Many analyses of `financial repression' the controls that most developing countries impose on international capital flows and on domestic financial intermediaries highlight the inefficiencies arising from these controls and propose clear-cut policy prescriptions that they be removed. Applications of the theory to policy design have generally neglected important additional distortions: the effects of government revenue constraints and the distortions associated with different forms of taxation.

In Discussion Paper No. 489, Research Fellow Alberto Giovannini and Martha de Melo estimate the revenues developing countries obtain from controls on financial markets, which they view as a form of taxation. Giovannini and de Melo measure this revenue conservatively as the difference between the foreign and domestic costs of borrowing by the central government (excluding the rest of the public sector), multiplied by the stock of government domestic debt. They measure the cost of foreign borrowing ex post and include the dollar interest rate of external debt, realized changes in the dollar exchange rate and liabilities in other currencies.

Giovannini and de Melo conduct their empirical tests on a sample of 24 countries that were `developing' at some point during 1972- 87, whose central governments had significant foreign commercial borrowing and for which data on the stock and cost of government domestic debt were available. They use the ex post difference between foreign and domestic borrowing costs. Unexpected devaluations will significantly affect the relative cost of domestic debt, even in countries with relatively free financial markets: their results are therefore biased towards finding financial repression and cannot formally test for its presence.

Giovannini and de Melo find an unweighted average revenue from financial repression of about 2% of GDP and 9% of total government revenue, which varies significantly across countries: from 0 in Indonesia to 6% of GDP in Mexico and Zimbabwe. Repression generated up to 40% of total central government revenue in Mexico and about 20% in India, Pakistan, Portugal, Sri Lanka and Zimbabwe. At such levels, governments should design means of coping with the shortfall that will result from liberalizing their financial sectors. They also find that countries with higher inflation and hence higher rates of currency depreciation tend to raise more revenue from financial repression.

Government Revenue from Financial Repression
Alberto Giovannini and Martha de Melo

Discussion Paper No. 489, January 1991 (IM)