The case for macroeconomic stabilization policy, for LDCs as much as for developed countries, rests in part on empirical judgements about the workings of the economic and political system. The economic judgements concern the presence and severity of market failure and the existence of policy rules which are technically feasible and which would enhance the performance of the economy or the welfare of the individuals in it. The political judgements concern the likelihood that these desirable and potentially feasible policy measures will in fact be implemented, given the actual political and administrative constraints on the design and conduct of economic policy. In Discussion Paper No. 93, Research Fellow Willem Buiter discusses the design of stabilization policy in developing countries, with a special emphasis on policy responses to external shocks.

Some forms of stabilization policy are a response to market failures, Buiter notes. Stabilization policies are undertaken precisely because of the existence of market failures and of persistent 'non-Walrasian' equilibria, which are thought to be endemic in the major industrial countries. Many of the distributions underlying such market failures may themselves be the result of past, present or anticipated government policies. A complete solution to such problems will require structural adjustment as well as stabilization policy: for example, a cure for repressed financial markets, according to Buiter, requires the abolition of artificial ceilings on interest rates and other rates of return. But for as long as these structural distortions remain part of the economic environment, before the necessary remedies have taken full effect, they will have major implications for the design of stabilization policy.

Buiter reviews the macroeconomic policy options available to developing countries and discusses sensible stabilization policy responses to external shocks, such as a deterioration in the terms of trade, a slowdown in the rate of growth of export demand, an increase in the interest rates at which developing countries can borrow abroad and an increase in the external rate of inflation. It is by no means self-evident that stabilization policies designed in response to industrial countries' business cycles are appropriate for the developing countries, whose labour market structures and output composition are very different. Stabilization policies derived from the experiences of industrial countries are likely to perform better in a developing country, the larger is the share of the modern industrial and service sectors in its total GDP. This implies that the more developed 'Southern Cone' countries should not ignore the industrial countries' experiences of demand management.

He notes several modifications of the industrial country paradigm which deserve special attention in discussions of developing countries' policies. Traditional macroeconomic stabilization policies are likely to be inappropriate in countries with a small modern industrial sector, a large informal urban sector and a large share of GDP derived from subsistence agriculture. Demand management is also unlikely to play a useful role when the production of agricultural cash crops for a competitive global commodity market accounts for a major share of GDP. Production is unlikely to be constrained by insufficient domestic aggregate demand, and overall economic performance will be determined by 'supply-side' measures such as policies to maintain a competitive real exchange rate or to reduce real marginal labour and capital costs.

Buiter draws attention to the dangers faced by developing countries which engage in structural financial reform or trade liberalization without first considering the budgetary and monetary consequences of such measures. Domestic financial liberalization or an increased degree of international financial openness may lead the private sector to substitute foreign assets for domestic currency in its portfolios ('dollarization'). This will tend to reduce the monetary base from which the government derives revenue, through seigniorage or an inflation tax. If government revenues fall in this fashion, the inflation rate will increase unless the government undertakes an appropriate fiscal correction, which may itself have consequences for stabilization or distribution.

Buiter distinguishes between two uses of the nominal exchange rate as a policy instrument by developing countries. The first is a devaluation of the exchange rate that accelerates an ongoing real adjustment and is aimed at achieving a correction which would otherwise take place through an adjustment of domestic nominal prices and wages. The second use of a devaluation is to correct a persistent and well-entrenched real disequilibrium resulting from excessive domestic factor costs or an unsustainable fiscal programme. Under the first set of circumstances, devaluation will be successful, but the benefits will be modest. The benefits of devaluation in the second situation may well be major, but it will be unsuccessful unless backed by other measures. Without an effective policy to control real incomes or other labour market or industrial policies that change the balance of power between labour and private or public capital, a devaluation of the exchange rate is unlikely to have any lasting supply effects, claims Buiter.

Buiter concludes that while controlled access to international credit can be beneficial to developing countries, full and unrestricted financial integration into the world economy is undesirable while internal financial, goods and factor markets are underdeveloped and inflexible. This might be said of many industrialized countries as well, he notes.


Macroeconomic Responses by Developing Countries to Changes in External Economic Conditions
Willem H Buiter

Discussion Paper No. 93, February 1986 (IM)