Interest rates can have a substantial influence on the rate and pattern of economic growth by influencing the volume and productivity of investment, as well as the volume and disposition of saving. This is especially true for countries where financial markets are relatively well developed or where private investment constitutes a significant share of total investment. Even in countries with less developed financial markets or in those where investment is overwhelmingly the responsibility of the public sector, interest rates may have a significant effect on the mobilization of household savings and on investment decisions. In analyzing this set of issues, it would be useful to distinguish between the effects on saving and those on investment, bearing in mind, however, that while the volume of new investment that can be undertaken is related to the amounts of both foreign borrowing and domestic saving out of current income, it is domestic saving that is by far the more important source of investment financing in most developing countries. Finally, both the savings and investment aspects of interest rate policies influence income distribution, which is treated in the last part of this section.
Interest Rate Policies and Savings
A development strategy that relies on financial savings10 as a major source of investment finance requires price or nonprice incentives in order to stimulate saving in financial form and, where relevant, capital inflows. There is, however, considerable disagreement over the influence exerted by interest rates on the volume of saving: while an increase in interest rates may stimulate saving by making future consumption iess expensive relative to current consumption (substitution effect), it may also tend to reduce saving by lowering the amount of present saving necessary to buy a given amount of future consumption (income effect). The available empirical evidence on the relative importance of these two effects, based largely on the experience of Asian and Latin American countries, suggests that the substitution effect is more important than the income effect in developing countries, although not overwhelmingly so,11
It is sometimes argued, moreover, that the impact of a rise in interest rates on saving may be dampened by institutional influences. For example, in some middle-income developing countries, as in industrial countries, a substantial portion of saving takes the form of contributions to social security, retirement funds, and various other forms of contractual saving, which would be forthcoming regardless of the rate of return on them. Furthermore, in low-income countries there may be little scope for additional long-term private saving, and in countries where large segments of the population have little or no access to financial institutions it is unlikely that interest rates, at least in the organized financial sectors, are a significant determinant of the rate of saving. But even in countries where most saving and investment decisions are either undertaken within the public sector or directed by a system of planning, properly chosen interest rates can perform a useful role in helping to make these decisions.
The view that interest rates play only a minor part in the determination of saving is sometimes supported by empirical tests that reveal only a weak response of saving to interest rates. In evaluating the empirical evidence, however, it must be borne in mind that most studies have focused on industrial countries that typically show a very narrow range of variation in real interest rates. Applicability of these studies to developing countries is questionable, particularly to those countries where real interest rates remain substantially negative for protracted periods.12 In such circumstances, attitudes toward saving are likely to be fundamentally modified; for instance, a preference for hoarding goods and inventories may be strengthened at the expense of accumulating financial assets. This effect may be less marked, however, when real interest rates are negative only temporarily—for instance, as a result of a sharp, once-for-all increase in prices—or are only mildly negative.
The form in which savings are held has been of crucial importance in many developing countries, where interest rate policies or political developments have encouraged the public to hold a large proportion of their wealth, and place an even larger proportion of their saving out of current income, in forms such as real estate, consumer durables, precious metals, gems, art works, and foreign currency holdings (cash or deposits). The flight out of domestic currency deposits into these inflation hedges reduces the amount of financial savings available for financing investment.13 This process has the additional result of either increasing upward pressure on prices (to the extent that the hedges are domestic goods) or weakening the home currency’s position in foreign exchange markets (to the extent that the hedges are foreign goods and assets). When interest rate repression has led to such developments, a substantial increase in interest rates can be expected to have both short-run and long-run effects on available financial resources.
(i) In the short run, an increase in interest rates has the immediate effect of a once-for-all rise in the demand for domestic interest-bearing financial assets and an accompanying fall in the demand for foreign financial assets, as well as in the demand for inventories and consumer goods that had previously served as inflation hedges. When a large change in interest rates is involved, this reallocation of the public’s wealth may be substantial. As a result of these changes, financial intermediaries are provided with additional resources that are potentially available to finance a higher level of investment expenditures. Moreover, if nominal interest rates corrected for expected depreciation of exchange rates exceed the rates prevailing in world markets, capital inflows can be expected to increase, enlarging further the resources available to domestic financial institutions.
(ii) In the long run, the elimination of financial repression and the improved incentives for holding domestic financial assets combine to raise the average rate of return on savings and may therefore result in a continuing increase in the flow of saving as a proportion of income—provided, of course, that these incentives are maintained. The same factors also tend to increase the proportion of new savings allocated to domestic financial assets rather than to consumer goods and foreign financial assets. An increase in interest rates is thus likely to increase the flow of funds to the domestic banking system even when the shift in the propensity to save is small and the flow of saving remains approximately unchanged.14 Owing to these factors, elimination of interest rate repression improves the long-run capacity of the economy to finance domestic investment.
There is evidence from the experience of a number of countries that the interest rate—or, more precisely, the real return on deposits—has a significant effect on the volume of financial savings, and that there is some association between a decline in financial savings and a poor growth performance.15 The following illustrative examples are intended to suggest effects of both repression and reform of interest rates.
The recent experience of Ghana provides an instance of the effects of interest rate repression. During the period 1976–80, interest rates never exceeded 13 percent, while inflation was at or near the triple-digit range. The strongly negative real interest rates caused a precipitous flight from banking deposits. Financial savings in real terms consistently declined during this period, as shown by a steady fall in the real level of broad money.16 This trend was interrupted only in 1979, when time deposits actually grew slightly in real terms as a fall in the rate of inflation moderated the loss to depositors.
Similar developments occurred in Jamaica during 1977–80, when interest rates much below inflation rates resulted in a steady erosion of time and savings deposits in real terms. In 1978, for instance, when the inflation-adjusted rate of interest was -29 percent, deposits fell by 22 percent in real terms. The decline of real financial savings may have been one factor in the negative growth performance of 1977–80.
Brazil provides another interesting example of the effects of a change to highly negative real interest rates, especially since it has historically maintained interest rates commensurate with inflation. A brief departure from its historical policies in 1979–80 brought an immediate response from savers. The authorities allowed the interest rate to become sharply negative in real terms during 1979–80, causing a decline in the real value of both financial savings and foreign borrowing.
Two other examples illustrate how a steady policy of positive inflation-adjusted interest rates can lead to steady growth in financial intermediation. Malaysia maintained such a policy during 1975–80 and experienced double-digit growth in real time and savings deposits. Korea also followed this policy, although somewhat less consistently, and experienced a slightly more erratic growth in real financial savings, which rose steadily over the last decade, except for two periods (1974—75 and 1979–80) when interest rates were permitted to become negative in real terms. The healthy development of financial intermediation in these two countries has apparently contributed to the maintenance of high rates of growth.
There are also examples of striking results that can be obtained by liberalizing repressed interest rates. Before interest rates were liberalized in Argentina, beginning in 1976, real interest rates were strongly negative. As a result, time and savings deposits in real terms practically disappeared, declining by over 70 percent in 1975, Over the next few years, however, interest rates were allowed to rise to remunerate savers more adequately for the effects of inflation. Deposits rose rapidly in response, more than doubling in real terms in 1977. The recent experience in Turkey also confirms the response of savers to changes in real interest rates, as time and savings deposits nearly tripled in real terms in 1981 after a shift in policy that resulted in a substantial improvement in real interest rates, which had previously been sharply negative.
Interest Rate Policies and Investment
A principal motive for pursuing a policy of low interest rates in developing countries is the desire to stimulate investment. The two main questions that will therefore need to be addressed in this section concern the impact of interest rates on (i) the volume of and (ii) the productivity of investment.
Volume of Investment
In discussing the impact of interest rates on the volume of investment, the focus is on “fixed investment”—that is, investment in plant and equipment—rather than on investment in inventories. It is true that in some situations the ability of the economy to operate at full capacity may be limited by a shortage of working capital to finance minimum inventories. Such a situation, however, is not likely to prevail after a protracted period of repressed interest rates. Moreover, long-run growth, which is our principal concern here, is determined ultimately by the quantity and productivity of fixed investment.
Leaving aside the issue of the productivity of investment, which will be taken up subsequently, interest rates can influence the quantity of fixed investment by their influence on both the volume of total investment and the allocation of that total between fixed capital formation and inventory accumulation. The effect that interest rates have on the volume of investment by their impact on the availability of financing has already been discussed in the previous section: consequently, the focus here will be on the manner in which interest rates influence the demand for such financing and the uses to which the financing is put.
The primary objective of economic policy in developing countries is the promotion of growth and economic development—the former implying the expansion of the productive capacity of the economy, the latter implying the improvement of production techniques and organization, as well as of the quality of the factors of production themselves. Achievement of both objectives requires that developing countries allocate a substantial portion of their resources to investment. Consequently, the authorities of these countries place heavy emphasis on policies to increase the buoyancy of investment, both public and private. One such policy, common to many developing countries, is to maintain low interest rates.
The arguments in favor of low interest rates in developing countries link the need for a high level of investment to theoretical explanations of investment behavior that posit an inverse relationship between the interest rate and investment demand. One way to establish this inverse relationship is to view the profitability of investment as depending upon whether the present value of the series of expected returns, discounted by the rate of interest, exceeds the purchase price of the capital asset. In a competitive environment, entrepreneurs will desire to undertake all those investment projects that are deemed profitable. At a lower interest rate, the present value of returns will be higher for each project, thus rendering profitable some projects that would have been unprofitable at a higher rate of interest. Consequently, the volume of desired investment will be higher the lower is the interest rate. This inverse relationship between the interest rate and the desired investment has often been used as a justification for low interest rates.
Conversely, it is often argued in developing countries that a rise in interest rates will reduce the level of desired investment by raising the cost of financing certain investments above their expected rates of return. This argument not only neglects the crucial distinction between desired and actual levels of investment but also ignores the fact that expected rates of return are influenced by the attractiveness of the general economic environment. Since, as argued in Section IV of the paper, interest rate repression may lead to capital outflows and foreign exchange shortages (which could hamper the implementation of future investments), expected returns to investment projects may be negatively affected by a policy of maintaining interest rates that are low relative to inflation.
Moreover, when interest rates are low relative to the prevailing rate of inflation, firms will find it profitable to borrow, or use their internally generated resources, not only for fixed investment but, just like households, for their own inflation hedges-—inventories, imported goods, real estate, and foreign currency deposits,17 Borrowing for all these purposes is typically incorporated into the demand for “working capital.” This additional demand for funds—just like financing the government deficit—tends to “crowd out” financing for fixed investment.
While a change in the rate of inflation generally leaves unchanged the relative profitabilities of inflation hedges and fixed investment, an increase in interest rates is likely to induce firms to economize on their use of borrowing for working capital. This can be explained by several considerations of an institutional nature: interest rates on credit for working capital may be increased more sharply; investment plans for plant and equipment may be less sensitive to variations in interest rates, particularly in the public sector; and credit for working capital is likely to be subject to stricter official limits than credit for fixed investment. It must be added, however, that the opposite effects could prevail if bank lending were heavily biased toward the short-term credits and if long-term credits for investment did not bear substantially lower interest rates than those on short-term credits for working capital.
The argument that lower interest rates raise the rate of investment implicitly assumes that additional resources for investment will somehow be obtained. This outcome depends on additional saving being created as the initial investment results in an increase in output and income. There is good reason to believe, however, that in many developing countries such a flexible response of supply in the short term may be impossible because of bottlenecks in the supply of certain factors of production, such as skilled labor, that are essential complements to increased investment.
If the market mechanism is permitted to operate and is properly functioning, additional investment and saving may be generated without an increase in output, as the initial excess demand for credit will tend to push up lending rates and thereby stimulate an increase in deposit rates.18 As discussed earlier, the higher deposit rates will attract more resources to the banking system. While there will be some reduction in desired investment as interest rates rise, actual investment will increase as long as the interest elasticity of saving is greater than zero.
In contrast, when the market mechanism is prevented from functioning by extensive regulation of interest rates, the necessary real credit expansion will not be forthcoming to finance the increased investment demand and the financial system must then resort to various rationing schemes to allocate the available credit. Indeed, as suggested in an earlier section, it is conceivable that low interest rates, while possibly inducing a high level of desired investment, could actually reduce realized investment because of adverse effects on available credit.19
It is difficult, however, to assess empirically the significance of interest rates as a determinant of investment expenditures, because of the substitution of alternative mechanisms to raise funds for investment when the regular channels become inoperative. First of all, savings might be generated through “forced saving,” by which the inflationary process resulting from increased aggregate demand brings about a redistribution of real income from low savers to high savers20 and, in particular, from wage earners to profit recipients. In planned economies, forced saving may take the form of planned discrepancies between the wage bill, in real terms, and the volume of consumer goods made available, at fixed prices, to the public. In the presence of shortages of these goods, those households whose consumption plans have not been fulfilled have no choice but to place their unspent earnings in savings deposits.
Furthermore, under financial repression there is a tendency for various forms of quasi-financial intermediation, such as unofficial credit markets21 and self-financed investment, to develop. The interest rates prevailing in unofficial credit markets are almost always higher than those in the official markets, thus indicating the existence of highly profitable investments that are not receiving financing from the established financial institutions. To the extent that unofficial credit markets perform the functions of financial intermediaries without being subject to interest rate regulations, they tend to offset and conceal the detrimental effects of financial repression. It has to be borne in mind, however, that unofficial credit markets tend to be relatively inefficient substitutes for a properly functioning banking system. Similar considerations may also apply to self-finance.
For countries with access to international capital markets another source of additional financing is foreign borrowing, either directly by those seeking the financing or indirectly by domestic banks for on-lending. Foreign credits of this sort tend to be characterized by high interest costs (in part reflecting risk premiums), to be channeled only to larger firms with an international repuration, and to be used for working capital or trade financing (because of their short maturities) rather than fixed investment. Moreover, recent experience has shown plainly the dangers of heavy reliance on foreign borrowing when accompanied by large fluctuations in the terms of trade and world demand.
Finally, increased investment can be financed by an increase in government saving. There are a number of ways in which government saving can be raised: by increasing tax effort, by reducing noninvestment expenditures, and by raising the prices of goods and services provided by the public sector. Indeed, in planned economies and other countries where a large proportion of total investment is undertaken by the public sector, the rates of interest may not significantly influence either the government’s ability to mobilize resources or its investment decisions.
In some circumstances, it is thus possible to finance additional investment without an increase in interest rates, but this either may require further fiscal effort or may be accomplished in ways that diminish the efficiency of financial intermediation. If additional real saving is not forthcoming through any of these channels, some of the desired investment will simply not take place.
Productivity of Investment
The rate of economic growth depends not only on the volume but also on the productivity of investment. Sometimes the latter is measured in terms of the private rate of return; but from the standpoint of the authorities, whose primary concern is to secure maximum benefits from investment for the entire society, the appropriate measure is the social rate of return. This rate, which is difficult to quantify, is often proxied by the rate of economic growth realized from a given volume of investment, and it is in this sense that the impact of interest rate policies on investment productivity will be discussed here.
In market-oriented economies, private rates of return generally guide a large part of investment decisions, and it is the divergence of private and social rates of return that impels the authorities in many developing countries to adopt policies designed to affect the allocation of investment. When interest rates are repressed, this divergence may widen as activities with low marginal rates of social return, such as the accumulation of inventories for hedging purposes and speculative commodity or real estate transactions, are relatively more profitable. This reaction to interest rate repression helps explain why in some developing countries strenuous efforts to increase aggregate investment by a policy of subequilibrium interest rates have resulted in only mediocre rates of growth.22
There are several reasons why a policy of maintaining interest rates that are low relative to the rate of inflation tends to lower the productivity of investment.
(a) Interest rate policies are normally accompanied by administrative arrangements, such as selective credit policies, to ration scarce financial resources. The success of selective credit arrangements depends crucially on the ability of the policymakers to correctly identify priority sectors and to control the end-use of funds allocated to them. Neither of these tasks, however, is accomplished easily.23
First of all, it is not easy or even possible to identify correctly the priority sectors. There is a lack of reliable data on many important economic activities; the existing structure of relative prices, both within the domestic sector and between the domestic and the foreign sector, becomes highly distorted as a result of widespread use of regulations and controls; and predictions regarding the future developments in prices of primary commodities and of manufactures, as well as in exchange rates and international interest rates, are subject to substantial uncertainties. As economic judgments are, in practice, often subordinated to the social and political pressures exerted upon policymakers by various groups competing for scarce financial resources and demanding “priority” status, an increasing number of sectors or activities are, in fact, granted such status. Moreover, there is a strong tendency to include the government itself and all public sector activities in the priority category, even when the economic justification for doing so is slender or nonexistent.24
Even when specific priority projects with high returns are correctly identified, selective credit policies still tend, over time, to lower the overall rate of economic growth. This occurs because, by biasing the allocation of funds toward certain sectors and away from others, such policies tend to support indiscriminately all projects, even those with low social rates of return, being financed in the priority sectors, at the expense of nonpriority projects with greater benefits for the economy as a whole. This tendency results in an excessive buildup of capacity in the favored sectors of the economy. The resulting underutilization of capacity effectively lowers the rate of return to investment in those sectors, even if an individual project merits being undertaken when reckoning its rate of return on the assumption of full capacity utilization.
(b) A policy of interest rates that are low relative to the expected rate of price increase tends to reduce the necessity for careful evaluation of projects by enterprises and the economical use of resources in the implementation of these projects. Again, this is especially true for priority sectors.
(c) Even if the allocation of credit under conditions of financial repression were not subject to government regulation but were determined entirely within the banking system, there is reason to doubt that it would be undertaken in an efficient manner. Well-established firms having close connections with banks would tend to be favored over smaller or newer borrowers, without regard to the rates of return on the respective investments, because of personal connections and of risk aversion by the banks. Short-term credit for inventory accumulation might well be favored over long-term credit for investment, especially in inflationary conditions. Indeed, it is such considerations that in part motivate governments to interfere in the process of allocating credit.
(d) As indicated earlier, whenever the cost of credit is below the rate of inflation, it will be profitable for businesses to use their credit facilities to carry large inventories of raw materials, intermediate goods, and finished products.25 These transfers of financial resources into inflation hedges further aggravate the scarcity of funds for productive investment and add to inflationary pressures, and (for imported inventories) waste scarce foreign exchange resources.
(e) Policies that lead to subequilibrium interest rates, especially when accompanied by a policy of maintaining high real wages, result in a bias in investment in favor of capital-intensive techniques. The resulting rise in unemployment tends to lower the rate of social return on the investment and exacerbates an already severe social problem. Furthermore, capital-intensive techniques of production generally require large amounts of working capital and foreign exchange, the demand for which aggravates existing shortages of both credit and foreign exchange.
Interest Rate Policies and Income Distribution
Low interest rates, either across the board or for specified types of credit, are often defended on the grounds that they bring about a redistribution of income from the upper echelons of society to the poorer classes. which are supposed to benefit from rates applied to credit for farmers, small businesses, and also sectors like housing, where output is thought to meet the needs of low-income households. There is reason to doubt, however, that the overall distributive impact of overall repression of interest rates, or even preferential interest rates, is substantial—or indeed that it is even in the intended direction.
With respect to agricultural credit, repressed interest rates may have perverse effects in countries where there are a number of large units—plantations, collective farms, or private estates—together with many small holdings, because there is a tendency for the larger units, which are often more productive, to receive the lion’s share of preferential agricultural credits. Smaller farmers must then often depend on local moneylenders, who lend at much higher (often exorbitant) rates, or on large landowners, whose credit is related to crop-sharing arrangements that are disadvantageous to the borrowers. It is only in countries where smallholders are the dominant group in agriculture and organized credit institutions reach into rural areas that either repressed interest rates or preferential rates on agricultural credits are likely to serve the purpose of assisting the majority of agricultural producers. In this instance, however, the elimination of financial repression, by giving the rural population access to organized banking, may also obviate the apparent need for interest rate repression with regard to rural loans.26
Low interest rates are often considered to encourage local, small-scale enterprises and at the same time to stimulate employment by the expansion of labor-intensive activities. Conversely, it is argued that a substantial increase in interest rates would drive small-scale producers out of business and encourage a more monopolistic structure of production and trade. If. however, a rise in lending rates permits increases both in deposit rates and in the funds available for lending, the typical borrower in this sector will also benefit to the extent that he holds financial savings and has previously had an unsatisfied demand for credit; small-scale enterprises are often constrained by the availability of credit rather than its cost.
When housing construction is subsidized through negative real interest rates in an inflationary environment, there is a widely observed tendency for credit to be used in substantial part for accumulating unused inventories of building materials as an inflation hedge. In general, it is firms and wealthier individuals, rather than low-income households, that have the easier access to such credit and the greater ability to avail themselves of this type of hedging opportunity. The ultimate result of this activity is that the price of building materials is driven up, new housing becomes more expensive to acquire, and the share of lower-income groups in the acquisition of new housing diminishes.
The overall effect of negative real interest rates on the distribution of income and wealth thus seems likely to favor the upper-income groups in a society. Low-income households and producers have little opportunity to invest their savings other than in thrift institutions or in their own enterprises. On the borrowing side, they often have only limited access to credit from financial institutions or choose not to seek such credit. High-income households and owners of large productive units have wider scope for placing their savings in different assets, and. if nominal deposit rates are lower than the expected rate of inflation, they will tend to place them in deposits abroad, in real estate, or in the variety of other inflation hedges mentioned previously. At the same time, large firms or individuals in higher social echelons tend to have well-established relationships with financial institutions and therefore relatively easy access to credit for working capital (including inflation hedges) as well as for investment. On the basis of these considerations, there is some reason to suppose that negative real interest rates, especially in an environment of rapid inflation, have the net effect of transferring income from small savers to large borrowers. Indeed, this may be true in industrial as well as in developing economies.
Even if it were true, however, that lower-income groups or weaker sectors were in fact assisted by low interest rates—for instance, by preferential rates specifically designed for those groups or sectors—it may be asked whether special policies on interest rates and credit are the best available means for extending such assistance. The costs of doing so by these means include the wasteful use of investible resources (through indiscriminately encouraging both high-return and low-return projects in these sectors) and the diversion of resources from more productive uses. Direct government assistance—such as public infrastructural investments and social expenditures—is likely to be a more efficient means of assisting low-income groups.
As distinct from government revenues or investors own internal sources of finance.
See, for example. Fry (1978, 1980), Leff and Sato (1980). McDonald (1983). and Sundararajan and Thakur (1980).
Moreover, it has been argued that even in developed economies, restrictions on net yields to savers adversely affect gross costs of borrowing due to reduction in financial intermediation. See, for example, Boskin (1978). where he discusses these issues in relation to taxation of interest income.
With the possible exception of foreign currency deposits held in domestic financial institutions. However, a situation in which such deposits grow to a large amount can seriously endanger the stability of the domestic financial system.
To the extent that a substitution of financial savings for consumer durables takes place, it will be reflected in the national income accounts as an improvement in the propensity to save, because purchases of consumer durables, and hence total consumption expenditures, will decline.
Data relevant to the cases mentioned in the following paragraphs are provided in Appendices I and II, which also contain a more detailed discussion of the experience of Argentina, Brazil, Korea, and Turkey. Appendix III presents some data relevant to the relationship between growth of financial savings and growth of output and discusses briefly the question of causality between these two variables.
In all the cases cited here, financial savings are, for reasons of data availability, measured by broad money, consisting of monetary and quasi-monetary deposits with the banking sector, which is ordinarily the preponderant component of financial savings in developing countries.
Uncertainty about future changes in the real rate of interest, or about real rates of return on fixed investment, may also contribute to a preference for these types of investment.
When competitive conditions do not prevail in the banking sector, however, deposit rates may rise more slowly than lending rates, and their increase may have to be buttressed by official guidelines or legal floors.
The disequilibrium created by an excess of investment over saving has several implications for macroeconomic adjustment in developing countries. Several of these issues are discussed in Section IV of this paper, for an alternative presentation, see Leff and Sato (1980).
The greater ability of high savers to place their accumulated savings in various inflation hedges is dealt with below.
This term refers to the “curb markets” or “informal markets” through which funds are lent outside the established financial institutions, either by a series of bilateral arrangements between ultimate lenders and ultimate borrowers or by agents who operate outside government regulation of financial operations.
The beneficial impact of equilibrium interest rate policy on the efficiency of investment and growth is empirically confirmed in several recent studies, including Fry (1980a and b). and Sundararajan and Thakur (1980).
For a discussion of the necessary conditions for the successful implementation of selective credit policies, see Khatkhatc and Villanueva (1978) and references cited therein.
It should be noted, however, that often (he reasons given for priority are social rather than economic.
There may. in some circumstances, be other important motives for large inventory buildups, e.g.. uncertainties regarding the future availability and price of imports.
See McKinnon (1973). pp. 77–79.