CHAPTER 21 Factors in Stabilization Programs

International Monetary Fund
Published Date:
February 1996
  • ShareShare
Show Summary Details
Emil G. Spitzer

The purposes of stabilization programs and the evolution of stand-by arrangements as a means by which the Fund is able to assist members undertaking such programs have been described in Chapter 20. It has been shown that in considering members’ proposals, the Fund focuses attention on certain critical factors by which the success of a program can be tested.

The main considerations in the selection of the criteria of performance have been (1) that they are capable of being expressed in quantitative or objective terms, (2) that actual performance can be compared with the criteria without delay due to time lags, and (3) that the criteria are significant enough to serve as a basis for an appraisal of the economy as a whole. The present chapter examines typical policy objectives in detail and describes the difficulties experienced in defining them.


Domestic credit ceilings

In the majority of financial programs the control of domestic bank credit has been of particular importance. The major instrument used for this purpose is a ceiling on the expansion of the domestic assets of the central bank during the period of the stand-by arrangement. An over-all ceiling is capable of being defined and administered easily; it also has the advantage of providing operational flexibility to the central bank. Frequently, however, separate ceilings have been applied to different categories of central bank credit. For example, in some programs there has been one ceiling on credit to the government and another on credit to the private sector; when budget deficits have been the primary cause of financial instability, this has been useful in indicating the magnitude of the government’s share in the total credit expansion. Sometimes there are separate ceilings for central bank credit to the government, to various official agencies or state-owned enterprises, and to the banking system; the use of this technique has been based on the view that separate ceilings on each major category of central bank credit would facilitate the control of over-all credit expansion. Sometimes ceilings on credit to particular official agencies have been not so much an element in the over-all program of credit control as a means of encouraging changes in the financial operations and pricing policies of the agencies concerned or of ensuring that compliance with the over-all ceiling does not involve undue restraint on credit to the private sector.

Credit ceilings have frequently been phased in order to take account of seasonal changes in the demand for bank credit; it has also been necessary to provide some accommodation for the government where revenues or expenditures are subject to significant seasonal variation.

A ceiling on the domestic assets of the central bank has usually been supplemented by provisions regarding the policy to be followed in respect of changes in the legal requirements governing the minimum reserves of the commercial banks. Where, at the beginning of a program, the commercial banks were in a highly liquid position, programs have provided for an increase in reserve requirements in order to reduce excess liquidity and keep the expansion of credit within the desired limit. In some other programs, marginal reserve requirements (i.e., requirements applying only to an expansion of bank deposits after a certain date) have been introduced to increase the control over monetary expansion. On some occasions the amount of rediscounts to the commercial banks has been limited, in order to achieve the desired effect on commercial bank credit. Some programs, however, have aimed at a more comprehensive control of credit by establishing additional ceilings on commercial bank credit to the government or even broader limits on government indebtedness to commercial banks and to the private sector.

Provisions of a more selective nature, regulating the flow of credit to particular purposes or industries, have been part of a number of programs but only rarely have they been quantified. The central bank has exercised such selective controls either by establishing quantitative limits on the amount of the various types of bills which are eligible for rediscounting or by changing the eligibility requirements, or by fixing subceilings for credits to particular industries or purposes under the broader ceiling on credit to the private sector.

In some situations an increase in the central bank’s rediscount rate has been used as a supplement to quantitative limits on the supply of credit. However, where the rate of inflation is rapid, it is difficult to use interest rate policy as a credit control device; in such conditions, an adequate interest rate in real terms would necessitate a very considerable increase in the nominal interest rate.

The policy instruments mentioned above are aimed at controlling the total expansion of bank credit indirectly by reducing the liquidity of the commercial banks. Because of particular institutional arrangements, some countries have preferred to regulate the amount of credit extended by the banking system as a whole. This regulation has taken different forms, depending largely upon the degree of flexibility desired by the monetary authorities and the legal powers they possess for controlling commercial bank operations. For example, in some programs a ceiling has been placed on total credit extended by the banking system. In others, a ceiling has been placed on credit extended by the commercial banks, either with or without a ceiling on credit by the central bank. In some circumstances a ceiling has been established for each bank, allowing it to expand its lending by a specified percentage over the level existing as of a base date.

The choice between a ceiling on credit of the banking system as a whole and a ceiling on central bank credit only is influenced largely by institutional factors. Use of the former has usually been confined to countries where commercial banks are relatively few in number and are subject to strong central bank control. Another important consideration in deciding whether a ceiling should cover commercial bank assets has been the promptness with which the relevant data become available. Credit ceilings covering commercial bank credit may prove ineffective as a guide to policy if there are substantial lags in reporting data.

In the formulation of credit ceilings, various problems have arisen with regard to the inclusion or exclusion of certain central bank assets. The practice of covering credit to the central government, to state and local governments where applicable, and to other official entities has been fairly uniform. Further, the inclusion in credit ceilings of loans or discounts to commercial banks and other banking institutions, such as development banks and agricultural banks, has posed no particular problem. In many countries central banks do not lend directly to the private sector, but where they do (i.e., where they engage in commercial banking activities) the ceilings have covered such credit.

In several instances credit for the marketing of major agricultural crops has had to be excluded from domestic credit ceilings because the financial requirements depended on the size of the crops and therefore could not be predicted. For example, in a program for Ceylon the ceiling excluded finance for rice marketing, and in a program for Ghana it excluded financing of cocoa purchases by the Cocoa Marketing Board. In several other programs (e.g., those for the Sudan and the United Arab Republic), such problems have led to the establishment of separate ceilings on the financing of agricultural crops. The advantage of this technique is that it prevents the use of such credit for other purposes. Sometimes the letter of intent has specified the estimated size of the agricultural crop on which the ceiling on marketing finance is based and has provided for an adjustment in the ceiling, if necessary, after consultation and agreement with the Fund.

In the majority of programs, domestic credit ceilings have not covered items which in the balance sheets of the central bank are included in “other assets” (fixed assets, etc.). The smallness of such “other assets” has been sufficient reason for their exclusion. In addition, delays in the availability of data covering “other assets” and their tendency to exhibit random short-term fluctuations would have complicated the task of setting ceilings.

With regard to the coverage of central bank assets, it should be noted that two different approaches have been adopted in defining the assets subject to domestic credit ceilings. The more common method, called the “selected assets” approach, has been to specify for inclusion certain balance sheet items which cover the main domestic assets of the central bank. The alternative method, the “residual” approach, has been to define domestic assets as the difference between the total assets of the central bank and specified assets, primarily foreign assets, which have been excluded from the ceiling. The “residual” approach has been used mainly in programs for countries in the Western Hemisphere. Both the “selected” and the “residual” approach have also been applied to the liability side of the balance sheet. Where the “residual” approach has been adopted on both sides of the balance sheet, the ceiling has sometimes been defined as the difference between the currency issue and the net foreign assets of the central bank.

“Residual” ceilings have tended to be more comprehensive, covering in most cases “other assets” as well as the main credit operations of the central bank. The main argument for adopting the “residual” approach has been that it is likely to reduce the chances of credit being extended outside the coverage of the ceiling; any new type of domestic operation of the central bank, unforeseen at the time of the establishment of the ceiling, is automatically covered. On the other hand, ceilings on “selected assets” have usually been more meaningful to the authorities of the country concerned; when the ceilings are applied to such specific central bank assets as credit to the government and credit to the commercial banks, which are subject to the direct control of the authorities, there is a greater chance of compliance. The “residual” approach, however, does not necessarily preclude the use of additional ceilings on specific components of domestic credit. In some programs, for example, a global ceiling arrived at on a “residual” basis has been supplemented by a ceiling on credit to the government sector, or by a number of separate ceilings each covering an important component of central bank credit.

Some problems in the formulation of credit ceilings have been related to the fact that monetary expansion or contraction may result not only from movements in the domestic assets of the central bank, but also from movements in its domestic liabilities. For instance, the monetary consequences of a reduction in the government’s deposits with the central bank are the same as those of an increase in the central bank’s advances to the government. Netting of assets and liabilities has therefore been an important aspect of the formulation of credit ceilings.

In establishing most domestic credit ceilings, the government’s deposits are netted against its liabilities to the central bank. This is not done, however, when government deposits are low, with little scope for further reduction, or when the government has given other undertakings directly related to its revenue and expenditure.

In the majority of instances, deposits of the commercial banks with the central bank have not been netted, particularly in those programs in which the ceilings have been related to over-all credit operations of the banking system, interbank claims being excluded. However, the freedom of the commercial banks to reduce their deposits with the central bank has been limited by undertakings to maintain specified reserve ratios. This has also been the practice in programs where the ceilings have been applied only to the central bank. Sometimes, commercial banks’ deposits with the central bank have been excluded from the ceiling even without a commitment on reserve requirements, either because these deposits were relatively small or because an excessive expansion of commercial bank credit was considered unlikely. In some instances, these deposits have been netted because of the unsatisfactory record of the commercial banks in complying with reserve requirements. In other instances, the authorities of the country concerned have preferred this arrangement because of its greater operational flexibility compared with a commitment to maintain specified reserve requirement ratios.

Other arrangements involve the compulsory holding of deposits with the central bank. The most common practice requires the commercial banks to make advance deposits for imports. Although these deposits are maintained primarily for balance of payments reasons, they also have some monetary effects: an increase in them represents a contractionary factor while a decrease is a source of monetary expansion.

In most programs for countries with compulsory deposit arrangements such deposits have not been netted against liabilities, mainly because the amounts involved were small or were not likely to change substantially. In several programs, however, where compulsory deposits have not been so netted, they have, nevertheless, been taken into account: for example, some items, such as losses on official foreign exchange transactions, have been excluded from the ceilings in the expectation that they would be a source of monetary expansion offsetting the contractionary effects of expected increases in advance deposits for imports.

Foreign transactions

The majority of countries that have stand-by arrangements associated with stabilization programs are recipients of foreign aid as well. As the commodities imported under the various aid arrangements are sold, they generate counterpart funds which in some countries have a significant effect on the monetary situation. While counterpart funds are being accumulated the effect is contractionary, but when the funds are spent they represent a source of monetary expansion. The monetary effects of significant movements in counterpart funds are usually taken into account in the formulation of credit ceilings by netting the counterpart fund accounts. In a few countries where counterpart funds are not deposited with the central bank but with commercial banks, they have not been covered by a ceiling because the amounts involved have not been significant.

In some countries characterized by exchange rate instability, foreign exchange operations have been the source of substantial central bank profits or losses. These may be divided into two groups, those arising from the revaluation of foreign assets and foreign liabilities in connection with exchange rate adjustments, and those arising from a variety of other foreign exchange operations.

As a result of devaluing the exchange rate, the domestic currency value of the central bank’s foreign assets and liabilities is increased. If the assets exceed the liabilities, a profit is earned, and if the liabilities are greater, there is a loss; but as revaluation of the assets and liabilities involves only book entries, the profits and losses arising during the period of the stand-by arrangement are not included in credit ceilings. Profits arising from revaluations may, however, pose a problem if they are transferred to the government; in the absence of provisions to the contrary, this increases the margin available for credit expansion under the ceiling. In such cases the stand-by arrangement may specify that the ceiling will be reduced by the amount of any transfer of profits from the exchange revaluation account to the government.

The other category of profits and losses arises from a variety of foreign exchange operations other than revaluations of foreign assets and liabilities. Sometimes, there have been substantial losses on swap arrangements entered into at contractual exchange rates below those prevailing at the time of repayment of the swap. Losses on liquidation of forward exchange contracts have also occurred. In a number of countries, profits or losses (generally the former), have resulted from the existence of multiple currency practices. In some countries, for instance, such practices have yielded substantial profits owing to the application of penalty exchange rates to major export items.

An important characteristic of these profits and losses is that they involve real monetary effects; they have, therefore, usually been covered by the ceiling. In some instances, while the exchange profit accounts as such have not been included, profits yielded by the exchange system have been credited directly to the government’s accounts with the central bank and in this way have been covered by the ceiling. In a few instances, where exchange operations have been a source of loss, the accounting system has not permitted an accurate identification of the items involved, but it has been possible to take account of their estimated expansionary effects in constructing the ceiling.

There have usually been excluded from the ceiling domestic loans by a central bank which represent the counterpart of certain foreign loans, such as development loans from the World Bank and the Inter-American Development Bank. In these instances the central bank becomes involved in a credit operation only as an intermediary. Accordingly, the exclusion of such credit from the ceiling has the same result as if the foreign credit had been extended directly to the government or the private sector. However, for the counterpart of the loan to qualify for exclusion from the ceiling, the foreign credit must exceed a specified maturity, generally five years. This minimum maturity limit serves to ensure that exclusion from the ceiling does not extend to the lending of the counterpart of short-term foreign borrowings undertaken by the central bank for balance of payments reasons. The absence of provisions regarding the treatment of loans representing the counterpart of foreign credits channeled through the central bank has generally reflected a judgment that no such credits were likely during the period of the stand-by arrangement, rather than a different view on how they should be treated. However, if the amount of foreign credits to be channeled through the central bank is known when a financial program is drawn up, the ceiling can be set correspondingly higher and the need to provide for the exclusion of counterpart loans will not arise.

Where the government or the private sector has been engaged directly in short-term borrowing abroad, the level of the ceiling has sometimes varied inversely with the amount of foreign borrowing. Thus, the letter of intent has, for example, provided that the ceiling on credit operations of the central bank is to be reduced by the amount of any Treasury bills placed abroad. Some programs have contained provisions setting ceilings on specified forms of short-term and medium-term foreign indebtedness.

The Fund has applied such limitations where a rapid accumulation of foreign debt has threatened to disrupt financial stability and the burden of servicing has become critical, and in the context of multilateral debt renegotiations. In other instances such limitations have been applied in order to reinforce the effectiveness of domestic credit and fiscal policies. The insertion of such provisions has, however, more often been motivated by the desire to avoid external debt problems than by the need to limit domestic credit expansion.

Provisions concerning foreign debt obligations were first included in stand-by arrangements in 1958, when Argentina and Chile decided to curtail authorizations of government imports under deferred payments arrangements. While in early stand-by arrangements no quantitative commitments were undertaken, in later arrangements specific commitments were made with regard to the amount of certain categories of foreign credits that could be contracted, or for which official guarantees could be given during the period of the stand-by arrangement. Access to the Fund’s resources was made contingent on the fulfillment of these commitments.

The degree of restrictiveness applied to borrowing abroad has varied considerably. The most restrictive debt limitation used in stabilization programs has been the prohibition of certain categories of foreign credits for a given period. Its effect has been to reduce the level of such foreign indebtedness by the amount of repayments falling due during the period of the stand-by arrangement. Less restrictive have been ceilings permitting new credits, but limiting them to less than net repayments (i.e., repayments less refinancing). In some other programs, ceilings on new credits equal to repayments have been used to ensure that the outstanding level of indebtedness would not increase. In some instances when this method has been used, however, the level of indebtedness has still tended to decline because disbursements have lagged behind the authorization of new credits. The least restrictive limitation on foreign credits has been one which has permitted the total outstanding debt subject to control to increase by a specified amount during the period of the program.

The prohibition of certain types of credit has been used in two situations. First, where an excessive debt servicing burden has emerged, or has threatened to emerge, as a result of a rapid accumulation of debts, a prohibition has been applied to provide a breathing spell in which the magnitude of the debt problem could be assessed and guidelines for the contracting of new debts as well as procedures for the orderly repayment of past debt obligations could be established. The second kind of prohibition has been related to the enforcement of domestic credit ceilings or the strengthening of fiscal measures. This has been necessary when the public or private sector, by borrowing abroad, has tended to circumvent a ceiling imposed on domestic credit expansion.

Sight drafts and letters of credit at the lower end of the time spectrum, and long-term credits at the upper end, have generally been excluded from limitations. The credits most frequently subjected to limitations have been short-term and medium-term suppliers’ credits and, to a lesser extent, short-term bank credits not related to commercial transactions. Experience has shown that because of the relatively easier circumstances under which suppliers’ credits can be obtained, fairly substantial indebtedness can be incurred in a short time. Moreover, the fact that such credits are frequently handled through private channels on both sides often conceals their magnitude from the authorities unless contingent government liabilities are involved.

Stabilization programs supported by stand-by arrangements have sometimes contained limitations on certain new foreign credits contracted by both the public and private sectors, but in most instances the limitations have been applied only to credits contracted by the government and its agencies. Credits contracted by the private sector have usually been controlled by denying them the benefit of official guarantees. For example, in Chile’s 1959 stabilization program the contracting of new suppliers’ credits by the public sector was prohibited, while the contracting of such credits by the private sector without official guarantee was permitted. However, in connection with Chile’s 1965 multilateral debt rearrangement, the contracting of new suppliers’ credits by both sectors was made subject to limitations. The 1965 and 1966 Brazilian stabilization programs prohibited the Bank of Brazil and the commercial banks from guaranteeing short-term and medium-term foreign loans unless these were used to finance foreign trade.

Unguaranteed private sector credits have been subject to reporting in some programs; this has helped the authorities to secure better information on the foreign debt situation of the private sector.

The types of suppliers’ credits to be controlled have usually been identified in terms of the number of years to final maturity. The maturity limit has often been placed at eight years on the ground that this period is long enough to account for the bulk of credit transactions which are to be brought under control. The fixing of a maturity limit has, however, created some problems.

The terms traditionally applied to exports of light capital goods and durable consumer goods tend to be of shorter duration than those relating to heavy capital goods. There has been some concern that the exclusion from ceilings of long maturities geared to imports of heavy capital goods may exert pressure on suppliers to extend the terms for other types of goods beyond their typical shorter maturities. It has also been felt that limitations applied to suppliers’ credits of less than a certain number of years may discriminate against suppliers who are not able to offer longer terms. The risk of curtailing competition or discriminating against certain suppliers exists in any arrangement prescribing upper time limits on credit.

Limitations on foreign credits applied for the purpose of making domestic credit controls more effective have been placed only on short-term borrowing abroad, defined as credits with maturities up to five years. Longer-term credits not subject to limitations are often extended by foreign governments or government agencies, or by international institutions after scrutiny and evaluation whereby the chances of an unduly large accumulation of such long-term debts are reduced.

Moreover, an important objective of stabilization programs is to create conditions conducive to an orderly growth of the economy, which requires the inflow of long-term capital. Consequently, far from placing limitations on the inflow of long-term capital, stabilization programs are designed to encourage the financing of development by long-term loans. Development loans usually involve relatively small debt servicing obligations in the early years because the terms of such loans usually include grace periods. Hence, there is less immediate danger of a sharp rise in debt servicing obligations resulting from long-term development loans than from short-term and medium-term suppliers’ credits.

The complete prohibition of new foreign indebtedness has been regarded as an exceptional debt control procedure to be reserved for extreme situations. Even though short-term commercial credits from abroad (up to one year) are usually exempted, such prohibitions impede the financing of the movements of goods by longer-term credits. Also, domestic enterprises often depend on foreign credits for supplementing their working capital. Some degree of flexibility has been provided in some arrangements by permitting exceptions from the prohibitions through consultation with the Managing Director of the Fund. However, in dealing with such requests, difficult problems are likely to arise. For example, it may be necessary to determine whether the project to be financed will make a direct contribution to the balance of payments and whether the amortization schedule proposed is consistent with the project’s earning capacity. It may also be necessary to determine whether financing at lower cost could be obtained from other sources.

Limitations aimed only at some debt reduction are less onerous. Limitations which permit the contracting of new credits up to the amount of repayments during the period of the stand-by arrangement are no more than a standstill device, designed to prevent the level of indebtedness in the restricted category from rising. For administrative and planning purposes, a limitation permitting a specified amount of new credits may be preferable to one which links the contracting of new credits to the irregular flow of repayments. Limitations permitting a certain amount of new credits to be contracted as old credits are repaid have the advantage that they enable the authorities to keep the debt situation under control without disrupting the country’s established commercial relations with its creditors.

Some member countries have undertaken, with the support of the Fund, to implement a stabilization program designed to eliminate inflationary conditions at a time when their external debt was large but on the whole sustainable. Limitations on foreign credits have then been applied to strengthen confidence in the creditworthiness of the country by keeping foreign borrowing under careful control. Such limitations have not aimed at reducing foreign indebtedness but rather at restraining its rate of growth, while the main stabilization effort has been directed toward restoring domestic financial stability.

In most stabilization programs containing limitations on foreign credits these limitations have been applied to the public sector as a whole, including public enterprises. As their borrowing capacity is enhanced by the implicit or explicit guarantee of the government, public enterprises have tended to shift to foreign borrowing in order to escape the pressures exerted by domestic financial restraints. A limitation on foreign borrowing by public enterprises is therefore necessary in most instances.

Sometimes domestic credit ceilings have been adjusted for subscriptions to international organizations. This problem has arisen mostly in connection with a prospective increase in a country’s quota in the Fund. If the government makes the subscription, the ceiling excludes credit to the government to finance it. If the central bank makes the subscription, there is no need for an adjustment in the ceiling as the subscription to the Fund affects only the foreign assets and liabilities of the central bank and has no effect on the ceiling on domestic assets.

In programs for some Central and South American countries, it has also been necessary to take into account in constructing the ceilings subscriptions to such international organizations as the Central American Bank for Economic Integration (CABEI) and the Inter-American Development Bank (IDB). As far as the gold or foreign exchange component of any subscription has been concerned, the view has been taken that the making of such a subscription should lessen the permissible domestic credit expansion. Accordingly, the usual practice has been to include in the ceiling the credit to the government to finance the subscription where the government rather than the central bank subscribes.

On occasions, the net position on local currency subscriptions to international organizations has been of some importance. The making of a subscription in local currency by the central bank raises both its assets and its liabilities without affecting its net position. However, the net position is affected where the international organization subsequently uses the local currency subscription to finance projects or to lend to other borrowers in the subscribing country. The use by CABEI of the local currency component of subscriptions has been an important source of monetary expansion. Accordingly, in some arrangements with members of that institution (e.g., Costa Rica, Guatemala, and Honduras) the net position on local currency subscriptions has been included in the ceiling. Sometimes the net position on local currency subscriptions to the IDB has also been included in the ceiling, but only when domestic financing from this source has been significant.

Drawings on the Fund or repurchases may have to be taken into account in the formulation of credit ceilings. In the majority of countries, such transactions are carried out by the central bank. Their effect is simply to raise the foreign assets and liabilities of the central bank, and they have no implications for credit ceilings covering domestic assets of the central bank. It makes also no difference whether the central bank records its liability to the Fund in its balance sheet as a noninterest-bearing note or as a deposit. Neither does the meeting of repurchase obligations have any implications for the credit ceiling when effected by the central bank.

In some countries, however, it is the government rather than the central bank which draws on the Fund, and the foreign exchange drawn is usually sold to the central bank for national currency. Some governments leave the proceeds on deposit with the central bank. This may reflect either a voluntary decision on the part of the authorities concerned or the requirements of the national legislation governing transactions with the Fund. In order to cover this aspect, a provision may be included in the letter of intent that the counterpart of Fund drawings will be sterilized. The alternative to leaving the currency on deposit with the central bank is for the government to substitute noninterest-bearing notes, thereby making the national currency counterpart available to the government. Where this has been the practice, one of two measures has been adopted in the relevant programs to take account of this additional source of funds to the government. In some programs where it was considered likely that the amount of drawings available under the stand-by arrangements would be fully utilized, the ceilings have been set at levels lower by this amount than they would have been if the counterpart funds had not been available to the government. Another approach has been to permit the local currency counterpart to be used to meet the needs of the government but to word the letter of intent so that, in effect, the limit on other forms of credit expansion is reduced by the amount of any use of counterpart funds. This has served to ensure that drawings on the Fund do not affect the margin for credit expansion under the ceiling but at the same time has left the government free to use the counterpart funds rather than borrow from the central bank to meet its financial needs. The need to maintain this freedom of choice may exist, for example, because of a statutory limitation on government borrowing from the central bank.


As mentioned before, major inflationary pressures frequently originate in public sector deficits. In such situations it has not been sufficient to place global ceilings on the expansion of bank credit, or separate ceilings on central bank credit to the government or the public sector as a whole. These indirect methods of controlling public finances do not go to the root of the problem; moreover, being subject to strong pressure from the government, the central bank authorities cannot always enforce these credit ceilings. Therefore, more direct fiscal measures have been stipulated in many programs (in addition to limiting access to bank credit).

In order to keep the deficit of the public sector within limits compatible with the monetary objectives, these fiscal measures have been designed to increase budgetary receipts, to reduce the growth of current or capital expenditures, and to strengthen the financial position of public enterprises.

For the most part, measures designed to increase revenue have involved additional taxation, mainly in areas where taxes can be introduced quickly and where they can be easily collected. The most common measure has been the imposition of taxes on foreign trade transactions, especially in connection with exchange rate adjustments. In situations of prolonged inflation, devaluation of the currency (or, where a multiple exchange rate system had existed, unification of the rates at a realistic level) has been a major policy instrument, frequently leading to the removal or the liberalization of restrictions on trade and payments.

Such exchange rate adjustments have often been accompanied by the introduction of new, or the increase of existing, export taxes or import subsidies. For certain primary export commodities the supply is inelastic in the short run, and devaluation may therefore generate excessive profits. An export tax designed to absorb part of these profits has frequently been imposed as a temporary measure. In some instances, the liberalization of quantitative restrictions following the unification of multiple rates has been facilitated by an increase in import duties or, pending an adjustment in the tariff structure, by the imposition of import surcharges. On the other hand, import subsidies for certain items of general consumption have sometimes been granted for a transitional period to minimize the impact of the devaluation on the cost of living; but in view of the distorting effects of import subsidies on the production and consumption of the commodities concerned, and of the burden on the budget, the use of such subsidies has been far less common.

Another frequently used tax measure has been the imposition of indirect taxes on nonessential commodities. Action in the field of direct taxation has been less frequent because it has been difficult to obtain quick and effective results. However, various programs have aimed at improvements in the income tax structure or stricter enforcement of existing income taxes, and the collection of tax arrears.

Measures to restrain the growth of government expenditures have been an important feature of many stabilization programs. For example, limits have been placed on government expenditures by establishing a ceiling on the fiscal deficit, or on current or total government spending. Sometimes also a certain revenue target has been stipulated. These measures have aimed not only at improving the current budgetary situation but also at providing additional resources for the financing of public investment expenditures. This has been done in some programs by establishing minimum targets for the achievement of savings in the public sector, i.e., minimum targets for surpluses of current revenues over current expenditures. In some instances, the program has provided that, in order to comply with the over-all credit policy, the government would refrain from initiating new investment projects unless foreign aid was actually forthcoming, or some other form of noninflationary financing was available. Sometimes new taxes have been imposed to provide the financial resources required to cover the domestic costs of investment projects without undue resort to bank financing.

In several instances an improvement in the over-all budgetary situation has been achieved by bettering the financial position of public enterprises. Some programs, for example, have introduced substantial upward adjustments in the charges for the services of various public utilities, such as electricity, gas, and water, state-owned railroads, or urban transportation, thereby reducing their dependence on budgetary support. Again, some programs have specified measures directed toward raising the efficiency of these enterprises through better management control, more efficient work rules, or the discontinuation of certain uneconomic operations.


Sometimes, where a substantial adjustment in the exchange system has been necessary, a gradual approach has been adopted to ease the problems of transition. For example, the devaluation of exchange rates to realistic levels has been carried out in more than one step; in the intervening period subsidies have been given to provide added incentives to some exports, and the demand for imports has been limited by maintaining quantitative restrictions, which have on occasion been reinforced by import surcharges. Similarly, some multiple rate systems have been gradually unified through a succession of simplifying measures.

In a number of instances, the political and administrative difficulties encountered in adjusting a highly complex exchange system have been so great that only a partial or selective adjustment could be made at the initial stage, although the ultimate objective has always been the achievement of a single realistic exchange rate. In these circumstances, the existing rate of exchange has been maintained for certain transactions such as traditional exports, essential imports, and some invisibles, while the rate for other transactions has been devalued either directly or indirectly through exchange taxes and subsidies. Alternatively, a complex multiple rate system has been simplified so as to consist of only a few rates; this has usually involved a devaluation on both the buying and selling sides through the elimination of the more appreciated rates. In general, the provision of a subsidy through the exchange system, i.e., through the maintenance of a more depreciated rate for exports than for imports, has been avoided.

Adjustment of the exchange system to a realistic single rate has not always involved the immediate establishment of a fixed rate. Many programs have provided for a fluctuating exchange rate as a temporary device, because of the difficulty of determining in advance the rate that would be appropriate after the stabilization measures have taken effect. When a combination of restrictions and multiple rates has existed, the effectiveness of the immediate measures to bring inflation under control has frequently been in doubt. A fluctuating rate used in these situations has carried with it a commitment that the rate would be allowed to move in accordance with the market forces, and that the authorities would intervene in the exchange market only to maintain orderly market conditions.

In some programs, the purpose of such a commitment was to ensure that gross foreign exchange reserves would not fall below the level maintained at the beginning of the program. The assumption was that observance of a minimum level for gross reserves would cause the exchange rate to be determined in the market, free from government intervention. This commitment was, however, not sufficiently precise and sometimes proved to be ineffective, because the gross reserve position could be improved by external borrowing. It was also very difficult to distinguish between temporary fluctuations and a long-term trend, especially in countries where export earnings are subject to wide fluctuations.

In recent years, as an alternative to fluctuating rates, exchange rate flexibility through periodic rate adjustment has been introduced in some programs which could not aim at an early restoration of stability. In several countries, inflation has been so acute and prolonged that it has not been feasible to restore financial stability in the immediate future. In such circumstances, the stabilization programs have been aimed at the gradual deceleration of inflation. Continuing domestic inflation, however, is likely to wipe out rapidly the beneficial effects of an exchange rate adjustment, and the exchange rate must therefore be adjusted periodically in order to counteract the adverse effects of domestic price increases on the balance of payments.

Experience has shown that some mutually agreed objective criteria are required to judge the compliance with the commitment on the flexibility of the exchange rate during the period of the stabilization program. Such programs have, in most instances, included certain “balance of payments tests” to ensure that official support does not cause the exchange rate to diverge significantly from the basic market trend. By these tests the minimum level at which the net foreign exchange reserves of the central bank or the banking system are to be maintained during the period of the program has been specified, thus setting limits within which the authorities would intervene in the market to smooth out fluctuations. The minimum limits to the net foreign exchange position of the monetary authorities, as defined in the arrangement, have usually been phased over the period covered by the program. The commitment not to let the net foreign exchange reserves fall below the agreed level (or levels) during the period of the program is binding, and compliance with this commitment is checked periodically.

In some programs the periodic exchange rate adjustments have been linked to movements in the domestic price index rather than to balance of payments performance, i.e., the programs have provided that a specified increase in the price index would be followed by an appropriate adjustment in the exchange rate.

The commitment to follow a flexible exchange rate policy has the advantage of being applicable to countries maintaining a fixed exchange rate. The setting of a minimum level for net reserves aims at preventing the authorities from maintaining an exchange rate which does not yield, in the context of the stabilization program, the desired balance of payments performance. The balance of payments adjustment through exchange rate action becomes necessary when other means of adjustment are judged to be insufficient. Thus, the level of net foreign exchange reserves is used both as a guide for the determination of the exchange rate adjustment and as a test of the effectiveness of this action.

The inclusion of a balance of payments performance test as an additional binding commitment in a stand-by arrangement increases the stringency of the conditions applied to the use of the Fund’s resources, and has important policy implications. The main problems involved in the formulation of the test are the definition of net foreign exchange reserves, the choice of the minimum level for reserves, and the linking of the test to the required exchange rate adjustment.

The definition of net foreign exchange reserves has to draw a line between those operations which do not influence the performance under the test, and those which should yield the desired performance. If a foreign credit facility is included in the definition of net reserves as a liability of the monetary authorities, the performance under the test is not affected by the receipts from this source, as the minimum reserves level covers both assets and liabilities of the monetary authorities. If the proceeds from such a foreign loan are redisbursed to meet current obligations, there is a net loss in the reserves as defined. If the proceeds are redisbursed in order to repay an obligation listed as a liability of the monetary authorities, there is no change in the net reserve position. Similarly, any movement in gross reserves may or may not affect the performance under the test depending on whether there is a simultaneous change in the liabilities of the monetary authorities. This does not mean, however, that foreign assistance cannot be used without impairing the results of the test, except to repay a liability of the monetary authorities. To the extent that the amount of foreign assistance was known in advance, it may have been taken into account by lowering the performance target by an equivalent amount when it was fixed. For example, if a balance of payments surplus would have been required in order to meet short-term obligations coming due, a foreign credit (or a drawing on the Fund) may allow the authorities to aim at equilibrium rather than a surplus in the balance of payments. However, once a performance target has been chosen, and unless this target is a deficit, resources from foreign borrowing can be used only to be added to gross reserves or to consolidate the debt of the monetary authorities.

Generally, the performance target has covered the amount of foreign assets and short-term liabilities of the central bank. In one arrangement the liabilities of the central bank were defined as “its liabilities to nonresident banks and other financial institutions, excluding loans with a maturity of over five years granted for development purposes.” Sometimes, however, the over-all amount of foreign assets and liabilities of the Treasury and other official agencies has been included when they have formed an important part of foreign indebtedness. In one instance, the performance target covered the net position of the whole banking system, including the commercial banks, because the commercial banks had tended to borrow heavily abroad in order to ease the tightness of domestic credit restrictions. Suppliers’ credits have on occasion been taken into account, either by a separate commitment to limit the amount of such credits, or by a clause providing for a reduction or increase in the target in accordance with any decrease or increase in suppliers’ credits.

The choice of a level for net reserves involves a projection of the payments and receipts of the monetary authorities as defined; the receipts are estimated as accurately as possible and a certain amount is allowed for payments in order to yield a performance considered as desirable and feasible. If heavy short-term obligations fall due during the program period, the level of gross reserves and of foreign assistance available may indicate that a certain minimum surplus will be necessary in order to meet the payments coming due. Some basic elements in the balance of payments projection, such as the prospective foreign exchange receipts and payments of the government, may be known with some degree of certainty, while some other elements, such as exports, imports, and private capital movements, are difficult to estimate. It is on these elements that certain assumptions have to be made and targets have to be established through negotiation and agreement with the authorities. In general the stabilization program aims at achieving a balance of payments surplus or at least an equilibrium over the whole period, and the rest of the financial program has to be consistent with that target.

After an over-all net reserves target has been chosen, it may be necessary to phase it according to the seasonal pattern of receipts and payments, taking into account any predictable elements that may temporarily affect the level of reserves. The establishment of phased net reserves targets, usually on a quarterly basis, is a difficult task, as foreign exchange reserves may be subject to wide short-term fluctuations, especially in countries where receipts from exports are largely dependent on agricultural crops. The seasonality of such receipts may not be clearly apparent or there may be some leads or lags from one year to the other. In some instances, a fixed margin of deviation from the target for seasonal or accidental factors has been allowed or a more elaborate phasing has been used.

Unexpected circumstances such as strikes, droughts, or fluctuations in world market prices for major export commodities may also temporarily affect the level of receipts. Consequently, some arrangements have included a safeguarding clause for situations where receipts from exports rose above or fell below the estimated level as a result of a change in the export volume or in export prices. The target could then be adjusted downward by the amount of the shortfall, or adjusted upward by an amount equivalent to the excess of receipts over the projected level. These safeguarding clauses, however, change the nature of the test by eliminating certain types of transactions from the balance of payments performance which is tested. If, for instance, compensation is given for some major receipts, the test primarily affects payments, and failure to pass the test may occur only if actual payments differ significantly from estimated payments irrespective of the level of receipts. Therefore, such safeguarding clauses, like the definition of foreign exchange reserves, are closely related to the problem of deciding for which type of transaction a certain performance is desired.

Agreement to submit to a balance of payments performance test implies that appropriate corrective action has to be taken before rather than after the points in time when the performance is checked, since failure to pass the test makes the country ineligible for drawings under the stand-by arrangement. Although the type of appropriate action is not always specified, it is generally understood that the achievement of the target should essentially result from an exchange rate adjustment. Thus, the staff report concerning the stand-by arrangement with Uruguay, of May 25, 1966, contained the following statement:

  • Attainment of the balance of payments performance targets will clearly depend on the implementation of the other policies embodied in the program. … Any necessary corrections will have to come through movements of the exchange rate, a tightening of monetary policies, or a combination of the two.

Other types of action, such as an increase in quantitative import restrictions or in exchange restrictions, may be specifically barred, but a tightening of monetary policy or of fiscal policy is never specifically ruled out. The timing and magnitude of the corrective action are left to the authorities to the extent that their action is successful in keeping the net foreign exchange reserves above the stipulated minimum limit. In the case of failure to pass the test, their decisions become a subject of the consultations with the Fund in order to agree on “the terms on which further drawings may be made.”

If the balance of payments performed better than the net foreign exchange reserve target required, the authorities could, in theory, either accumulate additional reserves or let the exchange rate appreciate. However, sometimes other conditions in the stand-by arrangement prevent any appreciation of the exchange rate, or the further improvement in the net reserves position may be considered preferable in view of the general indebtedness of the country. Thus the flexible exchange rate policies involving a balance of payments performance test have, in practice, resulted in the periodic devaluation of fixed exchange rates.


In most programs appropriate monetary and fiscal measures have been sufficient to restore internal financial stability. In some instances, however, direct action in the field of prices and wages has been considered necessary. Thus, some programs have provided for the immediate or gradual removal of price controls where these were hampering production and distorting the allocation of resources. Some other programs have included specific price adjustments, such as the increases in public utility rates mentioned above, within the context of commitments made in the fiscal field.

Wage policies have frequently been of particular concern because the effectiveness of monetary and fiscal measures designed to combat inflation is greatly affected by wage movements. However, the ability of the authorities to influence wage movements differs widely, depending on the institutional framework within which labor unions operate. Consequently, stand-by arrangements have frequently contained general declarations of wage policies, but rarely are these policies made binding commitments. Where the stabilization measures have resulted in a significant upward adjustment of prices, the programs have sometimes provided for an initial increase in wages in the public sector to be followed by a period without change. In the 1959 stand-by arrangement with Bolivia, the adjustment of prices of commissary goods to market levels and certain specific wage increases for workers employed by the Mining Corporation were made binding commitments (and subsequently modified). In the 1965 Brazilian arrangement, the fiscal program included an undertaking not to increase wages and salaries of government employees beyond the budgeted amount, unless further financial resources should be made available through additional taxation. With regard to the private sector, in countries where the government exercises control over wage policy through its role as arbitrator in wage negotiations, some undertakings on wage policy have been included. Where the government has little direct role in connection with wages, some programs have attempted to implement wage policy by moral suasion.

    Other Resources Citing This Publication