Stabilization programs supported by use of Fund resources typically contain limitations on domestic credit expansion by the banking system. In many instances, the programs also include limits on foreign borrowing by the public sector which aim, on the one hand, at the prevention or alleviation of debt service difficulties and, on the other hand, at limiting the possibilities for the substitution of foreign financing for domestic credit. Through the combination of limitations on domestic and foreign credit, stabilization programs endeavor to contain overall credit expansion to a magnitude consistent with the balance of payments, growth, and price objectives. 1


Stabilization programs supported by use of Fund resources typically contain limitations on domestic credit expansion by the banking system. In many instances, the programs also include limits on foreign borrowing by the public sector which aim, on the one hand, at the prevention or alleviation of debt service difficulties and, on the other hand, at limiting the possibilities for the substitution of foreign financing for domestic credit. Through the combination of limitations on domestic and foreign credit, stabilization programs endeavor to contain overall credit expansion to a magnitude consistent with the balance of payments, growth, and price objectives. 1

Stabilization programs supported by use of Fund resources typically contain limitations on domestic credit expansion by the banking system. In many instances, the programs also include limits on foreign borrowing by the public sector which aim, on the one hand, at the prevention or alleviation of debt service difficulties and, on the other hand, at limiting the possibilities for the substitution of foreign financing for domestic credit. Through the combination of limitations on domestic and foreign credit, stabilization programs endeavor to contain overall credit expansion to a magnitude consistent with the balance of payments, growth, and price objectives. 1

In the formulation of financial programs, attention has been focused primarily on the aggregate demand effects of the financial variables and their implications for the balance of payments. In contrast, this paper is concerned with the analysis of the implications of credit policies on output and growth and how they relate to the development of the current account and overall balance of payments. The framework chosen for the analysis is one in which availability of financing is a direct and major determinant of current and future production. In particular, the paper addresses the following issues: (1) How do limits on domestic credit and on foreign borrowing affect current output and investment and, hence, future economic growth? (2) Are there differential effects associated with restricting foreign rather than domestic credit, and investment or working capital credit versus consumption credit? (3) Can an overly restrictive credit policy reduce current and future output so that over time the current account position will deteriorate rather than improve? (4) Finally, to what extent does it matter, for current account developments and debt servicing capacity of a country, whether credit expansion leads to supply responses in the traded goods sector (i.e., export sector or import substitution sector) as opposed to other sectors of the economy?

The paper identifies three channels through which credit policies can affect production in the economy: (1) the indirect link between credit and overall aggregate demand in a Keynesian setting of unused capacities and unemployment; (2) the direct link between working capital availability and current production; and (3) the link between credit, investment, and future production. Although closely related, (2) and (3) differ in the time horizon involved. After a brief review of the first link, the paper concentrates on an analysis of the second and third links, which have not received extensive treatment in balance of payments literature, although they have been explored recently in the context of growth, development, and stabilization efforts of closed economies. 2 The principal conclusions are as follows:

  • Limiting the overall level of credit is not a panacea for balance of payments problems; considerations regarding the distribution and the use of credit are important because of the impact on current and future production and, hence, the developments over time in the current account of the balance of payments.

  • Under a fixed exchange rate system, the origin of credit (i.e, whether from domestic or foreign sources) does not matter for the achievement of demand management objectives such as output, employment, and the current account, 3 but it is important for the overall balance of payments, that is, exchange reserve developments.

  • Under a flexible exchange rate system, domestic and foreign credit have different exchange rate effects, and, hence, the aggregate demand, production, and current account implications of a credit expansion depend in part on the origin of credit.

  • The current account objectives are best served by permitting credit expansion and investment to take place in the sector with the highest productivity, independent of whether this sector produces traded or nontraded goods; unless distortions prevail, and interest rates are held too low to fulfill their allocative function, this does not require deliberate policy measures, such as sectoral credit allocation.

  • Tight credit policies can lead to negative effects on current and future production and, hence, endanger the current account objectives when prevailing distortions lead to a “crowding out” of productive uses of credit. Reduction of price distortions and, particularly, the adoption of appropriate exchange rate and interest rate policies must, therefore, be part of any program that contains limitations on credit.

I. Possible Links Between Output and Credit

credit, domestic demand, and output

It is well known that credit expansion is one of the factors that determine final demand in the economy. To the extent that domestic production of goods and services, in turn, responds to changes in aggregate demand, one could characterize these output effects as being indirectly induced through credit policies. The role played by domestic bank credit, and the capital market and foreign financing, respectively, will depend on the structure of the economy and the creditworthiness of the country. Naturally, domestic supply potential will depend on the overall employment situation prevailing in the economy, the degree of capital utilization, the level of technology, and the degree to which imported goods are substitutes for domestic output in satisfying aggregate domestic demand. In full employment situations or when employment is constrained by the available fixed capital stock and the prevailing fixed coefficient technology in production, credit-induced demand expansion will result primarily in price increases, reduced exports, and induced imports rather than in domestic supply expansion. However, even when technology and employment conditions permit a significant domestic production response, and even if demand increases are assumed to be directed primarily toward nontraded or home goods, some additional import demand can be expected, at least to the extent that domestic production requires imported raw materials and intermediate goods. In an open economy this implies that demand-induced changes in domestic value added must fall short of the changes in domestic absorption (which cause the supply response), and their difference will be reflected in movements of the current account of the balance of payments. This conclusion remains necessarily true as long as the marginal propensity to import final or intermediate goods is positive.

These supply responses to changes in credit via aggregate demand changes and their implications for the current account of the balance of payments are well known; they have been recognized in the research on which the Fund’s financial programs are based (Polak, 1957; Polak and Argy, 1971) and are also in part implicit in the theory of the import multipliers (compare, e.g., Machlup, 1965). This paper will therefore not pursue further this important “indirect” link between global credit policies and supply effects.

relationship between credit, working capital, and investment

A direct connection between the level of current production and credit expansion exists because of the need for firms to finance the cost of inputs for the duration of the production process, including the time required to sell the product. In contrast to many models that directly include money balances in the production function, 4 the argument is not that credit from foreign or domestic sources constitutes a productive input per se; rather, it is based on the proposition that the availability and/or cost of financial working capital will determine, within limits set by other economic parameters, the volume of production that can be financed. 5 Production plans of the firm, the number of workers hired, and the size of the inventory of finished and intermediate goods and raw materials held by the firm are affected by the amount and terms of credit. In this sense, credit acts as a catalyst to mobilize and transfer real resources from domestic and foreign savers to the productive sector. Together with self-financing through the retention of profits or the sale of equity, the amount of credit extended (i.e., the level of financial intermediation) determines the quantity of factor services that can be marshaled by the entrepreneur to carry out current production. In an open economy, these resources can be of domestic or foreign origin. In this context, the role of interest rates is of particular importance, that is, the extent to which interest rates function as a relative price in channeling real resources from the household sector or from abroad to the productive sector, as well as between industries. Problems could arise, on the one hand, through inappropriately fixed interest rates and, on the other hand, through “financial repression” (McKinnon, 1973) generated by, for example, excessive financing needs of public sector consumption, which, thus, pre-empt both credit and real resource availability to the private sector. The role of bank credit (or credit extended by nonbank financial intermediaries) in financing current production depends also on the institutional arrangements in the economy, such as the existence and depth of capital markets, the prevailing degree of self-financing from retained profits, and the overall liquidity of the economy. Even where current production is financed through the accumulation of past profits or original capital endowments of the firm, production expansion may depend, entirely or in part, on credit availability and/or the cost of credit.

High real interest rates or quantitative limitations on the availability of credit for working capital could lead to reduced levels of inventories, which then limit the ability of the firm to produce and supply final goods smoothly. Moreover, considering that workers are often a “quasi-fixed” factor of production for the firm (oi, 1964), a shortage of working capital may be reflected in a cutback in hiring. While part of the financial working capital will be used to build up inventories of raw materials, and semifinished and finished products, other parts of working capital will take the form of increased money balances held by the firms. Nadiri (1969) has shown that for the United States increased production volume requires additional liquidity reserves for the firm. This is probably even more true in developing countries, with fragmented financial markets, an absence of regular overdraft facilities, and long time lags for financial transactions. As the penalty for illiquidity of a firm may be quite severe, that is, bankruptcy or high costs associated with selling illiquid assets, sufficient cash reserves may be a precondition for production expansion, even if money per se does not enter the production function. Availablility of financial working capital is not only crucial for the production of industrial firms or enterprises engaged in retail and wholesale operations but also for small-scale farming, an important source of output in many developing countries. Farmers producing cash crops often depend on credit from moneylenders and, where available, on bank credit for their working capital needs to cover the necessities of life and the cost of seeds and other inputs during the time preceding the harvest and sale of the agricultural produce. 6 Also, the successful transport of export crops will often depend on sufficient financing being available to marketing boards. To the extent that the crop is consumed or exported at the end of the harvest cycle, this involves seasonal financing rather than the financing of a permanent accumulation of inventories. 7

The liquidity situation of firms, and their recourse to credit from various sources, determines not only the size of the current volume of production that can be financed but also the extent of fixed capital formation undertaken by the firm and, hence, future production potential. Although, in general, funds are fungible, the argument can be advanced that, within certain limits, investment credit can be separated from credit for working capital, primarily because credits extended for working capital typically are of short maturity and hence unsuitable for financing investment projects with often long gestation and amortization periods. Moreover, lending institutions prefer to maintain a close relationship between the maturity of credits extended and the maturity of the project financed. 8 The availability of credit for investments and, depending on the interest rate policy pursued in any particular country, the cost of such credit will affect future production potential. Hence, supply will respond differently over time, depending on whether credit is extended for investment or for working capital, since the first affects future and the latter current production.

II. Credit and the Balance of Payments

For the analysis in this section it is necessary to define clearly the time period under consideration, because certain credit policies may lead to a deterioration in the balance of payments in the short run while actually improving the external position in the longer run. Moreover, it is also necessary to distinguish between private profitability of the uses of credit in a microeconomic sense and overall profitability, as measured by changes in the value added in the economy, in a macroeconomic sense. This distinction becomes crucial in countries where there are widespread distortions in the economy. Particularly in situations where artifically low lending rates of interest exist (as is characteristic of financially repressed economies), market signals do not fulfill their allocative function and give incorrect information regarding the profitability of certain uses of credit. As is shown below, the assessment of profitability has to be undertaken in light of the cost of servicing the interest burden of increased foreign borrowing or reduced earnings on foreign exchange reserves. Initially, it is assumed that the exchange rate remains fixed. The formal analysis is in the Appendix.

effects of investment credit

First, attention is focused on the current account of the balance of payments. By definition, the current account must equal the difference between national income and domestic absorption. National income, in turn, equals, again by definition, the sum of the value of domestic output and the net balance on factor earnings. Consider first the case where credit—from either domestic or foreign sources—is increased ceteris paribus, that is, without a change in domestic financial savings, to finance additional fixed capital formation. Because of the (sometimes quite lengthy) gestation period of fixed investments, the immediate impact of the increase in credit would be the addition to domestic absorption of the amount of investment without adding to current output. 9 This implies, however, that the current account of the balance of payments must deteriorate pari passu with the increased investment expenditures. 10 Moreover, if the increased level of investments was financed through domestic credit expansion, there would be an overall balance of payments deficit and with it a loss of interest-bearing foreign exchange reserves. If the financing was achieved through a fresh inflow of foreign funds, there would be an overall balance of payments equilibrium, but there would also be a need to pay interest on the increased level of foreign debt in the future. If one assumes that there is no significant difference between the interest rate paid on foreign loans and the interest rate earned on foreign exchange reserves, the short-run and long-run current account implications of domestic and foreign credit financing of investments are identical. This again underscores the need to control foreign borrowing for the same demand management reasons as for domestic credit. The immediate significant difference between the use of domestic credit and foreign credit is, however, the effect on the level of the country’s international liquidity, although net foreign financial wealth (defined as the difference between exchange reserves and foreign debt) remains unaffected.

Once the new investment comes on stream it will add to domestic output. Assuming that the investment was profitable from the country’s point of view, then the net addition to domestic output will be larger in value terms than the interest cost implied by the reduction in foreign exchange reserves or the actual interest to be paid on the increased level of foreign debt. With the coming on stream of the investment project, the current account will improve. If, however, the proceeds from the credit expansion were used to finance an investment that was not profitable for the economy as a whole, the initial current account deficit that reflected the cost of investment would be followed by smaller, but persistent, deficits resulting from the excess of the interest costs over the rate of return of the project.

investment in traded goods sector versus nontraded goods sector

In some of the literature, the argument has been advanced that the resources borrowed from abroad need to be channeled into investments in the import substitution and export industries to achieve an improvement in the current account performance and to avoid future debt servicing difficulties, thereby alleviating the foreign exchange constraint that limits economic growth. As the model (see the Appendix) demonstrates, this argument is not generally valid if prices are sufficiently flexible and widespread distortions are absent. The conclusions of the model may appear startling at first: if new investments in the home goods sector make a relatively larger contribution to income growth than those in the traded goods sector, it may be best to permit credit expansion for investment in the home goods sector rather than in the export sector or import substitution sector in order to improve the current account. In the absence of distortions, this does not require deliberate allocative credit policies, since market forces and, particularly, interest rates would correctly channel investment credits into the sector of highest growth.

An intuitive explanation of this result is that the current account reflects the differences between savings and investments in the economy. Promoting faster income growth leads to higher personal savings, greater self-financing out of profits, and, ceteris paribus, also to higher savings by the public sector via increased tax revenue. Looking at the issue from the micro rather than the macro point of view, the key argument is that for consumption (or absorption) purposes home goods and traded goods are substitutes. It follows, then, that, if there is an expansion of home goods production and relative prices are permitted to adjust, this increase in supply will satisfy a part of aggregate demand that would otherwise be directed toward the acquistion of imports, import substitutes, or goods that can be exported as well as domestically consumed. This deflection of demand implies an improvement in the trade balance and the current account. If nominal prices and wages exhibit downward rigidities, the adjustment process may be advanced through exchange rate adjustment to bring about the required relative price movements.

Finally, a twist can be added to this conclusion by considering the implications for the current account of the balance of payments if tax rates and savings propensities differ between the traded sector and the nontraded sector. In this case, the straight-forward recommendation that it is best to invest in the highestyielding sectors and projects is no longer true, since the current account improvement takes place only to the extent that part of the increased income will not be reflected in higher absorption (consumption) but will be transformed into national savings through savings by households, firms, or the government.

In these circumstances, balance of payments policies may require investments to be channeled into sectors where private savings propensities are high or where, through appropriate tax measures, the government can capture and sterilize part of the increased income. It may well be that the high-growth sector may not coincide with the sector where private and public savings potential is highest.

In summary, the profitability 11 of investment in terms of adding to aggregate domestic supply, the gestation period of the projects, the costs of foreign borrowing, and the time horizon chosen will determine whether credit extended for fixed investment will improve or lead to a deterioration in the current account of the balance of payments.

effects of credit for working capital

Turning to the other possibility that domestic or foreign credit is extended for short-term working capital, other than permanent inventory accumulation, and choosing a time horizon for the analysis long enough to cover the production period, a more or less simultaneous stream of increased output will parallel the increased wage bill and other factor payments financed through the credit expansion. This does not require that there be a perfect synchronization between increased output and absorption if only seasonal or other transitory inventories are financed, such as in agricultural production. 12 Again, under the plausible assumption that additional domestic or foreign credits will be utilized only to finance a profitable expansion of production, it follows that the additional wages paid and other factor payments cannot exceed the value of the change in output.

If all newly generated income is spent, the increased level of absorption will just be balanced by the increased value of output, and the current account, as measured over the relevant time period, remains unchanged. If it was foreign credits that were utilized for working capital, an equal build-up of foreign exchange reserves must result, with additional interest earned on the increased holdings of reserves roughly balancing the interest on the increased foreign debt.

Assuming, however, that at least some of the additional income is saved by either households or firms or as the result of government taxation, the current account of the balance of payments must improve. The conclusions are similar to the case of fixed capital formation discussed above: so long as credits (domestic or foreign) are used for profitable purposes, the current account of the balance of payments is likely to improve. In the extension of credit for working capital, gestation periods are absent, so that no initial deterioration of the balance of payments will result, but the current account will reflect the differences between the increased flows of income and domestic absorption. As private or public sector savings determine the difference between these two flows, they also determine the current account performance. Restricting domestic credit to force firms to take recourse to foreign funds can again be used in certain circumstances to improve the reserve position of the country. However, so far as current account policies and overall demand management policies are concerned, domestic and foreign credit can be treated as close substitutes. Moreover, both kinds of credits also have identical effects on the money supply under a fixed exchange rate system.

III. The Case of a Flexible Exchange Rate

As most countries’ currencies are pegged, formally or informally, to either a major currency or to a basket of currencies, the analysis so far has been based on the assumption of a fixed exchange rate. This section briefly sketches how the conclusions would have to be modified if a country was to pursue a flexible exchange rate policy. For reasons of analytical clarity, it is assumed that the country is a “clean floater,” that is, that aside from day-to-day fluctuations, which except in the very short run cancel each other out, there are no changes in the level of foreign exchange reserves and that the foreign exchange market is always cleared through movements of the exchange rate.

Under this specification, the current account and capital account balances must always sum to zero, which implies that if foreign borrowing remains unchanged the change in the current account balance must also be zero. Under a fixed exchange rate system the domestic price level was tied via the exchange rate to the foreign price level (although relative price movements of home goods were permitted), while under the flexible exchange rate regime exchange rate movements will be reflected in movements in the price of the traded goods, expressed in domestic currency units. The analysis must, therefore, take account of the relationship between price level changes and production and savings behavior; moreover, conclusions will depend also on the assumptions made regarding wage behavior.

foreign borrowing and the exchange rate

Any increase in foreign borrowing will tend to generate upward pressures on the exchange rate. However, to the extent that in the short term the credit expansion leads to a pari passu increase in absorption but without a matching increase in production, the emerging current account deficit will tend to offset the exchange rate effects of the inflow of foreign capital. Thus, in the cases of investment credit, or working capital credit used for permanent inventory accumulation, or credit for government consumption, there will be, under the conditions of the model, a perfect offset at the initially prevailing exchange rate. Hence, no exchange rate change will take place, and the conclusions derived above for the case of a fixed exchange rate will continue to hold. Over time, on the one hand, with the need to service the increased level of indebtedness and, on the other hand, with increased production flows, if any, from the use of credit, the exchange rate will tend to appreciate to the extent that the value of additional production outweighs the interest charges on the increased foreign debt and to depreciate if the opposite holds true. The latter will be the case if government consumption or unproductive investments were financed.

If a working capital expansion, other than for permanent inventory accumulation, takes place, there will be no initial change in absorption at the prevailing exchange rate, so that the exchange rate must appreciate as a result of the capital inflow. This appreciation will lead to a reduction in the nominal price of traded goods. Depending on wage rigidities, this would lead to either unemployment or reduced nominal wage rates. If wages decline, further production increases will result, since the real volume of working capital has increased. On the demand side, households will find that the appreciation of the exchange rate has increased the real value of their savings by lowering traded goods prices. Households start to dissave, so that a current account deficit will emerge, and the new appreciated equilibrium exchange rate will equate this deficit and the capital account surplus.

domestic credit expansion and the exchange rate

Any increase in domestic credit expansion that leads to increased absorption by the recipient of the credit, such as investment or government credits, will at the prevailing exchange rate put pressure on the current account. However, in the absence of foreign borrowing, including induced capital flows, the current account must remain in balance if the reserve level is to remain unchanged; hence, the exchange rate must depreciate. Rising prices of traded goods result from the exchange rate adjustment, and households will find the value of their savings reduced. As households cut back on consumption in an attempt to restore the real level of their savings, absorption declines. Exchange rate movements leading to rising prices force households to increase their nominal savings and thus to release the real resources to finance investment or government consumption. This contrasts with the fixed exchange rate case discussed above, in which these additional resources were supplied through the current account deficit. For domestic financing of increased government consumption, part of the real resources will also be released from the productive sector, since the real value of investment and working capital credit declines with rising product and factor prices. Inflationary government finance will thus result in a cutback of present and future production, an effect which in the recent literature is often referred to as stagflation.

IV. Characteristics of Credit Policies in Fund Programs

In addition to limitations on official foreign borrowing, Fund programs typically contain an overall ceiling on domestic bank credit and a subceiling on credit to the government from either the domestic banking system or the central bank. The limitations on official foreign borrowing13 and domestic credit to the government are aimed at avoiding internal or external financing of excessive government expenditures, which would prejudice the achievement of the balance of payments objective. In general, the objective of credit ceilings is not to limit the extension of credit for working capital to the private sector and only rarely to contain credits for productive investment projects, except perhaps for low-yielding investments by the government (e.g., certain infrastructure projects) with often long gestation periods. Support for this conclusion comes also from the fact that foreign borrowing with long maturities (say, over 10 to 12 years), which is typically project-related, usually obtained favorable treatment under the foreign debt ceiling in the past and, more recently, is no longer subject to external debt limitations. Although the overall demand management objectives could in most cases be achieved by limiting overall credit extended by the banking system, the subceiling on domestic credit to the government can be used to assure that this is not brought about through a crowding out of private sector investment and working capital requirements. It therefore appears that the structure of a typical Fund program is potentially consistent with the need to reduce overall credit expansion while safeguarding “productive” credit requirements to assure investment and growth, on the one hand, and high levels of current production and nongovernment employment, on the other hand.14

Throughout this paper it has been argued that, at least in the longer run, increased levels of profitable credit will strengthen the current account performance rather than place a burden on the balance of payments. Clearly, profitability must thereby be measured in terms of actual increases in domestic value added. If cost-price distortions are widespread, some private sector investment may be profitable so far as the entrepreneur is concerned but may not result in an adequate addition to domestic value added. It is, therefore, important to note that financial stabilization programs supported by the use of Fund resources were frequently complemented by measures relating to exchange rate, interest rate, and other pricing policies designed to reduce or eliminate existing distortions, and particularly the prevailing financial repression, so as to promote appropriate resource allocations. Such measures must be considered to be an integral part of growth and balance of payments policies, since in the presence of widespread distortions tight credit policies would lead to an excessive reduction in current and future domestic supply. Finally, although Fund-supported stabilization programs contain quantitative limitations on global credit aggregates, this does not imply that such programs encourage rationing of credit in the economy through credit allocation rather than through an appropriate interest rate policy.


A very simple and abstract theoretical model of an open economy characterized by the presence of financial repression is developed in this section. The salient features of this model are that current and future production levels are linked directly to the volume of financial intermediation through the domestic banking system and to the availability of foreign financing.

First, it is assumed that the economy produces and trades in one composite good.15 This assumption is relaxed later, when output is disaggregated into traded and nontraded goods. For convenience and without loss of generality, units are defined so that both the price level and the exchange rate are equal to unity.

credit, production, and income

The model distinguishes financing by origin and use. Total credit is denoted by Cr, the superscripts * and d denote foreign and domestic sources, respectively, and the postscripts W, I, and G denote the use of the credit for the financing of working capital, investment, and government needs, respectively, so that

  • Total credit (source) Cr = Cr* + Crd

  • Total credit (use) Cr = CrW + CrI + CrG

    • and

  • Foreign credit (use) Cr* = CrW* + CrI* + CrG*

  • Domestic credit (use) Crd = CrWd + CrId + CrGd

    • and

  • Working capital (source) CrW = CrW* + CrWd

  • Investment credit (source) CrI = CrI* + CrId

  • Government credit (source) CrG = CrG* + CrGd

Also, changes in the level of credit outstanding are denoted by a dot, so that, for example, CṙI* denotes a flow of investment credit from foreign sources, while CrG denotes the level or stock of overall (i.e., domestic and foreign) outstanding credit to the government.

Production, X, is assumed to depend on the size of the existing real capital stock, K, and the amounts employed of the variable factor, V. While the stock of real capital is given at any moment in time, employment of the variable factor will depend on its price relative to the price of the final product and the financing possibilities open to the firm. Without loss of generality, it is assumed that, at the prevailing prices of inputs and outputs, firms would be willing to expand production but are effectively constrained in their utilization of the variable factor by the availability of financial working capital. In the absence of self-financing, financial credit available from domestic and foreign sources for working capital, CrW, limits the maximum employment of the variable factor; in this assumption the model is similar to the Ricardian specification of a wage fund.16


It is assumed that all changes in financial working capital are reflected in changes in money balances held by firms. The stock of real capital, K, including permanent inventories, is assumed to change over time, through investment, K˙ In addition to the possibility of self-financing of investment, SF, fixed capital formation also depends on the amount of total investment credits, CrI, available. It is also assumed that at the prevailing domestic interest rate all new investments are profitable, as seen by the entrepreneur.17


National income, Y, equals the sum of the value of production, X, and the earnings, iZ, on net claims on foreign economies, Z; i thereby denotes the prevailing rate of interest in international financial markets.18Z is defined as the difference between reserves, R, and net foreign indebtedness: Z = R -Cr*. In the absence of differential interest rates on foreign debt and foreign reserves, iZ will be positive if the country’s foreign reserves are higher than its debts (Z > 0) or negative if debts to other countries exceed the level of reserves (Z < 0).


aggregate demand and the external sector


Savings by the private sector can take the form of increasing the level of self-financing of investment needs either through a complete or partial retention of profits by the firms, or through an accumulation of financial wealth in the household sector. For convenience, it is assumed that money is the only financial asset available, so that the overall flow of private savings, S, can be expressed as


where MHH denotes the level of cash balances of households and MḢH the flow.

It is assumed that both the desired level of financial wealth of the household sector, MHHD, and the desired level of self-financing by firms, SFD, are increasing functions of income.


It is further assumed that the flows of household savings and self-financing are proportional to the difference between desired and actual levels, the latter denoted by a bar:


In long-run equilibrium, MHHD=MHH¯andMH˙H=0,andSFD=SF¯andSF˙=0. Starting from this position, a marginal increase in income has the following effect on savings and self-financing:


and obviously the flow of savings generated by a change in income, dY, must then be S = (shh + sf)dY.

Government sector

As an identity, government expenditure must be equal to the tax revenue plus the change in net credit from domestic and foreign sources to the government, CṙI, which can be positive or negative:


where t denotes the tax rate on income. For convenience, it is assumed that all government expenditures take the form of consumption and that investments by the public sector, including those of public sector enterprises, are included in the private sector.


In the model, domestic demand or absorption, A, has three components: private consumption, C; fixed investment, K; and government expenditures, G.


From equations (2), (4), and (9), equation (10) can be rewritten as




In an open economy, the current account of the balance of payments, CA, can always be expressed as the difference between income and absorption, so that




Under the specification of the model, a current account deficit emerges if credit for investment or government consumption exceeds financial savings of households.

One also notes that CA=Z˙. As changes in foreign credit are denoted by Cṙ*, the overall balance of payments, that is, reserve changes, , can be expressed as




Equation (14) shows that under a fixed exchange rate system the current account of the balance of payments is determined by the overall credit expansion and the use made of credit 19 but independent of whether credit originates from foreign or domestic sources, while equation (16) shows that, in addition to the overall level of credit and its sectoral distribution, the distinction between bank financing from domestic and from foreign sources is very important as far as reserve developments and the overall balance of payments are concerned.

Finally, it is possible to derive the monetary survey of the model economy; changes in money supply come about through a change in foreign exchange reserves or in domestic credit. On the demand side, changes in money holdings are divided between the changes in the money balances held by the private household sector and changes in the financial working capital held by firms, so that20


policy exercises: credit expansion and the current account

This section analyzes the effects on the current account of the balance of payments of expansions in credit for working capital, investment, and government expenditure, respectively. The starting point is equation (14). Note that in long-run equilibrium, S,MH˙H,SF˙,Cr˙I,andCr˙G=0, so that the current account balance must also be zero.

Totally differentiating equation (14) yields dcA = dMḢH; − dCṙI − dCṙG; therefore,


Starting from stock equilibrium, dCrW = dCṙW, dCrI = dCṙI, dCrG = dCṙG, and dMHH = dMḢH; that is, as these flows were originally zero, changes in these flows must be equal to the new flows. Also note that MHH¯CrW,MHH¯CrI,MHH¯CrG=0, since credit expansion to firms and governments does not have any initial effect on the existing monetary balances of households. (It will be shown below that cash balance effects, reflecting price changes, play a significant role in the flexible exchange rate case.)

First, consider the effects of an expansion in credit for working capital on the current account of the balance of payments, holding investments and government expenditures (dK˙,dG=0) constant rather than investment credit and credit to the government. This implies that changes in tax revenue (tdY) and in retained profits (sfdY) are not automatically spent but are reflected in changed credit needs of these respective sectors: dCrI = -sfdY; dCrG = -tdY. From equation (18) and making use of the investment function and government budget constraints, dCrl = dK - sfdY and dCrG = dG - tdY. One therefore obtains


since YCrW=XCrW>0 and shh, s f, t > 0.

An increase in credit for working capital initially generates additional demand for factor inputs but not for produced goods; therefore, there is no initial impact on absorption and hence on the current account, and, therefore, Z remains initially unaffected. As marginal savings propensities in the private and public sectors are positive, the expansion of production and income through an increase in credit for working capital will lead to an unambiguous improvement in the current account. Aside from larger savings by households, rising income levels, and with that increased self-financing and higher tax revenue, lower the credit needs for the financing of investment and government expenditures. Note that if, however, firms and governments were to spend automatically any increase in retained profits and in tax revenue, the improvement in the current account would be significantly smaller and equal only to the increase in savings of private households.

Next, consider the effects of an increase in investment credit on the current account; note that working capital and government expenditures are held constant and that all increased self-financing brought about by changes in income is reflected in additional investment, so that dK = dCrI + sf dY and


Note that ∂Y/∂ = ∂X/∂K - i ≷ 0 and positive if the rate of return on investment exceeds the foreign rate of interest. If the investment financed through investment credit expansion is profitable (i.e., ∂X/∂K > i), then it becomes a matter of the time horizon chosen whether (shh + t) ∂Y/∂CrI exceeds unity. For very short time horizons, increased fixed capital formation results in an equally sized current account deterioration, reflecting the increased absorption level. Equation (20) shows that the higher the productivity of investment and the higher the marginal propensity to save and to generate national savings through taxation the shorter the time required for the overall effect of productive investment to make a positive contribution to the current account performance. In the long run, the current account effect of increased productive investment will unambiguously be positive. If, however, the rate of return on the initial investment is less than the interest rate on foreign debt (or on foreign exchange assets), the initial current account deficit associated with the expansion in investment will be followed by smaller deficits in the future. In this instance, the interest cost has exceeded the contribution of new investment to income; hence, savings, and with that the current account, will deteriorate.

Finally, consider the current account effects of a change in government credit. Note that the total change in government expenditures—although initially equal to the change in credit—will also depend on the effects of induced income changes. For the following results, working capital’ and investments, rather than investment credit, are held constant:


Note that ∂Y/∂G = -i, so that dCA = -[(shh + s f) i/1—it] dCrG—dCrG < 0. The increase in government credit is initially reflected in an equally sized current account deficit, which needs to be financed. Note that savings by households and self-financing by firms decline as the debt burden reduces income, so that credit needs to be expanded more to keep investment unchanged. The reduction in income leads to a persistent deterioration in the current account, after the initial impact of increased spending, which is also negative.

extension of analysis to two-sector model

Disaggregation of the model on the production side into traded and non-traded goods sectors is needed to determine whether promotion of investment and employment in the traded goods sector (i.e., export industries and import substitution industries) is preferable from a balance of payments view to investments in other sectors of the economy.

As above, it is assumed that output and employment depend on the capital stock and the amount of the variable factor employed, and the assumption is added that these factors of production are specific to each sector. Denoting variables of the home goods sector and traded goods sector with the superscripts h and t, respectively, equation (1) can be rewritten as


Using the price of the traded goods as the denominator and denoting the relative price of the home goods by q, equation (3) becomes


However, it still remains true that the current account balance must equal the difference between income and expenditures; or, putting it slightly differently, the current account deficit equals the difference between private savings by households and firms and the government deficit and private investment. Equation (14), therefore, retains its validity. What changes, however, is that equation (24) replaces equation (3), so that one obtains the following effects of changes in the investment level (rather than in credit):


where E=∂q∂XhXhq


As the two expressions are identical except for the term in the inner square brackets, the current account effects of investment in the nontraded and traded goods sectors can be analyzed by contrasting these two terms.

XhqKh(1+E)andXtKt measure the value added from increased investments in the home goods sector and traded goods sector, respectively. The inclusion of E (which is the inverse of the price elasticity of demand for home goods) reflects the fact that relative prices of home goods will change with an expansion of production, while traded goods prices remain fixed.

Clearly, if XhqKh(1+E)>XtKt, then investments in the home goods sector lead to larger income gains than do investments in the traded goods industries. As greater income levels, ceteris paribus, imply higher savings by households, firms, and governments, investment in the most productive sector will lead to the most favorable current account performance (see equations (25) and 26)). If investments in the home goods sector are more profitable (in terms of value added) than those in the traded goods industries, then the current account objective is best served by investing in the home goods sector, and vice versa if investments in the traded goods industries are more profitable. This result depends crucially on the assumption that the marginal savings and tax rates do not differ among sectors.

flexible exchange rate case

Under the fixed exchange rate specification the differentiation of equations (13), (15), and (16) with respect to credit expansion (of various use and origin) yielded the current account and the overall balance of payments effects. Clean floating requires, however, that the current account and capital account sum to zero, so that CACrd=0;CACr*=1;R˙=0

With the current account and capital account predetermined by the origin of a credit expansion, it is now possible to solve for the production, exchange rate, and price level effects of credit expansion. For simplicity, the analysis is conducted in the framework of the one-sector model.

Output, income, and the price level

As under a flexible exchange rate system the price of output in domestic currency may vary, it is necessary to rewrite equations (1) and (3) in real terms:




Note that the domestic and foreign price levels, p and p*, are linked by the “law of one price,” so that P=1ep* and that reserves, R, and foreign debt are denominated in foreign currency units and therefore are unaffected in real terms by domestic price level and exchange rate changes. To determine the effect of a domestic price level change on real income, equation (28) is differentiated with respect to e, and p* is treated as an exogenous constant:


For future reference, (Yep*)∂e is denoted by g (g > 0). It is important to note that a depreciation of the exchange rate (i.e., a reduction in e), which is equivalent to an increase in the domestic price level, ceteris paribus, reduces the real value of financial working capital and thus output.

Domestic and foreign credit and the exchange rate

Rewriting equation (7) in real terms, and assuming for convenience that the desired stock of real money balances is strictly proportional to real income, yields


where the proportionality factor, k, is the inverse of the long-run velocity of money. Transforming equation (14) into real terms and making use of equation (29), as well as abstracting from self-financing and changes in investment credits, yields


One notes that the current account is measured in units of foreign currencies.

Differentiating equation (30), and solving for the effects of a change in domestic credit to government, yields


Note that dCA must be equal to zero, since, in a flexible exchange rate system, the current account stays balanced in the absence of capital flows. Reordering the terms permits solving for the exchange rate adjustment brought about by the domestic credit expansion to government:


that is, the exchange rate depreciates. If real income was to remain unchanged (g = 0), this would simplify to


This is the Dornbusch result that the exchange rate adjustment required to restore equilibrium after a domestic credit expansion depends on the stock adjustment speed, α, and on the size of money balances, which are reduced in real terms through the increase in the price level. The more quickly households attempt to restore their level of real savings the smaller the required depreciation. The fact that real output and, hence, income decline with the reduction in real working capital (as measured by g) implies that the original excess supply of liquidity in the economy is even higher, and a larger depreciation of the exchange rate will result.

Finally, differentiation of equation (30) with respect to an expansion of foreign credit to the government yields


so that α[kgMHH¯p*]eCrG*=0, since, under a flexible exchange rate systern, dCA = -dCrG*. (Note that foreign borrowings are denominated in foreign currency units.) From equation (32) it follows that


While domestic financing of government deficits raises the price level and depresses output and the exchange rate, foreign financing will leave these variables initially unchanged. As the capital inflow makes foreign savings available for the financing of government expenditures, no exchange rate adjustment is required initially to free the necessary domestic resources. In the longer run, however, a depreciation will result from the need to free the required resources to service the increased foreign debt. Mutatis mutandis, the model can also be solved for domestic and foreign financing of private sector capital needs.


  • Bailey, Martin J., National Income and the Price Level: A Study in Macroeconomic Theory (New York, 1971).

  • Ben-Zion, Uri, and Vernon W. Ruttan,Money in the Production Function: An Interpretation of Empirical Results,Review of Economics and Statistics, Vol. 57 (May 1975), pp. 24647.

    • Search Google Scholar
    • Export Citation
  • Bottomley, Anthony,The Structure of Interest Rates in Underdeveloped Rural Areas,Journal of Farm Economics, Vol. 46 (1964), pp. 31322.

    • Search Google Scholar
    • Export Citation
  • Brunner, Karl, and Allen H. Meltzer,The Uses of Money: Money in the Theory of an Exchange Economy,American Economic Review (December 1971), pp. 784805.

    • Search Google Scholar
    • Export Citation
  • Dornbusch, Rudiger, and Jacob Frenkel,Inflation and Growth,Journal of Money, Credit and Banking, Vol. 5 (February 1973), pp. 14156.

    • Search Google Scholar
    • Export Citation
  • Fischer, Stanley,Money and the Production Function,Economic Inquiry, Vol. 12 (December 1974), pp. 51733.

  • Friedman, Milton,The Demand for Money: Some Theoretical and Empirical Results,Journal of Political Economy, Vol. 67 (August 1959), pp. 32751.

    • Search Google Scholar
    • Export Citation
  • Friedman, Milton, The Optimum Quantity of Money, and Other Essays (Chicago, 1969).

  • Fry, Maxwell J.,Money and Capital or Financial Deepening in Economic Development?Journal of Money, Credit and Banking, Vol. 10 (November 1978), pp. 46475.

    • Search Google Scholar
    • Export Citation
  • Galbis, Vicente,Financial Intermediation and Economic Growth in Less Developed Countries: A Theoretical Approach,Journal of Development Studies, Vol. 13 (January 1977), pp. 5872; reproduced in Finance in Developing Countries, ed. by P. C. I. Ayre (London, 1977), pp. 5891.

    • Search Google Scholar
    • Export Citation
  • Guitian, Manuel,Credit Versus Money as an Instrument of Control,Staff Papers, Vol. 20 (November 1973), pp. 785800.

  • Gurley, John Grey, and Edward S. Shaw,Financial Aspects of Economic Development,American Economic Review, Vol. 45 (September 1955), pp. 51538.

    • Search Google Scholar
    • Export Citation
  • Gurley, John Grey, and Edward S. Shaw,Financial Intermediaries and the Saving-Investment Process,Journal of Finance, Vol. 11 (May 1956), pp. 25776.

    • Search Google Scholar
    • Export Citation
  • Gurley, John Grey, and Edward S. Shaw, Money in a Theory of Finance, The Brookings Institution (Washington, 1960).

  • Kapur, Basant,Alternative Stabilization Policies for Less Developed Economies,Journal of Political Economy, Vol. 84 (August 1976), pp. 77795.

    • Search Google Scholar
    • Export Citation
  • Keller, Peter M.,Controlling Fluctuations in Credit,Staff Papers, Vol. 24 (March 1977), pp. 12853.

  • Khan, Mohsin S., and Pentti J. K. Kouri,Real Money Balances as a Factor of Production: A Comment,Review of Economics and Statistics, Vol. 57 (May 1975), pp. 24446.

    • Search Google Scholar
    • Export Citation
  • Levhari, David, and Don Patinkin,The Role of Money in a Simple Growth Model,American Economic Review, Vol. 58 (September 1968), pp. 71353.

    • Search Google Scholar
    • Export Citation
  • McKinnon, Ronald I., Money and Capital in Economic Development, The Brookings Institution (Washington, 1973).

  • Machlup, Fritz, International Trade and the National Income Multiplier (New York, 1965).

  • Moroney, John R.,The Current State of Money and Production Theory,American Economic Review, Papers and Proceedings, Vol. 62 (May 1972), pp. 33543.

    • Search Google Scholar
    • Export Citation
  • Mundell, Robert A., Monetary Theory: Inflation, Interest and Growth in the World Economy (Pacific Palisades, Calif., 1971).

  • Nadiri, M. I.,The Determinants of Real Cash Balances in the U.S. Total Manufacturing Sector,Quarterly Journal of Economics, Vol. 83 (May 1969), pp. 17396.

    • Search Google Scholar
    • Export Citation
  • Niccoli, Alberto,Real Money Balances: An Omitted Variable from the Production Function?—A Comment,Review of Economics and Statistics, Vol. 57 (May 1975), pp. 24143.

    • Search Google Scholar
    • Export Citation
  • 0i, Walter,Labor as a Quasi-Fixed Factor,Journal of Political Economy, Vol. 72 (December 1964).

  • Polak, J. J.,Monetary Analysis of Income Formation and Payments Problems,Staff Papers, Vol. 6 (November 1957), pp. 150.

  • Polak, J. J., and Victor Argy,Credit Policy and the Balance of Payments,Staff Papers, Vol. 18 (March 1971), pp. 124.

  • Prais, Zmira,Real Money Balances As a Variable in the Production Function,Review of Economics and Statistics, Vol. 57 (May 1975), pp. 24344.

    • Search Google Scholar
    • Export Citation
  • Prais, Zmira,Real Money Balances as a Variable in the Production Function,Journal of Money, Credit and Banking, Vol. 7 (November 1975), pp. 53543.

    • Search Google Scholar
    • Export Citation
  • Reichmann, Thomas M., and Richard T. Stillson,Experience with Programs of Balance of Payments Adjustment: Stand-By Arrangements in the Higher Tranches, 1963-72,Staff Papers, Vol. 25 (June 1978), pp. 293309.

    • Search Google Scholar
    • Export Citation
  • Shaw, Edward S., Financial Deepening in Economic Development (Oxford University Press, 1973).

  • Sidrauski, Miguel,Inflation and Economic Growth,Journal of Political Economy, Vol. 75 (December 1967), pp. 796810.

  • Sinai, Allen, and Houston H. Stokes,Real Money Balances: An Omitted Variable from the Production Function,Review of Economics and Statistics, Vol. 54 (August 1972), pp. 29096.

    • Search Google Scholar
    • Export Citation
  • Sinai, Allen, and Houston H. Stokes,Real Money Balances: An Omitted Variable from the Production Function?—A Reply,Review of Economics and Statistics, Vol. 57 (May 1975), pp. 24752.

    • Search Google Scholar
    • Export Citation
  • Tobin, James,Money and Economic Growth,Econometrica, Vol. 33 (October 1965), pp. 67184.


Mr. Keller, economist in the External Finance Division of the Exchange and Trade Relations Department, is a graduate of the University of Würzburg, Germany, and the University of Rochester, New York.


Compare, for example, the theoretical treatment by Friedman (1959 and 1969), Sidraquski (1967), Levhari and Patinkin (1968), Nadiri (1969), Bailey (1971), Brunner and Meltzer (1971), Mundell (1971), Dornbusch and Frenkel (1973), McKinnon (1973), Kapur (1976), and others. Of particular interest is Levhari’s and Patinkin’s (1968) criticism of Tobin (1965) for omitting in his growth model the role of real money balances as productive input. Empirical attempts by Sinai and Stokes (1972) to estimate the role of real money balances for production also sparked a heated debate. See, for example, the criticism by Ben Zion and Ruttan (1975), Khan and Kouri (1975), Prais (1975 a and 1975 b), Niccoli (1975), and others, and the rejoinder by Sinai and Stokes (1975).


Except to the extent that recourse to foreign borrowing enables a country to mobilize real foreign resources by financing a current account deficit that perhaps could not have been financed out of existing foreign exchange reserves.


The argument is, therefore, in a wider sense, akin to what Moroney (1972) terms the Austrian view of financial working capital as a fund from which advances are made to real factor inputs, including labor.


Compare also Bottomley (1964).


As discussed below, the balance of payments implications are quite different in the case of a permanent rather than transitory inventory build-up.


This is not to say that, as seen from an individual firm, there exists no scope for internal substitution between categories of credit.


This analysis applies also to the case of a permanent increase in inventory investment.


As pointed out above, for the purposes of this investigation we abstract from the possibility that adding to aggregate demand per se will increase output. This assumption is not as restrictive as it may appear, since, in the typical country with a current account problem, insufficient levels of domestic demand are not the problem.


With profitability appropriately assessed relative to increased interest cost of foreign borrowing to the economy rather than relative to domestic interest rates.


If, however, the increase in working capital was used to finance permanent increases in inventories of intermediate or final goods, this would not be the case, and the inventory build-up, just as with fixed capital formation, would lead to an initial excess of absorption over production and the current account would deteriorate.


Few programs contain limitations on private foreign borrowing.


It remains an empirical question whether financial programs supported by the use of Fund resources have resulted in relatively or absolutely faster real growth of private sector credit from domestic and foreign sources or, to use McKinnon’s (1973) terminology, whether a significant reduction of financial repression was achieved by limiting the crowding-out effect of financing excessive government expenditures.


This assumption does not imply that there is only one good produced, traded, and consumed in the economy, but that constant relative prices permit the aggregation of the various goods into one composite commodity.


It is assumed that all retained earnings are utilized for the financing of fixed capital formation and permanent inventory accumulation.


This does not imply necessarily that investments are profitable in the macroeconomic sense; on the contrary, it is typical of a financially repressed economy that part of the investment will not meet this criterion, since domestic interest rates are arbitrarily depressed.


No other factor payments (e.g., workers’ remittances) are assumed to exist.


Note that savings are, inter alia, a function of income, and income developments depend on the use made of credit.


Substituting equations (15) or (16) for Ṙ shows that this specification is indeed consistent with the behavioral assumptions of the model.

IMF Staff papers: Volume 27 No. 3
Author: International Monetary Fund. Research Dept.