks of money. The flexible price states that the Purchasing Power Parity (PPP) holds continuously both in the short-run and long-run. However, the sticky price model (Dornbusch, 1976) states that since the goods market reacts sluggishly in comparison to the money market, PPP holds only in the long run.

However, post-Keynesian economists often criticize some versions of the monetary approach for their robustness (Harvey, 1991). For instance, as per the monetary model, higher-income results in stronger currency; however, a higher interest rate induces weaker currency. This is in contradiction to the static version of the MF model that states that higher income induces imports and weakens currency due to capital outflow; a higher interest rate induces capital inflow resulting in stronger currency (Chinn, 2012). However, Taking Fisher’s relation into account, an increase in interest rate will result in a weakening of the currency, as inflation differential will feed the interest rate differential in the case of flexible price monetary models, as PPP holds both for the short-run and the long run. But, the impact of the money market gets adjusted immediately against the sluggish goods market, a reason that causes overshooting. Thus, the exchange rate is positively influenced by inflation while negatively influenced by the interest rate.

The main determining factor of the exchange rate as per portfolio balance models is risk premium due to imperfect substitutes of assets that are dominated in different currencies. Though nations have perfect capital mobility, this capital is not a perfect substitute. Thus, the main difference between monetary and portfolio models was the influence of variables on the long-run exchange rate. The formal stated macroeconomic fundamental is the main determinant, while later established capital flows have the most important role (Barbosa et al., 2018).

The role of financial variables in determining exchange rates has been more prominent than with increasing integration of cross-border holding of assets and liabilities (Mundell, 1963). In the case of a high portfolio recipient country like India, Sensex6 is the most influencing variable in determining India’s exchange rate (Raju & Sharma, 2013). Several other recent studies have taken variables like capital flows, BoP position, GDP at factor cost, foreign exchange reserves, forward exchange rate, and central bank intervention to identify determinants of the exchange rate. (Monica & Santhiyavalli, 2017; Dua & Rajan, 2010). However, in the case of Saudi Arabia, though oil constitutes a major component of the trade basket, it did not have any influence on exchange rate movement (Ben Dhiab et al., 2021).

In recent times, the role of externalities has been another important factor that explains exchange rate movement. (Gongkhonkwa, 2021) studies the influence of the COVID-19 pandemic on the Thai Baht exchange rate and concludes that, as confirmation of COVID cases increases in Thailand, the Baht depreciated with the Chinese yuan (CNY), Malaysian ringgit (MYR), Singapore dollar (SGD), Vietnamese dong (VND), Australian dollar (AUD) and Taiwan dollar (TWD) however, appreciated with the United States dollar (USD), Japanese yen (JPY), Indonesian rupiah (IDR) and Hong Kong dollar (HKD). Similarly, the author further examines the impact of the pandemic taking new cases and death casualties and their role in the movement of the Thai Baht with major other trading currencies.

There have been limited studies on Nepal’s exchange rate policy. Most of the studies focus on the relationship of the exchange rate with one or more economic variables such as GDP, trade, remittance, export performance, trade balance, and so on. For instance, a study empirically identifies that the Real Effective Exchange Rate (REER) affects economic activity through the aggregate demand channel in the Nepali economy. Thus, the authors emphasize a need to keep REER constant for a favorable trade situation (Thapa, 2002). Similarly, another study concludes that the exchange rate directly impacts both cash flows and trade flows; thus, depreciation of the nominal exchange rate results in improvement of foreign exchange reserve and trade deficit (Adhikari, 2018). However, in contradiction (Chaulagain, 2015) uses Johansen’s cointegration test to analyze the J-curve phenomena in the case of Nepal and conclude that there is no such relationship. The study rather indicates that there is no scope to address trade imbalance through currency devaluation. Unlike trade, however, the impact of the exchange rate on remittance has been found positive. Nepal’s remittance inflow is determined by the nominal USD-NPR exchange rate and economic activity in host countries, mainly in India and the Gulf countries, indicating spillover of economic growth. This implies that when exchange rate volatility has a direct impact on the liquidity of the economy due considerable size of remittance inflow (Budha & Pant, 2016).

The pegged regime with India has been another cornerstone of studies undertaken. (Maskay, 2001) uses the binomial probit model to analyze the probability of adjustment of Nepal’s exchange rate policy as against INR. The empirical study identifies that while relative interest rate has significant effects on Nepal’s exchange rate probability, money and output growth do not have significant effects. Similarly, Pandey (2014) uses the BEER approach of Clark and MacDonald and also Johansen integration VAR models to assess possible REER misalignment and empirically identify that the exchange rate has been overvalued consistently from 2005 to 2008.

3. Data and Methodology

3.1. Data Source

The time-series data of quarterly frequency has been taken for analysis. Data is obtained from NRB quarterly financial highlights from 2004–05 to 2019–20 based on the Nepali calendar7. However, the USD-INR nominal exchange rate is taken from the RBI Handbook of Statistics. The data obtained from NRB was not tabulated on a time series at the source, thus, each cross-sectional observation had to be extracted from a respective quarterly publication. Further, data of Nepal Stock Exchange (NEPSE) Index data for 2004 and 2005 were taken from the Securities Exchange Board of Nepal (SEBON) as these data were not included in the NRB quarterly report.

3.2. Model Specification

The causal analysis of the USD-NPR exchange rate has been analyzed using multiple regression based on the Ordinary Least Square (OLS) approach. The model is chosen as per the monetary approach to exchange rate determination. Monetary models in comparison to portfolio and hybrid are preferred because the latter models have poor empirical implications that many studies could not even find signs of coefficients of bond price correct. Thus, taking USD-NPR as a dependent variable, choices of variables have been ascertained predominantly as per the existing theories. Most of the previous studies have taken a combination of one or more variables such as money stock, total output, interest rate differential, inflation differential, aggregate trade, bond prices differential, and cumulative trade balance as explanatory variables to understand the determination of exchange rate in the long run based on macro approach (Chinn, 2012).

Among the variables, inflation rate, interbank rate as a proxy of interest rate, Nepal stock market (NEPSE) index, Gross Domestic Product (GDP), and total accumulated foreign exchange (forex) reserves are considered for this study. The rationale behind taking GDP instead of trade is that, primarily, net factor import abroad (NFIA), a component of GDP, proxies trade component, and in addition, it is a better proxy of overall economic performance. Further, the reason behind taking forex reserves is to capture the net transfers such as remittance, which is a substantial figure, nearly one fourth of Nepal’s GDP. Similarly, the stock market index influences the exchange rate in many developing economies as they invite capital inflow through portfolio investment. In the case of Nepal the Foreign Investment and Technology Transfer Act, 2019 has given a conditional leeway for foreign investors to participate in the capital market. Moreover, due to the porous border with the possibility of labor and capital from India, their impact on the stock market seems possible.

In the regression estimation, the USD-NPR exchange rate, foreign exchange reserves, GDP, and NEPSE index have been log-transformed to address the issue of skewness and Heteroscedasticity. However, the percentage figure data, namely, interest rate and inflation rate, have been taken as it is to capture the actual causation. The lag effect on the equation has been determined based on apriori information. An appropriate dummy (dum) for 2007–08 Q4 has been taken into account for the global financial crises.

4. Results

4.1. Unit Root Test

Time series data need to be stationary to avoid spurious results. Thus, the unit root testing of each of the time series variables has been examined using the Augmented Dickey- Fuller (ADF) test using the equation below.

\(\Delta Y_{t}=\psi Y t-1+\sum_{i=1}^{p} \alpha_{i}+\Delta Y_{t-1}+u_{t}\)       (1)

In the equation above, Yt is the relevant time series variable,  Δ is the first difference operator with ut stochastic error component.8 As per the unit root testing, the following results have been obtained (Table 1).

Table 1: Unit Root Test Results

4.2. Empirical Analysis Results

The OLS regression result is obtained as below:

Δ(lnnexr) = –0.315(–1.244) + 0.003inf***(2.639) + 0.027(Ingdpt–1) – 0.005inf** – 0.079Δ(In NEPSE)***(–2.639) + 0.062Δ(Infxrt–1) + 0.058063dum(1.862)

R2 – 0.294 \(\bar{R}^{2}-0.218\)  F-statistic–3.884*

D-W Stat–2.006       (2)

*** Significant at 1%.

From the regression result above, inflation, interest rate, and NEPSE index are significant explanatory variables at a 99 percent confidence interval. The expected signs of all the variables as per economic theory except for foreign exchange reserves. The equation above explains only 21.8 percent of exchange rate variation which means that the unexplained component of an exchange rate movement is beyond the variables chosen. In terms of statistical validity, any linear combination among two independent variables was not found to be highly correlated; hence the issues of multicollinearity do not exist in the equation. This is indicated by the models themselves, as there are 3 significant variables and the adjusted R-squared is also low. Further, the D-W statistic of 2.00 depicts that the issue of auto-correlation of residuals does not exist to influence the regression results significantly. In addition, as per the Breusch-Pagan-Godfrey test, the probabilistic chi-square value was 0.3809, which means that the equation above does not have issues of Heteroscedasticity. Further, the Root Mean Squared Error (RMSE) is 0.046, which indicates a high predictive capability of a model to the extent it explains causality.

The significant impact of the NEPSE index on the exchange rate could be due to the transmission effect arising from capital movement from India, increasing foreign direct investments (FDI) over the years, and unsterilized increase in money supply due to remittance receipts. The main reason why inflation is significant, though with a minimal impact, is the price differential with the rest of the world. Nepal’s trade competitiveness has been deteriorating over the years as domestic production has been replaced by foreign goods, causing an increase in imports and a decrease in exports. The steep rise in imports over the years has contributed to the depreciation of the Nepali currency. Similarly, the possible reason for the significance of interest rate can be attached to increased money supply from informal transactions through porous open borders and remittance receipts through informal sources10. One peculiar phenomenon that is observed in the case of Nepal is poor macro-economic linkages resulting in the oscillating behavior of variables. For instance, while credit has been increasing nearly around 18 percent annually over the last 20 years, this transmission in GDP is nowhere closer, as it has grown on an average at only 4.4 percent per annum. Thus, stimulated interest rate due to high credit flow in the unproductive sector must have impacted the exchange rate through the asset price channel11. Further, GDP and, accumulated foreign exchange in the regression equation are not significant variables, their impact exchange rate is not statistically valid to interpret.

In a floating exchange rate system, if the exchange rate is not determined by major economic variables, what determines the USD-NPR nominal exchange rate? Since NPR is pegged with its largest trading partner India, does it have any owing to the determination of Nepal’s exchange rate? To explore this question, a graph is displayed below that depicts the residuals of the OLS regression model obtained from equation (2) and the quarterly rate of change in the USD-INR exchange rate (Figure 6).

Figure 6: Residual Plot Vs. USD-INR Exchange Rate (Quarterly Rate of Change)

The graph above (Figure 6) displays a striking similarity between the movement of residuals of equation (2) above and the rate of change of the USD-INR rate. This states that an unexplained portion of the exchange rate movement has a very close relationship with USD-INR. A causal analysis taking USD-INR as an exploratory variable in equation (2) was carried out that showed significant influence ( R 2–0.87) with very high coefficient (0.89). However, the regression suffered from very high auto-correlation with the D-W statistic (3.21), questioning the statistical validity of the model.

4.3. Long-Run Relationship Between USD-INR and USD-NPR Rate

To identify the long-run equilibrium relationship between USD-NPR and USD-INR rate, Engle and Granger’s (1987) cointegration method based on residuals has been adopted. For this, we first ensured that both variables were of the same order integration. The test of stationarity was performed using the ADF test are identified that both the series are stationary at I(1). Then, the data in level form was taken to estimate static long-run relationship using the fully modified OLS regression as below:

lniexrt = B0 + B1lnnexrt + εt       (3)

From the equation above, the long-run relationship is obtained as below obtained:

The adjusted R2 is below the D-W stat, as per the thumb rule, indicates that regression may not be spurious. To ensure this, we performed a unit root test of the residuals. The residual series should be stationary at I(0) without any intercept and trend to conclude that regression equation (4) is not spurious and there exits is a long-run equilibrium relationship between the series. The stationarity result of residuals is presented as follows (Table 2).

Innexrt = 0.47 + 0.99 Iniexrt + εt       (4)

\(\begin{aligned} &(24.73) \\ &\bar{R}^{2}-0.99 \end{aligned}\)      \((2418) D-W Stat =2.16\)

Table 2: The Stationarity Result of Residuals

*MacKinnon (1996) p-values.

As per the result above (Table 2), we determine that since residuals are stationary, there exists a long-term equilibrium relationship between USD-NPR and USD-INR rate, which means India’s exchange rate determines Nepal’s exchange rate in the long run. This empirically concludes the graphical representation in Figure 6. Even if any of the market forces influence the USD-NPR rate in the domestic market, such change is likely to be emanating from India (Thapa, 2002).

In a dual regime, alignment in both peg and floating should be in accordance with economic fundamentals. Since the INR-NPR rate is a nominal anchor, the policy focus is evident on defending this rate with unlimited convertibility. Due to this, conventionally, it is believed that if the pegged rate is defended, the flexible rate is automatically adjusted. This holds in the case of an open economy with free capital flow. However, with restricted capital flow and strong qualitative measures like moral suasion to restrict the usage of pegged currency12, the exchange rate regime seems not to be in accordance with the economic fundamentals.

The regression results above prove that the USD-NPR rate has some anomalies with economic fundamentals13. Thus, two issues become evident from this study. Firstly, since USD-INR determines the floating rate, is the INR- NPR peg rate rightly valued? This question is more pertinent because, in an efficient system, the translated value from USD-INR to USD-NPR should be consistence with currency competitiveness that is intrinsic to economic performance. Secondly, is USD-INR fairly valued? The possibility of misalignment because India follows a managed floating regime, and the RBI actively intervenes in the foreign exchange market. Cheng and Orden (2006) have analyzed India’s exchange rate misalignment from 1980-to 2005 to assess its impact on the agricultural sector and identified episodes of misalignment, such that the gap between the actual and equilibrium exchange rate narrows in the later time period of the study. A floating regime without free capital flow and an efficient market system cannot give a fair picture of currency competitiveness.

4.4. Inter-Dynamics Among Macroeconomic Variables

The second aspect of this paper is to understand the macroeconomic interlinkages of major economic variables (Sims, 1980) of Nepal. Theoretically, macroeconomic variables influence each other. For instance, GDP is influenced by its lag, inflation, interest, rate, exchange rate, stock market performance, and so on. This interlinkage among economic variables is important for policy decisions as they directly ascertain policy transmission and even repercussions. Thus, to identify the inter-dynamics of variables, we have taken Vector Autoregressive Model (VAR) developed (Sims, 1980). As per Sims’ critique, when the 2 variables are contemporaneously correlated, the endogenous variables can be regressed as follows:

y1t = β10 + β11y1t−1 + α11y2t−1 + u1t        (5)

y2t = β20 + β21y2t−1 + α21y1t−1 + u2t       (6)

Where y1 and y2 are contemporaneously correlated variables. To understand Nepal’s macroeconomic environ- ment, eight stationary transformed variables namely interest rate and inflation in actuals, GDP, NEPSE index, forex reserves, remittance, USD-NPR exchange rate, and trade deficit (multiplied by –1 to convert into positive value) after log transformation, have been considered for analysis. Lag lengths of 5 have been chosen based on Akaike Information Criteria (AIC).

4.5. Endogenous Relationship

The endogenous relationship of lags based on VAR results has been populated as follows (Table 3).

Table 3: The Endogenous Relationship

*Significant at 5%.

From the table above (Table 3), it is observed that, while some variables are significantly influenced by the past performance of other variables, some do not have a strong relationship. For instance, inflation is not influenced by its own lags or any lag of any other variables. On the contrary, Foreign exchange reserves seem to have a strong relationship with all the variables. It is influenced by its own 3 and 4 lag, second lag of NEPSE index, 1 and 4 lag of interest rate, 3 and 5 lag of remittance, 1, 4 and 5 lag of inflation, 1, 3 and 4 lag of GDP, 1, 3 and 4 lag of USD-NPR exchange rate, 1, 2 and 4 and 1, 2 and 4 lag of trade deficit. This is consistent with the theory, as all forex reserves are majorly dependent on overall economic performance.

Among the other variables, GDP, and interest rate trade deficit are other variables that seem to be significantly influenced by other variables. GDP is influenced by its own preceding lag, second lag of interest rate and forex reserves, 2 and 4 lag of USD-NPR exchange rate, 4 and 5 lag of inflation, and fifth lag of remittance. Similarly, the interest rate is influenced by its own third lag, the first lag of forex reserves, and the fourth lag of exchange rate and trade deficit. Moreover, a trade deficit is influenced by all of its lag except for the fourth one, the first lag of GDP, the third lag of exchange rate, the fourth lag of interest rate, 2 and 4 lag of inflation, and 2 and 3 lag of NEPSE Index.

4.6. Impulse Response Function

As an extension of the VAR model analysis, we have further analyzed responses of Nepal’s exchange rate against each of the macroeconomic variables. The impulse response function is graphically represented as follows (Figure 7).

Figure 7: Response of Nepal’s Exchange Rate Against Each Impulse Variable

In an endogenous model with multiple variables and numerous lags, it becomes noisy to understand input-output shock among any two given variables. However, for this, the impulse response function is used to analyze the response of a variable when a positive shock of one standard deviation is given an impulse variable, by employing vector auto regressive models. In the figures above, we have tried to analyze the response of the USD-NPR exchange rate with an impulse on macroeconomic variables taken for VAR analysis.

In the first figure, the appreciation of the exchange rate as a response to a positive shock in interest rate is visible in the first quarter itself, this corrects slightly in the second quarter, overshoots in the third quarter and stabilizes gradually. As per the theory, the exchange rate reacts instantaneously to changes in the money market. Similarly, in the adjacent figure, a positive shock on inflation shows depreciation of the exchange rate as per the theory. It corrects within the third quarter and normalizes from then on. It implies that when inflation increases, the exchange rate instantaneously depreciates and takes to a quarter of the time to converge back to normal indolently.

In the third figure, when a positive shock is given to GDP, the exchange rate depreciates, negligibly instead of appreciation. It immediately converges back to normal in the second quarter and stabilizes, showing no significant causal relationship between GDP and exchange rate in the first four quarters. This is consistent with the result achieved from OLS regression, reiterating that GDP was no significant role determinant of the exchange rate in the case of Nepal. Similarly, the inverse relationship between the NEPSE index and the exchange rate is visible in the fourth figure (first figure in the second row). It is visible that the instant positive shock on the NEPSE index has a slight and steady effect on the exchange rate in the first three quarters, before converging in the fourth quarter.

In the fifth figure, the exchange rate shows no response to positive shock given to forex reserves in the first quarter. As per the theory, exchange rate and reserves should have an inverse relationship, however, in the second quarter, the relationship is positive that corrects gradually from the third quarter. Perhaps, this anomaly is due to a non-significant causal relationship between forex reserves and exchange rate in the case of Nepal. Similarly, in the sixth figure, the exchange rate depreciated instead of appreciation in the first quarter, when a positive shock is given to remittance flow. However, this was corrected to the large extent in the second quarter itself, indicating a lagged influence.

In the last figure, the exchange rate shows a positive response to one standard deviation positive shock on the trade deficit. Since, the trade deficit was a negative series originally, later positively transformed (by multiplying the series by –1) for log application, it is a negative response and contrary to the existing theory. Though the series corrects by the third quarter itself, the relationship is inconclusive in the case of GDP. Further, the J-curve phenomenon was not visible in the case of Nepal and this was found consistent with Chaulagain (2015).

5. Conclusion and Policy Implications

There have been very few studies on the exchange rate regime of Nepal. This paper attempts to explore a novel aspect of Nepal’s dual exchange rate regime by identifying major determining factors of the USD-NPR exchange rate. Further, the paper also empirically establishes the long term relationship between India and Nepal’s exchange rate, studies the inter-dynamic relationship among major macroeconomic variables, and draws some policy inputs through impulse analysis. The causal relationship through OLS regression demonstrates that though diminutive, interest rate, inflation rate, and NEPSE index are significant variables that determine exchange rate in the long run. However, the long-run cointegrating relationship between USD-NPR and USD-INR, along with graphical residual movement analysis concludes that India’s nominal exchange rate movement is the major force in determining USD-NPR exchange rate movement.

Further, there is a reasonable interaction among macroeconomic variables. Forex reserves are highly sensitive to inflation, GDP, exchange rate, and trade. The requirement of forex reserves in a pegged regime cannot be undermined, thus, its sensitivity is a matter of concern for Nepal. On the contrary, inflation not being influenced by its own lag or lag of any other variables indicate that it is a standalone variable that has only unidirectional causation to GDP, NEPSE index, forex reserves, and trade deficit. The exchange rate is a nominal anchor of monetary policy. Inflation, interest rate, and exchange rate do not show a strong relationship indicating a possible challenge for policymakers to obtain efficient transmission through monetary policy instruments. Further, the trade deficit predominantly influenced by its lags indicates the trade trap, more precisely the export competitiveness trap.

The exchange rate in long run must be majorly determined by the macroeconomic variables. However, when flexible exchange rate movements do not reflect fundamentals and are determined predominantly by the performance of a pegged currency even in the long run, it strongly indicates an anomaly in a currency. Thus, this study indicates that NPR might not be rightly valued. Primarily, the possible reason for such anomaly could be pegged rate alignment itself. Since USD-INR flexible exchange rate majorly determines USD-NPR, it is more to do with the rate at which USD gets translated from INR to NPR. The INR-NPR pegged rate has remained unchanged for the last 27 years, despite significant changes in the economic landscape. A separate study is required to identify possible misalignment of pegged rate, its sustainability, and policy implications to optimally determine a better regime. The sustainability of Nepal’s peg rate given that total foreign currency reserves hover only around $10–12 billion can be a question for discussion. Thus, when small economies peg with larger trading partners, if misaligned, it needs adjustment as per fair value reflected by fundamentals. Otherwise, the economic repercussions like an increased proportion of shadow economy and export competitiveness trap will prevail, as is persisting in Nepal, which can eventually lead to inevitable currency crises.

An immediate solution for policymakers is to ensure that macroeconomic variables and output shock are aligned with the pegged country to avoid economic ramifications. Further, the flexible exchange rate must have a very close translation difference to avoid arbitrage opportunities. This invites twin challenges for a country like Nepal. Firstly, to keep currency in alignment secondly, to keep economic performance and shock too in alignment. However, in the long run, there is no option but to constantly assess currency competitiveness and revalue/devalue as per fundamentals, change the regime itself to a softer peg or adopt a floating regime.

Finally, a peg with strongly intervened14 currency itself invites a deeper question. What if pegged currency itself is not fairly valued? If this holds, Nepal could succumb to importing misalignment from India, even if the pegged rate remains in equilibrium. While this may be stable for bilateral trade and transfer, it may not be optimal. This issue applies to other countries like Bhutan, Lesotho, Eswatini, and Namibia, who have a conventional pegged arrangement with managed floating currencies like Indian Rupee and South African Rand; only countries that are pegged with currencies other than USD, Euro, or composite basket15.

참고문헌

  1. Adhikari, D. (2018). Impact of exchange rate on the trade deficit and foreign exchange reserve in Nepal (NRB Working Paper no. 43). Kathmandu: Nepal Rastra Bank. https://www.nrb.org.np/contents/uploads/2019/12/NRB_Working_Paper-NRBWP-43-March_2018
  2. Barbosa, L. O. S., Jayme, F. G., & Missio, F. J. (2018). Determinants of real exchange rate in the long run for developing and emerging countries: A theoretical and empirical approach. International Review of Applied Economics, 32(1), 62-83. https://doi.org/10.1080/02692171.2017.1332017
  3. Ben Dhiab, L., Chebbi, T., & Alimi, N. (2021). The asymmetric effect of oil price shocks on economic growth and real exchange rate in Saudi Arabia. Journal of Asian Finance, Economics and Business, 8(12), 295-303. https://doi.org/10.13106/jafeb.2021.vol8.no12.0295
  4. Bhanumurthy, N. R. (2006). Macroeconomic fundamental and exchange rate dynamics in India. Economic and Political Weekly, 18, 1101-1107.
  5. Budha, B., & Pant, B. (2016). Remittances and exchange rate linkages: Experiences of Nepal (Working Paper No. NRBWP-33). Kathmandu: Nepal Rastra Bank. https://www.nrb.org.np/contents/uploads/2019/12/NRB_Working_Paper-NRBWP-33-July_2016
  6. Chaulagain, M. (2015). Devaluation and testing the J-curve hypothesis: A case of Nepal (NRB Working Paper No. W27). Kathmandu: Nepal Rastra Bankhttps://www.nrb.org.np/contents/uploads/2019/12/NRB_Working_Paper-NRB-WP27-Devaluation_and_Testing_the_J-curve_Hypothesis-Dr_Mahesh_Chaulagai.pdf
  7. Cheng, F., & Orden, D. (2006). Exchange rate misalignment and its effects on agricultural producer support estimates (PSEs) in India. Queensland, Australia: International Association of Agricultural Economists.
  8. Chinn, M. D. (2012). Macro approaches to foreign exchange determination. Hoboken, NJ: Wiley.
  9. Ding, X., & Wang, M. (2022). The effectiveness of foreign exchange intervention: Empirical evidence from Vietnam. Journal of Asian Finance, Economics, and Business, 9(2), 0037-0047. https://doi.org/10.13106/jafeb.2022.vol9.no2.0037
  10. Dornbusch, R. (1976). Expectation and exchange rate dynamics. Journal of Political Economy, 84(6), 1161-1176. https://doi.org/10.1086/260506
  11. Dua, P., & Rajan, R. (2010). Exchange rate policy and modeling in India. Oxford, UK: Oxford University Press.
  12. Engle, R. F., & Granger, C. W. J. (1987). Co-integration and error correction: Representation, estimation, and testing. Econometrica, 55(2), 251-276. https://doi.org/10.2307/1913236
  13. Evans, M., & Lyons, R. (1999). Overflow and exchange rate dynamics (NBER Working Paper no. 7317). Cambridge, MA: National Bureau of Economic Research. https://www.nber.org/system/files/working_papers/w7317/w7317.pdf
  14. Fleming, M. (1962). Domestic financial policies under fixed and floating exchange rates. Washington, DC: International Monetary Fund.
  15. Frankel, J., & Froot, K. (1995). Chartists, fundamentals, and trading in the foreign exchange market. Americal Economic Review, 80(5), 181-185.
  16. Frenkel, J. A. (1976). A monetary approach to the exchange rate: Doctrines aspect and empirical evidence. Scandinavian Journal of Economics, 78(2), 220-224. https://doi.org/10.2307/3439924
  17. Gongkhonkwa, G. (2021). COVID-19 pandemic: Impact on Thai baht exchange rate. Journal of Asian Finance, Economics, and Business, 8(7), 121-127. https://doi.org/10.13106/jafeb.2021.vol8.no7.0121
  18. Gylfason, T. (2000). Fix or flex? Alternative exchange rate regime in the era of global capital mobility. North American Journal of Economics and Finance, 11, 173-189. https://doi.org/10.1016/S1062-9408(00)00033-4
  19. Harvey, J. T. (1991). A post-Keynesian view of exchange rate determination. Journal of Post Keynesian Economics, 14(1), 61-71. https://doi.org/10.1080/01603477.1991.11489878
  20. Kaltenbrunner, A. (2015). A post-Keyensian framework of exchange rate determination: A Minskyan approach. Journal of Post Keynesian Economics, 38(3), 426-448. https://doi.org/10.1080/01603477.2015.1065678
  21. Maskay, N. M. (2001). The impact of Indian monetary policy and the official exchange rate on Nepalese monetary policy. NRB Economic Review, 15, 44-63.
  22. Maskay, N. M. (2007). A study of trends in Nepalese-Indian currency exchange rate during the period of 1932-1960. NRB Economic Review, 25, 11-23.
  23. Monica, S., & Santhiyavalli, D. G. (2017). Determinants of the exchange rate of the Indian rupee against the US dollar. International Journal of Commerce and Management Research, 3(1), 54-58.
  24. Mundell, R. A. (1963). Capital mobility and stabilization policy under fixed and flexible exchange rates. Canadian Journal of Economics and Political Science, 29(4), 475-485. https://doi.org/10.2307/139336
  25. Nepal Rastra Bank (NRB). (1981). NRB in 25 years. Kathmandu: Nepal Rastra Bank.
  26. Nepal Rastra Bank (NRB). (2018). Sixty years of Nepal Rastra Bank. Kathmandu: Nepal Rastra Bank.
  27. Pandey, A. (2014). Exchange rate misalignment in Nepal. Journal of South Asian Development, 9(1), 1-25. https://doi.org/10.1177/0973174113520581
  28. Paudel, R. C., & Burke, P. J. (2015). Exchange rate policy and export performance in a landlocked developing country: The case of Nepal. SSRN Journal, 11, 41. https://doi.org/10.2139/ssrn.2596165
  29. Raju, R., & Sharma, N. S. (2013). Analysis of India's exchange rate under the new economic policy regime. Arthshastra Indian Journal of Economics & Research, 16, 27-34.
  30. Sims, C. (1980). Macroeconomics and Reality. Econometrica 48(1), 1-48. https://doi.org/10.2307/1912017
  31. Thapa, N. B. (2002). An econometric analysis of the impact of real effective exchange rate on economic activities in Nepal. NRB Economic Review, 14, 17-36.