Frameworks for Monetary Stability

15 Money Versus Credit: The Role of Banks in the Monetary Transmission Process

Carlo Cottarelli, and Tomás Baliño
Published Date:
December 1994
  • ShareShare
Show Summary Details

A striking feature of many industrial economies during the past decade or so has been the marked cycle in credit markets. Beginning in the early to mid-1980s, many countries experienced rapid credit growth and debt accumulation, accompanied by a boom in asset markets, strong aggregate demand, and inflationary pressures. These conditions were sharply reversed around the turn of the decade as credit extension declined, asset prices tumbled, and economic activity stagnated. Furthermore, the balance sheet positions of households and firms deteriorated, banking systems in a number of countries came under severe stress, and concerns arose about the available supply of credit. These developments combined with changing financial market structures and practices to focus attention on the role of credit markets in the transmission of monetary policy actions to the economy at large, a role that for many years had been neglected as emphasis centered on the monetary aggregates.

The renewed interest in the macroeconomic role of credit, and especially in bank credit, in recent years has also stemmed from a number of institutional, empirical, and theoretical developments. These include: (1) financial innovations and deregulation that caused unpredictable shifts in the monetary aggregates and led to the abandonment of monetary targets in a number of countries, and generally to the downgrading of the importance assigned to monetary aggregates in the monetary policy process; (2) empirical evidence that changes in the availability of credit are associated with changes in the level of economic activity; and (3) advances in the theory of imperfect information applied to financial markets. This theory explains the role of banks in dealing with asymmetric information between borrower and lender, the allocation of credit, and why credit availability is likely to be important.1

This paper discusses the implications for the conduct of monetary policy of alternative views of the monetary transmission mechanism, with emphasis on the role of banks. Two issues are of particular concern: (1) whether the behavior of banks provides a distinct channel of transmission of monetary policy to aggregate real activity, additional to the normal liquidity effects stemming from the market for bank deposits; and (2) whether independent credit market disturbances and disruptions of the normal credit creation process can have important macroeconomic consequences for reasons separate from any effects they may have on the money supply. Much of the paper is related to the macroeconomic stabilization role of monetary management, but it also touches on what has historically been a primary responsibility of central banks—namely, ensuring the safety and soundness of the banking system.2

The paper is organized as follows. It begins by characterizing the monetary policy process in terms of instruments and targets of the central bank, and draws the distinction between an intermediate target approach and an indicator variable approach in the conduct of monetary policy. It then distinguishes between the “money view” and the “credit (or lending) view” of the monetary transmission mechanism. The focus is on financial market imperfections and the role of banks. It goes on to discuss the implications for the conduct of monetary policy of a separate credit channel in the transmission mechanism and of independent credit market disturbances when information is imperfect and banks are “special.” The empirical evidence bearing on the money and credit views is referenced in the relevant sections. Finally, the paper offers concluding remarks.

Instruments and Targets of Monetary Policy

The monetary transmission mechanism describes the channels through which monetary policy actions are transmitted to the ultimate objectives of policy. Policymakers’ views of these linkages influence the variables on which they will concentrate in formulating monetary policy and in assessing the stance of policy. Two alternative views of the transmission mechanism are presented below. Before doing so, however, it is useful to define some terms by sketching out the monetary policy process.

At its most general level, the monetary policy process may be characterized schematically as follows:

The instruments are prices or quantities under the direct and immediate control of the central bank—i.e., the components of its balance sheet and discount or official rates3—which the monetary authorities adjust so as to achieve their ultimate objectives. The latter can be defined broadly as the arguments appearing in the policymaker’s objective function, usually some macroeconomic aspects of economic activity. However, given monetary policy’s limited influence on real variables over the long run, but its dominant influence on the price level, the central macroeconomic goal of monetary policy is generally taken to be the maintenance of price stability over time.

Monetary authorities typically do not link their policy instruments directly to their ultimate objectives because the effects of policy actions are delayed and distributed over time and because there is uncertainty about the impact of policy actions on the ultimate goal. Instead, they often adopt one of two approaches: an intermediate target procedure, or an indicator or information variable procedure. Both are two-stage policymaking processes that involve intermediate steps between a central bank’s operations and its ultimate objectives. In the intermediate target procedure, a time path for some variable that is expected to lead to the achievement of the ultimate target is specified in the first stage. Then, in the second stage the instruments are set so as to keep the intermediate target on its specified path.

Depending on how rapidly and how directly changes in instrument settings affect the intermediate target, central banks may find it useful or necessary to adopt some variable as an operating target to guide their daily operations and some other variable as an intermediate target. For instance, the effects of central bank actions are reflected almost immediately on short-term interest rates and exchange rates, whereas the effects on monetary and credit aggregates are spread over time. The operating target is usually a very short-term interest rate, while a nominal variable such as a monetary or credit aggregate, nominal income, or the exchange rate is employed as an intermediate target in order to anchor the price level.4

It is not always feasible or appropriate to direct policy instruments toward the achievement of an intermediate target. This is only so if the intermediate variable bears a stable and predictable relationship to the ultimate target of policy. However, as the experience with monetary targeting has shown, such relationships may be subject to unpredictable shifts. The dynamic considerations that motivate the adoption of intermediate targets—namely, the behavioral and economic lags and the imperfect knowledge of the transmission of monetary policy actions—are nevertheless still operative. In these instances intermediate variables such as monetary or credit aggregates continue to play an important role in the policy process, but are used in a different manner: not as variables to be controlled, but rather as indicator variables. As such, they provide information about the current and prospective level of economic activity and serve to guide the setting of policy instruments so as to best achieve the policy goal.5

As is discussed below, what distinguishes the money and credit views is not differences regarding the objectives of policy or the instruments employed by the monetary authorities, but rather differences regarding the channels through which changes in the instruments affect real activity. This has implications for the two-stage policymaking procedure described above and for other aspects of the conduct of monetary policy.

Money and Credit in the Transmission of Monetary Policy

In the standard (“money”) view of the monetary transmission mechanism, the supply and demand for money determines the short-term money market interest rate, which feeds to other market interest rates and affects spending and output.6 Monetary policy works through the liabilities side of banks’ balance sheets, that is, by affecting bank deposits. Two conditions must be met: (1) the central bank must be able, by changing the quantity of available reserves to the banking system, to affect the supply of bank deposits; and (2) there are no perfect substitutes for money, so that agents are not able to offset perfectly the change in the quantity of money by switching into alternative financial instruments. If money substitutes are available, the impact of monetary policy is dampened, with the response of the short-term interest rate to a change in the money supply being smaller the more closely other financial instruments can substitute for money.7

In the money view there are effectively only two classes of assets: money and all other assets (“bonds”). Bank credit is indistinguishable from other nonbank financial assets. The unique character of banks is thus confined to the liability side of their balance sheets—that is, to their capacity to create money by issuing demand deposits.

The essence of the “credit view,” however, is that bank credit is not a perfect substitute for other forms of borrowing. A large group of borrowers, particularly households and small and medium-sized businesses, are very dependent on bank credit because it is impossible, or very costly, for them to borrow directly by issuing securities on the open market. For these borrowers bank loans have no close substitutes. The reason is that banks do not simply intermediate between savers and investors; they also deal with the problems of incomplete or asymmetric information in the credit market. By specializing in the evaluation and screening of loan applicants and in the monitoring of loan performance, banks can create credit more efficiently than individual agents or other institutions can.8 This enables them to bring together ultimate borrowers and lenders at a lower cost than the latter can achieve in direct transactions themselves, and to extend credit to borrowers who cannot obtain credit in the open market (bank loans are “special”). The “specialness” of banks provides an additional, independent channel of monetary transmission: the credit channel. Monetary policy works not only by affecting bank liabilities, but also by affecting bank assets.

Analogously to the requirements for the conventional money view to operate, two conditions must be met for the credit channel to work: (1) loans and securities must be imperfect substitutes as bank assets, so that monetary policy actions directly affect the supply of bank loans; and (2) bank loans and nonbank credit must be imperfect substitutes as corporate and household liabilities, so that a reduction in the availability of bank credit reduces real economic activity by constraining the spending of bank-dependent borrowers. The first condition implies that in response to a monetary-policy-induced change in the size of their balance sheet, banks do not simply adjust their holdings of securities and leave their loan supply unchanged, while the second condition implies that borrowers cannot offset without cost a reduction in the supply of bank loans by resorting to other sources of financing or using internal funds. Another implication of the imperfect substitutability of bank and nonbank credit is that bank loan rates and other borrowing rates may diverge. Furthermore, and importantly for the transmission mechanism, this implies that the spread between money market interest rates and bank loan rates may change in response to a monetary policy change.

If bank loans are “special,” in the sense defined above, then monetary policy may be more potent than the conventional money view suggests. Consider the case of a tightening of monetary policy. In this case, a reduction in banks’ deposits must be matched by a decline in banks’ holdings of loans and securities. To the extent that banks reduce their loans, the spending of bank-dependent borrowers declines. The decline in investment and aggregate demand will be greater than can be accounted for by the money channel alone because borrowers cannot fully offset the reduction in bank lending by nonbank borrowing as is assumed in the money-only view.9

The reduced supply of loanable funds may be allocated among borrowers through a market-clearing rise in interest rates on bank loans or through nonprice rationing, such as more stringent requirements on collateral, the size and maturity of loans, loan commitments, or the denying of loans to some classes of borrowers.10 At some interest rate levels, a rise in market interest rates caused by a tightening in policy may induce a further rise in loan rates. The increases in the margin between loan rates and market interest rates are risk premiums that reflect increased probabilities of default as economic activity weakens and as higher loan rates increase borrowers’ total debt burdens. Banks increase loan rates by more than is required to cover the higher cost of funds resulting from higher market interest rates.11 However, banks might not raise loan rates sufficiently to clear the market; the interest rate that equates supply and demand in the loan market may be greater than the interest rate that maximizes expected bank revenue, so that there will be an excess demand for loanable funds and banks will deny loans to some loan applicants.

To elaborate, a bank’s expected revenue from a loan is the product of the capitalized amount of the loan and the probability of repayment. The probability of default (or non-repayment) is, in turn, an increasing function of the real loan rate. As the loan rate rises from relatively low levels, the bank’s expected returns increase and it increases its loan supply. However, the supply of loans does not rise monotonically with the rate of interest because the probability of default increases and marginal expected revenue falls. At some interest rate level, expected revenue peaks and a higher interest rate causes expected revenue to fall as the increase in interest receipts is more than offset by the anticipated increase in defaults.12 Beyond some level, then, banks maximize profits by denying loans to some potential borrowers rather than by raising loan rates, so that at the prevailing loan rate there will be an excess demand for loanable funds. Two other points should be noted: (1) a credit rationing equilibrium exists in that the interest rate at which the demand for loanable funds equals the supply of loanable funds is not an equilibrium interest rate because a bank could increase its profits by charging a lower interest rate; and (2) the rejected loan applicants may seem indistinguishable from those who receive loans and if a larger supply of credit became available they would receive loans.

This equilibrium credit rationing occurs because the interest rate charged by a bank may itself affect the riskiness of the loan pool in two ways: (1) by raising the average “riskiness” of those who borrow, as the more risk-averse investors drop out of the loan market while the interest rate rises (an adverse selection effect); and (2) by changing the behavior of the borrower, as a higher interest rate induces a borrower to undertake riskier projects with potentially higher payoffs but with lower probabilities of success (an adverse incentive or moral hazard effect).13 It should be stressed, however, that banks need not ration credit in order for the credit channel to exist; all that is necessary is that bank credit and other forms of credit be imperfect substitutes.14

The credit view outlined above requires that the central bank be able to influence the flow of bank credit directly—the first of the two conditions noted earlier. Much of the debate on money versus credit has centered on this condition, but it is not necessary in order for credit market imperfections to be relevant to the transmission mechanism. A broader credit view emphasizes the role of credit market imperfections in the propagation of monetary and real shocks. The focus here is not on the impulse to the economy of the direct effect of central bank actions on the supply of loanable funds. Rather, the focus is on the propagation of the effects of monetary policy as credit market frictions may magnify the impact of economic disturbances on borrowers’spending.

According to this broader credit view, informational asymmetries between borrowers and lenders may introduce a wedge between the cost of (uncollateralized) external funds and the cost of internal funds. There is a premium for external funds that reflects the deadweight losses (“agency costs”) involved in optimal financial contracts under imperfect information, relative to the full-information equilibrium. The premium could reflect compensation to the lender for costs of evaluation and monitoring, or it could reflect a “lemons” premium when the quality of the loan is unobservable. The size of the premium for external funds depends inversely on a borrower’s creditworthiness (collateralizable net worth relative to the size of the loan), or equivalently, on the condition of firm and household balance sheets. Thus the stronger the potential borrower’s balance (i.e., the greater the level of net worth), the lower will be the premium for external funds. The borrower’s creditworthiness, in turn, depends on economic conditions; balance sheets are likely to be stronger in good times and weaker in bad times. This means that the cost of external funds will fall in booms, expanding investment demand and strengthening the upturn; conversely, in recessions the cost of external funds will rise, depressing investment spending and magnifying the downturn. Economic disturbances can thus raise or lower the real cost of borrowing in a way that magnifies the overall impact of the shock.15 This would include a monetary-policy-induced change in the open market interest rate. For instance, a tightening of monetary policy that raises open market interest rates tends to lower asset values and worsen a firm’s liquidity. The consequent deterioration in balance sheet positions will, in turn, tend to decrease the availability of internal funds and raise the cost of external funds, raising borrowers’ overall price of funds and thus amplifying the effect of changes in market interest rates on the cost of capital. Credit market imperfections thus give rise to a kind of “financial accelerator” effect on borrowers’spending.16

To conclude this discussion, the conditions necessary for the credit channel to exist are analogous to those for the money channel to exist, while the broader credit view introduces a potentially important interaction between real and financial factors in the operation of monetary policy. On theoretical grounds alone, then, there is no reason not to pay attention to the lending channel in the formulation and conduct of monetary policy, whereas not to do so may neglect a significant dimension of the monetary transmission mechanism. The extent to which central banks should do so, however, and how they should do so, will depend on the presumed strength of the lending channel, on how quantitatively important it is. This issue is addressed below in conjunction with some of the implications for policy of the credit view.

Implications of Credit Market Imperfections

The existence of credit market imperfections of the type described above has a number of implications for the analysis of economic developments and the conduct of policy. These range from the choice of indicator variables and the measurement of monetary conditions, to the development of market institutions, regulatory reform, and the economic performance of economies undergoing transformation.

Intermediate Policy Targets

If the monetary transmission mechanism is best characterized as operating through both a money channel and a credit channel, then credit could in principle serve as an intermediate target for monetary policy. Whether it should do so would depend on whether it satisfies the main criteria determining the suitability of a financial aggregate as an intermediate target, namely: (1) that it be closely and reliably related to the ultimate target of policy; (2) that it contain information about future movements of the goal variable; and (3) that it be closely related to the operating targets or the instruments over which the central bank can exert direct control. Since there can be only one explicit intermediate target at which to aim policy, the issue often reduces to which specific monetary or credit aggregate to target.17 This involves a comparison of the relative stability of the money-income and credit-income relationships, the relative importance of money-demand and credit-demand shocks, and the relative degree of control that the central bank can exert over credit aggregates and monetary aggregates.

These are empirical issues and as such may differ across countries and over time, and according to the choice of aggregates. In the literature many examples can be found of some particular monetary or credit aggregate being more highly correlated with income than alternative aggregates, or of credit-demand shocks being relatively bigger (or smaller) than money-demand shocks. Almost invariably, however, the estimated relationships have shifted over time, as changes in transactions technology, the introduction of new financial products, and changes in financial market structures have altered historical relationships. What has generally been particularly difficult to capture empirically has been the short-run dynamics of the demand for money and credit. Nonetheless, for many countries there is considerable evidence supporting the existence of a long-run relationship between the levels of real money balances, real income, and interest rates.

Although a priori there is little basis for the adoption of a monetary policy procedure that focuses exclusively on specific monetary aggregates, to the neglect of credit measures, in practice countries that have adopted intermediate targets generally have focused on monetary aggregates. This may be because the degree of control that the central bank can exercise over credit, including bank credit, in the short term is believed to be inadequate, or at least inferior to that over the (narrower) monetary aggregates.18 It may also reflect the finding that it is often the broader credit measures that are most closely related to economic activity and, moreover, that the monetary aggregates tend to be better predictors of economic activity—that is, the monetary aggregates (especially the narrower aggregates) tend to lead output whereas the credit aggregates tend to move roughly contemporaneously with output.19 The evidence on timing should not be interpreted as necessarily invalidating the credit view. The slower adjustment of credit to policy changes may reflect the fact that bank loans are quasi-contractual commitments, so that banks first adjust their security holdings but over time, as they rebalance their portfolios, they also adjust their loans. It could also reflect portfolio adjustment by businesses that view their assets and liabilities as buffer stocks of liquidity.20

Credit as an Indicator Variable

Even if credit does not serve as an intermediate target, it can still be an important indicator variable. A major function of an intermediate target is to serve as an explicit nominal anchor to prevent cumulative one-way policy errors and thus ensure that policy actions do not take the price level off course in the long run. However, the experience in operating monetary policy in the 1980s led a number of countries with floating exchange rates to the conclusion that there is no financial aggregate that can durably serve as an intermediate target.21 Nevertheless there remains a need for clearly expressed objectives for the conduct of monetary policy so as to anchor expectations by removing uncertainty in markets on how the monetary authorities will react to unforeseen contingencies, to remove a potential inflation bias by providing the monetary authorities with an added measure of self-discipline, and to hold the monetary authorities accountable. This need has led to attempts to establish a monetary policy framework that can combine the credibility and disciplinary benefits of rules with the flexibility, in an evolving financial environment, to base decisions on varied sources of information rather than on a single variable.22

The approach followed by some central banks has been to adjust instrument settings in response to expected future deviations of the goal variable (typically the inflation rate) from its target value. This approach is more demanding in terms of information than the intermediate target approach. In making judgments about the policy actions necessary to achieve the target path for the goal variable, the central bank needs information on how current economic trends are likely to affect the future direction of the economy, as well as information on how its own actions are likely to influence the projected outcome. In this respect, central banks have found it helpful to monitor a range of indicators that have potential predictive content. A range of indicators is preferable because no single indicator can adequately summarize the monetary stance, and because each of the indicators is subject to influences other than monetary policy so that it will be easier to identify the nature of economic disturbances by observing several indicators. Of course, the interpretation of the implications of movements in the various indicators is based on the understanding of the operation of the monetary transmission channels and the nature of the shocks faced by the economy and the financial system. As the “noise” around various indicators changes or as the transmission channels change in conjunction with changes in financial market structures and practices, the interpretation of the indicators and the importance assigned to them may have to change.

A focus on credit may be especially relevant during periods when financial market innovations or deregulation have a large impact on the monetary aggregates (e.g., via the introduction of money substitutes or the shift of deposits out of banks and into nonbank financial intermediaries). The opposite would be true when credit substitutes become available, or other changes reduce the central bank’s influence on the volume of lending. Attention to credit may also allow the early detection of credit market shocks, such as a changed willingness of banks to extend loans induced by a change in their perception of the risk associated with making loans or in their net worth, that too narrow a focus on monetary aggregates may be slow to signal. For example, when—for whatever reason—banks do not have the capital to support additional lending, their reluctance to issue managed liabilities may spuriously contract the broad monetary aggregates, overstating the degree of monetary tightness.

Measurement of Monetary Conditions

In implementing monetary policy in the absence of a formal intermediate target, the policy instruments must be set so that the thrust of policy will be consistent with the attainment of the ultimate goal (price stability). This requires an assessment of monetary conditions. Money market interest rates are usually taken as an important indicator of monetary conditions. But if monetary policy is transmitted through both a money channel and a credit channel, then money market interest rates may provide only a partial picture of monetary conditions, since it would omit the effect on private sector spending decisions of changes in the spread between loan rates and money market rates. Indeed, a number of studies have found that the relative availability of bank lending, as measured either by relative prices (interest rate spreads) or relative quantities (the ratio of bank loans outstanding to nonbank financing), has considerable information about future movements in investment and output.23

Interest rates, including loan rates, however, may not be a reliable indicator of the impact of financial variables on aggregate demand when credit is rationed; bank balance sheet variables (the quantity of credit extended to various sectors) may be more appropriate indicators. Whether or not equilibrium credit rationing is an empirically important phenomenon is difficult to assess because of the lack of data on the terms of individual bank loans. The evidence bearing on the issue has been almost exclusively generated using aggregate data and confined largely to testing a central implication of credit rationing, namely, that bank loan rates are “sticky” While a number of studies have reported findings consistent with the existence of equilibrium credit rationing, they do not exclude alternative interpretations.24

Richer Analysis

Consideration of the role of banks in the monetary transmission process allows a richer analysis and understanding of economic developments. In particular, shifts in credit market sentiments have implications for output, inflation, and exchange rates that cannot be fully explained by the conventional money view. An interesting case of relevance to a number of countries is the implication for the behavior of the real exchange rate of the swing in credit growth in the 1980s-1990s.

Consider an expansionary credit market shock, say an increased willingness of banks to extend loans, which narrows the spread between the interest rate on loans and on bonds or which results in a loosening of lending standards. The increase in lending stimulates demand and exerts upward pressure on interest rates and the exchange rate. All else remaining unchanged, the excess demand can be eliminated by an appreciation of the real exchange rate, either through an appreciation of the nominal exchange rate or through an increase in the domestic, relative to the foreign, rate of inflation. However, if the nominal exchange rate is fixed, the stimulative effect of the credit shock is greater since the bond interest rate is tied by the exchange rate constraint; there will be an additional monetary expansion in order to keep the interest rate fixed.

If the credit shock is reversed and banks become less willing to extend loans, the effects on demand and on the exchange rate will be magnified in the opposite direction. Credit market swings can thus result in sizable real exchange rate cycles and complicate the task of pegging the exchange rate.25

Asset Prices and the Credit Cycle

Recent cyclical experience in the industrial countries, most notably in the United States, Japan, and some Scandinavian countries, has given new impetus to a hypothesis regarding the macroeconomic role of credit earlier advanced by Irving Fisher and Hyman Minsky. According to this hypothesis, the workings of the credit market can magnify and perpetuate economic cycles of boom and bust. In the less extreme forms observed most recently, wide fluctuations in credit and asset prices, and the associated balance sheet adjustments, can exacerbate business cycles and complicate the achievement of monetary policy objectives. In a cyclical upswing, excessive credit expansion can support asset speculation and permit overleveraging of balance sheets. Subsequently, falling asset prices can generate debt deflation, while actions to reduce debt buildups and overhangs associated with overleveraging can curtail spending. In the most recent cyclical experience, booming asset prices (primarily for real estate and equities) have been closely related with the rapid expansion of credit that has followed the major deregulation of the financial system (e.g., in Norway, Sweden, and Finland) or the increase in competition between banks and nonbank financial institutions in a substantially deregulated environment (e.g., the United States).26 The close association between movements in asset prices and aggregate credit can be taken as prima facie evidence that credit “matters.”

Recent discussions of the monetary policy aspects of the credit cycle have focused primarily on macroeconomic issues.27 According to these studies, failure to account for the credit channel of monetary transmission can result in biased estimates of the effect and speed of transmission of a given monetary policy setting. Thus, the freeing of the credit channel through deregulation and enhanced competition is cited as a factor that has changed the information needs of central banks for conducting an effective monetary policy. The existence of the credit channel is generally viewed as a factor that has contributed to destabilizing the monetary aggregates and making them less useful as either an intermediate target or an information variable, to changing the degree of tightness or ease associated with a given interest rate setting, and to creating a need to focus more closely on credit developments and movements in asset prices. Movements in the latter in particular have been interpreted as a sign of impending inflationary pressures that are not captured in conventional price indices—which focus on prices of the flow of goods and services—and of the need to adjust monetary policy well before other signals are received. Indeed, these studies propose the construction of broader transactions-based price indices to be used, alongside standard indicators of inflation in goods and factor markets, as a guide for conducting monetary policy.

The proposal rests on distinguishing analytically between inflationary pressures and the rate of increase in the price index of a particular set of transactions. In particular, it rests on the notion that to properly measure inflationary pressures, it is essential to take into account all types of transactions: stocks (of assets) and flows (of goods and services). The analytical distinction is important because under certain conditions excess liquidity and credit can be concentrated in asset markets, so that inflationary pressures manifest themselves first in asset prices and perhaps only later in the prices of the flow of goods and services. A monetary policy that focuses on the average price of a narrow subset of transactions, namely, conventional price indices, may thus be slow in detecting the buildup of inflationary pressures.

While the suggestion that monetary authorities should pay greater attention to asset price developments has obvious merits, it is not without difficulties in practice. When the monetary control problem is cast in an intermediate target or indicator framework, recommendation of a focus on asset prices (or credit flows) may provide insufficient guidance to the policymaker in devising policy instrument settings that are consistent with the ultimate objectives of monetary policy. In the absence of a price index for wealth, policymakers will continue to rely on measures of current consumption service prices (consumer price index) and current output prices (gross domestic product deflator) as measures of inflation. The usefulness of the recommendation thus depends on whether a sufficiently close relationship can be estimated between asset prices (the information variable) and inflation, conventionally measured (the ultimate objective), and between asset prices and the monetary policy instruments. Furthermore, since asset prices tend to be volatile, it is crucial that the authorities should be able to distinguish changes in asset prices that reflect relative price adjustments stemming from nonmonetary factors (portfolio shifts, aggregate supply shocks, and factors specific to asset markets) from those that reflect monetary conditions. However, the relationship between monetary policy and sharp movements in asset prices has not been firmly established empirically.28 There has been little empirical work in this regard to draw on. While some central banks take account of asset prices when establishing appropriate instrument settings, their approach appears to be informal and heuristic.

Finally, it should be noted that, given the evident differences in the responsiveness of asset prices and the general price level to credit conditions, adjusting monetary policy settings to movements in asset prices could pose difficult dilemmas for central banks, which would have to choose between possibly failing to restrain speculative behavior and possibly causing an undesired contraction of the real economy.29

Rather less attention has been paid to the microeconomic dimension of the credit channel and the credit cycle.30 However, from the perspective of the credit view of money, there is an important interrelationship that could be exploited for the conduct of monetary policy. Indeed, to a significant extent, the macroeconomic effects of the credit cycle can be traced to the flow of funds and microeconomic developments—to the willingness of both borrowers and lenders to overleverage their capital positions, and for lenders (1) to finance speculative asset purchases by permitting the purchased assets to back the loans at collateral values that too fully reflect current market prices, or (2) to permit undue regional and asset-specific concentration in their loan portfolios. Once asset prices begin to turn down—possibly because of a slowing in the business cycle, a change in tax or monetary policy—the macroeconomic problem is amplified by credit effects, as borrowers are forced to sell assets at distress prices to meet their cash-flow borrowing commitments, and lenders are forced to allocate capital to cover realized loan losses and establish suitable contingent loss provisions. Weak banks and other distressed financial institutions may then curtail lending to creditworthy borrowers (who, according to the credit view, may lack alternative sources of credit), while forced asset sales may further exacerbate the problem. In the end, the process may generate major insolvencies within the financial system, ultimately risking the efficient allocation of financial capital or even the sound working of the payments system.

The Microeconomic Dimension of Monetary Policy

What the foregoing scenario clearly suggests is that the magnitude of the business cycle (and the responsiveness to monetary policy) depends in part on the soundness and competitiveness of the banking and financial system, which in turn can be influenced by microeconomic measures that seek to establish an appropriate supervisory and regulatory framework. By establishing adequate supervision based on suitable rules regarding capital adequacy, collateral valuation, portfolio concentration, etc., it may well be possible to limit the credit-induced amplification of the business cycle, both by limiting credit excesses during the upswing and, as a result, some of their consequences during the downswing. The aims of soundness and competitiveness in the banking and financial system should be integral aspects of monetary policy. Yet, as illustrated by the cyclical experience during the 1980s and early 1990s and the credit view of monetary policy, weak or overleveraged financial institutions can distort the transmission of monetary policy. It follows that macroeconomic measures need to be reinforced by appropriate microeconomic measures. Three elements appear to be involved.

First, from the perspective of preventive measures, there is a need for transparency in financial transactions, to be effected through adequate and timely financial disclosure requirements, in order to promote financial discipline. Without the distorting effects of blanket depositor protection through unlimited deposit insurance schemes or the effects of widespread acceptance of the doctrine of “too big to fail,” depositors have an incentive to allocate funds only to healthy financial institutions, which in turn have an incentive to maintain their viability through prudent lending policies and adequate management and control of risk. Adequate financial disclosure is needed to support the competitive forces that promote market discipline.31 Another possible preventive measure could involve the strict separation of the deposit-taking and lending functions of banks (100 percent reserve requirement, or “narrow banking”). However, if banks are “special,” if the lending function of banks is macroeconomically important, then such reform proposals may have important unintended consequences for macroeconomic performance.

Second, effective prudential regulation—in the form of minimum capital standards, loan classification and provisioning norms, exposure limits, and so on—backed by strong supervision can help to curtail the lending excesses and the severity of the subsequent workout. Whereas some commentators cite the imposition of the Basle risk-weighted capital standards as a factor that forced banks to curtail lending and therefore exacerbated the recent credit crunch in U.S. credit markets,32 it needs to be recognized that a significant expansion of credit had already taken place before the capital standards were imposed. It may be contended, therefore, that if the capital adequacy measures had been in place at the beginning of the credit cycle, they might have contained the lending excesses that generated the subsequent deterioration in lending portfolios.33

Finally, there is an argument for close association between the banking and financial supervisor and the monetary authorities.34 Whereas the credit cycle has implications for the safety and soundness of the financial system, it is also the case that disruptions of normal banking activity or the malfunctioning (or nonexistence of intermediation) may have greater impact on demand and output than is predicted by the money-only view. Moreover, important nonlinearities may be involved. The classic example is the role of the disruption of the banking system in exacerbating and prolonging the Great Depression.35 Similar problems may emerge in countries where the financial system is not well developed. For example, it has been argued that in Eastern European countries the negative output effects associated with monetary contraction have been magnified by the underdeveloped state of credit markets in those countries.36 An implication is that a central bank cannot content itself merely with controlling the rate of monetary expansion, but at times may have to act aggressively to protect lending and other functions of banks.

Development of Markets and Institutions

In developing countries, firms tend to rely more heavily on bank lending than on securities markets as sources of funds. In many developing countries, the banking sector is the only organized capital market; equity markets are often poorly developed, making bank loans the dominant form of external corporate financing. Thus, policies that improve the efficiency of the banking system may have beneficial effects on resource allocation and growth. For example, the establishment of well-defined property rights (e.g., in land) can facilitate bank lending by providing sources of collateral. Similarly, in the area of legal institutions, an efficient judiciary that can deal with defaults quickly and at low cost, allowing the lender to seize and dispose of the collateral, also enhances the willingness of banks to lend.37

In recent years a number of countries have undertaken financial sector reforms, moving from a system in which monetary policy operates mainly through direct controls to a market-based system. The liberalization of interest rates, the abolition of credit ceilings and directed credit, and related reform measures change the monetary policy transmission mechanism from one based primarily on quantity rationing (”exogenous” credit rationing) to one based chiefly on prices, through the operation of market-determined interest rates.38 The relative importance of these two rationing mechanisms and of the money and credit channels will depend on the degree of interest rate liberalization, the financial market structure and practices that influence banks’ behavior toward credit risk, and the macroeconomic environment.

Furthermore, as operating procedures change and as markets develop, it is important to take into account the degree of control exerted by the central bank over the monetary aggregates and the supply of bank credit, and the extent to which financial sector development affects the availability of money substitutes and bank credit substitutes, as this will affect the channels of transmission of policy and the information content of financial aggregates. For instance, reductions in reserve requirements and increased securitization, as the financial sector develops, make it easier for banks to insulate their sources of funds from the effects of open market operations; while the liberalization of the deposit interest rate, the reduction of reserve requirements and other measures which lower the cost of banking intermediation, insofar as they encourage the growth of the banking sector at the expense of the informal sector, will tend to increase the effectiveness of monetary policy.39

Concluding Remarks

In conclusion it should be stressed that the credit view is not an alternative to the conventional money view of the monetary policy process—it is a supplement. Paying greater attention to the role of banks, not only in creating money but also in extending credit, allows for a fuller and richer analysis of a wide range of monetary policy issues, extending from the use of financial aggregates as intermediate target or information variables, the assessment of monetary conditions, financial market innovations and deregulation, and the implications of a move from the use of administrative controls to a market-based system of monetary control. But perhaps most significantly, it allows for an integration of the microeconomic and macroeconomic dimensions of monetary policy.


    Bank for International SettlementsSixty-third Annual Report (Basle: Bank for International SettlementsJune1993).

    BattenD. and others “The Instruments and Operating Procedures for Conducting Monetary Policy in the Group of Five Countries,”IMF Working Paper No. 89/57 (Washington: International Monetary Fund1989).

    BergerAllenN. and Gregory F.Udell“Some Evidence on the Empirical Significance of Credit Rationing,”Journal of Political EconomyVol. 100No. 5 (October1992) pp. 10471077.

    BernankeBenS.“On the Predictive Power of Interest Rates and Interest Rate Spreads,”Federal Reserve Bank of Boston New England Economic ReviewNovember/December1990 pp. 5168.

    BernankeBenS.“Credit in the Macroeconomy,”Federal Reserve Bank of New York Quarterly ReviewVol. 18No. 1 (Spring1993) pp. 5070.

    BernankeBenS. and AlanBlinder“Credit, Money, and Aggregate Demand,”American Economic Review Papers and ProceedingsVol. 78No. 2 (May1988) pp. 43539.

    BernankeBenS. and AlanBlinder“The Federal Funds Rate and the Channels of Monetary Transmission,”American Economic ReviewVol. 82No. 4 (September1992) pp. 90121.

    BernankeBenS. and MarkGertler“Agency Costs, Net Worth, and Business Fluctuations,”American Economic ReviewVol. 79No. 1 (March1989) pp. 1431.

    BernankeBenS. and AlanBlinder“Financial Fragility and Economic Performance,”Quarterly Journal of EconomicsVol. 105 (February1990) pp. 87114.

    BlanchardOlivierJ.“Why Does Money Affect Output? A Survey,”inHandbook of Monetary Economicsed. by Benjamin M.Friedman and FrankHahn (Amsterdam: North-Holland1990).

    Blundell-WignallAdrian and MarianneGizycki“Credit Supply and Demand in the Australian Economy,”Research Discussion Paper No. 9208 (Reserve Bank of AustraliaJuly1992).

    BrunnerKarl and Allan H.Meltzer“Money, Debt and Economic Activity,”Journal of Political EconomyVol. 80No. 5 (September/October1972) pp. 95177.

    BrunnerKarl and Allan H.Meltzer“Money and Credit in the Monetary Transmission Process,”American Economic Review Papers and ProceedingsVol. 78No. 2 (May1988) pp. 44651.

    ButtigloeneLuigi and GiovanniFerri“Monetary Policy Transmission via Lending Rates in Italy: Any Lessons from the Recent Experience?” (unpublished Rome: Banca d’Italia1994).

    CalomirisCharlesW.“Financial Factors in the Great Depression,”Journal of Economic PerspectivesVol. 7No. 2 (Spring1993) pp. 6185.

    CalomirisCharlesW. and CharlesM. Kahn“The Role of Demandable Debt in Structuring Optimal Banking Arrangements,”American Economic ReviewVol. 81No. 3 (June1991) pp. 497513.

    CalvoGuillermoA. and FabrizioCoricelli“Output Collapse in Eastern Europe: The Role of Credit,”IMF Working Paper No. 92/64 (Washington: International Monetary Fund1992).

    CalvoGuillermoA. and Manmohan S.Kumar“Money Demand, Bank Credit, and Economic Performance in Former Socialist Economies,”IMF Working Paper No. 94/3 (Washington: International Monetary Fund1994).

    CaramazzaFrancescoDougHostland and StephenPoloz“The Demand for Money and the Monetary Policy Process in Canada,”Journal of Policy ModelingVol. 12No. 2 (Summer1990) pp. 387426.

    CurrieDavidThe Finnish Economic Crisis: Analysis and Prescription (Helsinki: Bank of Finland1993).

    DaleSpencer and AndrewHaldane“Bank Behaviour and the Monetary Transmission Mechanism,”Bank of England Quarterly BulletinVol. 33No. 4 (November1993) pp. 47891.

    DiamondDouglasW.“Financial Intermediation and Delegated Monitoring,”Review of Economic StudiesVol. 5 (July1984) pp. 393414.

    FamaEugeneF.“What’s Different About Banks?,”Journal of Monetary EconomicsVol. 15 (January1985) pp. 2939.

    FamaEugeneF.“Contract Costs and Financing Decisions,”Journal of BusinessVol. 63No. 1Part 2 (January1990) pp. 57191.

    FarmerRogerE.A.“A New Theory of Aggregate Supply,”American Economic ReviewVol. 74No. 4 (December1984) pp. 92030.

    FilcWolfgang“Credit as Intermediate Target of Policy?,”inMonetary Theory and Monetary Policy: New Tracks for the 1990sed. by Stephen F.Frowen (New York: St. Martin’s Press1993).

    FriedmanBenjaminM.“The Roles of Money and Credit in Macroeconomic Analysis,”inMacroeconomics Prices and Quantities: Essays in Memory of Arthur M. Okuned. by JamesTobin (Washington: Brookings Institution1983).

    FriedmanBenjaminM.“Targets and Instruments of Monetary Policy,”inHandbook of Monetary Economicsed. by Benjamin M.Friedman and FrankHahn (Amsterdam: North-Holland1990).

    FriedmanBenjaminM.“Intermediate Targets Versus Information Variables as Operating Guides for Monetary Policy”(unpublished paper presented at an International Conference of De Nederlandsche Bank and Tilburg University on “A Framework for Monetary Stability”AmsterdamOctober1993).

    FriedmanBenjaminM. and KennethN. Kuttner“Money, Income, Prices, and Interest Rates,”American Economic ReviewVol. 82No. 3 (June1992) pp. 47292.

    GertlerMark“Financial Structure and Aggregate Economic Activity: An Overview,”Journal of Money Credit and BankingVol. 20No. 3Part 2 (August1988) pp. 55988.

    GertlerMark and SimonGilchrist“The Role of Credit Market Imperfections in the Monetary Transmission Mechanism: Arguments and Evidence,”Scandinavian Journal of EconomicsVol. 95No. 1 (1993) pp. 4364.

    GoodhartCharlesA.E.The Evolution of Central Banks (Cambridge, Massachusetts: MIT Press1988).

    GoodhartCharlesA.E.“The Objectives for, and Conduct of, Monetary Policy in the 1990s.”London Stock Exchange Financial Markets Group Special Paper No. 49 (London: London School of Economics and Political ScienceAugust1992).

    GreenspanAlan“No Single Regulator for Banks,”Wall Street JournalDecember 151993.

    GreenwaldBruce and JosephStiglitz“Information, Finance Constraints, and Economic Activity,”inFinance Constraints Expectations and Macroeconomicsed. by MeirKohn and Sho-ChiehTsiang (Oxford: Clarendon Press1988).

    GuitiánManuel“The Role of Monetary Policy in IMF Programs”(unpublished paper presented at an International Conference organized by De Nederlandsche Bank and Tilburg University on “A Framework for Monetary Stability”AmsterdamOctober1993).

    GuitiánManuel“Rules or Discretion: National and International Perspectives,”inFrameworks for Monetary Stability: Policy Issues and Country Experiences (Washington: International Monetary Fund1994).

    GurleyJohnG. and Edward S.ShawMoney in a Theory of Finance (Washington: Brookings Institution1960).

    HargravesMonicaGarry J.Schinasi and Steven R.Weisbrod“Asset Price Inflation in the 1980s: A Flow of Funds Perspective,”IMF Working Paper No. 93/77 (Washington: International Monetary Fund1993).

    HillierBrian and M.V.Ibrahimo“Asymmetric Information and Models of Credit Rationing,”Bulletin of Economic ResearchVol. 45No. 4 (October1993) pp. 271304.

    HutchisonMichaelM.“Asset Price Fluctuations in Japan: What Role for Monetary Policy,”Working Paper No. PB93–11 (San Francisco: Federal Reserve Bank of San Francisco, Pacific BasinDecember1993).

    International Monetary FundWorld Economic Outlook (Washington: International Monetary FundMay1993).

    JamesChristopher“Some Evidence on the Uniqueness of Bank Loans,”Journal of Financial EconomicsVol. 19No. 2 (December1987) pp. 21735.

    JonesDavidM.“Comments on Credit in the Macroeconomy: The Role of Credit in Economic Activity,”Quarterly Review Federal Reserve Bank of New YorkVol. 18No. 1 (Spring1993) pp. 7176.

    KashyapAnilK.Jeremy C.Stein and David W.Wilcox“Monetary Policy and Credit Conditions: Evidence from the Composition of External Finance,”American Economic ReviewVol. 83No. 1 (March1993) pp. 7898.

    KashyapAnilK. and Jeremy C.Stein“Monetary Policy and Bank Lending,”NBER Working Paper No. 4317 (Cambridge, Massachusetts: National Bureau of Economic ResearchApril1993).

    KasmanBruce“A Comparison of Monetary Policy Operating Procedures in Six Industrial Countries,”Federal Reserve Bank of New York Quarterly ReviewVol. 17No. 2 (Summer1992) pp. 524.

    KhanMohsinS. and V.Sundararajan“Financial Sector Reforms and Monetary Policy,”IMF Working Paper No. 91/127 (Washington: International Monetary Fund1991).

    KingStephenR.“Monetary Transmission: Through Bank Loans or Bank Liabilities?,”Journal of Money Credit and BankingVol. 18No. 3 (August1986) pp. 290303.

    KonishiToruValerie A.Ramey and Clive W.Granger“Stochastic Trends and Short-Run Relationships Between Financial Variables and Real Activity,”NBER Working Paper No. 4275 (Cambridge, Massachusetts: National Bureau of Economic ResearchFebruary1993).

    KuglerPeter“Credit Rationing and the Adjustment of the Loan Rate: An Empirical Investigation,”Journal of MacroeconomicsVol. 9No. 4 (Fall1987) pp. 50525.

    LounganiPrakash and MarkRush“The Effect of Changes in Reserve Requirements on Investment and GNP,”Working Paper Series No. 91–21 (Federal Reserve Bank of ChicagoDecember1991).

    MiltonLeslie“An Analysis of the Information Content of Alternative Credit Aggregates,”Bank of Canada Technical Report No. 49 (Ottawa: Bank of Canada1988).

    MinskyHymanP.“Financial Crisis, Financial Systems, and the Performance of the Economy,”inPrivate Capital Marketsed. by IrwinFriendHyman P.Minsky and Victor L.Andrews (Englewood Cliffs, N.J.: Prentice-Hall1964).

    MironJeffreyA.Christina D.Romer and David N.Weil“Historical Perspectives on the Monetary Transmission Mechanism,”NBER Working Paper No. 4326 (Cambridge, Massachusetts: National Bureau of Economic ResearchApril1993).

    MishkinFredericS.“Asymmetric Information and Financial Crises: A Historical Perspective,”inFinancial Markets and Financial Crisesed. by R. GlennHubbard (Chicago: University of Chicago Press1991).

    MorganDonaldP.“Are Bank Loans a Force in Monetary Policy?,”Federal Reserve Bank of Kansas City Economic Review (Second Quarter1992).

    MullerPatrice“The Information Content of Financial Aggregates During the 1980s,”inMonetary Seminar 90 (Bank of Canada1992) pp. 183304.

    O’BrienP.F. and F.X.Browne“Credit Crunch? The Recent Slowdown in Bank Lending and its Implications for Monetary Policy,”OECD Working Paper No. 107 (Paris: Organization for Economic Cooperation and Development1992).

    RadeckiLawrenceJ.“A Review of Credit Measures as Policy Variable,”inIntermediate Targets and Indicators for Monetary Policy: A Critical Survey (Federal Reserve Bank of New York1990).

    RameyValerie“How Important Is the Credit Channel in the Transmission of Monetary Policy?,”Carnegie-Rochester Conference Series on Public PolicyVol. 39 (December1993) pp. 145.

    RomerChristina and DavidRomer“Credit Channel or Credit Actions? An Interpretation of the Postwar Transmission Mechanism,”NBER Working Paper No. 4485 (Cambridge, Massachusetts: National Bureau of Economic ResearchOctober1993).

    RosaRobertV.“Interest Rates and the Central Bank,”inMoney Trade and Economic Growth: Essays in Honor of John Henry Williams (New York: Macmillan1951).

    SamieiHossein and Garry J.Schinasi“Real Estate Price Inflation, Monetary Policy, and Expectations in the United States and Japan,”IMF Working Paper No. 94/12 (Washington: International Monetary Fund1994).

    SchinasiGarryJ. and MonicaHargraves“Boom and Bust in Asset Markets in the 1980s: Causes and Consequences,”IMF Staff Studies for the World Economic Outlook (Washington: International Monetary FundDecember1993).

    StiglitzJosephE.“Government, Financial Markets, and Economic Development,”NBER Working Paper No. 3669 (Cambridge, Massachusetts: National Bureau of Economics ResearchApril1991).

    StiglitzJosephE. and AndrewWeiss“Credit Rationing in Markets with Imperfect Information,”American Economic ReviewVol. 71No. 3 (June1981) pp. 393410.

    StockJames and MarkWatson“New Indexes of Coincident and Leading Economic Indicators,”inNBER Macroeconomics Annual 1989 by OlivierBlanchard and StanleyFischer (Cambridge, Massachusetts: MIT Press1989).

    TobinJames“Monetary Policies and the Economy: The Transmission Mechanism,”Southern Economic JournalVol. 44No. 3 (January1978) pp. 42131.

    TobinJames and WilliamC. Brainard“Financial Intermediaries and the Effectiveness of Monetary Controls,”American Economic Review Papers and ProceedingsVol. 53No. 2 (May1963) pp. 383400.

    VegaJuanLuis“The Role of Credit in the Monetary Transmission Mechanism,”Economic Bulletin Banco de España (July1992) pp. 5566.

    WilliamsonStephenD.“Costly Monitoring, Financial Intermediation, and Equilibrium Credit Rationing,”Journal of Monetary EconomicsVol. 18No. 2 (September1986) pp. 15979.

    WilliamsonStephenD.“Financial Intermediation, Business Failures, and Real Business Cycles,”Journal of Political EconomyVol. 95No. 6 (December1987) pp. 11961216.

    WojnilowerAlbertM.“Not a Blown Fuse,”Federal Reserve Bank of New York Quarterly ReviewVol. 18No. 1 (Spring1993) pp. 4649.

The authors wish to thank Garry Schinasi for helpful discussions and David Bigman and Burkhard Drees for comments.

The view that variations in the availability of credit can have significant effects on real economic activity has a long history. In the postwar period it has been emphasized, inter alia, by Rosa (1951), Gurley and Shaw (1960), and Minsky (1964). Brunner and Meltzer (1972), Tobin (1978), and Friedman (1983), among others, have also discussed the role of credit. A distinguishing feature of the recent literature is the implication of the economics of imperfect information for the credit creation process. The rapidly growing literature on asymmetric information and financial structure has been surveyed by Gertler (1988) and Bernanke (1993).

Goodhart (1988) refers to these two aspects of monetary management (the macroeconomic role and the supervisory and regulatory role of central banks) as the “macro function” and the “micro function” of central banks and notes that historically the macro function was grafted onto the micro function.

In countries in which money markets are not well developed, reserve requirement ratios may also function as instruments.

Monetary policy instruments and operating procedures in a group of industrial countries are described and compared in Batten and others (1989) and Kasman (1992).

The intermediate target and indicator (or information) variable approaches to the formulation and implementation of monetary policy have been extensively described by Friedman (1990, 1993).

The response of real activity to movements in nominal money balances is transitory; once all adjustments are completed money is neutral. The temporary real output effects of a monetary policy shock are a consequence of sticky prices or expectations. The stickiness, in turn, arises from the presence of informational imperfections or contractual rigidities. Both the money and credit views rely on some form of imperfect price adjustment for the nonneutrality of monetary policy. However, this aspect of the transmission of monetary policy is independent of the issue of whether there exist separate money and credit channels and is, therefore, outside the scope of this paper. For a review of the theoretical and empirical literature on the sources of nonneutrality, see Blanchard (1990).

See Tobin and Brainard (1963).

On this see, for instance, Diamond (1984), Fama (1985, 1990), Williamson (1986), James (1987), and Calomiris and Kahn (1991).

It should be noted that this direct credit channel would operate even if money market interest rates did not change, which would be the case if there are perfect substitutes for bank deposits. In this limiting case, monetary policy would affect real spending only through the credit channel.

Lending standards may be tightened simultaneously with an increase in loan rates.

In the money view this would not occur, as borrowers could switch to other forms of financing without cost.

That is, at interest rates above the rate that maximizes the bank’s expected returns, the loan supply curve becomes backward sloping.

See Stiglitz and Weiss (1981) for a credit market model in which the adverse selection and moral hazard problems associated with ex ante asymmetric information may lead to credit rationing. For a credit model in which credit rationing stems from costly monitoring by banks of the project returns of borrowers (ex post asymmetric information), see Williamson (1986, 1987). Recent theoretical literature on the consequences of asymmetric information in the credit market is reviewed in Hillier and Ibrahimo (1993).

For instance, in the model developed by Bernanke and Blinder (1988), there is no credit rationing; the credit channel operates only through the relative price of credit.

When information is asymmetric, the Modigliani-Miller theorem is inapplicable. Firm’s financial positions become relevant to their investment decisions, introducing an interdependence between financial structure and real factors.

See, for instance, Bernanke and Gertler (1989,1990), Brunner and Meltzer (1988), Farmer (1984), Greenwald and Stiglitz (1988), and Williamson (1987).

Indicative target growth rates can be set for more than one financial aggregate, but the central bank cannot try to hit them all. This is not an issue if, as discussed below, the intermediate variables serve as policy guides instead of formal targets.

Radecki (1990) reviews the literature—bearing mostly on the U.S. economy—on the potential usefulness of broad credit measures and bank credit in the formulation and implementation of monetary policy. For some more recent evidence and evidence pertaining to other countries, see Kashyap, Stein, and Wilcox (1993), Ramey(1993), Milton (1988), Muller (1992), Blundell-Wignall and Gizycki (1992), O’Brien and Browne (1992), Filc (1993), and Vega (1992). The analysis has assumed that the money supply is exogenous. Under a fixed exchange rate regime, or when there is a balance of payments constraint, the money supply is endogenous. The appropriate intermediate target in this case would be net domestic assets (or domestic credit). This aspect has been discussed thoroughly by Guitián (1993).

This is a generalization, and does not mean that one cannot find evidence of credit measures leading output or that some credit aggregates are better predictors of economic activity than some monetary aggregates in specific periods.

See Bernanke and Blinder (1992), Morgan (1992), Dale and Haldane (1993).

For a discussion of the Canadian experience and the role of money as an indicator variable, see Caramazza, Hostland, and Poloz (1990).

On the issue of rules versus discretion in the conduct of policy, see Guitián (1994).

For supporting evidence see Bernanke (1990), Bernanke and Blinder (1992), Buttiglione and Ferri (1994). Dale and Haldane (1993). Friedman and Kuttner (1992). Kashyap, Stein, and Wilcox (1993), Mishkin (1991), and Stock and Watson (1989). For evidence challenging the credit view see Konishi, Ramey and Granger (1993), Miron, Romer and Weil (1993), Ramey(1993), and Romer and Romer (1993). For a review of the recent literature, see Kashyap and Stein (1993). For evidence in support of the credit view based on the effect of changes in reserve requirements (instead of financial spreads) on real activity, see Loungani and Rush (1991).

For evidence on the empirical significance of credit rationing, see King (1986), Kugler (1987), Berger and Udell (1992), and the studies cited therein.

This analysis, which follows from the Bernanke and Blinder (1988) extension of the IS/LM model, was used by Currie (1993) to explain developments in the Finnish economy.

See, for instance, International Monetary Fund (1993), and Schinasi and Hargraves (1993).

For instance, Hutchison (1993) concludes that monetary factors have played a relatively small role in real estate price fluctuations in Japan over the 1956–1993 period, while Samiei and Schinasi (1994) find support for the hypothesis that monetary variables had an important influence on real estate prices in the United States and Japan and, more importantly, that the impact of monetary variables on property prices changed in the 1970s compared with the 1980s. For other evidence linking asset price inflation and monetary policy, see Bank for International Settlements (1993) and Schinasi and Hargraves(1993).

As a case in point, consider the possibility of reacting now to the recent sharp rise in German equity prices (and European equity prices more generally) by tightening monetary conditions.

An exception is Hargraves, Schinasi, and Weisbrod (1993).

Indeed, there is a school of thought that argues that when the financial system is truly competitive and risks of moral hazard are suitably contained so that depositors and borrowers must accept the consequences of loss and bankruptcy, adequate disclosure requirements can be an effective substitute for prudential supervision and regulation by the state.

Goodhart (1992) pushes this argument to its logical conclusion, noting that cyclical variations in prudential regulations could be used to help dampen the business cycle. Such use would require extremely close coordination between the supervisor and the monetary authorities.

Responsibilities for supervision and regulation of the banking system vary widely across countries, extending to the format assignment of such responsibilities to a separate body in some countries even though in practice close cooperation with the central bank is necessary. Goodhart (1992) has argued against the separation of the “macro” and “micro” functions of central banks and Alan Greenspan (1993) has recently reiterated the U.S. Federal Reserve’s position that for a central bank to effectively carry out its macroeconomic function it must have a microeconomic function as well.

The role of financial factors in the Great Depression is reviewed in Calomiris (1993).

See Calvo and Coricelli (1992) and Calvo and Kumar (1994).

This does not necessarily exclude the existence of the type of equilibrium (or “endogenous”) credit rationing described above under “Money and Credit in the Transmission of Monetary Policy.”

For an overview of the implications of financial sector reform on the objectives, instruments, and operating procedures of monetary policy, see Khan and Sundararajan (1991).

    Other Resources Citing This Publication