Effect of Exchange Rate Volatility on Exports: An Empirical Analysis of Disaggregated Data of the Indian Manufacturing Sector

Abstract

The effect of real exchange rate volatility on real exports is explored empirically in this article. The study uses disaggregated Indian manufacturing sector data consisting of seven categories: leather & leather manufactures, chemicals and related products, engineering goods, electronic goods, textiles (excluding readymade garments), readymade garments and other manufactured goods. The analysis is focused on India’s quarterly data from 2004Q2 to 2018Q2 using the ARDL bound test procedure. The ARDL bound test results show that real exports are co-integrated with fluctuations in real exchange rates and world real GDP. The study also reveals that exchange rate volatility has mixed effects on manufacturing exports both in the long run and short run. The impact of real-world GDP (WGDP) on real exports has been found positive and significant in the long run except for leather and leather manufacturers where it is negative and significant. In the case of the short-run, real-WGDP has mixed effects on exports.

JEL Classification: C22, F14, F31, F1

Keywords

ARDL Cointegration Exchange rate volatility Exports

I. Introduction

The high degree of volatility and uncertainty of exchange rate movements since the beginning of the generalised floating in 1973 have led policymakers and researchers to investigate the nature and extent of the impact of such movements on various macro-economic performance indicators. Since then, one of the most debatable issues is the impact of exchange rate fluctuation on trade in both developing and advanced countries. Exchange rate refers to the price of one country’s currency expressed in terms of the currency of another country. It is a key financial variable that affects decisions made by exporters, importers and policymakers. Advocates of fixed exchange rates argue that volatile and unpredictable exchange rates may lead to many harmful macroeconomic consequences such as volatility of prices and output, deterioration of total exports, as well as worsening the external competitiveness. On the other hand, proponents of floating exchange rate regimes believe that exchange rate flexibility helps the balance of payment adjustments in response to external shocks and positively influences the trade volume and economic growth. Before 1991 Indian Rupee was pegged to a basket of currencies dominated by the US Dollar. Due to the Balance of payment crisis in the 1990s, the Indian economy started opening up itself and the Reserve Bank of India (RBI) was forced to implement a set of market-oriented financial sector reforms and a paradigm shift from fixed to market-based exchange rate regime. The result of these reforms can be seen as an increased share of India in the global merchandise exports which was 0.6% in 1993 and 1.7% in 2018 (Ministry of Commerce and Industry, 2019). Share of manufacturing sector export was 74.9% in 1993 which reached 80.90% in 1999 though it started declining after that but still managed to remain at 73% of total exports (CMIE Economic Outlook database).¹ The introduction of the generalised floating system of exchange rate attracted researchers and policymakers to investigate the impact of exchange rate volatility on trade. Thereafter, so much discussion and research have been going on the subject but a consensus on the impact of exchange rate volatility on trade is still an issue of debate.

Theoretical Deliberation Regarding Impact of Exchange Rate Volatility on Trade Performance

Hooper and Kohlhagen (1978) and others did a theoretical investigation of the association between increasing exchange-rate volatility and international trade transactions. According to their argument, risk-averse traders reduce trade due to the higher cost of risk arrived through a more volatile exchange rate. Reasons might be firstly the difference between exchange rates at the time of order placement and delivery and payments. Since unpredictable changes in exchange rates made it uncertain to measure the exact profitability of the chosen transaction, resultantly international trade has lost its benefits and attractiveness, especially for risk-averse traders. Secondly, hedging facilities are not accessible to each country equivalently and wherever available it varied for each trader and investor within the country adding further costs and limitations.

Other theoretical findings imply that exchange rate volatility can have both negative and positive effects on trade volume depending on the situation. According to De Grauwe (1988), the primacy of income effects over substitution effects can lead to a positive link between trade and exchange-rate volatility. This is because, if exporters are sufficiently risk-averse, an increase in exchange-rate volatility increases the projected marginal utility of export revenue, causing them to increase exports. He claims that the impact of exchange rate uncertainty on exports should be proportional to risk aversion. When the risk is higher, a risk-averse exporter who is concerned about income loss may export more. A less risk-averse individual, on the other hand, may not be concerned with the worst-case scenario and, considering the return on exports is less appealing, may elect to export less when risks are higher. According to Bailey and Tavlas (1988), traders may be able to predict future exchange rate changes better than the typical foreign exchange market participant, and the gains from this information may be sufficient to cover the risk of exchange rate unpredictability.

Increased uncertainty from excessive volatility in currency rates can also influence foreign trade, according to theoretical models of hysteresis in international trade Baldwin and Krugman (1989). Effects that persist after the factors that caused them have been removed are referred to as hysteresis. Although these theoretical models illustrate that trade flows have hysteresis, they also suggest that hysteresis in trade flows can be explained by a mix of sunk costs (exit costs that are not recoverable by a foreign firm) and exchange-rate uncertainty. The ability of a foreign corporation to enter or exit is connected to exchange rate levels. Their findings imply that exchange-rate uncertainty can alter trade behaviour, including risk-neutral enterprises, when considerable sunk costs are involved in foreign transactions.

However, predicting how trade would be affected is challenging. Froot and Klemperer (1989) illustrate that when market share matters in an oligopolistic market structure, exchange-rate uncertainty can affect the price and quantity of trade, either positively or negatively, regardless of risk choices. In the presence of sunk costs, Dixit (1989) shows that the hysteresis band or zone of inaction grows as the exchange rate becomes more volatile and uncertainty promotes a wait-and-watch attitude among agents in this situation. As a result, trade can be affected due to no entry and exit of the firms in the market. In conclusion, because theory alone cannot establish the sign of the relationship between foreign trade and exchange-rate volatility, the impact of exchange-rate volatility on foreign trade is an empirical matter.

Empirical Deliberations Regarding Impact of Exchange Rate Volatility on Trade Performance

Some researchers have found no significant relationship between exchange rate fluctuations and trade, for example, in their empirical investigation, Hooper and Kohlhagen (1978) and De Grauwe (1988) examined the effects of exchange rate uncertainty on the volume of trade among developed industrialised countries using aggregate export data and found no significant relationship. Belanger et al. (1992) investigated sectoral US imports from Canada and found no statistically significant impact of exchange rate volatility on trade. Frankel and Wei (1993) investigated the relationship between exchange rate fluctuation on trade in the Pacific region using the gravity model and found a weak causal effect. Qian and Varangis (1994) found an ambiguous relationship between exchange rate volatility on trade. Haider and Adil (2017) used disaggregated data of Indian manufacturing sector exports and revealed that the real exchange rate has no significant impact on exports. Bhattacharyya and Rit (2018) examined the effect of nominal exchange rate volatility on Indian export and worked out that there is no direct effect but through domestic prices, the nominal exchange rate affects exports indirectly. Jyoti (2021) used aggregated Indian manufacturing export data and found that real exchange rate volatility has negative but insignificant impacts on real exports.

On the other hand, some researchers have found a significant negative relationship between exchange rate volatility on trade. Arize (1997) and Arize et al. (2000) revealed a negative significant impact of exchange rate volatility on trade in some less developed countries and some developed industrialised countries respectively. Chou (2000) used disaggregated data on industrial material export from China and worked out a significant negative effect of exchange rate volatility on trade. Cheung and Sengupta (2013) found a negative and strong significant impact of exchange rate volatility on trade using disaggregated data of Indian non-financial sector firms. Panda and Mohanty (2015) suggested a negative significant effect of exchange rate volatility on the real export of India. Baak et al. (2007) and Pino et al. (2016) explored a significant negative impact of exchange rate volatility on trade both in the long and short-run in some East-Asian countries.

Some authors worked out a positive relationship in their analysis. Koray and Lastrapes (1989) showed a positive weak causal effect of exchange rate volatility on US bilateral export. Asseery and Peel (1991) investigated a significant positive effect of exchange rate volatility on exports. McKenzie (1999) reviewed the literature for the last quarter of the twentieth century and found a mixed impact of exchange rate volatility on trade and noted that most of the empirical literature has focused basically on the measure of exchange rate volatility instead of trade data. The result related to the impact of exchange rate volatility using disaggregated trade data remains obscure.

After considering both theoretical and empirical research in this area it can be stated that exchange rate volatility can affect trade directly through uncertainty and adjustment costs and indirectly, through its impact on the structure of output and investment, as well as government policy. A large number of empirical research have already been carried out on the subject. An increase in exchange rate volatility is expected to harm trade flows and, as a result, the global economy’s overall health. However, neither theoretical models nor empirical investigations offer a definitive answer, leaving the produced results vague and inconsistent. From the review of literature, it is clear that most of the studies have focused on aggregated data on export and very few works have been done on India, especially in the manufacturing sector which has a share of approximately 73% of total export. In view of this, the study aims to examine the impact of exchange rate volatility on India’s manufacturing export.

There are four parts to this article. The first section is an introduction including a review of the literature on the effect of exchange rate fluctuations on trade. The study data and method are discussed in the second section. The third section analyses the findings. The concluding observations of the study are presented in the final section.

II. Methodology

Model Specification

In order to analyse the effects of real exchange rate volatility on real manufacturing exports of India, the study has used a simple time series model followed by Panda and Mohanty (2015) where real export of India is taken as a dependent variable, and the exchange rate volatility of India and world GDP (WGDP) as independent variables. The model can be described as:

(X_{t}) = Ω_{0} + Ω_{1} \ln W_{t} + Ω_{2} V_{t} + ε_{t}

(1)

Where In(X_t) is the natural logarithm of real exports of India at time t

In w_t is the natural logarithm of real-WGDP at time t

V_t is the real exchange rate volatility of India at time t

𝜖_t is an error term

Ω₀, Ω₁ and Ω₂ are the coefficients.

Data Sources and Variables

The analysis is based on quarterly data and covers a period from 2004Q2 to 2018Q2. Disaggregated data of manufacturing sector exports which consisted of seven categories include leather & leather manufactures, chemicals & related products, engineering goods, electronic goods, textiles (excluding readymade garments), readymade garments and other manufactured goods. The real effective exchange rate has been taken from Economy Outlook database of the Center for Monitoring Indian Economy (CMIE). Export Unit Value Indices² data have been taken from the Handbook of Statistics on Indian Economy published by the RBI. World income as a proxy of foreign economic activity for export demand equation is calculated using the geometric mean of GDP data of India’s 16 major export destination countries (USA, China, Hong Kong, UAE, UK, Bangladesh, Germany, Australia, Singapore, Belgium, Netherlands, Nepal, Republic of Korea, France, Saudi Arabia, Malaysia, Viet Nam) and has been taken from FRED (Federal Reserve Bank of St. Louis) database. WGDP and Unit value indices of export data have been collected on a yearly basis and then converted into quarterly data using the quadratic interpolation method following Arize et al. (2000, p. 16) due to the unavailability of quarterly data. The study selected 2011–2012 as the base year for all the series used in the model. All variables have been transformed into logarithms, except exchange rate volatility, to compress the scale on which variables are measured Gujarati (1995).

The variables used in this study are Real exports as a dependent variable and GDP data of India’s major export destination countries and real exchange rate volatility as an independent variable.

E_{x} = (\frac{e x}{e x u v i} \times 100)

Where Ex is the real exports of the domestic country to the world; ex is the quarterly nominal exports of the domestic country to the world measured by the US dollar; exuvi is the domestic country’s export unit value index.

Real foreign income is measured by geometric mean of GDP data of India’s 16 major export destination countries.

Real exchange rate volatility is calculated using the Moving Average Standard Deviation formula:

V r e e r = [\frac{1}{2} £_{i = 1}^{2} {(l r e e r}_{t + i - 1} {l r e e r}_{t + i - 2})^{2}]^{(\frac{1}{2})}

Where Vreer denotes the real effective exchange rate in log returns and the moving average at order 2. Moving average standard deviation has been chosen over other volatility measures as it is a time-varying volatility measure and is suitable for low-frequency data (Arize et al., 2000; Poon & Granger, 2003).

Estimation Strategy

The study employed the ARDL³ bounds test approach of co-integration developed by Pesaran and Shin (1998) and Pesaran et al. (2001) to estimate the impact of real exchange rate volatility on India’s real export of the manufacturing sector. The advantages of the ARDL model are: first, it lies in the fact that whether the series is stationary at a level I(0) or stationary at the first difference I(1) we can employ the ARDL model to test the long-run relationship between variables, while the other cointegration testing procedures (Engle & Granger, 1887; Johansen, 1988; Johansen & Juselius, 1990) requires the underlying variables to be I(1) or stationary at the level. Second, the ARDL model is more suitable for small samples (Boutabba, 2014), and a dynamic error correction model (ECM) which integrates the short-run dynamics with the long-run equilibrium can be derived from ARDL through a simple linear transformation (Mohanty & Bhanumurthy, 2014).

As mentioned earlier in the case of the ARDL bound testing approach it does not matter whether the variables are I(0) or I(1) or a mixture of both but, variables should not be I(2) (Pesaran et al., 2001). The study employed a standard Augmented Dickey–Fuller (ADF) test (Dickey & Fuller, 1981) of unit root to test the stationary properties of each variable. In addition to this test, we have performed the Multiple Breakpoint test developed by (Bai & Perron, 1998). This test is useful when there are multiple unknown breaks in the series (Balakrishnan & Parameswaran, 2007) for all the series in the model. For the selection of optimal lag length, the study has used the Akaike information criterion (AIC) (Akaike, 1974) as AIC lag length criteria are good for a small sample (Burnham & Anderson, 2002).

Results of the Co-integration

The computed F-statistic of the ARDL bound test is compared to the critical tabulated value by Pesaran et al. (2001). The critical value of the F-statistic is based on two sets, one set assumes that all the variables of the model are I(0), while another set assumes that the variables are I(1). Based on F-statistic, there are three conditions: the null hypothesis of no co-integration can be rejected if the calculated test statistic is greater than the upper bound. Second, the co-integration test becomes inconclusive, if the calculated F-statistic falls into the bounds. The error correction term would be useful for testing co-integration if the bound test result is inconclusive, according to Banerjee et al. (1998).

The third condition is that the null hypothesis of no co-integration cannot be discarded if the measured F-statistic is less than the lower bounds value.

With the ARDL method, Equation (1) is rearranged as follows to find the long-run relationship between variables:

Δ \ln (X_{t}) = Ω_{0} + \sum_{i = 1}^{j} Ω_{1} Δ {(l n X)}_{t - 1} + \sum_{i = 1}^{j} Ω_{2} Δ {(\ln W)}_{t - 1} + \sum_{i = 1}^{j} Ω_{3} Δ {(V)}_{t - 1} + q_{0} {(l n X)}_{t - 1} + q_{1} {(l n W)}_{t - 1} + q_{2} {(V)}_{t - 1} + ϵ_{t}

(2)

In Equation (2), ∇ denotes the first difference operator, and the terms with summation signs represent the short-run error correction dynamics, while the second part of the equation with q corresponds to the long-run relationship and the other variables of the equation are as defined earlier. The time trend variable is included in this equation to capture the autonomous time-related shifts. The ARDL model calculates the optimal lag period for each variable by estimating (j+1)k regressions, in which j is the maximum number of lags to use and k is the number of variables in the model. The following short-run ECM, Equation (3), is calculated after establishing the long-run relationship among variables in Equation (1):

Δ \ln (X_{t}) = Ω_{0} + \sum_{i = 1}^{j} Ω_{1} Δ {(l n X)}_{t - 1} + \sum_{i = 1}^{j} Ω_{2} Δ {(l n W)}_{t - 1} + \sum_{i = 1}^{j} Ω_{3} Δ {(V)}_{t - 1} + π E C M_{t - 1} + ε_{1}

(3)

Where ECM stands for error correction mechanism. The speed of convergence from the short-run to the long-run model is determined by the error correction term.

III. Results and Discussion

Unit-root Tests

The study employed the ADF unit root test to check the stationary properties of all variables (see Table A1). The tests are performed with both trend and intercept and only with intercept (without trend). The results of the ADF test indicate that all the variables are a mixture of stationary either at a level I(0) or with a first difference I(1) so, we can apply the ARDL model.

Structural Break Tests

As mentioned earlier the study conducted a Multiple Breakpoint test developed by (Bai & Perron, 1998) to test the unknown structural breakpoints present in all the models selected for the study. When we are unable to easily describe a structural break by referring to a well-known, major and destructive occurrence, the Bai–Perron multiple breakpoint tests provide an endogenous way to understand multiple breaks at unknown dates. The Bai and Perron multiple breakpoint test results are seen in Table A2.

Structural breaks around 2008–2009 were the reason for increased export in all the categories used for the study. Although 2008 was just the starting of the global meltdown luckily India was in the balance getting affected by the crisis due to its measured policy actions. Industrial production and GDP growth were positive and exports shifted to neighbours in developing Asia (Roach, 2010). On the contrary Structural breaks during 2015 was the outcome of a decline in exports in the manufacturing sector due to the dormant of contemporary measures to overcome the global recession (Chandrashekhar & Ghosh, 2016). A structural break in December 2010 was caused by a sharp increase in the export of electronic goods due to an improvement in demand from both the EU and the USA along with a successful entry into new markets by Indian exporters (Economic Times Bureau, 2011). Another structural break in September of 2013, for the export of leather and leather manufacturers, is the outcome of increasing shifts in the data as the government allowed the export of finished leather, wet blue and EI tanned leather through the airports in addition to the existing permitted land and sea ports (Ministry of Commerce and Industry, 2013). A structural break in March 2013 in the export of other manufacturing goods category was the reason for a reduction in gems & jewellery export as the major portion of other manufacturing category consist gems and jewellery industry. According to Gems and Jewellery Export Promotion Council (GJEPC), the decline was the reason for unavailability of the gold and softening of the global prices of precious metals (Ghosal, 2014). Structural break in December 2011 in the export of engineering goods was the reason for sharp decline trend due to less demand from the US and EU due to global meltdown as their share in total engineering goods export is approximately 40% (RBI, 2012).

Results of the Bound Tests

To estimate the long-run relationship among all the variables, the study used the bound test and ADF (AIC) criteria for the selection of optimal lag length for all the models. The calculated value of the F-statistic is more than the upper bound and significant at the 1% level for all the models which means there is a long-run relationship among the variables (see Table A3).

Results of the Long-run Effects

The outcomes of the long-run results are represented in Table A4. The results show that the long-run coefficient of real exchange rate volatility (VREER) is positive and statistically significant for exports of leather & leather manufacturers, chemicals & related products, exports of engineering goods and textiles (excluding readymade garments) exports. On the other side, real exchange rate volatility is negative and statistically significant for the exports of electronic goods, and other manufactured goods. For the export of readymade garments, it is negative but insignificant. Thus, the study finds out the almost positive impact of real exchange rate volatility on disaggregated data of real manufacturing exports.

In the case of real-WGDP effect on export, results show that it is positive and statistically significant at 1% level for the exports of leather & leather manufactures, chemicals & related products, exports of engineering goods, textiles (excluding readymade garments) and other manufactured goods. Only for the exports of readymade garments, the coefficient of WGDP is negative and significant. Thus, the effects of real-WGDP as a proxy of foreign income on real exports of manufacturing goods are positive and statistically significant. Accordingly, the results of this study are in line with other studies earlier on this issue.

Results of the Short-run Dynamics

The coefficient of the error correction term has a negative sign and is statistically significant at a 1% level for all the models, indicating that there is evidence of a long-run relationship between the variables for all the models used in the study (see Table A5). The coefficient of ECM is –0.7609 when the dependent variable is Rexleather and –0.7929, –0.7469, –0.2181, –0.6045, –0.8892 and –0.9768 when dependent variables are Rexchemical, Rexengineering, Rexelectronics, Rextextiles, Rexreadygar and Rexotherman, respectively. Thus, based on the coefficients of the error correction term, disequilibrium in the current period will be adjusted by 0.77, 0.80, 0.75, 0.22, 0.61, 0.89 and 0.98% in the following period for their respective models.

In the short-run real exchange rate volatility is showing a negative and significant impact only for Rextextiles and Rexotherman, and for other categories it is insignificant. WGDP is positive and significant for Rexchemical, Rextextiles. Only for Rexreadygar WGDP has negative and significant effects.

Diagnostic Tests

The study employed the Lagrange multiplier test of residual serial correlation; a heteroskedasticity test and the Jarque-Berra test for normality to check the reliability of the estimated model. The results suggest that all models selected for the study are free from serial correlation, heteroskedasticity and non-normality. In three models, the study could not reject the null hypothesis of non-normality as the p-value is less than 5% but non-normality is not a very big issue especially when other diagnostic tests are consistent with the model accuracy (see Table A6). Finally, to examine the stability of the models, the study used the cumulative sum of the recursive residuals (CUSUM) and cumulative sum of squares of the recursive residuals tests (CUSUMSQ).

CUSUM and CUSUMSQ of Squares Tests

The test results indicate they were located between the two red lines and could not exceed the critical value at a 5% level of significance. As a result, it may be concluded that the ARDL models calculated are accurate and stable (see Figures A1 –A14).

IV. Conclusion

The main objective of the study was to investigate the long-run and short-run relationship between real exchange rate volatility and real exports of manufacturing goods by India. The results reveal that real exchange rate volatility has almost positive effects on real manufacturing exports in the long run. In the short run, the result of exchange rate volatility and WGDP is almost insignificant. The real-WGDP has an almost positive and strong significant impact on real manufacturing exports in the long run which is similar to previous studies. The positive benefits of exchange rate fluctuation on manufacturing exports may be the driving force behind the effective use of hedging, which allows exporters to avoid foreign exchange losses while securing the targeted return. Short-term negative consequences may have been the source of uncertainty, but these were addressed in the long run, and the effects became positive. Consequently, the study finds no strong evidence of the harmful effects of exchange rate volatility on manufacturing exports in India. As a result, RBI intervention in the Indian forex market to smooth out exchange rate fluctuations cannot be justified on the grounds that it promotes manufacturing exports.

Appendix A

Table A1.

Estimated Statistics of Unit Root Test.

Augmented Dickey–Fuller Test Statistics
Variable	Test Type	Level	First Difference	Order of Integration
Rexleather	With intercept only	–0.1466	–3.6877*	I(1)
Rexleather	With trend and intercept	–1.3087	–3.8109**	I(1)
Rexchemical	With intercept only	–3.2183**	–8.3875*	I(0)
Rexchemical	With trend and intercept	–3.5532**	–8.5080*	I(0)
Rexengineering	With intercept only	–2.7509***	–10.1251*	I(0)
Rexengineering	With trend and intercept	–2.7600	–10.1998*	I(1)
Rexelectronics	With intercept only	–1.7381	–3.9615*	I(1)
Rexelectronics	With trend and intercept	–1.6501	–4.5271*	I(1)
Rextextiles	With intercept only	–3.3430**	–4.0446*	I(0)
Rextextiles	With trend and intercept	–3.5017**	–4.0490*	I(0)
Rexreadygar	With intercept only	–0.5243	–3.2242**	I(1)
Rexreadygar	With trend and intercept	–2.5665	–3.2719***	I(1)
Rexotherman	With intercept only	–1.9658	–7.9535*	I(1)
Rexotherman	With trend and intercept	–1.4666	–7.9456*	I(1)
WGDP	With intercept only	0.8882	–4.0009**	I(1)
WGDP	With trend and intercept	–2.1419	–3.7529**	I(1)
VREER	With intercept only	–4.4949*	–6.0825*	I(1)
VREER	With trend and intercept	–4.4665*	–6.0902*	I(1)

Source: The authors.

Notes: ***, ** and * denote the level of significance at 10%, 5% and 1%, respectively.

Rexleather = Leather & leather manufactures; Rexchemical = Chemicals & related products; Rexengineering = Engineering goods; Rexelectronics = Electronic goods; Rextextiles = Textiles (excluding readymade garments); Rexreadgar = Readymade garments; Rexothman = Other manufactured goods.

All variables are in natural logarithmic form.

Table A2.

Results of Bai–Perron Multiple Breakpoint Test.

Dependent Variable	Break Test	F-statistic	ScaledF-statistic	Critical Values	Break Dates
Rexleather	0 vs 1*	7.0571	21.1712	13.98	September 2013
	1 vs 2*	6.6768	20.0302	15.72	December 2008
	2 vs 3	4.2396	12.7188	16.83
Rexchemical	0 vs 1*	23.6661	71.0010	13.98	March 2015
	1 vs 2*	6.8368	20.5104	15.72	December 2007
	2 vs 3*	10.8261	32.4782	16.83	December 2009
	3 vs 4	1.5051	4.5150	17.61
Rexengineering	0 vs 1*	16.6122	49.8365	13.98	December 2011
	1 vs 2*	16.1457	48.4369	15.72	June 2015
	2 vs 3	1.4474	4.3421	16.83
Rexelectronics	0 vs 1*	76.7466	230.2396	13.98	June 2008
	1 vs 2*	51.1309	153.3925	15.72	March 2015
	2 vs 3*	5.7565	17.2695	16.83	December 2010
	3 vs 4	5.5352	16.6055	17.61
Rextextiles	0 vs 1*	9.9741	29.9223	13.98	June 2015
	1 vs 2*	7.8738	23.6214	15.72	December 2008
	2 vs 3	3.9216	11.7647	16.83
Rexreadygar	0 vs 1	2.8411	8.5235	13.98	No break date
Rexotherman	0 vs 1*	18.0310	54.0928	13.98	March 2011
	1 vs 2*	7.6250	22.8749	15.72	June 2015
	2 vs 3*	7.2010	21.6030	16.83	March 2013
	3 vs 4	2.8566	6.2570	17.61

Source: The authors.

Note: * denote the level of significance at 5%, Bai–Perron (2003).

Table A3.

Estimated Statistics of ARDL Bound Test.

Dependent Variable	Lag Order of Model	Optimal Lag	Calculated F-statistic
Rexleather	ARDL (5, 5, 5)	5	7.6663*
Rexchemical	ARDL (1, 2, 0)	2	21.265*
Rexengineerin	ARDL (1, 0, 0)	1	16.5418*
Rexelectronics	ARDL (5, 4, 1)	5	4.3517***
Rextextiles	ARDL (8, 8, 7)	8	7.6756*
Rexreadygar	ARDL (5, 3, 0)	5	8.5193*
Rexotherman	ARDL (4, 0, 2)	4	8.6165*

Source: The authors.

Notes: The table reports the F-statistics value of the cointegration test.

*** and * denote the level of significance at 10% and 1%, respectively.

The critical values bounds are according to Pesaran et al. (2001) and range 3.17–4.14, 3.79–4.85 and 5.15–6.36 at 10%, 5% and 1%, respectively.

Independent variables are world real GDP (WGDP) and real exchange rate volatility (VREER) in all the models.

Table A4.

Estimated Long-run Coefficients.

Dependent Variable	Regressor	Coefficient	T-statistic	P-value
Rexleather	WGDP	–2.2843*	–5.1330	0.0000
Rexleather	VREER	5.7322*	3.3653	0.0021
Rexchemical	WGDP	1.7222*	7.8428	0.0000
Rexchemical	VREER	1.8723**	2.4093	0.0200
Rexengineering	WGDP	4.4524*	7.5105	0.0000
Rexengineering	VREER	2.5408***	1.7541	0.0857
Rexelectronics	WGDP	2.2726	–1.3584	0.1824
Rexelectronics	VREER	–14.965***	–1.7457	0.0890
Rextextiles	WGDP	0.9586	1.4146	0.1726
Rextextiles	VREER	4.4015***	1.7718	0.0917
Rexreadygar	WGDP	–0.8857*	–5.8578	0.0000
Rexreadygar	VREER	–0.4861	–0.5833	0.5629
Rexotherman	WGDP	2.2434*	3.1102	0.0004
Rexotherman	VREER	–5.3609*	–2.7919	0.0079

Source: The authors.

Note: ***, ** and * denote the level of significance at 10%, 5% and 1%, respectively.

Table A5.

Short-run Error Correction Model.

Dependent Variable	Regressor	Coefficient	T-statistics	P-value
Rexleather	D(WGDP)	3.3319	1.1568	0.2565
	D (WGDP (–1))	3.7699	1.3374	0.1912
	D (WGDP (–2))	–3.9412	–1.3691	0.1811
	D (WGDP (–3))	–3.4668	–1.2111	0.2353
	D (WGDP (–4))	8.3898	3.0074	0.0053
	D(VREER)	0.0531	0.0729	0.9423
	D (VREER (–1))	–2.9858	–3.2146	0.0031
	D (VREER (–2))	–1.9999	–2.5475	0.0162
	D (VREER (–3))	–0.8471	–1.1257	0.2692
	D (VREER (–4))	–1.2375	–1.7182	0.0967
	D0812	0.04313	1.5090	0.1418
	D1309	0.1634	3.8587	0.0006
	CointEq (–1) *	–0.7609	–4.9530	0.0000
	R-squared: 0.8272, DW- statistics: 1.9859
Rexchemical	D(WGDP)	5.8618	2.4874	0.0166
	D (WGDP (–1))	3.6379	1.5640	0.1247
	D1503	–0.1697	–6.1966	0.0000
	D0712	0.1440	3.5804	0.0008
	D0912	–0.0838	–2.4433	0.0184
	CointEq (–1) *	–0.7929	–8.1588	0.0000
	R-squared: 0.6178, DW- statistics: 2.3143
Rexengineering	D1112	–0.2885	–5.7756	0.0000
	D1506	–0.3639	–5.7890	0.0000
	CointEq (–1) *	–0.7469	–7.1868	0.0000
	-squared: 0.5095, DW- statistics: 2.4117
Rexelectronics	D(WGDP)	–7.5788	–1.6794	0.1013
	D (WGDP (–1))	5.9773	1.2071	0.2348
	D (WGDP (–2))	–9.8707	–2.0153	0.0510
	D (WGDP (–3))	12.106	2.7878	0.0082
	D(VREER)	–0.6321	–0.5463	0.5880
	D1503	–0.2079	–3.7769	0.0005
	CointEq (–1) *	–0.2181	–3.7070	0.0007
	R-squared: 0.4662, DW- statistics: 1.8518
Rextextiles	D(WGDP)	8.0409	3.3736	0.0030
	D (WGDP (–1))	7.8973	2.8380	0.0102
	D (WGDP (–2))	–6.1404	–2.3741	0.0277
	D (WGDP (–3))	–1.9141	–0.6735	0.5083
	D (WGDP (–4))	–0.5023	–0.1669	0.8691
	D (WGDP (–5))	7.4912	2.4520	0.0235
	D (WGDP (–6))	3.8847	1.3638	0.1878
	D (WGDP (–7))	–8.8228	–3.7097	0.0014
	D(WGDP)	0.5923	1.0506	0.3060
	D (VREER (–1))	–3.5297	–4.6432	0.0002
	D (VREER (–2))	–3.2029	–3.9868	0.0007
	D (VREER (–3))	–2.5736	–3.0412	0.0064
	D (VREER (–4))	–2.4522	–3.5633	0.0019
	D (VREER (–5))	–1.9474	–3.1614	0.0049
	D (VREER (–6))	–1.6578	–2.9318	0.0082
	D0812	–0.0856	–3.0643	0.0061
	D1506	–0.1752	–5.8837	0.0000
	CointEq (–1) *	–0.6045	–5.0328	0.0001
	R-squared: 0.8837, DW- statistics: 2.2515
Rexreadygar	D(WGDP)	–6.1966	–1.6735	0.0256
	D (WGDP (–1))	6.8409	2.2040	0.0332
	D (WGDP (–2))	–4.4679	–1.6194	0.1130
	CointEq (–1) *	–0.8892	–5.1772	0.0000
	R-squared: 0.8168, DW- statistics: 1.9736
Rexotherman	D(VREER)	–2.9519	–2.6891	0.0103
	D (VREER (–1))	2.3833	2.3449	0.0240
	D1103	0.1523	2.8614	0.0066
	D1303	–0.2306	–3.3284	0.0019
	D1506	–0.4009	–4.7970	0.0000
	CointEq (–1) *	–0.9768	–5.2067	0.0000
	R-squared: 0.6382, DW-statistics: 1.8333

Source: The authors.

Note: D0812 = Dummy variable for break date December 2008, D1309 = Dummy variable for break date September 2013, D1503 = Dummy variable for break date March 2015, D0712 = Dummy variable for break date December 2007, D0912 = Dummy variable for break date December 2009, D1112 = Dummy variable for break date December 2011, D1506 = Dummy variable for break date June 2015, D1103 = Dummy variable for break date March 2011, D1303 = Dummy variable for break date March 2013.

* denote the level of significance at 1%.

Table A6.

Diagnostic Test.

Dependent Variable	Test Type	LM-Version	P-value	F-version	P-value
Rexleather	Serial correlation	2.3829	0.3038	0.5875	0.5642
	Heteroscedasticity	13.3928	0.8179	0.5777	0.8935
	Normality	0.7984	0.6709	(Not applicable)
Rexchemical	Serial correlation	3.08610	0.2137	1.3078	0.2807
	Heteroscedasticity	12.728	0.1215	1.7314	0.1165
	Normality	1.8225	0.4020	(Not applicable)
Rexengineering	Serial correlation	5.2254	0.0733	2.4670	0.0958
	Heteroscedasticity	5.2242	0.3891	1.0285	0.4114
	Normality	6.2051	0.0449	(Not applicable)
Rexelectronics	Serial correlation	1.2684	0.5304	0.4500	0.6411
	Heteroscedasticity	6.8618	0.9091	0.4443	0.9415
	Normality	69.936	0.0000	(Not applicable)
Rextextiles	Serial correlation	10.285	0.0058	2.4545	0.1141
	Heteroscedasticity	24.811	0.5850	0.7926	0.7172
	Normality	3.1775	0.2041	(Not applicable)
Rexreadgar	Serial correlation	8.4584	0.0763	1.7969	0.1502
	Heteroscedasticity	5.2180	0.8761	0.4573	0.9075
	Normality	0.8152	0.6653	(Not applicable)
Rexotherman	Serial correlation	4.6737	0.0966	1.8858	0.1653
	Heteroscedasticity	4.1130	0.9665	0.3135	0.9787
	Normality	28.4197	0.0000	(Not applicable)

Source: The authors.

CUSUM and CUSUM of Squares Tests of F(Rexleather/WGDP, VREER)

Figure A1.

Source: The authors.

Figure A2.

Source: The authors.

CUSUM and CUSUM of Squares Tests of F(Rexchemical/WGDP, VREER)

Figure A3.

Source: The authors.

Figure A4.

Source: The authors.

CUSUM and CUSUM of Squares Tests of F(Rexengineering/WGDP, VREER)

Figure A5.

Source: The authors.

Figure A6.

Source: The authors.

CUSUM and CUSUM of Squares Tests of F(Rexelectronics/WGDP, VREER)

Figure A7.

Source: The authors.

Figure A8.

Source: The authors.

CUSUM and CUSUM of Squares Tests of F(Rextextiles/WGDP, VREER)

Figure A9.

Source: The authors.

Figure A10.

Source: The authors.

CUSUM and CUSUM of Squares Tests of F(Rexreadygar/Wgdp Vreer)

Figure A11.

Source: The authors.

Figure A12.

Source: The authors.

CUSUM and CUSUM of Squares Tests of F(Rexothman/WGDP VREER)

Figure A13.

Source: The authors.

Figure A14.

Source: The authors.

Footnotes

Acknowledgement

We gratefully acknowledge the valuable comments and suggestions of the two anonymous referees and the editor of this journal for constructive feedback on an earlier version of this article. The usual disclaimer, however applies.

Declaration of Conflicting Interests

The authors received no financial support for the research, authorship and/or publication of this article.

Funding

The authors declared no potential conflicts of interest concerning the research, authorship and/or publication of this article.

Notes

ORCID iD

Jyoti

References

Akaike

(1974). A new look at the statistical model identification. IEEE Transactions on Automatic Control AC , 19(6), 716–723.

Arize

C. A.

(1997). Conditional exchange rate volatility and the volume of foreign trade: evidence from seven industrialized countries. Southern Economic Journal , 64, 235–254.

Arize

C. A.

, Osang

, & Slottje

J. D.

(2000). Exchange rate volatility and foreign trade: evidence from Thirteen LDCs. Journal of Business and Economic Statistics , 18(1), 10–17.

Asseery

, & Peel

D. A.

(1991). The effects of exchange rate volatility on exports: Some new estimates. Economic Letters , 37(3), 173–177.

Baak

S. J.

, Al-Mahmood

M. A.

, & Vixathep

(2007). Exchange rate volatility and exports from East Asian countries to Japan and the USA. Applied Economics , 39, 947–959.

Bai

, & Perron

(1998). Estimating and testing linear models with multiple structural changes. Econometrica , 66(1), 47–78.

Bai

, & Perron

(2003). Computation and analysis of multiple structural change models. Journal of Applied Econometrics , 18, 1–22.

Bailey

M. J.

, & Tavlas

G. S.

(1988). Trade and investment under floating rates: The US experience. Cato Journal , 8(2), 421–449.

Baldwin

, & Krugman

(1989). Persistent trade effects of large exchange rate shocks. The Quarterly Journal of Economics , 104(4), 635–654.

10.

Balakrishnan

, & Parameswaran

(2007, July). Understanding economic growth in India: A prerequisite. Economic & Political Weekly , 42, 2915–2922.

11.

Banerjee

, Dolado

J. J.

, & Mestre

(1998). Error-correction mechanism tests for cointegration in a single-equation framework. Journal of Time Series Analysis , 19, 267–283.

12.

Bélanger

, Gutiérrez

, Racette

, & Raynauld

(1992). The impact of exchange rate variability on trade flows: Further results on U.S. imports from Canada. North American Journal of Economics and Finance , 3, 888–892.

13.

Bhattacharyya

, & Rit

(2018). On the relationship between the nominal exchange rate and export demand in India. South Asian Journal of Macroeconomics and Public Finance , 7(2), 1–23.

14.

Boutabba

M. A.

(2014). The impact of financial development, income, energy and trade on carbon emissions: evidence from the Indian Economy. Economic Modelling , 40, 33–41.

15.

Burnham

P. K.

, & Anderson

D. R.

(2002). Model selection and multimodel inference (2 ed.). Springer.

16.

Chandrashekhar

C. P.

, & Ghosh

(2016, 21 November). Understanding India’s export collapse . The Hindu Business Line.

17.

Cheung

Y.-W.

, & Sengupta

(2013). Impact of exchange rate movements on exports: an analysis of Indian non-financial sector firms. Journal of International Money Finance , 39, 231–245.

18.

Chou

W. L.

(2000). Exchange rate variability and China’s exports. Journal of Comparative Economics , 28, 61–79.

19.

De Grauwe

(1988). Exchange rate variability and the slowdown in growth of international trade . IMF Staff Papers, 35, 63–84.

20.

Dickey

D. A.

, & Fuller

W. A.

(1981). Likelihood ratio statistics for autoregressive time series with a unit root. Econometrica , 49, 1057–1072.

21.

Dixit

(1989). Entry and exit decisions under uncertainty. Journal of Political Economy , 97(3), 620–638.

22.

Engle

R. F.

, & Granger

C. W. J.

(1887). Co-integration and error correction: Representation, estimation, and testing. Econometrica , 55(2), 251–276.

23.

Economic Times Bureau. (2011, 9 January). India’s exports at 33 month high, surge 36% in December. The Economic Times . https://economictimes.indiatimes.com/dateline-india/indias-exports-at-33-month-high-surge-36-in-december/articleshow/7244784.cms?from=mdr

24.

Frankel

J. A.

, & Wei

S. J.

(1993). Trade blocs and currency blocs [Working Paper No. 4335]. National Bureau of Economic Research.

25.

Froot

K. A.

, & Klemperer

P. D.

(1989). Exchange rate pass-through when market share matters. Americal Economic Review , 79(4), 637–654.

26.

Ghosal

(2014, 29 November). Gem and jewellery exports decline by 11 % in FY14. The Economic Times . https://economictimes.indiatimes.com/industry/cons-products/fashion-/-cosmetics-/-jewellery/gem-and-jewellery-exports-decline-by-11-in-fy14/articleshow/34382603.cms

27.

Gujarati

D. N.

(1995). Basic econometrics (3rd ed.). McGraw-Hill.

28.

Haider

, & Adil

M. H.

(2017). An analysis of impact of exchange rate volatility on the Indian manufacturing exports. The Empirical Economics Letters , 16(12), 1375–1382.

29.

Hooper

, & Kohlhagen

S. W.

(1978). The effect of exchange rate uncertainty on the prices and volume of international trade. Journal of International Economics , 8(4), 483–511.

30.

Johansen

(1988). Statistical analysis of cointegration vectors. Journal of Economic Dynamics and Control , 12(2–3), 231–254.

31.

Johansen

, & Juselius

(1990). Maximum likelihood estimation and inference on cointegration-with applications to the demand for money. Oxford Bulletin of Economics and Statistics , 52, 169–210.

32.

Jyoti . (2021). Impact of exchange rate fluctuations on India’s manufacturing exports: An empirical investigation on long-run relation. Journal of Asian Economic Integration , 3(1), 61–73.

33.

Koray

, & Lastrapes

W. D.

(1989). Real exchange rate volatility and the US. Bilateral Trade: A VAR Approach. Review of Economics and Statistics , 71(11), 708–712.

34.

McKenzie

M. D.

(1999). The impact of exchange rate volatility on international trade flows. Journal of Economic Surveys , 13(1), 71–106.

35.

Ministry of Commerce and Industry. (2013). Indian foreign trade . Public Information Bureau, Government of India.

36.

Ministry of Commerce and Industry. (2019). Indian global trade share . Public Information Bureau, Government of India.

37.

Mohanty

, & Bhanumurthy

N. R.

(2014). Exchange rate regimes and inflation: Evidence from India. International Economic Journal , 28(2), 311–332.

38.

Panda

, & Mohanty

R. K.

(2015). Effects of exchange rate volatility on exports: Evidence from India. Economic Bulletin , 35(1), 305–312.

39.

Pesaran

M. H.

, & Shin

(1998). An autoregressive distributed lag modelling approach to cointegration analysis. In Storm

(Eds), Econometrics and economic theory in the 20th century: The Ragnar Frisch Centennial Symposium (pp. 371–413). Cambridge University Press.

40.

Pesaran

M. H.

, Shin

, & Smith

R. J.

(2001). Bounds testing approaches to the analysis of level relationships. Journal of Applied Econometrics , 16(3), 289–326.

41.

Pino

, Tas

, & Sharma

S. C.

(2016). An investigation of the effects of exchange rate volatility on exports in East Asia. Applied Economics , 48, 2397–2411.

42.

Poon

S. H.

, & Granger

C. W. J.

(2003). Forecasting financial market volatility: A review, Journal of Economic Literature , 41(2), 478–539.

43.

Qian

, & Varangis

(1994). Does exchange rate volatility hinder export growth? Empirical Economics , 19, 371–396.

44.

Reserve Bank of India. (2012). India’s Foreign Trade 2011–12 .

45.

Roach

S. S.

(2010 12 February). Indian economic recovery in terrific shape: Stephen Roach, Morgan Stanley. The Economic Times . https://economictimes.indiatimes.com/opinion/interviews/indian-economic-recovery-in-terrific-shape-stephen-roach-morgan-stanley/articleshow/5561104.cms?from=mdr