V Implications for the Adoption of Indirect Instruments
- Tomás Baliño, Charles Enoch, and William Alexander
- Published Date:
- July 1995
The experiences of industrial and nonindustrial countries confirm the essential validity of the two basic reasons for introducing indirect instruments: to establish effective monetary control and to improve the efficiency of financial intermediation by allowing an allocation of financial resources on a market-determined basis.
As to effective monetary control, there is little doubt that, transitional problems aside, industrial countries have been able to exercise firm monetary control through the exclusive use of indirect instruments. As the literature on the subject is large and readily available, this paper does not focus on it.42 The previous section noted, however, the experience of some industrial countries in which the usefulness of direct instruments declined precipitously following the liberalization of the capital account. That experience provides unambiguous evidence that there is no alternative to adopting indirect methods if monetary control is to be sustained. Capital account convertibility greatly increases the means for circumventing direct methods of monetary control.43
As regards nonindustrial countries, the evidence presented in the preceding section also suggests that once the transition period is complete, the volatility of both money multipliers and key monetary aggregates declines to manageable levels, tending to make them more predictable. Also, interest rates generally move to positive levels in real terms and, as reflected in increased volatility of short-term rates, become more sensitive to both market conditions and central bank actions after the transition. Thus, once the transition period is complete, the technical conditions exist for effectively implementing monetary policy by using indirect instruments.
Similarly, the evidence strongly suggests that the efficiency of financial sector intermediation improves. Interest rate spreads narrow sharply, and the growing share of bank loans and deposits in overall financial sector activity suggests that reintermediation through the banking system increases markedly following the removal of the direct controls. Positive real rates of interest and the increase in the private sector’s share of total credit—while not unambiguous indicators—can be viewed as presumptive evidence of a more rational allocation of credit as direct controls cease to be binding.
That said, the country experience also points to the substantial difficulties and costs that are often encountered during the transition to indirect methods of monetary control. These need to be recognized and thoroughly addressed. While some countries have been able to make the transition rapidly and rather smoothly, transition has been a lengthy process for many others. About half the countries in the nonindustrial sample encountered reversals, and many experienced severe financial crises—although such setbacks were not always related to the shift to indirect instruments. Moreover, in most countries, unsettled conditions during the transition temporarily made it technically harder to maintain monetary control.
Thus, stable economic conditions and a sustainable fiscal position are desirable when introducing indirect instruments of monetary policy. They are, however, neither necessary nor sufficient: adopting strong adjustment measures seems to be more important in the long run than the initial macroeconomic conditions. Countries that start with large fiscal deficits and high levels of inflation but achieve fiscal and monetary consolidation have a better chance of a relatively smooth transition process. Conversely, countries with favorable initial conditions but that undertake expansionary policies and those with extreme initial macroeconomic instability (such as triple-digit inflation and a fiscal deficit larger than 10 percent of GDP) are more likely to experience reversal of financial liberalization, even severe financial crises.
The question arises whether the evidence for indirect instruments is strong enough to support the case for introducing them in all circumstances, particularly for a small isolated economy where direct instruments seem to be working reasonably well and where the conditions for the successful use of indirect instruments appear to be lacking. Two points need to be considered. First, even if an economy lacks a well-developed domestic financial market, its residents will have some access to overseas financial markets (see Mathieson and Rojas-Suárez, 1993). Thus, the extent of financial competition that is relevant for the analysis cannot be established simply by counting the number of domestic banks. Second, monetary instruments can be designed to take into account the limitations of the domestic banking system. For instance, if insufficient banking competition prevents the running of an efficient auction of government or central bank securities, the categories of institutions that can participate can be broadened. Thus, institutional investors, large enterprises, and even individuals can be allowed to present bids.
In fact, the introduction of instruments like government securities can enhance competition by providing the public with a quality paper that can compete with bank deposits and by facilitating other operations—like collateralized lending. Moreover, the indirect instruments discussed in this paper are wide ranging; some of them can be effective when others might not function well. For instance, reserve requirements can often substitute for credit ceilings and improve the efficiency of financial intermediation, even in cases in which more market-oriented instruments (such as open market operations) may not function properly owing to poor competition. In sum, a pragmatic approach to the design, use, and phasing in of indirect instruments is warranted. That the adoption of indirect instruments will entail costs should hardly be surprising. As a key component of the broader process of financial sector liberalization, adopting such instruments has been associated with instability and overheating in particular markets.44
The risk that countries will incur heavy transition costs suggests that implementation should be carefully planned and comprehensive in order to avoid major pitfalls. As with the more general case of financial sector liberalization, however, theory offers little guidance on such practical aspects as key supporting measures, the pace of reform, and sequencing (including the important issue of the point at which to abandon the use of direct instruments). Country experiences may offer some guidance.
Analysis of the sample reported in the preceding section, as well as a detailed examination of some selected case studies included in Part II, strongly suggests that certain supporting actions, listed below, facilitate the transition to indirect methods of monetary control.
Insulate Monetary Policy from Deficit Financing
When the fiscal deficit is large, the achievement of monetary objectives is likely to come under strain because it becomes difficult to sell a large volume of securities to finance the deficit; such sales raise the interest rate. Moreover, the budget would need to make room for the additional interest cost of borrowing at market rates. Thus, strains arise if the authorities are reluctant to accept the interest rate consequences of an expansionary fiscal policy or if concern over the cost of securities sales leads to pressures to reduce the volume and monetize the government financing shortfall.
As a practical matter, fiscal balance is probably sufficient to remove this pressure and allow the transition to go forward uninterrupted, but the evidence presented in the preceding section is that very few countries start from this position. Instead, a program of strict fiscal control and progressive deficit reduction was shown to be closely correlated with successful transition. Deficit reduction by itself, however, will not prevent such pressures from emerging at a later stage, leading the authorities to revert to direct methods of monetary control (the U.K. experience).
Additional measures are therefore desirable, including the setting of strict limits on monetary financing of the government’s fiscal deficit by the central bank.45 A comprehensive program for public debt management can facilitate the observance of those limits. Such a program would involve widening the range of debt instruments and holders, adopting market-based selling techniques, and strengthening secondary-market arrangements and coordination with monetary management (see Sundararajan and others, 1994). It is also necessary to recognize the potential conflict between the goals of debt management and monetary policy and to prepare to resolve possible conflicts in favor of monetary policy. Thus, there must be a willingness on the part of the government to refrain from pressuring the central bank to keep interest rates low to minimize fiscal costs. Many countries have been addressing these issues by increasing central bank independence.46
Strengthen and Integrate Money Markets
Because control by the central bank over the supply of reserve money is the fulcrum of indirect monetary control, such methods will not be fully effective unless the “market” for short-term bank liquidity (either an interbank or a money market) can signal and transmit the central bank’s actions rapidly and transparently to all market participants. Thus, fairly seamless money and interbank markets are essential to the full use of indirect instruments. The relatively developed state of these markets was an important factor in explaining the smooth transition in most of the industrial countries. Conversely, segmented markets and a relatively slow pace of development are important factors in explaining why comparatively few nonindustrial countries have yet made the transition from open market type to full open market operations (the principal exceptions among the nonindustrial countries in the sample being Chile, Indonesia, and Mexico).
There are two key aspects to market development. First, measures to improve market infrastructure must be implemented at an early stage. As shown by the case of Poland, reform of the payments system to streamline the clearance and settlement of financial transactions is essential. Other infrastructure that has to be developed includes an appropriate legal framework to permit securities trading (covering such issues as settlement procedures, collateral arrangements, trading rules, and the regulatory framework for securities markets) and suitable market instruments and techniques (such as banker’s acceptances, commercial paper, and repos). These conditions facilitate interbank transactions and active liquidity management.47 In addition, the recent emphasis given by many countries to refining their payments systems demonstrates that the development and steady evolution of market infrastructure must be ongoing concerns.
Second, the central bank has to play an active role in market development. Perhaps somewhat paradoxically, the introduction and experimental use of indirect instruments at an early stage—even if they cannot be fully effective—can be essential to developing the conditions necessary for their becoming effective monetary instruments. As shown by the case studies of Indonesia and Mexico, by adopting market-based indirect instruments and transacting at market-related interest rates at an early stage, central banks can play a catalytic role in developing financial markets in general, and money markets in particular.
Restructure the Banking System and Foster Competition
A healthy and competitive banking and financial system is also a key element in ensuring that central bank actions to control the supply of liquidity are fully and rapidly transmitted. If the commercial banks do not respond to the signals given by the central bank by altering interest rates or liquidity conditions, indirect instruments will not have the desired effect on monetary and credit conditions and hence on the final economic objectives.48 Commercial banks may respond sluggishly to changes in their reserve position because their management is not used to a commercial environment, because they are not subject to hard budget or liquidity constraints, or because their financial position is so bad that they are unable to respond. Some instruments may be more appropriate than others in such cases of financial system fragility. For instance, using treasury bill auctions to reduce excess liquidity would create fewer problems for banks than increasing reserve requirements, especially if the latter are unremunerated. Responses of weak financial institutions may even be perverse; for example, weakened financial institutions may bid up interest rates to extremely high levels as they are forced to seek high-risk, high-return investments. This points to the need for clearly defined sanctions on management and owners of improperly operated banks.
The use of indirect instruments can be hampered by monopolistic or collusive behavior. It is important, therefore, that, as part of the process of introducing indirect monetary instruments, the authorities encourage competition in the banking sector through measures such as privatizing state-owned banks, removing barriers to entry, deregulating the domestic sector, and opening the market to foreign banks.49 Banks should be subject to hard budget and liquidity constraints and be financially strong enough to operate commercially under the new monetary arrangements.
The case studies suggest that, without appropriate restructuring to deal with problem loans and problem banks, weaker segments of the banking sector may be unable to adapt to the newly competitive environment, raising the risk of a financial crisis.50 Existing weaknesses in banks’ asset positions become increasingly difficult to manage as the economy becomes liberalized and debtors can no longer accumulate arrears. For example, banks holding old government paper issued at low rates may face capital losses when interest rates increase. The imposition of hard budget constraints and the ending of interest rate subsidies, which frequently accompany financial liberalization and the introduction of indirect monetary instruments, put pressure on banks’ financial positions.
Adapt Supervisory and Regulatory Framework to Market Conditions
In an environment of liberalized interest rates and unrestricted credit allocation, the ongoing solvency of particular financial institutions hinges on the ability of those institutions to manage new credit and market risks. Safeguards—in the form of minimum capital standards, provisioning for doubtful loans, limits on loan concentration, collateral requirements, collateral valuation standards, and adequate enforcement mechanisms—are needed to foster prudent behavior by financial institutions. Financial reporting and disclosure standards are needed to guarantee transparency in the operations of financial institutions and provide a basis for the public to assess the creditworthiness of particular financial institutions. For instance, Chile and New Zealand have strengthened requirements on information disclosure, with a view to facilitating the role of the market in assessing the situation of individual banks.
The too frequent experience has been that financial liberalization—in the absence of such measures—leads to financial crisis and subsequent reversion to direct methods of monetary control. Some experiences (such as Chile and Indonesia) suggest that these problems may surface long after the transition to indirect instruments has occurred.
Bolster the Technical Capacity of the Central Bank
Regardless of the instruments they use, central banks need to build up their technical capacity to maintain monetary control in an increasingly sophisticated financial world. Reliance on indirect instruments requires that the central bank have the capacity to project the demand and supply of reserves and their effect on broader credit and monetary aggregates; it also assumes that the central bank has the legal capacity to utilize indirect instruments, which may require changes in central bank legislation. Thus, the central bank will need a programming framework and some idea of the money multiplier relationship to estimate how much reserve money to add or withdraw to achieve the required effect on broader money and credit aggregates. This can be particularly difficult during the transition period when several of the key behavioral relationships tend to become unstable, at least temporarily, thereby greatly diminishing the information content of past observations. In those circumstances, central banks have to adjust their implementation strategies and tactics accordingly. The case of New Zealand, discussed in more detail in Part II, illustrates this point.
As part of their programming framework, central banks need to develop a framework for managing the liquidity they provide to the market in order to ensure that the short-run instrument setting is consistent with the policy objectives. Specifically, such a program provides the central bank with indications about the timing and the size of its interventions, which helps in making indirect instruments of monetary policy most effective.
To be useful, a reserve money program requires timely and accurate data on the central bank balance sheet and on financial sector developments. Timely and accurate reporting, as well as a short-term information system to provide early indications of monetary and interest rate developments, is helpful in implementing a system of indirect instruments.51 (See, for instance, the case study on Poland in Part II.)
Although market development will require further refining, the basic elements of an indirect monetary instrument strategy, frequently involving open market type operations, can be introduced fairly quickly as suggested by the experience of countries in the sample.
Pace of Transition
Chart 1 summarizes the experience of the sample countries in making the transition from direct to indirect instruments of monetary policy. It shows clearly that the sample divides almost equally into two distinct camps: those that were able to make the transition within one or two years and those that took much more time (many on the order of five or more years). For the sample as a whole, the average length of transition was close to four years, and the median was three years. If the industrial countries are included in the sample, the evidence suggests a clear tendency for a lengthy transition.
Chart 1.Transition to Indirect Monetary Instruments
1The average length of transition is 3.7 years.
While there are examples of both industrial and nonindustrial countries that have made successful transitions without subsequent reversals or major difficulties (Israel and New Zealand), the experience summarized in the preceding section suggests that a gradual pace makes a transition smoother and reversals less likely. The clear preference for gradualism is likely linked to the fact that such an approach provides more time for the necessary concomitant measures. It takes time to develop the infrastructure requirements (legal and regulatory framework, payments systems, and so forth), while the market-oriented nature of indirect instruments implies that the requisite development of financial markets and financial institutions may also take time. The central bank, too, must refine its operational capacity, often in circumstances where it is competing with the evolving financial sector to retain highly skilled personnel. Moreover, time is often needed to contain large fiscal and quasi-fiscal imbalances that constrain the operational autonomy of the central bank. Indeed, when the lengthy list of conditions needed to support full reliance on indirect instruments is considered, the case for gradual transition is strengthened.
However, circumstances do not always permit a country to undertake reforms at a gradual pace. Macroeconomic instability may create an urgent need for new instruments to strengthen monetary control to a level not achievable using direct instruments except at extreme cost (measured in terms of a loss of efficiency in the allocation of economic resources). For instance, direct instruments may have become ineffective or made obsolete by financial innovation or greater openness to international capital flows.
In these cases, the experiences of the sample countries show that rapid transitions can succeed, provided that the approach is sufficiently comprehensive and encompasses reforms in prudential regulation, payments, clearing and settlement systems, accounting and reporting systems, and the legal framework for financial transactions.52 In this regard, early action in the area of bank supervision and regulation is critical to control the risk of financial crisis, which has often precipitated reversals.
Although country experiences vary in regard to the specific instruments used in making the transition from direct to indirect instruments, certain broad patterns in the sample suggest the following stylized path for introducing indirect instruments of monetary control.53 The various concomitant measures to make the indirect instruments effective have to be introduced in parallel, taking into account their impact on monetary control and market development, as well as technical linkages among individual measures (Sundararajan, 1992).
Typically, a country introducing indirect instruments will need to absorb a liquidity overhang as a first step. However, even in the absence of an overhang, a central bank will need an instrument to absorb liquidity at its own initiative. Reserve requirements may be the only way to fulfill both needs at this stage of the transition and should be implemented as needed.54 By increasing the reserve requirements, a central bank can ensure the required sterilization. If the sterilization is attempted using other means alone—such as sales of securities—which banks may be unwilling to buy if markets are underdeveloped, it may be less certain. Moreover, reductions in reserve requirements can be used to inject funds in the market.
In addition, and particularly in cases of underdeveloped interbank markets, the central bank will need a mechanism to provide temporary accommodation (as lender of last resort) to individual banks having difficulties meeting their interbank settlement obligations. Thus, an overdraft or Lombard facility is likely to be needed as well.
The central bank will require a mechanism to provide funds to the market. This is particularly the case for economies in transition, in which commercial banks usually depend on the central bank for substantial amounts of financing.55 The central bank needs a credit facility to supply liquidity at its own initiative and to provide for the secular growth in domestic credit. In the initial stages, credit auctions can accomplish that purpose, as they provide a way to inject desired amounts of funds in a market-determined way. As noted earlier, though, this instrument needs to be designed carefully, to minimize adverse selection and distress borrowing.56
At this point, the central bank should begin to eliminate interest rate controls and initiate technical and legal preparations that will foster money and government securities markets and strengthen competition in banking. While some countries have maintained certain controls (such as maximum or minimum interest rates on some loans or deposits, and ceilings on spreads), there will be little room for indirect instruments to operate unless all short-term interest rates have been freed and banks have some scope to determine deposit and lending rates.
At the second stage, the authorities should introduce open market type operations in the form of auctions of short-term government or central bank securities and bring about a greater degree of interest rate flexibility. Sometimes, central banks will prefer to issue their own securities. Among the reasons for such a preference would be the government’s inability or unwillingness to issue the types of debt needed for monetary management and the risks of conflict between monetary policy and public debt management.
The issuance of primary debt has proved to be a powerful instrument that allows central banks to gain many of the advantages of open market operations even before their secondary markets are sufficiently developed for the central bank to operate in them. The issuance of government or central bank paper also provides a strong impetus to the development of secondary markets, which then have a basically risk-free paper to trade and a basis for pricing it. Moreover, that paper can also serve to underpin other market operations, such as repos, or serve as collateral.
The authorities should begin to reduce reserve requirements in the second stage, as operations with securities can serve to sterilize any excess liquidity that might result from those reductions. At this stage, auctions of government securities (or central bank securities) will often have to be combined with other instruments—such as short-term credit auctions, a Lombard window, or bill rediscounting—to facilitate both monetary control and money market developments.
In the third stage, the authorities should press ahead with the reforms to develop the money market further, in particular the secondary market for securities; this will provide the central bank with the opportunity to operate flexibly and continuously and will require further adaptations and refinements in instruments. However, experience shows that this process could take considerable time. Progress depends on how quickly the institutions, instruments, and infrastructure of the market are developed. The central bank can help in accelerating the necessary reforms.
As shown by the wide variety of experiences summarized in the case studies, the sequencing outlined above is anything but rigid. Variations are both inevitable and highly desirable. For example, depending on the degree of market development, a country may be able to undertake the first and second stages simultaneously. Other variations will depend on factors such as the degree of openness of the capital account and whether the government has a need to issue securities.
The question arises as to when a country should abolish direct instruments and rely solely on indirect instruments. No general answer can be given. Widespread interest rate controls, if binding, allow little room for the development of indirect instruments and need to be eliminated early in the process of reform. While it would be unwise to introduce bank-by-bank credit ceilings (if they were not already being used), countries may choose to maintain such ceilings temporarily while they develop indirect instruments. Such a choice could be appropriate, for instance, if the volatility of the demand for reserve money made it difficult to control monetary conditions solely through reserve money management. In those cases, ceilings can be made gradually less binding until they are phased out completely.
More generally, however, it may be useful to adopt a “belt and braces” approach: although usually only one set of instruments is binding at a time, direct and indirect instruments can coexist for a period until the financial markets are sufficiently developed to ensure the effectiveness of indirect instruments. Thus, it is important to stress that a country need not initially rely solely on indirect instruments. In fact, the transition can involve introducing indirect instruments even before the banking and financial markets are developed to the point where those instruments can be fully effective, because, as observed earlier, the use of these instruments can powerfully contribute to market development.57 Therefore, the case for a gradual transition to full reliance on indirect instruments of monetary control should not be interpreted as a case for delaying the process of transition. Indeed, the earlier the process begins, the more time, and hence discretion and flexibility, will be available to make the full transition. The value of this approach is that it allows banks (and the central bank) to familiarize themselves with the operational modalities of market-based instruments.
Many central banks, including those in countries with developed banking systems, have used moral suasion as a supplement to other instruments. It is hard to judge the advantages and disadvantages of this technique, since it can take many forms, sometimes facilitating and at other times preventing market reactions. Insofar as it represents an interference with the normal functioning of markets—by encouraging banks to carry out actions that go contrary to normal market incentives—central banks should eschew this technique once other instruments are in place.