Financial sector reforms are policy measures designed to deregulate the financial system and transform its structure with the view to achieving a liberalized market-oriented system within an appropriate regulatory framework. The pace of financial sector reform and innovation began to accelerate in the late 1970s in many industrial countries and in the early 1980s in a number of developing countries of the Pacific Basin and Latin America. Currently, major financial reforms are under way in many African countries and in Eastern Europe. The initial situation in many developing countries and in the formerly centrally planned economies of Europe was characterized by direct controls on interest rates and credit allocation, the absence of well–developed money and securities markets, and underdeveloped and highly regulated banking systems. With reform of the financial sector, this situation is giving way to a greater flexibility in interest rates, an enhanced role for market forces in credit allocation, a gradual deepening of money and securities markets, and increased autonomy for commercial banks. Alongside these developments, the framework of monetary policy is also undergoing major changes. Bank-specific credit ceilings and selective credit allocations are being replaced by market-based instruments for implementing monetary policy, and prudential supervision systems are being put into place to foster sound credit decisions.

Mohsin S. Khan and V. Sundararajan1

Financial sector reforms are policy measures designed to deregulate the financial system and transform its structure with the view to achieving a liberalized market-oriented system within an appropriate regulatory framework. The pace of financial sector reform and innovation began to accelerate in the late 1970s in many industrial countries and in the early 1980s in a number of developing countries of the Pacific Basin and Latin America. Currently, major financial reforms are under way in many African countries and in Eastern Europe. The initial situation in many developing countries and in the formerly centrally planned economies of Europe was characterized by direct controls on interest rates and credit allocation, the absence of well–developed money and securities markets, and underdeveloped and highly regulated banking systems. With reform of the financial sector, this situation is giving way to a greater flexibility in interest rates, an enhanced role for market forces in credit allocation, a gradual deepening of money and securities markets, and increased autonomy for commercial banks. Alongside these developments, the framework of monetary policy is also undergoing major changes. Bank-specific credit ceilings and selective credit allocations are being replaced by market-based instruments for implementing monetary policy, and prudential supervision systems are being put into place to foster sound credit decisions.

In support of such a transition, financial sector reform has involved measures to achieve the following objectives: greater independence for central banks in macroeconomic stabilization and adjustment efforts; enhanced competition in the banking system; stronger balance sheets and a higher quality of bank portfolios; an effective banking supervision system; and an efficient clearing and settlement system for payments. The relative importance of these objectives and the specific details of the supporting measures have varied from country to country, depending on the prevailing conditions, the commitment to reform on the part of the government authorities, and the speed of the reform process.

In all cases, however, such reforms have been motivated by the need to pursue stabilization and broader structural reform objectives in an efficient and effective manner. It is apparent that the linkages among financial sector reform, the monetary policy framework, and stabilization and other structural reform policies are close and complex. Successful pursuit of stabilization policies is necessary for the effectiveness of broader structural reforms—price reform, fiscal reform, exchange and trade system reform, and industrial restructuring policies—and for financial sector reform. At the same time, it is clear that stabilization policies may not be sustainable unless they are supported by rapid structural changes in key areas. In particular, wide–ranging structural changes in the financial sector are often needed for the effective and efficient conduct of monetary policy; without them, progress toward macroeconomic and financial stability would be difficult to achieve. Indeed, a growing literature suggests that the structure of the financial sector can have a significant influence on macroeconomic performance.2

Furthermore, the appropriate sequencing of different types of structural reform has been found important for overall stability.3 These complex interactions pose special challenges in the design of reform measures in the financial sector and in the sequencing of these measures in coordination with stabilization and other structural policies.

This paper provides an overview of the linkages between financial sector reform and the monetary policy framework—in particular the objectives, instruments, and operating procedures of monetary policy—based on the recent experiences of developing countries. The paper’s primary focus is on how financial sector reform affects the design and conduct of monetary policy and not on the financial sector reforms themselves.4 The paper presents a conceptual framework that highlights the role of financial sector reform in facilitating the transformation of a monetary policy framework. It also deals with the implications of financial sector reform for the objectives and design of monetary policy by considering the effects of reforms on money demand, the money supply process, and the transmission mechanism between monetary policy and the principal macroeconomic aggregates. Finally, the paper addresses some issues surrounding the implementation of monetary policy in the context of financial sector reform, with a particular focus on the problems of transition.

Financial Sector Reform from a Monetary Policy Perspective

The transition from direct controls on interest rates and credit aggregates toward indirect management of these variables, increasingly through market-based instruments, can be seen as an interactive and evolutionary process, whereby certain financial sector reforms facilitate the transformation of the monetary policy framework and vice versa. As illustrated in Chart 1, this process can be divided conceptually—not necessarily chronologically—into several stages distinguished by their degrees of interest rate flexibility. The transition from one stage to the next typically involves parallel reform in many central banking functions and in the broader financial sector.

Chart 1.
Chart 1.

Model of Reforms of the Financial Sector and Monetary Control

1Includes actions to deal with fixed-rate loans, nonperforming loans, capital adequacy, and subsidized selective credits.

Country experiences clearly illustrate the need for preliminary or parallel reforms of the financial sector when moving toward greater flexibility in interest rates and their eventual liberalization. In general, successful interest rate liberalization—that is, the successful management of interest rates through indirect instruments—is associated with several conditions: a relatively stable macroeconomic environment; an interest rate structure that is not in serious disequilibrium before liberalization; adequate competition in the banking sector; reasonable financial strength in the banking sector; an active and well-functioning money market; monetary policy instruments that are able to influence the marginal cost of funds to banks; and sufficiently strong bank supervision policies and instruments.5

Although theoretically an optimal sequence of reform is plausible, in a majority of developing countries reform of monetary management and the development of money markets must be pursued simultaneously. This is because, in practice, there are strong policy and operational linkages between the development of interbank money markets and markets in short-term instruments, on the one hand, and the reform of monetary management and the development of market-based instruments of monetary policy, on the other. Owing to these linkages, money market development has been emphasized early in the reform sequences of many countries (for example, Malaysia, India, and Indonesia).

The absence of well-developed money and securities markets implies that market-based instruments rely on operations in primary securities markets—typically government securities—and in the unsecured interbank market. Recognition that such operations must be carried out at market-related yields is a key step in stimulating interbank and money markets. Such markets, in turn, facilitate the efficient redistribution of surpluses and deficits of short-term funds, which is necessary for the smooth operation of indirect instruments. These technical linkages have to be exploited early in the reform program to provide adequate structural and operational support for implementing efficient and effective monetary policy. An early start on an interactive reform process is also required, because such reform can take some time to implement owing to the need to simultaneously establish various complementary changes in forecasting and other information systems as well as in the underlying accounting and interbank settlement systems themselves. Moreover, the reliance on primary markets in government securities as a means of implementing monetary policy generally calls for supporting changes in domestic public debt management policies and procedures. Thus, many parallel reforms—in central banking functions, money market structures, and public debt management—are involved in the transition from direct to market-based monetary management (from Stage 1 to Stage 2 in Chart 1).

As money and securities markets continue to develop, the range of monetary policy instruments and operations could be widened further. This widening enables the central bank to act primarily at its own initiative, with more sophisticated instruments and information systems, and to manage financial market liquidity at its own discretion (Stage 3 in Chart 1). The range of indirect instruments would now extend to include—in addition to reserve requirements and conventional refinance facilities at preannounced interest rates—arrangements for greater discretion in the use of refinance, special refinance facilities at the initiative of the central bank (such as refinance auctions and refinance quotas), operations in the primary-issue market (such as central bank auctions of its own paper or treasury bill auctions), operations in the unsecured interbank market (such as operations in government deposits, intervention in the interbank market through a specific lead bank, and special deposits with central banks), repurchase operations in specified securities (and in the foreign exchange market), and direct interventions in secondary markets.

In general, the role of primary-market operations in managing bank reserves would decline in relative importance, while that of other market-based instruments would rise. The level of reserve and liquid asset ratios would be reduced owing to the availability of market–based instruments. The importance of “defensive” monetary policy operations—operations to smooth out short-term fluctuations in bank liquidity and short-term interest rates—would also rise in order to prevent excessive interest rate volatility and to ensure that changes in trend are not obscured by day-to-day volatility.6 Such defensive monetary policy operations help to speed up the transmission of the effects of monetary policy and enable the smooth functioning of financial markets.

In moving toward full interest rate flexibility, parallel reform of prudential regulations and supervisory systems in addition to financial restructuring policies to deal with problem loans and enterprises have been found particularly important. Sound prudential policies and their proper enforcement are critical for minimizing major disruptions to growth and stability. Inadequacies of prudential policies and enforcement procedures together with the failure to correct macroeconomic imbalances in a timely fashion in the course of financial reform have contributed to excessive increases in interest rates (for example, in Chile and Uruguay), partly as a result of excessive risk-taking by banks. To prevent such outcomes, in addition to setting up adequate banking supervision, appropriate measures to recapitalize banks and deal with problem loans may have to be taken in order to ensure the effectiveness of adjustment policies and the success of interest rate liberalization. In many countries, decisions on interest rates have been constrained by a large share of low-fixed-rate loans and nonperforming loans in bank portfolios. Some initial financial restructuring policies to deal with these problems would greatly facilitate more active interest rate policies and more efficient credit allocation.7

Implications of Financial Sector Reform

The various financial sector reforms discussed in the previous section—in particular, removal of credit ceilings, liberalization of interest rates, strengthening of prudential regulations, and development of financial and securities markets—can have substantial and important effects on the demand for money, the money supply process, and the transmission mechanism between monetary policy and key macroeconomic aggregates, such as prices, output, and the balance of payments. This section examines the potential effects of financial sector reform on the demand and supply of money and then describes how the transmission mechanism is likely to change in the wake of financial liberalization.

Money Demand and Money Supply

The quantitative formulation of monetary policy typically assumes the existence of a relatively stable money demand function.8 Financial reforms that affect the determinants of the demand for money—income, prices, interest rates, and exchange rates—may lead to significant changes and possible instability in the demand for money. Such changes in the demand for money, particularly if they are unpredictable, obviously make it more problematic to ascertain the liquidity needs of the economy, thereby creating greater uncertainty in the supply of credit and money needed to achieve the policymaker’s ultimate objectives. In many countries undertaking financial sector reform, it has been observed that the ratios of money and credit to gross domestic product have risen, while the ratio of currency to deposits has declined.9

The effects of financial liberalization on money demand may be divided into one-time shifts of monetary stocks and changes in the elasticities of money demand with respect to its determinants.10 Onetime shifts in the stock of monetary aggregates may result from portfolio reallocations that arise from (i) interest rates that better reflect the risk-return characteristics of financial assets, (ii) the removal of certain quantitative restraints in the types of assets that can be held, and (iii) the development of new assets that would compete with money. Structural changes in money demand—changes in the effects of the variables determining money holdings—may be caused by some of the same factors, by the move toward indirect monetary control instruments when administrative credit ceilings are replaced by the voluntary portfolio behavior of agents responding to changes in income, wealth, and interest rates, and by policies to improve and deepen financial markets. The consequent competitive pressure and lower transaction costs in financial markets, and a reassessment of the risks of holding financial assets, are likely to lead to portfolio shifts and a heightened sensitivity of money demand to changes in income, wealth, and interest rates.

Consider, for example, the case of interest rate effects on the demand for money. In many developing countries interest rates are seldom considered an important determinant of money demand for two main reasons. First, the lack of alternative financial instruments tends to restrict agents to holding only monetary claims—principally currency and bank deposits. Second, interest rates are typically under official control and are often set below their equilibrium level. Since there are few interest-bearing financial assets and interest rates are infrequently changed, it is not surprising that the effect of interest rates can be ignored or discounted. Financial reforms that expand the menu of financial assets that bear market interest rates are likely to lead to an increased sensitivity of money demand to interest rates. Furthermore, as reform of the financial sector is a dynamic process, changes in the demand for money may occur continuously over time. Thus, shifts in the demand for money are not only likely in the short run but are also a possibility in the long run.

Turning now to the money supply process, in the pre-liberalization period characterized by direct controls on credit, interest rates, exchange rates, and capital flows, the traditionally measured money supply can often be specified simply as a function of reserve money.11 This assumes no change in the behavior of the public or the banks; the money multiplier is thus constant. In the post-liberalization period, both the public and the banks are free to choose their asset-liability portfolios according to market-determined rates of return and relative risks. As such, the components of the money multiplier can no longer be considered constant, rather they are endogenous variables that respond to changes in income, wealth, and interest rates. Among these components—namely, the currency-deposit ratio, the excess reserve ratio, and the required reserve ratio—only the required reserve ratio is directly influenced by policy. The other two ratios are behavioral functions of income, wealth, and the various interest rates (reflecting opportunity costs) facing both the public and the banks. When interest rates are liberalized and portfolio restrictions lifted, these component ratios change and by changing the money multiplier have an impact on the money supply. For example, the elimination of credit and interest rate ceilings facilitates a decline in excess reserves, reinforced by higher lending rates, while an increase in deposit interest rates generally leads to higher demand for bank deposits. The reduction in reserve requirements raises the value of the money multiplier and, for a given level of reserve money, the money supply as well. The increase in the money multiplier is also supported by the fall in the currency-deposit ratio as the opportunity cost of currency holdings rises12 and by a lower excess reserve ratio prompted by higher bank lending rates and a deepening of money markets. Thus, the interdependence of reserve money operations and the money multiplier, absent during the pre-reform period, is a major result of financial liberalization.

The predictability of the money supply hinges not only on the actions of the monetary authorities, through changes in reserve requirements and reserve money operations, but also on the predictability of the demands for currency, bank deposits, and excess reserves. Any variations in these demands mean that forecasting the money supply following financial liberalization will be difficult. Thus, it is readily apparent from a theoretical standpoint that financial liberalization—by altering the behavior of the public and banks—will change both the demand for money and the money supply process. By how much is an empirical question.

Several empirical studies have examined the effects of financial liberalization on the demand for money. For example, a recent study of nine Asian countries in different stages of financial sector reform reveals instability in the short-run demand for money in five of the nine cases and stability in the long-run demand for money in all cases.13 In addition to one-time shifts and changing income and interest rate elasticities, instability is reflected in the statistical significance of interest rate variables that were unimportant in the pre-reform period. Evidence on the effects of financial innovations on the demand for money—which are possible once financial sector reforms are undertaken—also bears on this issue. For example, Arrau and others (1991) show in a sample of 10 developing countries that the role of financial innovation (modeled in alternative ways) was quantitatively important in determining money demand.14 Their paper goes on to argue that, while it may be difficult to measure and forecast financial innovations, it is necessary to take them into account. Failure to do so may lead to inaccuracies in the projection of the demand for money and in problems identifying the transmission mechanism for monetary policy.

In contrast to the availability of empirical work on financial liberalization and the demand for money, empirical studies on the effects of liberalization on the money supply process are lacking. While there is evidence that the money multiplier and its component ratios do tend to change over time,15 demand equations for currency, bank deposits, and excess and borrowed reserves (or free reserves) have yet to be estimated for periods that cover pre- and post-financial liberalization. In theory, in the absence of offsetting changes in reserve money, increases in real interest rates and reductions in reserve requirements following financial reform would tend to raise the money supply through an increase in the money multiplier, which, in turn, is brought about by shifts out of currency and into bank deposits, by declines in excess and free reserves, and by lower required reserve ratios. Empirical validation of the theory is needed, although this task will prove more difficult than in the case of the demand for money if only because a larger number of demand functions need to be estimated and tested for stability.

Monetary Policy Transmission Mechanism

In industrial countries with highly developed financial markets, a broad consensus on the nature of the transmission mechanism has existed for some time.16 In brief, an expansionary monetary policy—undertaken, for example, through an increase in bank reserves supplied by the central bank through an open market purchase—leaves the private sector with too much money in its portfolio relative to other assets. In reestablishing portfolio equilibrium, banks increase credit, and agents bid up the price of securities and durable assets, thereby lowering their respective rates of return. Since the market price of securities rises, and that of durable assets such as physical capital increases relative to their replacement costs, agents attempt to increase their stocks of such assets by increasing their demand for newly produced units of these assets. In this way, an open market purchase of securities results in an increase in aggregate demand. Essentially, interest rates represent the key link between monetary actions and macroeconomic variables. The central bank alters interest rates indirectly, which, in turn, affects the interest-sensitive components of aggregate demand—housing, consumer durables, and investment expenditures—and thus prices, output, and the balance of payments. With a floating exchange rate, the central bank’s actions also affect the exchange rate, with further effects on the components of aggregate demand through changing wealth and relative prices.

In developing countries, however, the picture differs somewhat.17 In the first place, the menu of assets available to private agents is very limited. Organized securities markets in which the central bank can conduct open market operations scarcely exist in many developing countries. By and large, individuals can hold currency and deposits issued by the banking system, and they can borrow from commercial banks, although informal markets often emerge resulting in financial disintermediation through “curb” markets for deposits and loans. Durable goods, such as land and physical capital, can be held directly, but organized equity markets are small or nonexistent. Capital controls and prohibitions on the holding of foreign exchange limit the extent to which foreign assets may be acquired by domestic residents, although parallel markets for foreign currency surface in response to such regulations, thereby allowing agents to circumvent official controls to some degree. Finally, even in the case of those assets and liabilities available to individuals, such as demand, time, and savings deposits and bank credit, official restrictions often determine the interest rates paid or charged by financial institutions, although a variety of methods for avoiding interest rate controls typically emerge.

Monetary policy in such a controlled environment operates primarily through a credit-rationing channel, with the role of interest rates depending upon the extent to which agents circumvent or avoid direct controls through less formal and less regulated markets. Changes in the supply and control of credit—typically brought about through administrative means—have direct effects on aggregate demand. In simple terms, those agents who have credit made available to them are able to expand demand. Insofar as the effects of interest rates on the economy are concerned, two basic schools of thought exist. Adherents of the McKinnon (1973) view maintain that raising controlled bank interest rates need not be contractionary, because in a rationed regime the induced increase in saving will result in an increased supply of credit, which facilitates the financing of private investment and working capital. By contrast, the “neo-structuralist” view, as represented by van Wijnbergen (1983), emphasizes the importance of informal loan markets when bank interest rates are subject to legal ceilings. In this case, it is argued that increases in bank interest rates will draw funds away from such markets, thereby increasing the marginal cost of funds and exerting contractionary effects on the economy from both the demand and supply sides.18 For the McKinnon school, the primary channel of transmission is the direct effect of the controlled bank deposit interest rate on private saving, whereas the neo-structuralist school focuses on effects that are transmitted from the loan interest rates (both formal and informal) to interest-sensitive components of demand and supply—much as in the case of developed economies.

The removal of credit ceilings, the liberalization of interest rates, and the financial sector reform change the transmission channel from predominantly direct credit rationing, as discussed above, to predominantly price rationing through market-determined interest rates.19 The transmission channel thus begins to approximate the developed country paradigm. In this regard, the key market is the one for bank reserves, which the monetary authorities attempt to influence through control over the stock of reserve balances of banks. For example, a decrease in bank reserves brought about by open market sales of securities would raise short-term interest rates, which, if perceived as a lasting effect, would create a ripple along the yield curve, thus influencing the medium- and long-term rates as well. In addition, in an open economy, capital flows and exchange rates provide additional transmission channels for monetary policy. In countries with relatively open foreign exchange markets but fixed exchange rates, capital flows would prevent the domestic interest rate from deviating substantially, and for any length of time, from international rates. For instance, a decrease in bank reserves through open market sales would raise the domestic interest rate. The resulting capital inflows would offset both the initial reduction in reserves and the initial increase in the domestic interest rates. In these circumstances, the monetary authorities do not fully control the supply of money, which is partly determined by balance of payments deficits or surpluses. Full control in this case requires sterilization of balance of payments imbalances. Absent adequate amounts of securities and foreign exchange, such sterilization would be difficult to sustain while maintaining the fixed exchange rate. On the other hand, where capital flows are free and exchange rates flexible, the capital flows induced by the increase in domestic interest rates tend to cause the domestic currency to appreciate, which gives a measure of independence to short-run domestic interest rates and reinforces the restrictive monetary policy by lowering demand for domestically produced tradable goods and reducing inflation.

In a liberalized financial environment, the term structure of interest rates is also an important aspect of the monetary transmission process, since financial liberalization is likely to affect the yield curve.20 Whether the relationship between short-term rates and long-term rates is affected or not depends on two interdependent factors: the direct liquidity effect and the effect of the inflationary expectations- monetary policy credibility nexus. Under the first effect, short–term rates are negatively related to discretionary monetary operations. As for the second effect, a credible restrictive monetary policy, which is perceived to be anti-inflationary, will likely reduce the inflation premium built into long-term rates. On the other hand, a contractionary monetary policy that lacks credibility will do relatively little to affect long rates. Since in a liberalized financial regime the private sector has greater freedom to manage its portfolio, including its maturity structure, any policy-induced increase in short-term interest rates will be dampened by substitution away from long-term assets in the expectation of higher yields on short-term holdings, given that policy is seen to be credible.21

In response to changes in reserve money, the short-term rates and the relationship between short and long rates may also be affected by the endogenous behavior of components of the money multiplier in response to initial changes in the short rates. Changes in the interest rates on treasury bills and securities associated with open market operations affect bank deposit and lending interest rates, and hence the currency and excess reserve ratios.22 Changes in bank deposit and lending rates and in other interest rates also influence private decisions to accumulate real and financial assets and to purchase goods and services via income and wealth effects.

The relative importance of the liquidity effect in financially liberalized industrial countries is evident in the results from empirical tests of the relationship between long-term rates and both short-term and foreign interest rates.23 For most of the major industrial countries, the sensitivity of long rates to short rates is reduced during financial liberalization, while the responsiveness to foreign rates increases. This suggests that financial reforms increase the relative importance of inflationary expectations and of foreign rates in domestic interest rate determination, while the impact of domestic liquidity on the slope of the yield curve becomes weaker.

Even in financially liberalized economies, variations in the availability of credit have been considered an important transmission channel for the effects of monetary policy by various economists. This “credit school” has gathered strength in recent years owing to recent analytical and empirical developments in explaining credit rationing and examining credit market disturbances. These studies have espoused that changes in credit availability can have large effects on the real economy, both on the level of aggregate output and on its sectoral distribution. Their observation has implications for the choice of financial aggregates to be used as targets of monetary policy and calls for more explicit attention to bank regulations, bank behavior, and credit market conditions in formulating and analyzing monetary policy. 24 Thus, the relative importance of credit rationing and price rationing as channels of monetary policy are affected not only by the degree of progress toward interest rate liberalization but also by the prevailing macroeconomic environment and financial structure that govern bank behavior toward credit allocation.25

The discussion above shows that the relationship between financial aggregates and ultimate monetary objectives as well as the ability to control aggregates—that is, the relationship between instruments and operating targets of monetary policy, on the one hand, and financial aggregates, on the other—is likely to undergo changes in the process of financial liberalization. Therefore, the choice of financial aggregates that could serve as intermediate targets to guide monetary policy, as well as of operating procedures to ensure monetary control, has changed in most countries. First, the coverage of instruments that constitute different monetary aggregates and the information content of various aggregates have been reexamined in many industrial and developing countries. Second, central banks have increasingly relied on a basket of indicators rather than on a single intermediate target—such as a particular financial aggregate or interest rate—and pragmatically reset the chosen targets according to the latest information and developments. Third, the issue of making the appropriate choice between interest rate targets and financial aggregate targets has received more attention. Finally, the instruments and operating procedures of monetary policy have significantly changed in all countries that undertook financial liberalizations; new approaches were needed to improve monetary control in the changing environment.

Operational and Institutional Aspects of Monetary Policy During Financial Reform

In discussing monetary policy in a liberalized system, the previous section highlighted the roles of bank reserves and short-term interest rates in market-based monetary management. The procedures and instruments by which the central bank tries to influence the level of bank reserves and short-term interest rates (and exchange rates) depend on several institutional factors. Among these, the level of development of a country’s money markets, the nature of its clearing and settlement system (including the reserve accounting framework), and the type of domestic public debt management system in place are the most important. These factors influence the design of the monetary policy instruments and the instrument mix. Hence, reforming these factors is the first order of importance for monetary policy implementation. As already noted, the monetary policy framework itself affects the structure and depth of the money market, and the nature of the clearing and settlement system is another dominant force influencing the structure of the money market.26

The need initially to use primary markets in government securities—in part reflecting the lack of depth of secondary markets—for both monetary and debt management raises a variety of operational and institutional questions relating to the coordination of treasury and central bank policies. If the supply of government securities to the market based on debt management considerations proves insufficient or excessive and cannot be effectively coordinated with monetary goals, should the central bank use its own securities to manage bank liquidity? What are the respective roles of the treasury and the central bank in the primary issue of government securities, in managing the secondary market in those securities, in the clearing and settlement of transactions in the securities, and in short-term cash forecasting to facilitate debt and monetary management? If liquid asset ratios are used for monetary and debt management, what is the appropriate transition strategy for moving toward voluntary debt issue and market- based instruments? Such questions have been addressed in many countries during the course of monetary reform (including Poland, Nepal, New Zealand, and Malaysia).

Several aspects of the payments system governing money market transactions play a crucial role in the implementation process. First, whether the clearing and settlement of such transactions take place on the same day or with a lag (or whether there is a mixed system) affects the demand for excess reserves. The level and volatility of excess reserves depend not only on the accounting rules governing clearing and settlement between the central bank and private banks but also on those governing the private bank and its customers.27 These rules, together with the level of communications technology, could cause large variations in the size and variability of the interbank float, with major implications for the design and effectiveness of indirect instruments. The lags and other features of the settlement system for payments, the terms and conditions of access to central bank credit (to settle payments), and the rate of remuneration on excess reserves are some of the key elements that determine the demand for excess reserves and short-term money market conditions.

The structure of the reserve requirement system—particularly reserve averaging and the length of period over which the averaging is conducted—also interacts with the clearing and settlement system and affects the volatility of short-term interest rates and the operating costs to banks. For example, reserve averaging reduces interest rate volatility caused by uncertainties and lags in the settlement system. Lags in the reserve requirement as well as the coverage and uniformity of such requirements across institutions and instruments are important factors affecting monetary control. Reform of reserve requirement systems has figured prominently in many financial reforms (two examples are Malaysia in 1989 and the United States in 1984); some countries have even instituted zero reserve requirements (New Zealand, Canada, the United Kingdom, and Mexico), supported by adaptations in their interbank settlement and refinance systems.

Refinance systems—the rules and procedures by which a central bank provides credit to financial institutions—have also changed dramatically over the course of financial reforms. Refinancing and rediscounts provided at the initiative of the central bank have increased, while the amounts provided through conventional credit facilities at preannounced rates have declined. In particular, the use of sale and repurchase transactions in securities and bills and of auctions of refinance has increased as a means of supplying reserves at the initiative of the central bank. Such procedural changes have facilitated greater flexibility and control over short-term interest rates and fostered greater activity in money markets.

The choice of an operating system for monetary policy influences the pace of structural reform. First, active liquidity management by the central bank will induce banks to be active, thereby contributing to the development of markets. Active liquidity management, however, requires strong policy research and information systems to anticipate developments in bank reserves. It is only the central bank that can operate to anticipate situations of protracted excess liquidity or conversely a shortage of reserves when most agents are on one side of the market—and thereby promote conditions conducive to market development. Second, the instrument mix of the central bank should be consistent with market development goals as well as the goals of monetary policy. For example, a central bank could readily inject liquidity into the system by not replacing the securities being redeemed. However, in some circumstances, the central bank may choose to maintain the outstanding volume of securities in the market and inject additional liquidity through other means, so as to better sustain market development. Finally, the specific design of individual instruments can contribute to the structural development of the financial sector. The choice of securities that are eligible as collateral for refinance or repurchase operations, the specification of eligible liquid assets, and the structure of the reserve requirement system are some of the factors affecting the demand and supply of securities, the interest rate structure, and the depth of various segments of the market.

In addition to such structural considerations, the appropriate mix of instruments to absorb and supply reserves depends on the choice of operating targets (borrowed reserves versus unborrowed reserves, net domestic assets, and so on) and the projected path of autonomous factors affecting bank reserves. This projected path, in relation to the desired path of an operating target, determines the mix between those instruments that absorb reserves and those that supply them and also determines the maturity of various monetary operations.

Moreover, the choice of operating targets (and intermediate targets) should depend on the nature of the shocks affecting the monetary system. Conventional wisdom, based on analysis contained in Poole (1970), states that in times of major financial sector reforms resulting in significant shifts in the behavior of various money and credit aggregates, it is preferable to target interest rates. In times of major shocks to the real economy, it is appropriate to target money or credit aggregates. Similar considerations affect the detailed design of operating procedures. The choice between targeting some short-term interest rates (overnight interbank rates or auction rates for treasury bills) or some quantity measure (excess reserves, free reserves, borrowed reserves, or various definitions of domestic credit) also depends on the relative size of disturbances affecting the financial system.28 In times of major macroeconomic adjustment and supporting financial sector reform, it is hard to assess the relative magnitude of various shocks to the real sector and the financial system. Therefore, as noted earlier, central banks often tend to rely on a basket of indicators and pragmatically readjust the chosen target on the basis of the latest available information. Nevertheless, at any one point in time, the choice between prices and quantities becomes secondary to setting the right target for the prices or quantities, given the prevailing financial and nonfinancial indicators, and to achieving the target through an efficient set of monetary policy instruments that support both stabilization and structural objectives.

Concluding Remarks

Financial sector reform and monetary policy reform in both developed and developing countries during the past two decades have proven, on balance, to be highly beneficial. Reduced credit rationing and greater interest rate flexibility have led to an improved allocation of financial resources, while increased competition among banks and other financial institutions has lowered the cost of intermediation. At the macroeconomic level, growing evidence suggests that countries with more liberalized financial systems have benefited from increased savings, better and more efficient investment performance, and faster rates of economic growth.

Significant adaptation of the targets and operating procedures of monetary policy have been needed over the course of financial liberalization and innovation. The adaptations have included the pragmatic resetting of chosen targets—whether financial aggregates, interest rates, or exchange rates—based on up-to-date financial and nonfinancial indicators, and improved, flexible operating procedures both to influence the marginal cost of funds to banks and to foster active money markets.

Nevertheless, in a number of cases such reforms have created serious problems for the financial system. The failure to correct macroeconomic imbalances and implement an appropriate supervisory framework have been the primary reasons that such reforms have failed. The interaction between macroeconomic instability and inadequate bank supervision has encouraged banks to undertake risky lending in the presence of deposit insurance and loan guarantees, which has often resulted in an immediate and steep increase in real interest rates, with adverse consequences for investment and growth.

To prevent such outcomes, economic stabilization and improved bank supervision should generally accompany a full liberalization of interest rates. Indeed, a better flow of information, adequate disclosure, strong supervision of the banking system, and macroeconomic stability are key features of successful experiments in financial liberalization. Based on country experiences, a reassessment of deposit insurance and loan guarantee schemes is also necessary to sharpen the risk analysis performed by banks. Excessive short-term interest rate volatility could be minimized by the appropriate design of specific monetary policy instruments and operating procedures. For example, reserve averaging, modifications to refinance policy, and improvements in clearing and settlement systems reduce the interest rate volatility caused by uncertainties and lags in the settlement system and thus improve control over short-term interest rates. Most important, the elimination of fiscal imbalances and timely actions to recapitalize banks and deal with problem loans and enterprises would help relieve pressures on interest rates. Together, these measures would ensure the success of both stabilization and financial liberalization and thus lead to a more efficient financial system.


I.A. Hanfi

Let me start by congratulating Mohsin Khan and V. Sundararajan for their comprehensive and high-quality treatment of this subject. I share the main message of their paper that a stable financial and monetary framework is an important prerequisite for noninflationary and sustained economic growth.

In the past, public policy in much of the developing world was characterized by confidence in the capacity of the state to promote development by channeling resources to priority sectors. Monetary policy was seen as contributing to growth through a system of direct credit allocations and administered interest rates rather than through aggregate demand using traditional tools of monetary policy such as open market operations, changes in reserve requirements, and variations in the discount rate. Various factors, mostly of a structural and institutional character, strengthened the preference for direct control mechanisms. The narrow market for government securities did not allow the use of open market operations. Government securities were held by a limited number of financial institutions, basically as part of liquidity ratio requirements. The financial markets were narrow, unintegrated, and noncompetitive. The overall level of credit could theoretically have been controlled by changes in reserve requirements, but their effectiveness was often counteracted by the existence of windows for concessional finance. Whatever the virtues of selective credit control for monetary management in a particular developing country, its use tended to create vested interests and entailed large efficiency and social costs even when administered properly.

Of late, there is a growing appreciation in many developing countries of the merits of market forces for determining the allocation of resources, including credit. This thinking has been reinforced by the collapse of the centrally planned economies and their movement toward market-oriented, outward-looking systems. This historic conversion has been instrumental in encouraging similar changes in many parts of the world as long-hidden costs of centrally directed investment, production, and resource allocation have become better known. As a consequence, many governments are now undertaking deregulation, denationalization, and privatization in various sectors of the economy, including the financial sector.

One should remember that there are several preconditions to the success of any program of financial liberalization. Macroeconomic stability is one of the most important. A complete liberalization of interest rates in countries with high and unstable rates of inflation can lead to high real interest rates, which may retard investment and growth. Furthermore, macroeconomic instability makes it difficult to maintain the stability of the real exchange rate or the interest rate. The removal of controls on cross-border flows of capital may lead to volatile capital flows and undermine monetary control in situations of large macroeconomic imbalance. Where prices are distorted, financial liberalization may not improve the allocation of resources. Indeed, deregulation may make matters worse. Thus, financial reforms should start by controlling the fiscal deficit and establishing macroeconomic stability.

Second only to the maintenance of macroeconomic stability is the issue of central bank independence as a precondition for the reform process. With very few exceptions, central banks are generally not independent of government influence and, in some cases, of government control. Yet, the priorities of central banks and finance ministries can often diverge with respect to the weights to be assigned to growth, inflation, and the stability of the currency. As such, it is generally recognized that unlike other financial institutions, the central bank must have special status in order to perform its task without being unduly influenced by noneconomic considerations of a transient nature. For the central bank to take an objective stand on economic issues and to maintain a steady course, it must have the right to act independently and speak freely. The pursuit of an independent monetary policy presupposes both the willingness of the political authorities to allow the central bank the freedom to perform its statutory tasks and their preparedness to accept the cost of monetary policy conducted in a market-related environment, especially in the area of debt management.

Empirical studies suggest that market forces need to be supported by consistent public policies. While there are many instances of misdirected intervention by authorities, the fact remains that practically no country that has modernized its financial sector in recent decades has pursued purely market-determined policies. No single model will suit all countries at all times. Policy must be pragmatic and based on a realistic understanding of how monetary and financial variables interact with the real sector. Latin American experience suggests that a sudden dash to financial liberalization without appropriate institutional and legal frameworks can end in a crash, thus requiring even greater government intervention than before the reform. It is fashionable to call for doing away with the regulation of interest rates and the limitations on competition among financial institutions. But this line of thinking could in certain situations push up the cost of finance in unpredictable ways, with a devastating effect on the growth performance of the real sector. Hence the third precondition for the success of a financial reform program is managing the transition in a careful and prudent manner so as not to impose on the economy any avoidable burdens and risks in the wake of deregulation.

Besides the economic requisites, some institutional requirements also need to be met in order to support a program of financial sector liberalization. Rationalization of corporate and commercial laws and their efficient administration are necessary for a liberal financial system. Laws and procedures relating to disclosure of information, restructuring, bankruptcy, enforcement of contracts, foreclosures, and auditing of accounts are all relevant in this respect. No less important is the need to strengthen the machinery for effective supervision of the financial system. With the globalization of the financial markets, the need for much closer cooperation among national supervisory authorities is being increasingly felt. Some progress has been achieved toward such cooperation but much remains to be done.

Pakistan is in the process of financial sector reform. Together with a progressive deregulation of the economy, the Government of Pakistan has undertaken, on a priority basis, the speedy implementation of a comprehensive program of financial sector reform. The objectives of the program are to strengthen the role of the formal financial sector, contain the unofficial market, improve saving incentives, enhance the allocative efficiency of bank credit, and minimize intermediation costs through competition. The reforms initiated in the financial sector broadly cover the areas of public debt management, privatization of financial institutions, increased reliance on market-related monetary instruments, and a strengthening of the supervision of the banking system.

In the area of debt management, the main thrust of the reforms has been to improve and refine the pricing and marketing of government debt. Toward this end, auctioning of the public debt instruments of short-, medium- and long-term maturity is fully in place. Considerable progress has been made in reducing distortions in the interest rate structure. Policy changes in this area are designed to provide a secure return to small savers, achieve a balanced maturity structure of government debt, and develop a secondary market for financial instruments.

The scope of reforms relating to monetary policy extends to eliminating distortions in the interest rate structure and reducing the incidence of subsidized and directed credit. To achieve these objectives the market orientation of policy instruments has been encouraged and sharpened. Increasing reliance has been placed on the indirect control of money and credit, with the eventual goal of eliminating credit ceilings which have served as an instrument of credit regulation for a number of years. Also, progress has been made in the rationalization of rates of return together with reducing the size of subsidized and directed credit.

In the area of bank organization and supervision, the thrust of the reform measures taken so far has been to improve the performance of banks, including the development of financial institutions. New prudential regulations for all banks on loan classification, income accrual and provisioning policy, exposure limits, and audit requirements have been introduced. A program is being implemented to strengthen the capital base of banks. The State Bank of Pakistan is being strengthened to adequately perform its supervisory functions as well as those functions related to monetary policy and public debt management. The supervision of nonbank financial institutions will also be taken over shortly by the State Bank. A credit bureau is being set up to pool information from all financial institutions on their exposures and on the payment records of borrowers.

In conclusion I would like to say that liberalization of the financial sector accompanied by appropriate fiscal and exchange policies and competitive market conditions should be beneficial for the economy. However, a program of financial sector reform should be carefully sequenced over a reasonable, rather than a compressed, time frame. Also important is the need for effective supervision of the financial sector to ensure its viability and maintain public confidence. In this respect one cannot overemphasize the imperative need for ensuring that financial institutions are managed by persons of high integrity and competence.


The authors are grateful to I.A. Hanfi, Governor of the State Bank of Pakistan, Delano Villanueva of the International Monetary Fund, Mark Swinburne of the Reserve Bank of New Zealand, and participants of the seminar for helpful comments and suggestions. The views expressed in this paper are the sole responsibility of the authors and do not necessarily reflect those of the IMF.


See, in particular, Gertler (1988) for a survey of this literature.


See, for example, Genberg (1991).


For case studies and analysis of financial sector reforms, see World Bank (1989).


See, for example, Leite and Sundararajan (1990).


Although most central banks have historically sought to minimize interest rate volatility, analytical and empirical discussions of the impact of such volatility is scarce.


See Sundararajan and Balino (1991) for a discussion of the macroeconomic implications of weak bank portfolios and financial restructuring policies to address the weaknesses.


This requirement is sometimes interpreted incorrectly as implying that income velocity must be constant. What is required, however, is that the demand for money, or velocity, be a predictable function of a few variables.


See Bisat, Johnston, and Sundararajan (1990).


For details on the effect of financial liberalization on the demand for money, see Tseng and Corker (1991).


The money supply can be related to the money multiplier and reserve money as follows:M=mRM(1)

where M is the stock of money, m is the money multiplier, and RM is the stock of reserve money. The money multiplier can be expressed as: m=(1+C/TD)/(C/TD+RE/TD+RR/TD),(2)

where C is the stock of currency, TD is total deposits, RE is excess reserves of banks, and RR is required reserves. Insofar as the ratios in equation 2 are constant owing to the fixity of their determinants in the pre-liberalization period, equation 1 is a useful representation. In practice, the combination of credit ceilings and unfettered growth in bank reserves (for example, the liberal provision of refinance to support selected priority sectors) could lead to a large build-up or variations in excess reserves, causing the money multiplier (m) to vary, but possibly in a predictable way.


The currency-deposit ratio is also likely to fall for another reason: the process of the reintermediation of financial flows back to the previously controlled banking system and away from less regulated intermediaries and financing channels, which are unlikely to have been measured in monetary statistics previously.


See Tseng and Corker (1991). The countries in the sample are Indonesia, Korea, Malaysia, Myanmar, Nepal, the Philippines, Singapore, Sri Lanka, and Thailand. The sample period is from the early 1970s to 1989, and both narrow and broad money concepts are used.


The countries include Argentina, Brazil, Chile, India, Israel, Korea, Malaysia, Mexico, Morocco, and Nigeria.


See Bisat, Johnston, and Sundararajan (1990).


See, for example, Laidler (1978).


See Montiel (1991) for a detailed discussion of the transmission effects of monetary policy in developing countries.


In the latter case, this occurs because of the need to finance working capital.


It should be noted that even in liberalized markets an important element of quantity rationing by banks is likely to remain optimal—because of uncertainty and information costs involved in credit analysis, for example. See Stiglitz and Weiss (1981).


See Organization for Economic Cooperation and Development (1990).


For example, if an increase in short-term interest rates is seen as a clear shift toward tightness and reduces expected inflation, then long-term rates would decline, and the substitution effect would dampen short-term rates. Therefore, for a temporary period, a credibly restrictive monetary policy could produce a negative yield curve. Similarly, if the authorities add to bank reserves and attempt to reduce short-term rates, and if this is seen as a weakening of the anti-inflationary stance, then long-term rates would tend to increase (reflecting higher inflationary expectations), which would pull shorter rates back up through substitution effects.


The impact of short-term rates (which can be controlled by the central bank) on bank deposit and lending interest rates (which are based on banks’ pricing and portfolio decisions) could be complex and depends on the extent of competition in the banking industry, the regulatory constraints and quality of bank assets that affect portfolio choice and marginal costs (such as capital adequacy and provisioning rules, reserve and liquid asset requirements, and the proportion of nonperforming loans), and the degree of openness of the economy to capital flows.


See Organization for Economic Cooperation and Development (1990).


See Jaffee and Stiglitz (1990) and Modigliani and Papademous (1980, 1987) for discussions of credit market behavior and its role in the transmission of the effects of monetary policy.


See, for example, Villanueva and Mirakhor (1990).


For further elaboration of monetary policy implementation procedures, see Freedman (1990) and Kneeshaw and Van den Bergh (1989) for the case of industrial countries and Johnston and Brekk (1989) for the case of developing countries.


See Summers (1991) for a discussion of the role of central banks in payments system reform.


See Bryant (1983) for an analysis of the choice of operating targets under alternative sources of disturbances.

  • Arrau, P., and others, “The Demand for Money in Developing Countries: Assessing the Role of Financial Innovation” (Washington: International Monetary Fund, unpublished, 1991).

    • Search Google Scholar
    • Export Citation
  • Bisat, Amer, Johnston, R. and Sundararajan, V. “Issues in Managing and Sequencing Financial Sector Reform” (Washington: International Monetary Fund, unpublished, 1990).

    • Search Google Scholar
    • Export Citation
  • Bryant, Ralph C, Controlling Money: The Federal Reserve and Its Critics (Washington: Brookings Institution, 1983).

  • Freedman, C., “Implementation of Monetary Policy,” in Monetary Policy and Market Operation (Sydney: Reserve Bank of Australia, 1990).

    • Search Google Scholar
    • Export Citation
  • Genberg, H., “On the Sequencing of Reform in Eastern Europe” (Washington: International Monetary Fund, unpublished, 1991).

  • Gertler, Mark, “Financial Structure and Aggregate Economic Activity: An Overview,” NBER Working Paper No. 2559 (Cambridge, Mass.: National Bureau of Economic Research, April 1988).

    • Search Google Scholar
    • Export Citation
  • Jaffee, Dwight, and Stiglitz Joseph “Credit Rationing,” in Handbook of Monetary Economic Vol.2, edited by Friedman Benjamin M. and Hahn Frank H. (Amsterdam: North-Holland, 1990).

    • Search Google Scholar
    • Export Citation
  • Johnston, R.B., and Brekk, Odd Per “Monetary Control Procedures and Financial Reform: Approaches, Issues, and Recent Experiences in Developing Countries” (Washington: International Monetary Fund, unpublished, 1989).

    • Search Google Scholar
    • Export Citation
  • Kneeshaw, J.T., and Van den Bergh, P. “Changes in Central Bank Money Market Operating Procedures in the 1980s,” BIS Economic Papers No. 23 (Basle: Bank for International Settlements, 1989).

    • Search Google Scholar
    • Export Citation
  • Laidler, David, “Money and Money Income: An Essay on the Transmission Mechanism’,” Journal of Monetary Economics, Vol.4 (April 1978), pp. 151 -91.

    • Search Google Scholar
    • Export Citation
  • Leite, Sérgio P., and Sundararajan, V. “Issues in Interest Rate Management and Liberalization,” Staff Papers, Vol. 37 (December 1990), pp. 735 -52.

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

  • Modigliani, Franco, and Lucas, Papademous “The Structure of Financial Markets and the Monetary Mechanism,” in Controlling Monetary Aggregate Vol. lll (Boston: Federal Reserve Bank of Boston, 1980).

    • Search Google Scholar
    • Export Citation
  • Modigliani, Franco, and Lucas, Papademous “Money, Credit, and Monetary Mechanism,” in Monetary Theory and Economic Institution, edited by Cecco Marcello De and Fitoussi Jean-Paul (London: McMillan, 1987).

    • Search Google Scholar
    • Export Citation
  • Montiel, Peter J., “The Transmission Mechanism for Monetary Policy in Developing Countries,” Staff Papers, Vol. 38 (March 1991), pp.83 -108.

    • Search Google Scholar
    • Export Citation
  • Organization for Economic Cooperation and Development, Macroeconomic Consequences of Financial Liberalization, Report of Working Party No. 1 of the Economic Policy Committee (Paris: OECD, 1990).

    • Search Google Scholar
    • Export Citation
  • Poole, William, “Optimal Choice of Monetary Policy Instruments in a Simple Stochastic Macro Model,” Quarterly Journal of Economics, Vol. 84 (1970), pp. 197 -216.

    • Search Google Scholar
    • Export Citation
  • Stiglitz, Joseph, and A. Weiss, “Credit Rationing in Markets with Imperfect Information,” American Economic Review, Vol.71 (June 1981), pp. 393 -410.

    • Search Google Scholar
    • Export Citation
  • Summers, Bruce, “Clearing and Payment Systems: The Role of the Central Bank,” Federal Reserve Bulletin (January 1991).

  • Sundararajan, V., and Balino, J.T. Banking Crises—Cases and Issues (Washington: International Monetary Fund, 1991).

  • Tseng, W., and Corker, R. Financial Liberalization, Money Demand, and Monetary Policy in Asian Countries, Occasional Paper (Washington: International Monetary Fund, 1991).

    • Search Google Scholar
    • Export Citation
  • van Wijnbergen, Sweder, “Interest Rate Management in LDCs,” Journal of Monetary Economics, Vol.12 (September 1983), pp. 433 -52.

  • Villanueva, Delano, and Abbas Mirakhor, “Strategies for Financial Reform: Interest Rate Policies, Stabilization, and Bank Supervision in Developing Countries,” Staff Papers, Vol. 37 (September 1990), pp. 509 -36.

    • Search Google Scholar
    • Export Citation
  • World Bank, World Development Report 1989 (Washington: World Bank, 1989).