Article

How Mobile is Capital in Developing Countries?

Author(s):
International Monetary Fund. External Relations Dept.
Published Date:
January 1991
Share
  • ShareShare
Show Summary Details

Much more than is commonly assumed, reveals a new study

Nadeem U. Haque and Peter J. Montiel

Research Department, IMF

Controls over the international movement of capital, often assumed to be pervasive and stringent in developing countries, are neither, according to the results of our recent research into the effective mobility of capital in 15 diverse developing countries. As a result, capital moves in and out of countries, both legally and by evasion of controls, with much greater frequency and ease than one would expect. This has important consequences for economic policy-making.

The degree of capital mobility has an important bearing on the short-term effects of stabilization policies—including monetary, fiscal, and exchange rate policies—in developing countries. A high degree of mobility in a fixed exchange rate regime reduces the ability of a government to devise an effective monetary policy, which depends on controlling demand through monetary aggregates. By allowing external finance to move into an economy, a high degree of capital mobility may nullify the contractionary effects of a nominal devaluation, or limit the crowding out of private investment by an expansionary fiscal policy. The strength of the effects varies with the degree of capital mobility. That, in turn, depends on the effectiveness of capital controls.

Existing controls

The vast majority of developing countries maintains significant legal restrictions over capital movements, apparently motivated by the desire to facilitate monetary control and the management of their balance of payments. According to the IMF’s Annual Report on Exchange Arrangements and Exchange Restrictions (1989), many developing economies do not allow the free flow of financial capital. As a result, developing countries are often assumed to have closed capital accounts, except for borrowing by the public sector.

The reality is different. Massive episodes of capital flight and fairly routine evasion of controls—for example, through over- and under-invoicing of imports and exports respectively, smuggling, and bribery—appear to reflect the ineffectiveness of such controls. Moreover, by keeping large amounts of their earnings or flight capital abroad, residents reduce the ability of domestic authorities to control the use of those resources with local instruments, such as interest rates or credit ceilings. At times, financial pressures (either for inflows or outflows) may be strong enough to lead to the official relaxation of controls.

But not much is known with certainty about the nature and extent of capital controls, and hence capital mobility, in developing economies. Empirical research has produced meager and rather mixed results. Much of the earlier work to find out how far changes in domestic credit policies affected decisions by individuals and firms to move financial resources in or out of a country has typically shown that developing economies have limited domestic monetary autonomy, a finding consistent with a high degree of capital mobility. But little work has been done to establish the extent of mobility across different economies.

The study

To test for capital mobility, our study used annual data from 15 developing countries for the period 1969-87. The size of the sample was determined by data availability and the need to make the analysis manageable. Save for the series on debt, the data were drawn from the International Financial Statistics files of the IMF. Debt data were taken from the World Bank’s World Debt Tables. The countries included in the sample were Brazil, Guatemala, India, Indonesia, Jordan, Kenya, Malaysia, Malta, Morocco, the Philippines, Sri Lanka, Tunisia, Turkey, Uruguay, and Zambia.

The choice of countries was guided by two considerations: first, internally consistent time series of reasonable length were available in these countries in the two data sources we employed; second, among the countries that satisfied the first criterion, the sample had to maintain a geographical balance and represent a wide range of developing countries. Our sample consisted of six Asian countries, four African countries, three Latin American countries, and two European countries. It also included four low-income countries and three heavily indebted countries. The results should, therefore, be fairly representative of developing countries in general.

In order to test for capital mobility, we began with the following two well-known propositions: (1) the interest rate in the domestic market in a completely open economy (i.e., when the capital account is open) is determined by the uncovered interest rate parity (i.e., the sum of the foreign interest rate and the expected depreciation of the exchange rate); and (2) the interest rate in a closed economy is determined purely by domestic money market equilibrium. Consequently, we assumed that the domestic market-clearing interest rate in developing countries can be expressed as a weighted average of the uncovered interest parity interest rate and the domestic market-clearing interest rate that would be observed if the private capital account were completely closed.

To measure the effective degree of capital mobility in our diverse sample of developing countries, we estimated money demand equations by replacing the interest rate variable with the weighted average described in the box on methodology. This yielded an estimate of the weight attached to the uncovered parity interest rate, which represents an index of effective capital mobility.

Results of study

The results showed that 10 out of the 15 sample countries had high capital mobility; that is, the polar case of a completely financially closed economy can be ruled out, but not that of a completely financially open economy (see box). The governments of these ten countries have little control over domestic interest rates and the money supply. Four other countries fall in the intermediate category of partial capital mobility—neither a completely open nor a completely closed economy. In these countries, the domestic interest rate would be subject to partial control by the government, at least in the short run. In only one country, India, do the results suggest that capital is immobile.

These results indicate that, on average, over the period 1969-87, domestic market-determined interest rates for this rather diverse group of developing countries tended to move quite closely with external rates and were relatively less influenced by domestic financial developments—except to the extent that such developments were expected to lead to exchange rate adjustments. The effective degree of capital mobility may well have varied over time for some of the countries in the sample. However, scarcity of data prevents us from verifying this.

… and their implications

The finding of a high degree of capital mobility in developing economies has important implications for macroeconomic policy in these countries. First, this means that fiscal policy has relatively powerful effects, compared to monetary policy, on domestic demand and the trade balance. Second, monetary policy is relatively ineffective in a fixed exchange rate regime, with variations in domestic credit lending to be offset by capital movements. Monetary policy, therefore, is effective in influencing capital flows and the balance of payments.

The results also suggest that capital controls tend to be ineffective, or at least their effects are of limited duration. In spite of prevalent capital controls, capital flows proved to be sufficient to maintain fairly close interest parity relationships with external interest rates over this period.

Estimated degree of capital mobility in selected developing countries, 1969-87

A note on methodology

In a completely open economy, the domestic interest rate would be equal to the interest rate that prevails abroad, plus the expected rate of depreciation of the exchange rate (a situation known as interest parity). On the other hand, in a completely closed economy, the domestic interest rate adjusts to clear the domestic money market. In an economy where capital controls are less than perfect, but not completely ineffective, interest rates would be influenced both by interest parity and by domestic money market conditions. Consequently, the domestic interest rate can be viewed as a weighted average of these two polar extremes.

If the three interest rates—the open economy rate, the closed economy rate, and the rate that actually prevails—were available, then the weighted relationship could be estimated, and the weighting factor would provide an estimate of the degree of effective capital mobility. An estimate of 1 as the weight for the interest parity rate would indicate perfect capital mobility, while an estimate of 0 (zero) would suggest complete capital immobility. Values between 0 and 1 would indicate the degree of capital mobility.

Since the interest rate that would prevail if the capital markets are completely closed is a hypothetical construct, and since data for the market-clearing interest rate are not available in most developing countries, our method for estimating the degree of capital mobility relies on the estimation of money demand, where the demand for money is a function of the interest rate, income, and the lagged money stock. Substituting the weighted average for the domestic market-clearing interest rate in this specification gives us money demand as a function of the interest rate parity, the unobservable closed economy interest rate, and other variables.

In the presence of completely effective capital controls, the interest rate would adjust to clear the money market. The unobserved closed economy interest rate can, therefore, be calculated as a function of income, lagged money, and the closed economy money supply (i.e., the money stock net of the monetary effects of private capital flows). Since all these variables are observable, substitution of this result for the unobserved closed economy interest rate in the money demand equation gives us variables that are all observable and in which the degree of capital mobility can be directly estimated as the coefficient of the interest parity variable.

For a detailed explanation of the methodology and results, see “Capital Mobility in Developing Countries—Some Empirical Tests,” available as IMF Working Paper WP/90/117 from the authors.

The empirical effectiveness of this kind of interest arbitrage undermines a key argument used to support policies of “financial repression” in developing countries. Maintaining low administered interest rates in the formal financial system will not stimulate domestic investment, since the marginal cost of funds will be given by the market-determined rate in informal loan markets, and this rate will effectively move in line with external interest rates. Below-market controlled interest rates will, under such circumstances, lead to inefficient allocation of capital, reducing growth, and transferring income from domestic savers to favored domestic borrowers.

The results also underline the importance of avoiding exchange rate misalignment in developing countries. Anticipated exchange rate adjustments will quickly be reflected in capital outflows—as has in fact been the case recently in many developing countries—and in high domestic market-determined interest rates. The latter are, in turn, likely to imply high real interest rates, with adverse consequences for the level of economic activity and for medium-term economic growth. High capital mobility coupled with expectations of devaluation may thus help explain, at least to some degree, the unsatisfactory performance of private investment in many highly indebted countries over the past decade.

Although the empirical finding of a high degree of capital mobility implies a loss of policy autonomy, governments should not try to limit the degree of capital mobility for at least two reasons. First, there are familiar efficiency reasons for allowing domestic prices to reflect external opportunity costs. Second, as indicated previously, our findings suggest that measures to limit the degree of capital mobility tend to be ineffective in developing countries. The best policy under these circumstances would be to avoid exchange rate overvaluation through appropriate fiscal, monetary, and exchange rate policies.

Other Resources Citing This Publication