تاثیر نوسانات نرخ ارز بر عملکرد تجارت اندونزی در دهه 1990

Whether a real devaluation ultimately proves to be expansionary or contractionary depends on whether the boost given to the exportables sector offsets any possible output-depressing effects that may accompany the expenditure-switching policy. Failure of the exportables sector to adequately respond to the price incentives is a virtual guarantee that devaluation will be contractionary. This appears to have been the experience of Indonesia, the country worst hit by the crisis of 1997–1998. This paper explores whether the increased exchange rate variability of the Indonesian rupiah post 1997 may have been a cause for the country's poor export performance. J. Japanese Int. Economies18 (2) (2004) 218–240.

The spate of financial crises in emerging economies over the last decade has often resulted in the collapse of US dollar pegs. While pegs have sometimes been “hard,” more often than not they have been “soft” in the sense of not being backed by any institutional arrangements. This was the case in Southeast Asia in 1997–1998. In principle, Thailand and the other regional countries were supposed to have adopted basket pegged regimes, with the US dollar, Japanese yen and other currencies receiving weights consistent with their respective significance in economic linkages with the Southeast Asian countries. However, in reality, the US dollar had the overwhelming weight de facto, leading McKinnon (2001) and others to make frequent reference to the region's “dollar standard” (Table 1; also see Ito et al., 1998 and Rajan, 2002).A great deal of attention has been paid to the factors that have led to the crisis and eventual devaluation (i.e., are crises “self-fulfilling” or “fundamentals-based”?).1 There is also growing recognition of the need to better comprehend the post devaluation output dynamics (Rajan, 2001). While the first and second genre of models may disagree about why a crisis occurs, both are agreed that the devaluation signals the end of the crisis; the nominal devaluation, if translated into a real one, will give a much needed boost to the exportables sector and thus aggregate output.2 This was the case in Brazil, for instance, following the devaluation of the real in January 1999. On the other hand, the experiences of Mexico, East Asia and elsewhere have, by all indications, been quite painful, with severe output losses. These events have awakened us to the idea that there may be an intense recessionary threat associated with devaluation, at least in the short-term (Rajan and Shen, 2001). Thus, Dooley and Walsh (2000) have recently commented “(w)e are unsure why some crises are followed by… periods of economic recession while others are not” (p. 3). While there are a multitude of channels via which a devaluation could be contractionary (Bird and Rajan, 2001 and Rajan and Shen, 2001), whether a devaluation ultimately proves to be expansionary or contractionary depends on whether the boost given to the exportables sector offsets the output-depressing effects (Krugman, 1999). Any failure of the exportables sector to adequately respond to the price incentives is a virtual guarantee that devaluation will be contractionary. This appears to have been the experience of Indonesia, the country worst hit by the crisis of 1997–1998 (Fig. 1). Far from stimulating export growth, a severe depreciation of the rupiah against the US dollar in 1997–1998 resulted in an outright collapse of the country's exports (Fig. 2 and Fig. 3). Despite the fact that the rupiah fell by an average of 0.8 percent per day in nominal terms (against the US dollar) between July 1997 and January 1998, Indonesia's total exports of merchandise goods (in US dollars) declined by 8.5 percent at the end of 1998 compared to 1997.3 In volume terms, Indonesia's merchandise exports experienced an average annual drop of 14 percent between Q2: 1998 and Q1: 1999, with the worst annual decline occurring in the last quarter of 1998 (close to 20 percent).4There is, of course, the open question as to whether Indonesian exports comply with the Marshall–Lerner conditions. Studies for emerging economies have generally found foreign trade price elasticities to be sufficient to ensure an improvement in the trade account (Wilson, 2001). To the best of our knowledge, no study has focused specifically on the trade performance of Indonesia during the recent crisis.5 However, it is revealing to note that rupiah devaluations in the recent past have helped stimulate exports. For instance, despite a rupiah devaluation of around 28 percent against the US dollar in early 1983, the non-oil exports (in US dollar) grew by 27 percent that year and 17 percent in the next, compared to a decline of 13 percent in 1982. When the rupiah was again devalued by 31 percent in September 1986 to counter the export stagnation, non-oil exports (in US dollar) rose by 11 percent in that year and over 30 percent in each of the next two years (Rosner, 2000). So devaluations in Indonesia have historically provided the necessary export and growth impetus. There was no such export lift in 1997–1998. Why? There are two reasons that have most commonly been offered. First, Indonesia was not alone in devaluing its currency, other regional economies also simultaneously doing so. The rupiah devaluation may have failed to boost exports as no significant competitive price advantage may have accrued to Indonesia (i.e. phenomenon of “competitive devaluations”). Duttagupta and Spilimbergo (2000) find that competitive devaluation played a key role in exacerbating the real effects of the crisis in the East Asia through the trade channel. Second, given the recessionary conditions faced by the region, even if there was any positive price effect on exports, it may have been more than offset by negative income effects. However, these caveats ought to apply as much to the other crisis-hit Southeast Asian economies as they might to Indonesia. For comparison, merchandise exports of Malaysia, the Philippines, and Thailand experienced sharp falls in 1997 by 27 percent, 9 percent and 30 percent, respectively. But in 1998, exports rebounded in all three economies, with impressive rates of 33 percent for Malaysia, 67 percent for the Philippines, and 60 percent in the case of Thailand (all in US dollar terms) (Fig. 3). Indeed, the US dollar value of Indonesia's merchandise exports in 1999 was still below its level in 1996, unlike the three neighboring economies. This suggests a need for an alternative rationalisation for Indonesia's abysmal export performance post-devaluation. Two obvious explanations for this disappointing export performance appear to be favored by policy makers. First, that the collapse of the domestic financial sector which accompanied the currency collapse (due to the balance sheet effects as well as an outright bank panic) caused severe cuts in trade finance and prevented local producers taking advantage of the depreciated rupiah (Pardede, 1999). Second, there were adverse movements in the terms of trade of several of the country's key export commodities (Rosner, 2000). While there may well be an element of truth in both these reasons, another plausible explanation that has hitherto remained unexplored is the role of real exchange rate volatility. To be sure, past devaluations in Indonesia were all controlled ones in the sense that they involved a re-pegging of the rupiah at a new rate to the US dollar. In contrast, the devaluation of 1997–1998 was followed by a massive shift to one of relatively greater regime flexibility (Fig. 2).6 There has concomitantly been an intensification in the country's real exchange rate volatility which in turn may have had a detrimental impact of Indonesia's trade. As will be discussed in more detail, estimates of conditional variance confirm that the volatility of the real effective exchange rate of rupiah between February 1998 and July 2001 increased by more than thirty five times from its average in January 1994 to June 1997. The aim of this paper is to test these price, income, and volatility channels by estimating a set of export and import functions for Indonesia. In particular, we are interested in understanding the implication of the volatility of rupiah's real exchange rate on both the country's exports and imports. This is the basic question we try to answer in this paper. An important caveat is in order. The economic crisis faced by Indonesia has been accompanied by an acute political crisis and instabilities which in turn further deepened the overall economic crisis ( Rosner, 2000 and Siregar, 2001). The simultaneous economic and socio-political turmoil in 1998 invariably tends to contaminate the data and analysis, clearly making it extremely difficult to separate the role of exchange rate volatility and other crisis related factors in explaining the performance of exports and imports of Indonesia. Therefore, in order to address the important policy issue at hand, we choose to concentrate our analysis on the pre-crisis period between Q1: 1980 and Q2: 1997. We then extrapolate the conclusions reached for this period to the post-crisis period to answer the question as to whether the rise in currency volatility during the post-1997 crisis might be expected to adversely hamper the performance of Indonesia's trade, especially in 1998 and 1999. The more stable and conducive political environment in the pre-1997 period ought to provide us with a more reliable set of results that will be useful in understanding the post-crisis on goings in Indonesia. The remainder of the paper is organized as follows. The next section offers a brief overview of the main empirical literature on the impact of exchange rate volatility on trade. Section 3 is devoted to describing the data series and defining the various terms and variables to be used in the empirical analysis. We pay particular attention to defining and measuring exchange rate volatility. We construct two commonly used measurements of exchange rate volatility, viz. a Moving Average standard deviation introduced by Kenen and Rodrik (1986) and a GARCH model. Discussions on Johansen cointegration test results are provided in Section 4. The roles of the two volatility indices on Indonesia's total non-oil merchandise exports and imports are both considered. To further enhance our analysis on non-oil merchandise imports we decompose imports into capital and intermediate products. We also specifically test the impact of currency volatility on Indonesia's bilateral merchandise exports to Japan and imports from Japan (which is Indonesia's single largest trading partner). The final section offers a summary section and some concluding observations.

While there are a host of factors that could lead a crisis-induced devaluation (i.e., a devaluation following a currency crisis) to be contractionary, a necessary condition for economic recovery is that exports are boosted. Exports are supposed to be the engine of growth following such an expenditure switching policy. Nonetheless, despite the dramatic decline in the nominal value of the rupiah since mid 1997 (which has in turn been translated into a real devaluation), exports did not show a stable and strong pick up even four years after the break of the crisis; if anything, just the reverse happened. Indeed, the latest developments and outlook for Indonesia's external trade sector for 2002 continue to remain quite bleak (Siregar, 2001). The question we have explored in this paper is whether exchange rate volatility has had any detrimental impact on trade flows in Indonesia during the pre-crisis period. Our observation period has spanned Q2: 1980 to Q2: 1997. We have intentionally excluded the crisis period itself in order to circumvent problems related to structural breaks in the trade series which may be associated with various non-economic factors, like political uncertainty.15 Table 13 summarizes our regression results. Out of twelve regressions undertaken in this paper, nine cases indicate that exchange rate volatility adversely affected exports and imports performance of Indonesia during the pre-crisis period. Extrapolating these results forward, the rise in exchange rate volatilities should have played a critical role in explaining the poor performance of the trade sector in recent years. The adverse impact of exchange rate volatility on trade and the real sector may in part be the reason for the supposed “fear of floating” that has seemed to characterize many emerging economies.16 Recent financial crises involving emerging economies have called into question the wisdom of them adopting pegged exchange rates (be it “hard” or “soft”) and has strengthened the appeal of allowing for greater exchange rate flexibility. However, it is easy to overlook that flexible exchange rates bring with them their own problems. This surely has implications for the perennial issue of appropriate choice of exchange rate regime.