سیاست های پولی و مالی و هزینه های توقف های ناگهانی

This article investigates the effects of monetary and fiscal policies on output growth during sudden-stop balance of payments crisis in emerging markets and developing countries. Sudden stops in capital flows, and subsequent deep recessions, are a frequent occurrence in these countries but there is no professional consensus, and little systematic empirical evidence, shedding light on the macroeconomic policy mix most likely to limit output losses during these episodes. To address this issue, we investigate 83 sudden-stop crisis in 66 countries using a baseline empirical model to control for the various determinants of output losses during sudden stops. We measure the marginal effects of policy on output losses, and find strong evidence that monetary tightening (rise in the discount rate or unsterilized rise in international reserves) and discretionary fiscal contraction are significantly correlated with larger output losses following a sudden stop. Fiscal expansion is associated with smaller output losses following a sudden stop, but monetary expansion has no discernable effect. The macroeconomic policy mix associated with the least output loss during a sudden-stop financial crisis is a discretionary fiscal expansion combined with a neutral monetary policy.

The “sudden stop” of international capital inflows to developing and emerging-market economies has become a major disruptive factor in several recent financial crisis. The sudden-stop problem features an abrupt cessation in foreign capital inflows and/or sharp capital outflows leading to a balance of payments crisis. A growing literature suggests that the collapse of investment and financial intermediation resulting from sudden stops is the main component of the very dramatic output collapses that have periodically hit many developing and emerging-market economies. More than one hundred sudden stops in capital inflows may be identified over the past twenty-five years, with an average output loss by one measure approaching almost 10 percent of GDP.1 Calvo et al. (2002), for example, provide a sudden-stop interpretation for the recent crisis in Argentina in which the capital flow reversal together with dramatic real exchange rate depreciation significantly worsened the government's fiscal position, led to a debt default, and an output collapse. Hutchison and Noy (2006) show that sudden stops have severe consequences for the economy, as the abrupt reversal in foreign credit inflows in conjunction with a realignment of the exchange rate typically cause a sharp drop in domestic investment, domestic production and employment. In a broader historical examination, Bordo et al. (2001) argue that the sudden-stop problem has become more severe since the abandonment of the gold standard in the early 1970s. The IMF financial assistance programs signed by Thailand, Korea, and Indonesia during the 1997–1998 Asian financial crisis generated a very heated debate about the appropriate use of discretionary or activist fiscal and monetary policies during a crisis situation. The IMF policy recommendation, which was incorporated as an integral part of the conditionality agreements in their loan packages, called for fiscal and monetary tightening. This was articulated clearly by the IMF First Deputy Managing Director at the time, Stanley Fischer. One of the most prominent critics of this prescription was Joseph Stiglitz, then Senior Vice President and Chief Economist of the World Bank. This public disagreement on such a key policy issue among the leading economists at the two major Bretton Woods institutions was unprecedented. Fischer argues that the prescription of tight fiscal and monetary policy is justified by the fact that the governments that entered a crisis usually face large budget deficits and high inflation. When describing Thailand, Indonesia and Korea, Fischer writes that: “The macroeconomic parts of these programs consist of a combination of tight money to restore confidence in the currency and a modest firming up of fiscal policy to offset in part the massive costs of financial restructuring” (Fischer, 1998, p. 103). Providing further detail, he writes: “On the appropriate degree of fiscal tightening, the balance is a particularly fine one. At the onset of the crisis, countries needed to firm up their finances, both to cover the costs of financial restructuring, and—depending on the balance-of-payments situation—to reduce their current account deficits, which depend in part on the budget deficit” (Fischer, 1998, p. 105). Stiglitz, by contrast, agrees that the key monetary component is restoring confidence but argues that confidence arises out of a good macroeconomic environment and not tight policies in the midst of a financial crisis. A healthy growth rate is the best indicator, in his view, to bolster confidence and a prescription of tight money and high interest rates will do exactly the opposite. He notes that “… maintaining tight monetary policies has led to interest rates that would make job creation impossible even in the best of circumstances” (Stiglitz, 2002, p. 17). Thus, by making the recession even deeper, the policy ends up reducing confidence in the economy rather than enhancing it. Stiglitz, 1999a and Stiglitz, 1999b terms this the ‘beggar-thyself’ policy. Regarding the Asian financial crisis, he writes: “… contractionary fiscal and monetary policies combined with misguided financial policies led to massive economic downturns, cutting incomes, which reduced imports and led to huge trade surpluses, giving the countries the resources to pay back foreign creditors.” (Stiglitz, 2002, pp. 107–8). To date there is no professional consensus, based on theory or empirical studies, on which approach is more conducive to achieving growth targets following a sudden stop in capital inflows. Aghion et al. (2004) and Lahiri and Végh (2007), for example, in theoretical papers, examine the impact of monetary policy on currency crisis and conclude that contractionary monetary policy (an interest rate defense) might result in greater output contraction.2 In contrast, Christiano et al. (2004) conclude from their theoretical work that when there are frictions in adjustment in the traded goods sector, an expansionary monetary policy during a financial crisis might be welfare reducing. Similarly, Céspedes et al. (2004) and Cúrdia (2007) look at exchange rate policy during currency crisis and conclude that a flexible regime is Pareto superior.3Razin and Sadka (2004) offer an analysis of fiscal policy in a debt crisis and describe the conditions under which increasing the budget surplus might not help even if the original trigger for the crisis was government debt; while Mitra (2006) introduces an equivalent examination and concludes that the impact of fiscal policy on the growth outcome depends on the flexibility of production. Little empirical work has addressed the optimal policy response to a financial crisis. We attempt to fill this gap in the literature. In particular, we consider “sudden-stop” financial crisis and investigate the wide range of monetary, fiscal and exchange rate policy responses to 83 crisis that have been occurred over 1980–2003 in 66 countries. The paths of economies at the time of sudden stops vary widely (Table 1), with about 65 percent of sudden-stop crisis followed by an output contraction, and about 35 percent of the cases followed by an output expansion. In the extremes, at least by one measure (defined below), output varies from cumulative output loss (relative to trend) of over 80 percent of GDP to a cumulative gain of over 20 percent of GDP.

However, it is not clear what factors, and especially which government policies, have contributed to the wide diversity of outcomes. Some examples illustrate this point. During the 1982 Latin American crisis, Bolivia sharply contracted both money and fiscal policy, while Chile held monetary policy steady and only instituted a mildly contractionary fiscal policy. Both countries, however, experienced sharp declines in output – Bolivia on the order of 24% of GDP and Chile around 28% of GDP. In response to sudden stops, Malaysia pursued a fiscal expansion and a neutral monetary policy stance (1997) while Venezuela pursued a monetary expansion and a steady fiscal policy (1994). Both countries, despite the differences in their policy responses, experienced significant output declines. Moreover, it is not even obvious from casual observation of the aggregate data how policies are linked to output losses during sudden stops. The top panel of Table 1 shows the number of observations (frequencies) associated with output contraction and expansion, in the columns, against observations of fiscal contraction and expansion in the rows.4 About 2/3 of sudden stops are associated with output contractions. Of the 44 episodes of output contraction during sudden stops, 17 were associated with fiscal contraction and 27 with fiscal expansion. Of the 20 cases associated with output expansion, 8 cases were associated with fiscal contraction and 12 cases with fiscal expansion. Similarly, no simple story is apparent from monetary policy responses, shown in Table 1. The great majority of cases of either output contraction or output expansion were not associated with a significant change in monetary policy (73 percent and 89 percent, respectively). Monetary contractions were only followed in about 9 percent of sudden-stop episodes, and expansionary policy in about 13 percent of the cases. Only 5 episodes of the cases with output declined were associated with monetary contractions, and a similar percentage of monetary contractions were followed when output expanded during sudden-stop episodes. The summary statistics reported in Table 1, and illustrative cases discussed above, suggest that a number of factors, working simultaneously, have influenced the evolution of output following sudden-stop crisis. Our objective is to investigate the effects of macroeconomic policies on the path of output following sudden stops while controlling for a host of variables that are also likely to play an important role. No study to date has explored this issue using a broad range of crisis experiences. Rather, the extant literature typically considers a series of case studies. These provide very valuable insights but it is difficult to derive general conclusions from them. In our work we focus on sudden stops, since these are the crisis that have been most costly and the response to them the most controversial. The central issue we address is the effects on output from alternative macroeconomic policy responses to a sudden-stop crisis. We consider the effectiveness of monetary and fiscal policy responses, as well as various combinations of policy responses, in mitigating the output losses usually associated with financial crisis. We employ regression methods in our empirical investigations (cross-section of sudden-stop crisis episodes) to control for the wide variety of factors potentially affecting output paths of economies, and formally test several hypotheses on the effects of contractionary (expansionary) macroeconomic policy responses to financial crisis. We find that contractionary monetary and fiscal policies during a financial crisis are linked to larger economic downturns. Expansionary fiscal policy is associated with smaller output losses following a sudden stop, but expansionary monetary policy has no discernable correlation. Section 2 reviews the literature on sudden stops and highlights our contribution. Section 3 presents the basic empirical model. Section 4 discusses the data employed in the study, and Section 5 reports summary statistics on key macroeconomic variables and the primary empirical results of the study. Section 6 concludes.