سیاست های پولی در یک اقتصاد دلاری شده که در آن ترازنامه مهم است

Does the dollarization of liabilites and the resulting balance sheet vulnerability prevent monetary policy from serving its conventional countercyclical role? We study this question in a model of a small open economy in which domestic firms face an imperfect capital market, with risk premia depending on net worth as in Bernanke and Gertler [Am. Econ. Rev. 79 (1989) 14.]. In spite of the financial fragility channels present in the model, the conventional wisdom still holds: under a floating exchange rate, countercyclical monetary policy does help cushion the impact of foreign real shocks.

Monetary and interest rate policy remains the most contentious aspect of the response to the recent crisis in Asia and other emerging markets. Many analysts, led by the IMF's Stanley Fischer, contend that stopping the collapse of national currencies was priority number one; confidence, a reversal of capital flows, and growth would follow. Referring to the 1995 example of Mexico, Dornbusch (1998) wrote: Mexico fully implemented a stark US-IMF program of tight money to stabilize the currency and restore confidence. Starting in a near-meltdown situation, confidence returned and within a year the country was on the second leg of a V-shaped recovery. The IMF is unqualifiedly right in its insistence on high rates as the front end of stabilization. Not everyone agreed. The attack on high rates was spearheaded by Joseph Stiglitz, then the Chief Economist at the World Bank, whose objections included the “traditional criticism” of tough monetary policies to defend a fixed exchange rate, namely that they are too costly in terms of output or employment. But the concerns went further, since in East Asia and elsewhere such policies seemed not only to be painful, but also ineffective. Notably, the 1998 Global Economic Prospects published by the World Bank worried that high interest rates had little success in reducing pressure on currencies or stabilizing investor confidence, while at the same time imposing large output costs. This was the case whether the initial package entailed new agreements with the multilateral institutions (Indonesia, Korea and Thailand) or not (Malaysia and Philippines).2 That the Chief Economist of the World Bank should be disagreeing with his institution's own policies was peculiar. Even more peculiar was that this debate should be taking place at all. After all, monetary policies are supposed to be used countercyclically: in the Mundell–Fleming world, under flexible rates or an adjustable peg, a monetary expansion is called for to offset an adverse shock to productivity or world demand. But what the IMF and Dornbusch were advocating was a procyclical monetary policy: tightening in response to adverse shocks. They were not alone in this advocacy, for procyclical policies are apparently what many policy-makers prefer. Not only did the Asian countries eventually tighten in response to adverse shocks, both internal and external. In response to the 1997–1998 external shocks, most Latin American countries (including those that were nominally floating such as Mexico, Peru, and to a lesser extent Chile) also used tight money and high interest rates to prop up their currencies.3Gavin et al. (1999) and Calvo and Reinhart (2000) have documented this pattern in a systematic manner for most emerging markets. Why do policymakers behave the way they do, contravening conventional wisdom? Mostly because of the fear that devaluation would cause financial distress. Suppose that domestic firms have borrowed in dollars, and that at least some of them are in the non-traded goods sector and have earnings in local currency. Then a monetary expansion and the accompanying nominal devaluation can drastically increase the carrying costs of the dollar debt, generating a wave of bank and corporate bankruptcies. This danger has been stressed in some interpretations of the Asian crisis—particularly that of Corsetti et al. (1998) and Krugman (1999). In the coinage of the latter, it is balance sheet vulnerability that prevents a more relaxed monetary stance and the accompanying devaluation. While such balance sheet effects are plausible, economists are far from understanding them and their implications for monetary policy. Accordingly, this paper is an attempt to shed some light on this subject. We study whether, in a small open economy, balance sheet vulnerability can ever cause monetary policies to have the opposite effects than standard models predict. In particular, we ask whether contractionary balance sheet effects can more than offset the conventional Mundell–Fleming expansionary effect of unanticipated monetary growth. And whether, in the presence of balance sheet effects, monetary policy can be used in standard counter-cyclical fashion to cushion the domestic effects of real external shocks. The short answer is that, in the model we build, monetary policy can still do what it is supposed to do. Unanticipated increases in the money supply increase domestic output both on impact and, through an increase in investment, in the subsequent period. This feature of monetary policy can be used to offset real shocks, though not completely: there is no money supply response that can keep output constant in all periods. These conclusions hold in spite of the fact that devaluation has a harmful effect on the carrying cost of external debt. We study these issues in the context of a simplified, two-period version of the model in Céspedes et al. (2000, henceforth CCV). The setting is an open economy with home and foreign goods. The former is produced using the labor services supplied by wage-setting, monopolistic, labor suppliers. Capital accumulation is allowed and plays a crucial role in the analysis. To allow for balance sheet effects and related financial distortions, we assume that: ⋅ External debt is denominated in terms of foreign goods, and hence a real depreciation increases the debt burden in terms of home goods. This is the analytical equivalent of the “dollarization of liabilities” discussed by Calvo (1999) and others. ⋅ Foreign goods are inputs in domestic investment. As a consequence, a real depreciation makes domestic investment relatively more expensive and, as in Krugman and Taylor (1978) and Lizondo and Montiel (1989), may reduce investment and output. ⋅ Most importantly, domestic firms face an imperfect capital market, with risk premia depending on their net worth along the lines of Bernanke and Gertler (1989). Hence, a real devaluation could reduce the value of this net worth and thereby increase the risk premia faced by local firms when borrowing abroad. Monetary policy has real effects since money wages are set in advance. Assuming flexible exchange rates throughout, we focus on the transmission of unanticipated money supply changes to the rest of the economy in the presence of balance sheet vulnerability. This is in contrast with CCV, whose only concern was the choice between exchange rate regimes.4 Aside from its policy relevance, the theoretical interest of the exercise is that, in an open economy, the existence of dollarized liabilities means that monetary policy has effects on balance sheets that are absent in the closed economies studied by Bernanke and Gertler (1989) and others. In a closed economy an unanticipated monetary increase, by reducing real interest rates, raises asset prices (including the firm's own value) and boosts current income; both effects are expansionary, as they improve net worth. In an open economy, by contrast, monetary increases can potentially reduce net worth because the associated devaluation may increase the relative burden of existing dollar obligations. In our model, a temporary monetary expansion causes a temporary real exchange rate depreciation. The depreciation has the standard beneficial effects: it switches demand toward home goods and it reduces the nominal and real interest rate (more precisely, the own-rate of return on home goods), increasing investment and output. On the other hand, it has novel and contractionary effects: it causes financial distress because of the associated fall of net worth and it could potentially bring about an increase of the country risk premium. Nevertheless, these balance sheet effects are not sufficient to overturn the standard effects, and expansionary monetary policy turns out to be expansionary. We also consider the effects of monetary policy in the face of two kinds of shocks: temporary increases in the world real rate of interest and temporary declines in the rest of the world's demand for the home country's exports. In the absence of countercyclical policy, these shocks induce a persistent fall in investment and home goods output. We show that, in spite of the financial fragility channels present in the model, the conventional wisdom still holds: under a float, countercyclical monetary policy helps cushion the impact of the shocks. The paper is organized as follows. Section 2 presents the basic model. The model is highly nonlinear, but it is relatively easy to find a loglinear approximation; Section 3 discusses how. Taking advantage of such an approximation, Section 4 analyzes the effects of unanticipated monetary policy, and Section 5 analyzes monetary responses to real external shocks. Section 6 concludes.