Because the best place to begin discussing this broad topic is with a definition of financial sector reform, I start by trying to provide one. Next, I discuss how the process of developing a more efficient monetary control regime is related to the various components of financial sector reform. Then I consider some of the macroeconomic consequences of financial sector reform as they affect the formulation and implementation of monetary policy. I conclude with a discussion of how financial sec-tor reform can be used in a stabilization program to support macroeconomic policies.

V. Sundararajan

Because the best place to begin discussing this broad topic is with a definition of financial sector reform, I start by trying to provide one. Next, I discuss how the process of developing a more efficient monetary control regime is related to the various components of financial sector reform. Then I consider some of the macroeconomic consequences of financial sector reform as they affect the formulation and implementation of monetary policy. I conclude with a discussion of how financial sec-tor reform can be used in a stabilization program to support macroeconomic policies.

What Is Financial Sector Reform?

Financial sector reform can be viewed either as a set of policies or as a process of moving from one situation to another. The policies are those measures necessary to deregulate certain financial sector operations and transform the structure of the sector itself, with the objective of creating a liberal, market-based environment moderated by a good regulatory framework. Financial reform as a process is the movement from an initial situation of controlled interest rates, poorly developed money and securities markets, and an underdeveloped banking system toward a situation of flexible interest rates, an expanded role for market forces in resource allocation, increased autonomy for commercial banks (and possibly for the central bank), and a deepening of money and securities markets. This process is what I will discuss.

During such a process, the framework of monetary policy always undergoes a major transformation. The types of instruments that are used change dramatically. Bank-specific credit ceilings, controls on interest rates, and various mechanisms to direct or subsidize credit to end users are replaced by market-based instruments and new prudential supervisory systems. At the same time, the exchange and trade regime shifts from one with strong restrictions and limited foreign exchange markets to one that is moving toward convertibility and active interbank foreign exchange markets. Financial sector reform affects the objectives and design of monetary policy, the implementation of monetary policy, and the transmission mechanisms of monetary policy.

What are the components of financial sector reform? Box 1 provides a list of the major types of reforms. All are important and have received some attention in reforming countries. Interestingly, the group of measures affecting the payments clearing and settlement system has not been a particularly important aspect of overall financial reform in some of the Asian economies, probably because initial conditions were reasonably sound. Only in the later stages of the reform process have these countries taken measures to strengthen their clearing and settlement systems. On the other hand, in Eastern Europe and the former Soviet Union (FSU) these reforms have been extremely important from the outset and have been implemented before other basic changes in the financial sector. The policies that have been put in place in these countries include transitional reforms based on readily available technology, such as changes in accounting rules—for instance, clarifying when to debit and when to credit payments—as well as the development of interbank transfer systems and multilateral clearing arrangements based on more modern technologies and risk management systems. I will discuss the significance of these matters for monetary policy later.

Monetary Control Regimes and Financial Reform

Rather than listing the reforms a particular country may choose to implement and describing the ways in which they are linked, I will discuss the process of moving from a situation of undeveloped money and financial markets and direct controls on interest rates toward one involving more sophisticated money and financial markets and liberalized interest rates. Chart 1 shows that various reforms are necessary to facilitate the transformation of a monetary policy framework from one involving direct credit controls to one based on indirect instruments. The chart also shows that the way in which the monetary policy framework is transformed influences the way the financial sector evolves. Clearly, the process is an interactive one.

Chart 1.
Chart 1.

Stylized Model of Reforms of the Financial Sector and Monetary Control Regime

Source: M. S. Khan and V. Sundararajan, “Financial Sector Reforms and Monetary Policy,” IMF Working Paper WP/91/127, December 1991.1 Includes actions to deal with fixed-rate loans, nonperforming loans, capital adequacy, and subsidized selective credits, among other things.

Before proceeding, it is necessary to define two terms: interest rate liberalization and indirect instruments. Liberalizing interest rates does not mean that the authorities give up control of interest rates; rather, it is a process by which the authorities successfully manage key interest rates with indirect policy instruments. In other words, new monetary operations are introduced in order to influence the level of interest rates or another operating or intermediate target.

Components of Financial Sector Reform

  • Reforms of the interest rate regime, supported by the development of market-based monetary control procedures. These reforms involve liberalizing interest rates and revamping the operation of monetary policy.

  • The development of money and interbank markets, including the introduction of new money market instruments and efficient interbank settlement systems and the development of institutional arrangements for primary and secondary markets in treasury bills, bills of exchange, and the like.

  • Reforms of prudential regulations and the supervisory system, such as reevaluating the balance between off-site monitoring, on-site inspections, and external audits; modifying accounting and prudential regulations; strengthening enforcement options; and improving financial analyses of banks and establishing early warning systems.

  • Recapitalization and financial restructuring of weak financial institutions, supported by policies to restructure enterprises.

  • Measures to strengthen competition among banks. Such measures include relaxing entry restrictions, unifying portfolio regulations, privatizing and granting autonomy to state banks, and introducing policies on mergers and rules on disclosure of information.

  • Reforms of selective credit regulations. These reforms are aimed at reducing the scope of direct credit allocations based on restrictions on bank asset portfolios, removing or phasing out interest subsidies, and changing the operation of central bank credit policies, among other things.

  • The development of long-term capital markets that will strengthen public debt management, foster government securities markets, promote long-term corporate debt and equity markets through improved securities regulation, and improve the organization of trading and settlement in securities.

  • Reforms of the clearing and settlement system for payments. Such reforms involve reevaluating the role of the central bank in the payment system, developing an interbank settlement system for large payments, strengthening the rules governing net settlement systems, and introducing a book-entry settlement system for securities transactions.

  • The development of foreign exchange markets, supported by appropriate prudential regulations on foreign exchange exposure and the liberalization of exchange and trade restrictions. Reforms in this area include the setting up of institutional arrangements for foreign exchange trading.

Source: V. Sundararajan, “Financial Sector Reforms and Their Appropriate Sequencing,” in The 19th SEANZA Central Banking Course (Tokyo: Bank of Japan, 1992), pp. 1 77-203.

Through these operations, the central bank simply manages the level of balances that the commercial banks hold with the central bank and, in doing so, is able to achieve the desired target. Thus, indirect instruments are those with which a central bank manages some component of its balance sheet, producing an immediate impact on the level of balances the commercial banks hold at the central bank.

Clearly, the next question is whether this type of mechanism can be effective in influencing other variables, such as the level of interest rates in the market, the level of money supply, the level of credit in the system, and inflation. This question of how effectively key monetary aggregates can be controlled through indirect instruments has become a major issue in the financial liberalization process.

Reforms of the monetary control system comprise changes in the institutional arrangements for money markets; the design of market-based monetary policy instruments and operations, including supporting changes in information systems for monetary exchange policy management; and other supporting reforms, such as measures to improve clearing and settlement. These three components are closely linked, and reforms in these areas are mutually reinforcing. Experience suggests that reforms of the monetary control system should be given fairly high priority and should be implemented fairly early in the reform process, because (i) they give immediate support to stabilization objectives, and (ii) they lay the foundation for the development of financial markets.

The need for comprehensive reform

The process of interest rate liberalization—that is, the ability to manage interest rates through indirect instruments—requires a number of supporting conditions. Yet it is difficult to know a priori which type of reform should come first. Stable macroeconomic conditions are essential to successful interest rate liberalization. The initial structure of interest rates must be reasonable; if many end users of credit are benefiting from interest rate subsidies or very low interest rates, the process of liberalization becomes difficult to manage, because there will be too many winners and losers. Adequate competition in the banking system is also necessary and requires that banks be reasonably sound; otherwise they cannot compete and may respond to monetary policy in perverse ways. Money and interbank markets must be functioning reasonably well, and policy instruments for managing interest rates should be in place. Strong banking oversight policies must also be in place to control risktaking by commercial banks and to ensure that the banking system remains sound and able to respond flexibly to monetary conditions.

Ideally, then, a comprehensive set of reforms is needed in order to liberalize interest rates successfully. In reality, it is impossible to implement all these reforms at once or even to have them in place before rates are liberalized. Instead, reforming countries need a transition strategy to help them in the move toward greater interest rate flexibility, as Chart 1 shows. Typically, this strategy begins to take shape when the authorities decide to phase out direct controls and transform the way monetary policy is conducted by freeing interest rates to some degree. For this purpose, the major impediments to developing greater interest rate flexibility must be identified. Because these impediments differ across economies, the resulting action plan will be highly country-specific. In general, neither banks nor enterprises are able to handle increases in interest rates, banks because their loan portfolios are weak, and enterprises because they are heavily in debt. Therefore, some financial restructuring to improve both banks’ balance sheets and the financial position of enterprises may be necessary up front.

What are the other possible impediments to interest rate flexibility and indirect monetary management? The experience of many economies in transition reveals specific institutional impediments. First, the legal framework for transactions in new types of instruments, such as bills and fixed-income securities, may be weak or even nonexistent. Another common problem is the unreliable settlement system for payments, so that banks’ reserve balances show large fluctuations that reflect changes in the interbank payment float. These unforeseen fluctuations often completely overshadow other, more predictable influences on bank reserves, making it difficult for both the central bank and commercial banks to manage the reserve levels. In response to this problem, many countries have introduced changes in accounting rules relating to the timing of debits and credits and have made other transitional improvements in the clearing and settlement system that are aimed at reducing float and facilitating the development of indirect monetary instruments.

Designing monetary policy instruments and operations

Building on such initial reforms, more specific measures to develop markets and market-based instruments can be initiated (see Chart 1, Stage 2). The key question here is how to develop market-based instruments in the absence of markets. In fact, however, it is not necessary to have fully developed markets in order to develop open-market operations, because these types of operations can be designed according to the nature of existing markets. With very sophisticated financial markets, conventional open-market operations are possible. With very limited markets, open-market operations can initially be aimed at creating or strengthening a market. The easiest way to begin is with the central bank, which can develop simple, market-based instruments it then manages flexibly in order to create the conditions for interbank money markets to develop. These conditions set the stage for further developments in money and securities markets and enable further refinements in monetary operations, in line with developments in market infrastructure and practices.

The above is the typical interactive process that has emerged from country experiences. In the absence of well-developed money and securities markets, many central banks have relied initially on primary securities markets, auctions of central bank credit, or a combination of the two. The usual practice is to auction treasury bills or other government securities (a procedure needed in any case to provide noninflationary financing for budget deficits) and then to use these primary auctions as an instrument of monetary policy. The terms primary market operation and open market-type operation may be used to describe the use of this kind of monetary instrument, because open market operation is normally used to describe an intervention in a secondary market. But whatever the terminology, the operation must be conducted at market-clearing interest rates, something that requires a certain amount of commitment and coordination from the government, particularly if government securities are used to finance fiscal deficits and conduct monetary policy simultaneously. In this case, there are two goals but only one instrument, often necessitating strengthened arrangements for coordination between the ministry of finance and the central bank.

In principle, monetary management can be separated from public debt management if the instruments and markets are also separated. Many countries have in fact made this separation by issuing both central bank and treasury bills, even though it does create complications. In either case, open market-type operations can be used to even out fluctuations in bank liquidity and to avoid the protracted liquidity imbalances that occur when many banks are trying to invest excess reserves at once or are all facing seasonal liquidity shortages. The active management of liquidity by a central bank via simple open market-type instruments creates the conditions private sector agents and banks need to begin not only to manage liquidity but to establish the relationships necessary for this purpose.

However, the use of central bank or treasury bills alone is generally not sufficient to manage the ups and downs of bank reserves. These fluctuations are caused by so many factors, including capital inflows and unplanned government spending, that other instruments besides treasury bills are usually needed. In many economies in transition, the initial situation poses special challenges from this standpoint, because most of the banks are heavily dependent on the central bank. The banks’ assets are matched by a relatively small volume of deposits and a large amount of central bank credit, so that treasury bills are not the only way to introduce market-based instruments. In many of these countries, the central banks have introduced credit auctions to sell part of the outstanding central bank credit to banks, either on an unsecured basis or with collateral (if it is available).

In most countries where the authorities initially use central bank or treasury bills to manage bank reserves, other market-based instruments are eventually needed. Central bank or treasury bills are a very convenient instrument for withdrawing excess funds from the banking system. However, in open economies faced with unexpected capital inflows, trying to mop up all the liquidity in this way requires increasing the volume of central bank bills too suddenly. It is useful to have some additional instruments at hand to reduce the volume of credit to the system. Similarly, the authorities may also wish to supply additional reserves to the system. Some countries sell fewer central bank and treasury bills and do not replace them when they come to maturity. This practice is not an effective one, because a minimum level of securities should be maintained in the market to meet investors’ needs at all times. In short, there are practical constraints involved in raising the volume of issues too sharply, as well as limits beyond which the volume of treasury bills should not be reduced. Additional instruments—such as the short-term auctions of collateralized credit used in many countries (e.g., Indonesia)—are needed to slow the growth in reserves or to inject reserves into the system.

Even in the initial stages, and certainly in Stage 2, an appropriate instrument mix is necessary both to manage the level of bank reserves and to promote the development of money markets. The central bank can design these instruments to stimulate market development. For example, the use of collateral on central bank credit automatically increases the demand for such collateral, and various types of repurchase agreements not only provide some flexibility in managing reserves but also stimulate demand for different types of securities. With the help of a mix of instruments, the central bank can begin to deal with commercial banks on the basis of uniform, market-based criteria. With market-based methods, the distribution of demand for reserves among individual banks can be quite different from the distribution of supply, thereby setting up incentives to develop money markets. It is up to the money markets, not the central bank, to redistribute the reserves in a way that will eventually meet the needs of individual participants in the market.

In this way, the moment a central bank starts to use a simple mix of instruments to manage the liquidity in the system, commercial banks will be induced to become active, and interbank money markets will begin to deepen. These markets in turn facilitate further refinements in the operating procedures of monetary policy, leading to Stage 3 and fully flexible interest rates (see Chart 1). In this stage, the central bank operates mainly on its own initiative, using open-market operations, and the liquidity of the various securities in the system is not derived mainly from the central bank’s discount window but is provided primarily by the market. Defensive monetary policy operations to smooth out fluctuations in interest rates become progressively important as reforms proceed. These operations are needed to prevent interest rate volatility and ensure that the changes in the direction of interest rates sought by the central bank are not masked by such volatility.

Supporting reforms

The transition process described above requires many other reforms, some of which I have already mentioned. First and most important is the payments system. I mentioned earlier the accounting rules for debiting and crediting. Many countries have also introduced availability schedules to credit and debit payment transactions, a move that helps synchronize settlement transactions. This synchronization reduces fluctuations in bank reserves arising from variations in the amount of payments in transit. It is increasingly recognized that the development of at least a rudimentary large-value transfer system for interbank payments is key to the development of money and foreign exchange markets. Many countries are trying to develop their own large-value transfer systems that will allow interbank transactions to be settled on a same-day or even a real-time basis. These types of reforms, the design of which is heavily influenced by the details of monetary policy instruments and the behavior of central bank reserves, are important both to the effectiveness of monetary policy and to the management of risks in the payments system.

In addition, the central bank needs very strong information systems in the course of the transition to indirect instruments. It needs to know, on at least a weekly basis, what is happening to its balance sheet, and it needs a strong policy research program that will allow it to anticipate the evolution of its balance sheet. Depending on the prevailing exchange regime, foreign exchange market operations must be developed to foster well-functioning exchange markets and to coordinate monetary and exchange policies. A systematic program to promote government securities markets is also needed. The objectives of monetary policy, market development, and public debt management cannot be met with just a limited range of instruments (such as treasury bills) or on the basis of captive markets that rely on rigid liquid asset requirements. Instead, a comprehensive public debt management system should be developed based on a range of instruments that are targeted at a wide range of investors and that use a variety of different selling techniques. Simultaneously, the scope of captive markets based on liquid asset ratios should be progressively reduced and eventually eliminated.

Issues in the development of secondary markets in government securities need to be considered, including the appropriate microstructure of these markets. Should the government securities market be based on centralized stock exchange trading, or should it be a decentralized arrangement based on primary dealers supported by brokers? What are the privileges and obligations of dealers in the market? With a sufficient volume of securities and a wide range of securities holders, the central bank can operate at its own initiative and let the market provide liquidity to the dealers. The central bank manages the overall level of liquidity in the market through open market operations, however. Thus, an effective program of secondary market development involves, among other things, changes in operating procedures for the central bank that allow it to act on its own initiative.

This process of developing secondary markets is also an interactive one. The central bank licenses dealers and tries to operate through them rather than directly through its discount window. In this way, the central bank helps create conditions that promote the development of dealing skills and money markets, which in turn form the foundation for the further development of markets in long-term securities. What begins as the use of primary markets for monetary policy operations gradually gives way to the use of secondary markets. Moreover, the development of money markets and payment systems permits banks to maintain a reduced level of reserves, raising the demand for money market instruments.

In the transition process from Stage 1 to Stage 2 to Stage 3, parallel reforms of prudential regulations and the supervisory system are also very important. Why? Because as interest rates become more flexible, credit and market risks need to be managed with care. Some banks can assume large risk exposures. Other banks with a large volume of bad debts will not be able to allocate their credit resources appropriately in response to changes in monetary conditions, since the available resources must be used to roll over loans to nonviable borrowers. The situation requires policies not only to restructure bank finances but also to tighten bank regulations and supervision in order to improve credit allocation, reduce risks, and minimize potential disruptions to stability. As markets develop, the need to contain market risks—in addition to credit risks—assumes significance. For example, with new liquidity management instruments, supervising the asset-liability management of commercial banks becomes an issue. And with the development of foreign exchange markets, prudential regulation of banks’ foreign exchange exposures also becomes important.

Some Macroeconomic Consequences of Financial Reform

This discussion of the components of banking supervision provides a point of contact with my third topic, because prudential regulations may have a direct effect on macroeconomic outcomes. For example, a government may be trying to introduce stronger capital adequacy ratios or loan classification and provisioning rules to strengthen banking regulations. These policies can affect interest rate movements and the growth of credit. In trying to improve their capital adequacy ratios, banks may be restrained in expanding their assets, or they may adjust their interest spreads to build up reserves. In this case, prudential regulations may produce a monetary effect that complements or counteracts the effects of monetary policy. However, as a general principle, banking supervision should not be used as a tool of monetary policy but should be kept neutral. Nevertheless, the monetary effects of new prudential measures may dictate how the introduction of these measures should be timed.

Other macroeconomic aspects of financial sector reform can be grouped under three headings: effects on money demand, on money supply, and on the monetary transmission mechanism.

Effects on money demand

The stability of money demand provides the link between monetary policy instruments and the ultimate objectives of monetary policy, such as the rates of inflation and growth. The demand for money can change as a consequence of the reforms described above, through either one-time shifts or changes in its responsiveness with respect to one or more of the various determinants.

It is hard to generalize about the precise effects of financial reform on money demand. However, there is often an increase in the money-to-GDP ratio and a decrease in the currency-to-GDP ratio. Recent research suggests that most parameters of money demand do not change much in the long run but that the responsiveness of money to various underlying variables can fluctuate in the short run. One effect is universally accepted, however: the sensitivity of various monetary aggregates to interest differentials does increase substantially. At least in the initial stages of a reform process, there will be some instability in money demand.

Effects on money supply

The key variable to observe here is the level of excess reserves (i.e., the level of balances over and above the required reserves that the commercial banks hold with the central bank. As reform proceeds, the level of excess reserves typically becomes highly sensitive to short-term interest rates and to the details of the clearing and settlement arrangements. Some countries have consolidated all bank accounts into one account per bank in order to facilitate interbank settlements and centralized liquidity management. With multiple accounts, branches of banks were able to keep reserves in each location of the central bank offices. Consolidating these accounts reduced the level of reserves required for settlement and made a compensating tightening of monetary policy necessary. Much the same thing occurred when several countries made a change in the accounting arrangements for debiting and crediting payment transactions that led to a sudden reduction in the demand for bank reserves. Such factors cause shifts in the money multiplier, thereby affecting the controllability of monetary aggregates with indirect instruments.

In brief, many structural factors affecting bank reserves must be watched constantly during the reform process. In general, excess reserves tend to fall substantially during financial reform; in many cases, they have fallen from 20-25 percent of deposits to 2-5 percent in a matter of 12-18 months.

Typically, the excess reserves the system needs are constantly being economized, especially in the initial stages of financial liberalization. As a result, credit expands faster than money, assuming that credit controls are also liberalized. This phenomenon can put pressure on interest rates. The authorities must then either raise interest rates in order to control credit growth or encourage the fiscal authorities to make additional adjustments in fiscal policies. One way or another, the overall pressure on aggregate demand needs to be managed.

Of course, interest rates alone can be allowed to absorb the shock, but there is considerable danger in this course of action. For example, if nonfinancial companies are highly leveraged, as they often are in developing countries, and many borrowers are weak, high interest rates can cause distress borrowing and put pressure on banks to capitalize interest. This situation can create instability by causing further increases in interest rates. It may then become necessary to phase in interest rate reforms, with a view to avoiding excessive interest rate increases and to giving enterprises time to adjust their debt-equity mix. Additionally, it may be useful to continue using direct controls flexibly until the financial restructuring of the banks and enterprises is well under way.

These practical management issues highlight the fact that a wide range of reforms are needed in the financial sector to make interest rate liberalization effective. The speed with which interest rates are liberalized depends on the speed with which these supporting reforms can be put together. If there are delays in implementing the needed supporting reforms, then interest rate liberalization must be phased in and some direct controls continued to prevent excessive rate increases. In Korea, for example, the policy has been to phase in interest rate liberalization gradually. In adopting this policy, the authorities were motivated mainly by concern about the consequences of more rapid reforms for the highly indebted nonfinancial firms.

Effects on monetary transmission

What is the relationship between financial liberalization and the intermediate targets the authorities want to influence? A common problem is that the central bank may have the tools to exercise control over the day-to-day movements in money market interest rates but, in exercising this control, seems to produce only a delayed effect on bank deposit and lending rates. Similar problems arise in countries where the relationship between short- and long-term interest rates is subject to a number of influences. These issues need to be addressed in the context of financial sector reform.

The conventional transmission mechanism, in the textbook sense, operates through interest rates, and economic agents respond to the differentials in yields on various assets. Various interestsensitive components of expenditures, such as housing and consumer durables, are affected, with very clear consequences for levels of aggregate expenditures. Other types of transmission mechanisms—such as those that involve the availability of credit—are more typically associated with developing countries. Even in industrial countries, economists are increasingly emphasizing the role of credit rationing as a potentially important transmission mechanism. The argument is that monetary policy changes the availability of credit and thereby influences aggregate expenditures. However, monetary policy can also change the structure of the balance sheets of various economic agents, and this change can, in turn, influence real economic behavior. If, for example, the debt-equity mix of enterprises is influenced by a monetary policy shift, enterprises may try to adjust this mix.

The most important effects operate by way of the exchange rate. To the extent that the exchange rate is flexible, changes in monetary policy that influence interest rates can also affect the exchange rate and the levels of real expenditures. Both interest and exchange rates will have to be adjusted in some fashion to induce a tightening or loosening of monetary conditions. I must add here that the policy and operational issues of coordinating monetary control and exchange systems are beyond the scope of this discussion.

In the course of reform, policymakers must deal with potential instability in money demand. They operate with an incomplete knowledge not only of how the transmission mechanism of money policy is going to change but also of how the money supply process itself is going to evolve. These problems may seem to lead to the pessimistic conclusion that the financial reform process is difficult to manage during stabilization efforts. A more appropriate conclusion, however, is that these problems pose a challenge, which can be met if specific reforms in the financial sector are properly sequenced to support stabilization objectives (taking into account the interactions among different components of reform), with the caveat that continuous vigilance must be exercised in assessing monetary conditions. In practice, central banks monitor a set of indicators and then adjust their targets and instruments on a month-to-month basis. They try to achieve the newly set targets with the best instrument mix they have. In other words, monetary policy in the context of liberalized financial markets, or in the course of financial liberalization, operates via a kind of feedback mechanism. The authorities monitor a set of indicators, change the targets as needed, change the instrument mix in order to achieve the targets, and so on. At some point, the indicators themselves may have to be changed.

In conclusion, several aspects of financial sector reform need to be emphasized. First, a minimal system of prudential regulations must be introduced, or at the least, the system must be tightened up. Banks must be recapitalized and, in most cases, restructured financially. These measures greatly facilitate the overall effectiveness of reforms. Second, key reforms of monetary operations, money-market arrangements, and the clearing and settlement system need to be put in place early in the reform process to ensure that macroeconomic control is adequate while interest rates and direct controls are being liberalized. And the importance of supporting measures should not be overlooked. Seeking a proper role for monetary policy through central banking and financial sector reform remains an ongoing challenge.

Summary of Discussion

The first part of the discussion took the form of a question-and-answer session with V. Sundararajan. Initially, several discussants argued that a certain degree of macroeconomic stability is necessary before financial sector reform can effectively be undertaken. Conditions such as the hyperinflation that currently exists in Russia and Brazil hinder the effectiveness of financial reform and may prevent governments from initiating some measures. Mr. Sundararajan noted that specific financial sector reforms are possible even under unfavorable conditions and may in fact be needed to support stabilization policies—for example, reforms of the monetary control system. However, the rapid development of money and security markets is hindered by factors such as high inflation and unstable nominal interest rates.

The point was raised that a government, by issuing treasury bills, could crowd out the private sector. Discussants disputed this point, maintaining that the real source of crowding out in such circumstances is the size of the fiscal deficit, not the particular financial means used to fund that deficit.

Mr. Sundararajan was questioned about the merits of using central bank bills (as opposed to treasury bills) in open-market operations and the manner in which the cost of interest on these bills should be shared. Mr. Sundararajan’s view was that a single instrument is preferable for both public debt and monetary management, as this approach avoids the problem of a segmented market in the initial stages of reform. The sharing of interest costs should not be a major concern, as ultimate responsibility for the interest charges—and for the losses the central bank incurs in its monetary and exchange management functions (e.g., revaluation losses on foreign reserves)—lies with the government. As earlier speakers pointed out, a central bank often loses money because it is asked to take on functions that are properly the responsibility of the government.

Mr. Sundararajan was also queried about the management of large capital flows. He explained that the conventional wisdom is to accommodate these inflows if they result from a genuine increase in the demand for domestic money but to sterilize the monetary effects if they are merely speculative short-term flows. But he admitted that divining the true nature of such flows is, in practice, an extremely difficult matter.

He offered a similar answer in response to a question about targeting monetary aggregates during the reform process, when money demand may be volatile. Mr. Sundararajan noted that in situations involving significant fluctuations in money demand, the approach is to target an interest rate variable. When there is significant volatility in output and expenditures, or when an economy is experiencing external shocks, it is better to target a monetary aggregate and allow interest rates to find their own level. Again, however, determining the exact nature of a shock is extremely difficult, and economists must often rely on their own judgment. For this reason, there is no substitute for continued vigilance in assessing indicators of monetary conditions, and policymakers must always be willing to reset targets and instruments when circumstances change.

The discussion then turned to treasury bills. Discussants noted that the authorities should not allow the volume of bills in circulation to fall too low, as a shortage of treasury bills could hurt the markets and deter investors. If the number of treasury bills needed to finance the budget do not satisfy demand, it makes sense to overfund the budget in order to maintain a reliable supply of bills to regular investors. If a central bank chooses to issue its own paper, these bills should be used only as instruments of monetary management and should not be confused with the objective of debt management. In New Zealand, for example, where both treasury and central bank bills are used as instruments of monetary management, the bank bills are used to signal changes in the direction of monetary policy.

Finally, discussants were split on the issue of recapitalization and the financial restructuring of banks. Some felt that such actions send the wrong signals and that poorly performing banks, like any other loss-making business, should be allowed to fail. There was some agreement that the poor financial health of banks in the transition economies of Eastern Europe and the former Soviet Union is the result of government interference and not of inadequacies in the banks themselves. But discussants could reach a consensus neither on this question nor on the related subject of why banks should be regulated.

Proceedings of the Seminar Coordination of Structural Reform and Macroeconomic Stabilization, Washington, D.C., June 17-26, 1993