Monetary and Exchange Rate Policies for Transition Economies

Ten years into transition, the choice of an exchange rate regime remains nearly as controversial as it was at the outset. Sharply different regimes continue to coexist, from currency boards in Estonia, Lithuania and Bulgaria to (relatively) free floating in the Czech Republic, Poland and Russia, to a narrowly fixed rate in Hungary, Latvia and the Slovak Republic. Economic performance also remains heterogeneous. Most countries had returned to positive growth rates by 1998 but only a handful had restored their real GDP level of 1989. Although there is no clear link between growth and the exchange rate regime, it is obvious that exchange rate policy remains a vexing and crucially important problem.

Initially, exchange rate policy was largely dominated by the trade-off between disinflation and external competitiveness. With few exceptions (Romania and Russia, mainly) inflation is no longer dominating the policy agenda. International pressure (IMF, OECD and EU) has contributed to keep low inflation as an important objective of policy, but it is not clear that such a preoccupation is warranted. The countries that have achieved the best growth performance since 1989 are those that have settled for an inflation rate in the 10–15% range. Once the exchange rate is no longer exclusively guided by its role as a nominal anchor, a number of complex issues arise.

The fifth, and final, report in CEPR's Economic Policy Initiative addresses the macroeconomic challenges faced by the Associated Countries (ACs) as they prepare to join the European Union and eventually its single currency. Specifically, the report addresses a number of key questions for the transition economies: what approach should policy-makers adopt with regard to developments in their real exchange rate? What, if anything, can these countries do to avoid external crises? To what extent must any credible monetary policy be underpinned by sound fiscal policy? And what nominal anchor should be adopted?

What is the Equilibrium Exchange Rate?

Irrespective of the exchange regime chosen, the authorities must have a view of the appropriate exchange rate level. With few exceptions, after an initial fall at the time when markets were established, the real exchange rates in the transition economies have undergone massive real appreciations. This is due to both a catch-up, following the initial exchange rate collapse, and an equilibrium real appreciation, which results from the rapid gains in efficiency that the transition process implies. So when is the equilibrium level reached?

The report provides estimates of equilibrium real exchange rates for the following transition economies: Bulgaria, Czech Republic, Estonia, Hungary, Latvia, Lithuania, Poland, Romania, Russia, Slovak Republic, Slovenia and the Ukraine. The authors regress a number of variables on monthly US dollar wages using low-frequency cross-country data drawn from all five continents – 85 countries with observations taken every five years from 1970-95. The infrequency of the data and the large sample size provide justification that the result from this regression can be interpreted as representing a benchmark equation that estimates the equilibrium value of the real exchange rate. This approach rests on the assumption that the same process drives the dollar wage worldwide and is tested, and subsequently validated, by the use of regional dummy variables.

In the regression, dollar wages are used as they avoid the index problem that is associated with prices – i.e. price indices cannot be compared across countries but wages can once they are converted into the same currency. The dollar wage is explained by the following variables: GDP per capita, age dependency ratio, openness to trade, government consumption, non-bank and bank net foreign assets, credit to the private sector and regional dummies. These variables explain 86% of the total variance of dollar wages across time and countries.

The benchmark regression is then used to produce estimates of the evolution of the equilibrium real exchange rate over the period 1990–97. The results show that in most of the ACs the estimated equilibrium dollar wage rises during the transition period. However, this is not the case in Bulgaria, Hungary, Russia and the Ukraine where it is actually declining, often sharply. These reductions reflect the sharp fall in public spending, the deterioration of the net external asset position and falling credit to the private sector. Deeper structural problems (declining GDP per capita and a deterioration in the age dependency ratio) are also found to play an important role in Russia and the Ukraine.

More important for policy purposes is the value of the actual exchange rate in relation to its equilibrium level. Until 1996, there is little evidence of overvaluation in any of the countries studied, despite fear frequently expressed in the wake of the massive real appreciation witnessed since 1990. By 1996, however, the equilibrium real exchange rate estimates suggest that Hungary, Poland, Romania and Slovenia may well have reached a situation where they are close to overvaluation.

Capital Inflows and Crises Prevention

The aftermath of a currency crash has become a familiar scene. Monetary authorities blame financial markets for devious behaviour, while financial markets accuse the authorities of mistaken policies. Popular economists blame both authorities and markets for poor judgement and soon develop interpretations which explain the crisis that they failed to anticipate. In the aftermath of the Asian and Russian crises, much effort is being devoted to the quest for early warning indicators. Yet there are good grounds to believe that this effort is unlikely to succeed.

Currency crises often correspond to clear policy mismanagement. These crises, termed 'first generation' crises in the literature, are usually clearly foreseen by careful observers: Mexico in 1994, the Czech Republic and Thailand in 1997, Russia in 1998 and Brazil in 1999 are all thought to belong to this category. Sometimes, however, currency crises are self-fulfilling. These 'second generation' crises (i.e. self fulfilling, multiple equilibria crises) seem to characterize the 1997 crises of Indonesia, Malaysia and Korea as well as much of the after-effect of the Russian crisis, including what happened in Poland and Hungary. They are, almost by definition, unforeseeable.

The Report presents evidence that supports the view that currency crises are inherently difficult to detect with any degree of accuracy and that efforts to construct early warning indicators may be misguided. A conservative indicator will fail to signal most crises while occasionally provoking false alarms, which could be enough to precipitate a run. Lighter triggers will detect upcoming crises more often, but would even more often send unjustified warning signals.

This conclusion certainly applies to the transition economies that are starting to undergo the boom-bust cycles of capital inflows, real appreciation and sudden withdrawals of foreign capital. It is much more desirable to refocus policy on the fundamentals: fixed exchange rate regimes do not seem to coexist very easily with full capital mobility. A number of countries (the Czech Republic, Estonia, Poland and Russia) have quickly moved to full capital mobility, often under external pressure. An admittedly casual look at the evidence fails to unearth the growth-enhancing effects of full integration in the world financial markets. Using the EBRD index of capital liberalization, the Report shows that growth tends to be higher where FDI, as a percentage of GDP, is lower (the correlation coefficient is -0.19). There is just no apparent link between growth and capital liberalization in the transition economies (the correlation coefficient is 0.02).

Moreover, the emphasis on liberalization has not been matched by adequate regulation of the financial sector. Financial markets have inherent weaknesses (born out of unavoidable information asymmetries) which call for prudential regulation. Currency markets are the only financial markets not subject to elaborate regulation, but this does not mean that it is impossible to control them. Chile has shown how a country can use prudential rules to sort out capital inflows. By adopting encaje, a system of compulsory non-remunerated deposit requirements on all capital flows, Chile has considerably lengthened the maturity of its external liabilities. Although long maturity is not necessarily a guarantee of stability, the Chilean experience deserves close scrutiny in the transition countries.

Fiscal Policy and the Exchange Rate

Unless fiscal policy is sufficiently prudent, any fixed exchange rate policy will be unsustainable. Fiscal stabilization must be achieved before any type of exchange rate policy will work effectively. A useful distinction when thinking about fiscal stabilization is as follows. Fiscal dominance can be defined as a policy regime where monetary policy will always be called upon to solve fiscal unsustainability, through the monetization of government debt. In contrast, monetary dominance corresponds to the case where a debt build-up will have to be dealt with by fiscal means, through either a closing of the deficit or a debt default. In a monetary dominance regime the central bank cannot be coerced into bailing out an undisciplined government. Thus, fiscal stabilization can be defined as the move from fiscal dominance to monetary dominance. Once this is achieved, the central bank can control the price level and the exchange rate.

The link between fiscal policy and the exchange rate regime must be taken into account when considering the accession path. The tighter is the exchange rate commitment (e.g. adherence to an ERM 2 with narrow bands) the more fiscal policy must take on responsibility for its own sustainability, and for dealing with shocks and the costs of restructuring. Unless fiscal policy is under control, the central bank is likely to face strong pressure to monetize government debt. Central bank independence is part of the response to this, but the economies of Western Europe concluded that central bank independence alone was insufficient and imposed macroeconomic convergence criteria under the guise of the Stability and Growth Pact. According to the Report, incentives for responsible fiscal policy should be given at least as much weight as formal exchange rate agreements and nominal convergence criteria.

Exchange Rate Regimes

The question of why countries undergoing a similar shock (i.e. the transition shock) have adopted so widely differing exchange rate regimes remains a challenging puzzle. Country-specific factors have to be a large part of the explanation, but this raises a new and difficult question: what should the accession countries do in the run-up to EU and EMU membership? Will they be able to sustain these differences during the transition to the EU or should they converge on a single exchange rate regime? This is a question that needs to be tackled as part of the current discussions over accession.

A good starting point is the choice of an inflation rate, more precisely the speed of disinflation and the longer-run target. Compared with mature economies, lower inflation in transition economies may be more beneficial because of induced effects on financial deepening, inadequate inflation accounting (which is biased upwards) and uncertainty reduction. But a number of arguments pull in the opposite direction. Early in transition the social return on investment should be abnormally high. With acute capital market imperfections and poor tax compliance, the cost of raising public funds is high and the benefit is large. This suggests that the inflation tax may be of a higher social value than in mature economies. In addition, relative price adjustments are likely to be sizeable as the economic structure is rapidly being modified. Given downward nominal rigidities, some inflation may facilitate more rapid real adjustment.

It is hard to argue that a single speed for disinflation fits all countries. When unemployment is initially high, the marginal gain to slower disinflation is low, so rapid disinflation is desirable; when unemployment is initially low, slow disinflation is preferable since it avoids unnecessary persistence of recession. Except for the Czech Republic, transition economies now have high unemployment rates. Other things being equal, these countries should be encouraged to disinflate quickly.

For transition economies reputation building is a serious matter, therefore it is important that they announce targets that are achievable. Essentially, the ACs have the choice between a domestic nominal anchor (a monetary or inflation target) or an external nominal anchor (an exchange rate target). The question of which is best remains unresolved; different regimes have a comparative advantage in responding to different shocks. An exchange rate peg accompanied by unsterilized intervention may be suitable for coping with shocks to domestic money demand (money supply is automatically and endogenously supplied through the balance of payments), whereas domestic anchors may be better at coping with competitiveness shocks (the exchange rate can adjust). Transition economies face both kinds of shock, and the diagnosis of these shocks is never easy.

The general lessons learned from the experience so far seem to be that rigid monetary targets cannot be upheld while the determinants of money demand are changing. Inflation targets are unattractive until price liberalization and structural adjustment are more fully completed, the transmission mechanism of monetary policy is more reliable and adequate fiscal support exists. Exchange rate targets, in the absence of adequate fiscal support, invite substantial financial inflows that increase vulnerability to a subsequent speculative attack. No regime is a clear winner; otherwise, the debate over nominal anchoring in OECD countries would have been resolved decades ago.

Full adherence to the Stability and Growth Pact will be a requirement for entry into EMU. Against the background of this incentive for responsible fiscal policy, the report concludes that there are three possible transitional regimes for the years leading up to EU entry:

·         Domestic anchoring via an inflation target, in which a monetary indicator is one of several indicators used to assess the stance of policy; accompanied by a floating, albeit managed, exchange rate.

·         Unilateral crawling band, initially wide, whose rate of parity depreciation is preannounced for a period in advance but is expected to fall over time.

·         A currency board that maintains a fixed nominal parity by a commitment to allow unsterilized reserve flows to be fully reflected in the domestic money stock.

Which of these regimes is most appropriate may well vary from country to country.

Debates about monetary regimes need to be kept in perspective. The first and second priorities for transition economies should be structural adjustment and fiscal responsibility. Unless both exist, any rigidly set monetary policy will eventually fail. In the process, care should be taken to avoid establishing an uncompetitive real exchange rate. This requires a heavy dose of flexibility as well as, possibly, restrictions to capital inflows, which have repeatedly created unmanagable policy dilemmas, as South-East Asia and Latin America keep reminding us.

Economic Policy Initiative No. 5, 'Monetary and Exchange Rate Policies, EMU and Central and Eastern Europe' by David Begg (Birkbeck College, London, and CEPR), László Halpern (Hungarian Academy of Sciences, Budapest, and CEPR) and Charles Wyplosz (Graduate Institute of International Studies, Geneva, and CEPR).

The report was launched at Lunchtime Meetings held in London in October 1999 and Brussels in November 1999.