Current Developments in Monetary and Financial Law, Vol. 3

CHAPTER 21 Frameworks for the Resolution of Government-Central Bank Conflicts: Issues and Answers

International Monetary Fund
Published Date:
April 2005
  • ShareShare
Show Summary Details

It is fashionable to argue that the influence of central bankers has never been greater than it is today. But influence over what? It is perhaps better to say that, in comparing the relative influences of monetary and fiscal policies over aggregate economic outcomes, monetary policy is currently seen as both more powerful and, in recent years at least, more successful. But this was not always so. The current situation is the result of the experience with the stagflation of the late 1970s and early 1980s, when an emasculated monetary policy, combined with a profligate fiscal policy, delivered poor economic outcomes in the industrial world.

Nevertheless, current favorable sentiments about the success of monetary policy must be understood in the context of a broader, longerrun cycle whose origins can be traced back to the early part of the 20th century. The position of central banks has fluctuated from being essentially powerless to becoming the dominant force in economic policymaking.

The influence of central banks must also be assessed in light of the general economic environment, most notably the type of exchange rate regime in place, together with the extent to which international events spill over into the domestic policymaking setting. For example, the Gold Standard era during the first quarter of the 20th century severely limited the ability of central banks to formulate a made-at-home monetary policy while placing them front and center in the conduct of economic policy. The Great Depression of 1929-34 changed all that, and fiscal policy replaced monetary policy as the driving force behind policymaking. This phase continued through at least the end of World War II. The Bretton Woods era returned some limited influence to central banking, but it was still the case that monetary policy took a backseat to fiscal policy.

As pointed out above, the pendulum swung back in favor of monetary policy sometime in the 1980s. One notable development was the concern expressed by policymakers over defining more precisely than ever the role and influence of central banks in economic policymaking. In particular, it became fashionable to formally guarantee the autonomy of the central bank vis-à-vis the government, as well as provide central banks with explicit (or implicit) quantitative inflation objectives.

Through the decades the ebb and flow of the influence of monetary policy over the economic affairs of a nation reflected in part the omnipresent potential for conflict between government and the central bank. Yet, in spite of recent clarifications over the precise status of the central bank, the potential for conflict remains—if only because the current fashion of an independent central bank that focuses on some inflation objective need not be permanent if, for any reason, the policy fails to deliver good economic outcomes and loses public support. There is no better illustration of this than the current disputes raging between central banks and their governments in Central Europe. Indeed, in an attempt to turn back the clock, there are proposals to reduce the autonomy of central banks in Hungary and Poland before their respective parliaments.1 Therefore, as Alan Greenspan, the Chair of the Board of Governors of the U.S. Federal Reserve System, observed in 1996, “Despite waxing and waning over the decades, a deep-seated tension still exists over the government’s role as an economic policymaker.”2

The present chapter, then, is about frameworks for the resolution of government-central bank conflicts. It will also attempt to draw some broad conclusions based on historical experience and discuss a proposal to minimize such conflicts in the future.

The next section provides a typography of existing frameworks for the resolution of conflicts between the monetary and government authorities. The chapter then presents a discussion of the essential and enduring role of the statutory relationship between a central bank and the government in both fostering and mitigating the possibility of conflict. Also provided are some illustrative examples of central bank-government conflicts and how these were either resolved, or how they exposed a flaw in the formal relationship between the two main economic policymaking bodies. The penultimate section proposes some general principles central banks more generally ought to adopt in order to minimize the potential for conflict. The chapter concludes with a summary of the core elements of a central bank law that highlight how the overall relationship between the government and the monetary authority matters in properly defining a framework for the resolution of government-central bank conflicts. It should be emphasized at the outset that the evidence presented below focuses on the experience of industrial countries partly because government-central bank conflicts have had longer-lasting implications for the position of central banks, and partly because we can rely on a longer historical record to draw useful policy implications for all countries, not just for industrial countries. Further, the examples considered are highly selective and the description stylized in order to conserve space.

A Typography of Existing Statutory Frameworks

One can essentially delineate two types of conflicts between a government and the central bank. Typically, the relationship between the monetary and fiscal authorities is expressed through the Governor or President of the central bank—hereafter referred to as the CEO—and the Minister of Finance. Hence, a possible source of conflict is a clash of personalities. If the trigger for such clashes is simply the antipathy between two individuals, it is doubtful that any statutory framework will be able to deal with such occurrences.3

Implicit in this argument and, indeed, the premise of the chapter, is that the legal framework that defines the government-central bank working relationship ought to be capable of dealing with conflicts as they arise and in such a manner as to ensure that the public retains confidence that the central bank is able to carry out its mandate regardless of who leads the institution. Ideally, the statutes of the central bank should also be capable of reducing the likelihood that conflicts will emerge in the first place.

The type of conflict that can be most injurious to the reputation of the central bank and the government involves disputes over policies. The dispute, however, need not center only on questions of monetary policy. As we shall see below, conflicts often arise because the central bank is perceived, rightly or wrongly, to have involved itself in areas outside its jurisdiction. This suggests that one should examine how precisely the statutes define the mandate and limitations of the central bank’s authority in matters of economic policy, in order to isolate as best as possible the sources of government-central bank disagreements. In other words, it is precisely conflicts over policies that ought to be addressed in the statutes, via good legislative design.

Currently, legislation in industrial countries generally deals with conflict in one of three ways. First, in some countries such as the United States, there is no explicit recognition of the possibility of conflict. Part of the explanation lies in the fact that the U.S. Federal Reserve (“Fed”) was originally set up as an institution with a fairly narrow mandate in the Gold Standard era and little thought was given to questions of autonomy. As a result, there is no provision for dismissal. Indeed, the Fed was, as central banks go, a minimalist central banking institution, so questions surrounding autonomy from or conflicts with the President or Congress were not deemed to be necessary or relevant.

Elsewhere, most central bank statutes have some minimal recognition of the possibility of conflict between monetary and fiscal policy. Often, however, the provision for dismissal permits this eventuality only in the case of extreme misconduct. Examples of such provisions maybe found in the legislation governing the Japanese central bank as well as the European Central Bank (ECB). A partial explanation for this type of framework is that the authors of the legislation wanted to enhance the autonomy of the central bank by minimizing the conditions under which a dismissal would be permissible, or by altogether excluding the possibility of conflict by simply not defining it. Historical experience suggests, however, that this is a recipe for failure for the simple reason that, assuming governments have a preference for looser rather than tighter monetary policies, conflict is inevitable.

Finally, in the case of a few central banks, the device of the directive is used to deal with policy conflicts. Under this type of arrangement, statutes explicitly recognize the possibility of a disagreement over monetary policy between the government and the central bank, and the legislation may also outline how the conflict is to be ultimately resolved. A common feature in central banks in which a directive clause is in place is that the government is required to issue an order (i.e., a directive) that sets out the existence of a policy disagreement. It is not always mandated, however, that the directive be made public, nor do the particulars of the policy disagreement need to be announced.

An additional common feature of the directive approach, where it exists, is that it is intended to clarify where the ultimate responsibility for monetary policy lies, and ordinarily this is with the government.

Despite some similarities, there are a number of important differences in the manner in which conflicts are settled via the directive framework. First, even if the directive clarifies the location of the ultimate responsibility for monetary policy, statutes do not necessarily specify that, in the normal course of events, the central bank is solely responsible for carrying out monetary policy.

Second, there are differences in the manner in which conflict is ultimately resolved. In some instances (e.g., Netherlands pre-ECB, Germany pre-ECB, and Japan), government objections to policy get a hearing in the central bank council, and a monetary policy decision that the government wishes changed may be delayed, but, once a specified time has elapsed, the central bank may carry out its originally intended policy as planned if the objections are dismissed (optionally with a public and/or private explanation). In other instances (e.g., Australia, Canada), a policy conflict may ultimately require the intervention of the legislature in some form. In another variant of the directive scheme, as outlined in the 1946 Bank of England Act (since amended), there is an explicit requirement that the directive be carried out at once, and there are no grounds for appeal, nor can the monetary authority object to the ultimate decision. Interestingly, none of the directive schemes currently or previously in place explicitly mandates that the CEO of the central bank be expected to resign.4 Instead, such an expectation is said to be implicit, conditional, of course, on the nature or severity of the policy dispute.

One additional issue deserves further discussion. The effectiveness and consequences of the directive approach are also influenced by governance considerations. In some central banks, the senior board of the central bank acts as an intermediary in the conflict once the nature of the policy disagreement is announced. Indeed, in some cases, only the senior board of the central bank can formally reprimand or seek the dismissal of the CEO (e.g., New Zealand). At other central banks, as noted earlier, there can be more direct political involvement either from the executive branch, the legislature, or both (e.g., Australia). As we shall see, issues of central bank governance can play a crucial role in mitigating the negative consequences of policy disputes.

Table 1 provides some information relative to the varieties of existing frameworks for the resolution of government-central bank conflicts in the G-7 countries.5

Table 1.Government-Central Banks’ Conflict and Statutes
Explicit Government Override of Monetary Policy Permitted? (year introduced)Appeal and Conflict Resolution Procedures of Central Bank Decisions
Appeal ProcessType of Conflict Resolution
Minister responsibleLegislative bodyCentral bank boardNoneCentral Bank follows policy prescribed by governmentConflict made public by Central Bank government
CanadaYes (1967)XXX
FranceYes (1954)XX
GermanyNo (1957)X
ItalyNo (1936)XX
JapanNo (1957), Yes (1997)XX
United KingdomYes (1946), Yes (1998)X(1946)X (1998)X (1946)X(1998)
United StatesNo (1973)X
Source: P.L. Siklos, The Changing Face of Central Banking: Evolutionary Trends Since World War II.
Source: P.L. Siklos, The Changing Face of Central Banking: Evolutionary Trends Since World War II.

Do Central Bank Statutes Matter?

The previous section raises the obvious question: How important is it for central bank statutes to explicitly outline a conflict resolution procedure? First, to the extent that policy disputes have negative economic implications, clarity in dealing with conflict ought to mitigate these costs. Second, the experience of the 1990s suggests that central bank openness and accountability have improved monetary policy performance and may have, along with a more harmonized fiscal policy, contributed to the stellar economic performance of the past decade.6 Because these changes have also helped reduce the potential for policy conflicts, it would seem clear that the same wave of reforms leading to greater central bank autonomy, together with a clearer delineation of the sphere of responsibility in the area of monetary policy, would be further improved by having more explicit conflict resolution procedures where appropriate. As always, of course, there are costs and benefits to increasing the explicitness with which conflicts are recognized, and procedures for resolving them have been incorporated into law. This is the case, first, because not all events that can lead to policy conflicts can possibly be fully anticipated. Second, because it is not always possible to separate personality from policy conflicts, some flexibility is desirable in laying down procedures to resolve conflicts. This approach has the added advantage of reducing the temptation to alter legislation too frequently as the political and monetary authorities seek a reasonable amount of flexibility to deal with policy disputes.

Nevertheless, it must be underlined that there are clear benefits to providing conflict resolution procedures in central bank statutes. First and foremost, explicit procedures provide additional clarity regarding the limits of central bank autonomy. Such limits are consistent with notions of democratic accountability of governments for both fiscal and monetary policies. Second, explicit conflict resolution procedures limit the temptation for governments to manipulate the central bank, and monetary policy more generally, on a whim or for purely political advantage. History, as we shall see, provides powerful evidence to support this view. Prior to outlining the elements of what might be considered proper or adequate conflict resolution procedures, it is worthwhile to consider some examples of government-central bank conflicts. These examples can be informative as to the statutory requirements that ought to minimize the negative consequences from such conflicts.

Conflicts Between Governments and Central Banks: Some Examples from History

What follows is a selective survey of incidents of government-central bank conflicts.7 The aim is not to be exhaustive but instead to illustrate how certain aspects of central bank statutes may or may not have contributed to fanning the flames of the conflict in question. The examples serve to illustrate where the trade-off between explicitness and subtlety in conflict resolution procedures in central bank statutes might be located. What follows are brief accounts of the selected conflicts. Readers are referred to a variety of sources listed at the end of this chapter for additional details.

The Fed and the Treasury in the 1950s

The requirements of financing World War II led the U.S. government to effectively emasculate the Fed via the pegging of interest rates.8 Following the end of World War II, tensions between the Treasury and the Fed began to surface as the Fed continued to peg interest rates. Indeed, Marriner Eccles, then-Chair of the Board of Governors of the U.S. Federal Reserve, described the Fed’s task as “…merely executing Treasury decisions.”9 At the same time, Congress passed the Full Employment Act of 1946, a piece of legislation that effectively required the Fed to simultaneously aim for full employment and price stability. Following the end of wartime price controls, excess demand pressures began to mount (inflation rose sharply but began to subside by 1948), and Eccles called on the Treasury to abandon interest rate pegging and return effective control over monetary policy to the central bank. Because the legislation governing the Fed until 1946 did not explicitly define the relationship between the Treasury and the central bank,10 a separate accord had to be negotiated. Eccles, one of the eventual casualties of the process, ceded to McChesney Martin, the Chair of the Federal Open Market Committee (FOMC), which ultimately led to the Fed-Treasury Accord (the “Accord”) of 1951.11

Indeed, conflict existed within the Federal Reserve among Governors who held differing views about the degree of autonomy the Fed should have, its relationship with the Treasury, and the ultimate objectives of monetary policy.12 Shortly after the Accord, proposals for some form of inflation targeting were revived, then dropped, by Congress.13 Because there was considerable pent-up aggregate demand, the delay in clarifying the relationship between the government and the central bank would eventually lead to considerable inflation, at least by U.S. historical standards, and at least one fairly serious recession during the 1950s. Had the original legislation been clearer about the respective responsibilities of the Fed and the Treasury, FOMC members would perhaps not have chafed14 under Treasury’s control.

The Bank of Canada and the Coyne Affair

Canada, like the United States, experienced economic problems arising from pent-up wartime aggregate demand. As part of the war effort, the federal government issued short-term government bonds, called Victory bonds, that were to mature in the second half of the 1950s. As the time of maturity approached, the government wished to lengthen the average maturity of government bonds and embarked upon a so-called Conversion loan in 1957. The Bank of Canada stood ready to temporarily hold government bonds until the conversion was completed. While the Conversion loan program was carried out fairly successfully, the policy was being implemented at a time when the government decided to float the Canadian dollar following an economic boom during 1956 and 1957. Moreover, in 1956, the Bank of Canada changed operating procedures by permitting its discount rate, the Bank rate, to float with the yield on weekly auctions of 90-day Treasury bills.

Annual reports of the Bank of Canada for the years 1956 and 195715 emphasized the need for monetary policy to reduce inflationary pressures, though the reports denied that monetary policy was excessively tight. Since short-term Canadian interest rates drifted higher relative to those of the United States, and there were no capital controls, large capital inflows were a feature of the Canadian economic landscape of the 1950s.

The combination of these factors, and a looming recession in the late 1950s, set the stage for a conflict between the Bank of Canada’s Governor, James Coyne, and then-Finance Minister John Fleming.16 The federal government had been planning a budget to stimulate the economy at a time when elections were also expected, and this plan seemed to conflict with the efforts of the Bank to control inflation. Because price stability, together with other economic goals, is included in the Bank’s mandate,17 Coyne felt duty-bound to stem inflationary pressures. The fact that at the time there was relatively little communication between the Bank and the Finance Ministry, coupled with no statutory means to pressure the Governor to change course, resulted in the Minister of Finance eventually pressuring the Board of Directors of the Bank to remove the Governor.18 The Board, at first, did not carry out the Finance Minister’s wishes. Interestingly, during this period, Coyne himself pleaded with the government to consider legislative changes that would permit the removal of the Governor if the policies carried out by the central bank were at variance with the government’s wishes. The inspiration for this call came from the directive process incorporated in the Bank of England Act.19

Adding fuel to the fire was the unprecedented call by a group of economists for the Finance Minister to fire the Governor, citing incompetence.20 While there is insufficient space here to go into the details, it is clear that there were a number of factors that contributed to the conflict, but none more critical than the inability of the government to get the monetary policy it desired, the failure to contemplate formal means by which the central bank and the Treasury could communicate at regular intervals, and the almost total absence of procedures to resolve conflicts of a policy nature.

When the second legislative chamber, the Senate, came out in support of the embattled Governor, Coyne resigned, a mere six months before his term was due to expire. It took another six years, but the Bank of Canada Act was revised and a directive clause was added that clarified the location of the ultimate responsibility for monetary policy, namely with the government, while “in the ordinary course of events,” the Bank of Canada would be fully in charge of monetary policy21

The European Monetary System and German Reunification

Even if one accepts the main premise herein, namely that conflict resolution procedures are a vital, but too often neglected, aspect of the statutory relationship between the government and the central bank, there is one policy area generally considered to be outside the sphere of influence of a central bank that can lead to conflict.

It is generally accepted that governments have complete responsibility over exchange rate policy. Therefore, if policy, choices in the realm of exchange rates come into conflict with monetary policy, then the likelihood of a serious dispute between the monetary and fiscal authorities rises substantially.

Such was the situation when the German government took over responsibility for the former East Germany in 1989. Faced with the task of monetary unification, the federal government, headed by then-Chancellor Helmut Kohl, ignored the advice of the central bank, the Bundesbank, and opted instead for a conversion of East German marks for deutsche mark at an exchange rate deemed to be excessively inflationary, at least according to the Bundesbank.

A highly public dispute ensued, not surprising in view of the longheld practice of the Bundesbank to communicate in public its view about both fiscal and monetary matters.22 In any event, fears of inflation eventually proved to be correct, and the Bundesbank tightened monetary policy as promised. Nevertheless, in the meantime, then-President of the Bundesbank, Karl Otto Pòhl, resigned as a result of these events.23

The German experience illustrates that, even with autonomy guaranteed by statute and the minimal possibility of override of monetary policy by the government, conflict is still possible. The Bundesbank may have been the central bank that ruled Europe24 and the inspiration for what became the ECB,25 but it could not avoid conflict.26 Indeed, it is quite possible that while the authors of the Bundesbank law believed that an autonomous central bank represented an important counterweight to prevent profligate government spending, it is less clear whether these same legislators deliberately sought this outcome.27 Indeed, there is some debate about whether the Bundesbank law was clear enough about the ultimate responsibility for exchange rate policy, in particular, and perhaps monetary policy more generally. Nevertheless, the fact that there was no adequate escape clause in the event of such a conflict implies that considerable central bank autonomy can be just as much an incitement to conflict as too little autonomy.

The Reserve Bank of New Zealand and Inflation Targeting

New Zealand became the first country to formally adopt inflation targets in the late 1980s, following radical reforms to the structure and operations of the Reserve Bank of New Zealand (RBNZ). The original targets were expressed in terms of headline consumer price inflation and included plans to gradually reduce the targets until more permanent inflation control targets could be put into place.28 In 1994, as inflation fears grew in the United States especially, interest rates rose, and the RBNZ followed suit. This resulted in a breach of the inflation target that was publicly explained by the Governor as due to the effect of interest rate changes on the behavior of headline inflation. While the policy targets agreements (PTAs) between the Governor and the Treasurer that define the inflation objectives of the RBNZ made clear that the operational inflation target was to be understood in terms of core inflation, the distinction seemed lost to the public. Although future PTAs were to be written explicitly in terms of a core measure of inflation, this did not occur until an additional breach of the target in 1996 led to a more serious conflict that illustrated a deficiency in the operations of the RBNZ Act. Much publicity has been attached to a clause in the RBNZ Act that permits removal of the Governor in the event that inflation targets are not met. In principle, a breach in the target prompts the Treasurer to ask the Board of the RBNZ to provide advice about whether the Governor’s employment ought to be called into question.29 The difficulty is that the sought-after advice was provided largely by the staff of the RBNZ, and they reported directly to the Governor and not to the Board.30

In the end, the Board did not recommend removal of the Governor and, while the inflation target was eventually redefined in terms of core inflation, the incident called into question the manner in which a target breach should be addressed.31 No changes were made to the Bank’s governance, despite an extensive review that raised questions about the position of the Board vis-à-vis the Governor in the statutes.32 The case also illustrates that, despite the best attempts of the authors of the statutes to ensure strong accountability in delivering good monetary policy outcomes, mere inflation targets do not suffice in preventing conflicts from taking place. Governance questions, which, until recently, were largely ignored in discussions about the relationship between the government and the central bank, should be treated on an equal footing with the accountability and transparency considerations.33

The Bank of Japan and Deflation

Until the early 1990s, the Bank of Japan’s (BOJ) autonomy was very much in question. Like Germany’s law, the Bank of Japan Act of 1997 formally gave responsibility for monetary policy to the BOJ and it effectively permitted the government to delay but not overturn a decision of the monetary policy committee.34

Following the bursting of the property bubble, together with an essentially insolvent banking system, the Japanese economy faced stagnation and deflation, which has lasted on and off for over a decade.35 Conflicts arose as, on one occasion, the government asked the BOJ to reverse an interest rate decision (August 2000). The request was turned down by the Policy Board. Almost simultaneously, there were calls by Japanese officials, on repeated occasions, to impose an inflation target as well as to remove the present Governor. Political turmoil has not permitted successive Japanese governments to translate reform proposals into actual legislation, but the turmoil in Japan over the past decade, which arguably could be laid at the door of fiscal policy,36 illustrates the drawbacks of central bank statutes that fail to recognize the need for monetary and fiscal policy to work in harmony. While Japanese political leaders have complained about the conduct of monetary policy, possibly with good reason, there has been little recognition that monetary policy cannot alone deliver good economic outcomes without a coherent fiscal policy. The way in which the statutes of the BOJ are currently written, while protecting the autonomy of the central bank, does not provide for adequate communication between the Minister of Finance and the Governor.37 Additional drawbacks of the BOJ’s statutes permit conflicts to emerge as a result of tensions concerning the role and authority of the central bank over the exchange rate and foreign exchange reserves.38 On the one hand, the BOJ law (Article 15) permits foreign exchange purchases “…to stabilize the exchange rate of the national currency,” as required if a Ministry of Finance directive is issued. Yet the same law also stipulates (Article 3) that “…the BOJ’s autonomy regarding currency and monetary control shall be respected.” Obviously, the two articles are potentially in conflict with each other.39

The European Central Bank and Accountability

The ECB represents a novel experiment. It is a supranational central bank and, as a result, faces unique pressures that national central banks are unlikely to face. Given the relatively large number of member countries, and the prospect of more joining in the coming years, reasonably efficient governance requires that a subset of member countries must make monetary policy decisions. This requires careful attention to appointment procedures. The first illustration of pitfalls in this area took place with the appointment of the first President of the ECB, when it was rumored that the Dutch candidate, Wim Duisenberg, took the post on the condition that a French national would replace him before the first eight-year term was up. The first ECB President did eventually decide to retire before the end of his term, but the manner in which such appointments were made illustrates how political interference can create government-central bank conflicts.

A second source of conflict in the ECB case arises, again in part, from the unusual supranational nature of the institution. Unlike most national central banks, the ECB is effectively the guardian of the exchange rate regime. During the first year of the euro, as it plunged against the U.S. dollar, several national finance ministers pleaded with the ECB to intervene in foreign exchange markets. Nevertheless, as the ECB is first and foremost accountable to the European Parliament and second to the Council of Finance Ministers, the desire to make the ECB entirely autonomous can come into conflict with the need for accountability and transparency. Indeed, many commentators have complained that the ECB is insufficiently accountable (or transparent).40 The ECB case illustrates a fairly common thread in the examples considered in this section, namely that central bank autonomy alone does not prevent conflict. Instead, all aspects of central bank statutes must be called into question if a prime objective of the legislation is to minimize the likelihood of disputes between governments and their central banks.

Drawing Some Lessons

There is broad agreement that central bank autonomy and clarity of monetary policy objectives have contributed to improved economic performance. Table 2 vividly illustrates that one expression of the need for clarity has been the adoption of some form of inflation targeting as a principal central bank objective. Knowledge of the importance of price stability is not new:

As to the larger conclusion, we know this: Periods of serious price disturbances are periods of industrial and financial disturbance and social unrest. Practically never one without the other. And periods of price stability are periods of industrial and social equilibria and sanity.41

What is new is the recognition that some notion of price stability should be recognized, preferably in statutes.

Table 2.The State of Monetary Policy Objectives in the 1990s
Monetary Policy ObjectiveExchange Rate TargetMoney Growth TargetInflation Target
Number of countries that have adopted the relevant target313350
Number of countries that have dropped the relevant target1270
Sources: P.L. Siklos, The Changing Face of Central Banking: Evolutionary Trends Since World War II, and I. Mahadeva and G. Sterne, Monetary Policy Frameworks in a Global Context.
Sources: P.L. Siklos, The Changing Face of Central Banking: Evolutionary Trends Since World War II, and I. Mahadeva and G. Sterne, Monetary Policy Frameworks in a Global Context.

The case studies of the previous section demonstrate that independence alone, no matter how strongly guaranteed by statute, does not prevent government-central bank conflicts. Nevertheless, a survey of statutory objectives of central banks as shown in Figure 1, suggests that formal conflict resolution procedures, together with clearer monetary policy objectives, are present in only a small fraction of statutes. Hence, the potential for conflict remains, and, as has been known for some time, the drive for greater central bank autonomy must be tempered with recognition that governments and the public are ultimately responsible for monetary policy. The Report of the Royal Commission at the time the Bank of Canada was created (1936) said it best: “…while the Bank should resist temporary gusts of public policy, it must in the long-run show responsiveness to public opinion and be responsible to government.”

Figure 1.Varieties of Statutory Objectives

Sources: Siklos; and Mahadeva and Sterne. The vertical axis is the number of central banks with the specific objective.

An additional factor that contributes to raising the specter of conflict between governments and central banks arises because, despite improvements in core central banking principles, the fact remains that, as shown in Figure 2, governments are still typically not sufficiently accountable if they are not responsible for setting and quantifying the target for monetary policy.42

Figure 2.Who Sets the Objective of Monetary Policy?

Sources: Siklos; and Mahadeva and Sterne.

A selective survey of some government-central bank conflicts also suggests that an additional factor that has the potential for worsening disputes is poor communication either between the government and the central bank, or both with the public. Another approach introduces a detailed index of accountability43 and disclosure44 and suggests that both these requirements, when incorporated in central bank statutes, are correlated not only with each other but also with an index of central bank independence. Table 3 provides a ranking of central banks in 20 industrial countries along these three dimensions. Note that while the correlations in question are positive, as expected, they are not yet sufficiently high to entirely dismiss concerns expressed here over deficiencies in conflict resolution procedures.

Table 3.Central Banks in the Industrial World: How Open and Accountable?
Central BankIndex of Accountability

(maximum 1)
Index of Disclosure

(maximum 1)
Index ValueRankIndex ValueRank
United Kingdom.697.911
United States.5613.872
European Central Bank.716.5212
Source: RL. Siklos, The Changing Face of Central Banking: Evolutionary Trends Since World War II. Accountability and disclosure are defined in the text. More details can be found at The rank order is based on a sample of 20 Organization for Economic Cooperation and Development countries (not all are shown here).
Source: RL. Siklos, The Changing Face of Central Banking: Evolutionary Trends Since World War II. Accountability and disclosure are defined in the text. More details can be found at The rank order is based on a sample of 20 Organization for Economic Cooperation and Development countries (not all are shown here).

A lesser-known potential for conflict, and one that is also the direct result of the introduction of the principles of autonomy, accountability, and disclosure, has emerged in recent years. As explained by Alan Greenspan, Chair of the FOMC of the U.S. Federal Reserve, good conduct in monetary policy means that the central bank must be forwardlooking:

Because monetary policy works with a lag, we need to be forwardlooking, taking actions to forestall imbalances that may not be visible for months. There is no alternative to basing actions on forecasts, at least implicitly. It means that we often have to tighten or ease before the need for action is evident to the public at large…. This process is not easy to get right…and it is often difficult to convey.45

Hence, crucial to the successful implementation of monetary policy is the quality of the communication between the central bank and the public, an area that has only recently begun receiving much-needed attention.46 However, it is extremely difficult to envisage how central bank statutes can ensure that a desirable level of communication with the government and the public takes place. Therefore, once again it becomes even more important to ensure that all core elements of central bank statutes be written in such a way as to reduce as far as is practical the likelihood of disputes between the monetary and fiscal authorities, and this surely must include some attempt at ensuring that a variety of channels exist through which the central bank is able to express itself.

Finally, despite the apparent success of monetary policy regimes in most industrial economies during the 1990s, another potential source of conflict not easily dealt with in statutes is complacency about the merits of the existing monetary regime. Alan Greenspan has also made essentially the same point:

Whatever its successes, the current monetary regime is far from ideal. Each episode has had to be treated as unique or nearly so. It may have been the best we would do at the moment. But we continuously examine alternatives that might better anchor policy, so that it becomes less subject to the abilities of the FOMC to analyze developments and make predictions.47

A matter open to debate, then, is whether, as Mr. Greenspan contends, the impetus for change should be undertaken by the central bank alone, or whether statutes can make allowances for a regular review of monetary policy procedures.

The following section attempts to outline some core principles of central bank statutes that should contribute to delivering good conduct in monetary policy. While these principles are primarily meant to ensure that good monetary policy is delivered, the obvious corollary is to create conditions to minimize the likelihood of destructive government-central bank conflicts.

Core Elements of an Ideal Central Bank Law

The preceding sections have, it is hoped, made the point that the ideal framework to deal with government-central bank conflicts requires not only explicit conflict resolution procedures but a recognition that the statutes in their entirety also contribute to reducing the likelihood and severity of disputes. Consequently, what follows is a core list of elements of a central bank law that ought to ensure not only that good monetary policy is delivered but, in addition, that both the ultimate, as well as the day-to-day, responsibilities for the conduct of monetary policy are clearly assigned and consistent with the principles of democratic responsibility. The items on the following list are arranged roughly in order of importance. However, the various elements are not to be treated independently of each other; that is, one should not be able to pick and choose individual items to suit one’s purposes.

  • The sole objective of monetary policy should be price stability.

  • The central bank shall have instrument independence.

  • The details of an inflation target will be outlined in a separate agreement between the government and the central bank. The agreement shall be reviewed at regular intervals.

  • The central bank shall not have an exchange rate target and will intervene in foreign exchange markets only to maintain orderly markets. Any intervention shall be publicly announced and explained.

  • The central bank’s mandate shall include the maintenance of financial market stability. The central bank shall endeavor to maintain financial market stability in cooperation with an independent supervisory authority as well as via public announcements of any actions the monetary authority deems necessary for the maintenance of the financial stability objective.

  • The central bank will be entirely responsible and accountable for the day-to-day implementation of monetary policy. Monetary policy decisions shall not be subject to any direction from the government unless it makes public the reasons it wishes to direct the central bank to alter existing policies. From time to time, the government may, via periodic reviews of the conduct of monetary policy, wish to set a new course for the central bank. This may require amending the central bank law.

  • The central bank and the government shall agree on a funding agreement to last until the current government’s mandate ends or there are new elections. Changes in the funding agreement shall be publicly announced and explained by the Finance Minister.

  • All forms of government borrowing from the central bank shall be prohibited, even temporarily. Borrowing by the private sector shall be limited to extraordinary circumstances. Extraordinary circumstances shall be defined by the central bank and financial sector representatives, and the definition shall be the subject of periodic reviews.

  • Governance. The central bank shall have a Board of Governors (BOG) and a monetary policy committee (MPC). The MPC shall be responsible for monetary policy decisions, while the BOG shall be the oversight body for the central bank. The BOG shall be charged with, among other responsibilities, approving nominations to the MPC. Members of the BOG shall be appointed by the government, subject to legislative approval (for n years, nonrenewable). Only the BOG can remove the Governor or members of the MPC for just cause (nonpolicy reasons).

  • The central bank shall be subject to reporting requirements. These include an Annual Report, testimony in front of the appropriate legislative bodies and committees, regular outside audits, a research function with standards and expectations to be publicly announced, and provision of general information to the public about the conduct of monetary policy in regular public forums and publications.


Conflict between the government and the central bank is, arguably, inevitable given the natural tensions between monetary and fiscal policies. There is scarcely any objection nowadays to the notion that the central bank ought to be given responsibility for maintaining price stability. There is, however, somewhat less agreement about what is meant by price stability and the degree to which governments ought to be able to set the course of monetary policy and interfere in central bank decisions.

If conflict is indeed inevitable, central bank statutes should recognize the possibility. While not all eventualities can be anticipated by any set of laws, this chapter has argued, via historical case studies, that an ideal central bank law must be based on a small but comprehensive set of principles. An attempt was made to provide a list of core principles, and these highlight clarity in the relative responsibilities of the monetary and fiscal authorities, the importance of governance, and central bank accountability and disclosure. Even among the industrial countries there is still insufficient recognition of the importance of conflict resolution procedures. The situation is likely even more acute in developing and emerging markets. Now that there appears to be a growing consensus over the ultimate objectives of monetary policy, it is hoped that policymakers will begin to turn their attention to conflict resolution and governance issues.


See The Economist, 2002.

A. Greenspan, “The Challenge of Central Banking in a Democratic Society,” speech given at the annual dinner and Frances Boyer lecture of the American Enterprise Institute, Washington, D.C, December 5, 1996.

Virtually all central bank statutes contain a clause that permits the removal of the CEO in case of bad conduct, that is, usually on moral or ethical grounds. A difficulty is that unless bad conduct is precisely defined, or if the legislation fails to give examples of bad conduct, such a clause can be especially exploited when there is a personality conflict, as we shall see from the Canadian experience.

Louis Rasminsky, upon his appointment as the Governor of the Bank of Canada in 1961, suggested that the legislation explicitly consider resignation as the natural outcome of any serious policy conflict. The eventual revision of the Bank of Canada Act in 1967 included the directive clause but no clause requiring the Governor to resign under the circumstances laid out in the directive.

See P.L. Siklos, The Changing Face of Central Banking: Evolutionary Trends Since World War II (Cambridge: Cambridge University Press, 2002) for information concerning 13 other OECD economies.

See, e.g., J. Frankel and P. Orszag, eds., American Economic Policy in the 1990s (MIT Press, 2002), at 19-43.

Conflict of one sort or another between governments and central banks may not be so rare after all. A search on Lexis-Nexis, using the keywords “central bank” and “conflict or government” yielded over 1,300 hits in major world newspapers from 1992 to 2002 (July 2 to July 2) and over 2,800 hits over the same period in European newspapers.

See A.H. Meltzer, “Under Treasury Control, 1942-51,” in A History of the Federal Reserve, Ch. 7 (Chicago: University of Chicago Press, 2003),

See M. Eccles, Beckoning Frontiers: Public and Personal Recollections, S. Hyman, ed. (New York: Knopf, 1951).

The changes to the Federal Reserve Act in 1946 set the following goals for monetary policy: “… to promote effectively the goals of maximum employment, stable prices, and moderate long-term interest rates” (Federal Reserve, 1994).

Thomas McCabe served a brief interregnum between 1948 and 1951.

See Meltzer, supra note 8, and P.L. Siklos, “U.S. and Canadian Central Banking: The Triumph of Personalities Over Politics?” in C.L. Holtfrerich, J. Reis, and G. Toniolo, eds., The Emergence of Modern Central Banking from 1918 to the Present (Aldershot, U.K.: Ashgate Press, 2000), at 231-78.

See Siklos, supra note 5.

Meltzer, supra note 8, at 67.

See Bank of Canada, Annual Report (Ottawa: Bank of Canada, 1958), and Bank of Canada, Annual Report (Ottawa: Bank of Canada, 1957).

See T.K. Rymes, “On the Coyne-Rasminsky Directive and Responsibility for Monetary Policy in Canada,” in PL. Siklos, ed., Varieties of Monetary Reforms: Lessons and Experiences on the Road to Monetary Union (Dordecht: Kluwer Academic Publishers, 1994), at 351-66.

The preamble to the 1967 Act (still in force) states that the central bank is required to “…control and protect the external value of the national monetary unit and to mitigate by its influence fluctuations in the general level of production, trade, prices and employment, so far as may be possible within the scope of monetary action, and generally to promote the economic and financial welfare of Canada.”

See Siklos, supra note 5.

The Directive in the Bank of England Act of 1946 (section 4 (1), since repealed) reads as follows: “The Treasury may from time to time give such directives to the Bank as, after consultations with the Governor of the Bank, they may think necessary in the public interest.”

See S. Gordon, “The Bank of Canada in a System of Responsible Government,” Canadian Journal of Economics and Political Science (February 1961), at 1-22.

See B. Muirhead, Against the Odds: The Public Life and Times of Louis Rasminsky (Toronto: University of Toronto Press, 1999).

See, e.g., P.L. Siklos and M.T Bohl, “Do Words Speak Louder than Actions? The Conduct of Monetary Policy at the Bundesbank,” working paper, Wilfrid Laurier University, 2004,

See H. Tietmeyer, “The Bundesbank: Committed to Stability,” in S.F. Frowen and R. Pringle, eds., Inside the Bundesbank (London: Macmillan, 1998), at 1-10 and H. Tietmeyer, “Reflection on the German Treaty Negotiations of 1990,” in S.F. Frowen and R. Pringle, eds., Inside the Bundesbank (London: Macmillan, 1998), at 68-109.

See D. Marsh, The Bundesbank: The Bank That Rules Europe (London: Heinemann, 1992).

See M. Marshall, The Bank (London: Random House Business Books, 1999).

See Siklos and Bohl, supra note 22.

See generally, S. Lohmann, “Designing a Central Bank in a Federal System: The Deutsche Bundesbank, 1957-92,” in Pierre L. Siklos, ed., Variety of Monetary Reforms: Lessons and Experiences on the Road to Monetary Union (Boston, Mass.: Kluwer Academic Publishers, 1994), at 247-78, and Deutsche Bundesbank, Fifty Years of the Deutsche Mark (Oxford: Oxford University Press, 1999).

See L.E.O. Svensson, Independent Review of the Operation of Monetary Policy in New Zealand: Report to the Minister of Finance (February 2001),

See, e.g., Reserve Bank of New Zealand, “Letter from Sir Peter Elworthy to the Minister of Finance, 19 April 1996,” in Monetary Policy Statement (June 1996).

See P.L. Siklos, Submission to Independent Review of the Operation of Monetary Policy (New Zealand, 2000),

D. Mayes, Submission to Independent Review of the Operation of Monetary Policy (New Zealand, 2000),

See Svensson, supra note 28.

See also Siklos, supra note 5.

The Bank of Japan Law can be found at The relevant sections of Article 19 read as follows:

  • (2) The Minister of Finance (or a designated delegate) and the minister for economic and fiscal policy (or a designated delegate), when attending the Board meetings for monetary control matters, may submit proposals regarding monetary control matters, or request that the Board postpone a vote on monetary control matters until the next Board meeting of this type.

  • (3) If a request is made to postpone a Board vote under the provisions of the preceding Paragraph, the Board shall decide whether or not to accommodate the request, in accordance with the same voting procedures which apply to other matters.

For reviews, see T.F. Cargill, M.M. Hutchinson, and T. Ito, Financial Policy and Central Banking in Japan (MIT Press, 2000), and K. Okina, M. Shirakawa, and S. Shiratsuka, “The Asset Price Bubble and Monetary Policy: Japan’s Experience in the Late 1980s and the Lessons,” Bank of Japan Monetary and Economic Studies, 19 (S-l): February 2001.

A. Posen, Restoring Japans Economic Growth (Washington D.C: Institute for International Economic, 1998).

This state of affairs appears to continue to this day. For example, while the Cabinet office (2001) still argues that inflation targeting ought to be formally considered, the Bank of Japan continues to suggest that such a policy is not a panacea for its current deflation problem (Yutaka Yamaguchi, Deputy Governor, Bank of Japan, 2002).

See B.T. McCallum, “Japanese Monetary Policy Again,” mimeo, Carnegie-Mellon University (October 2001).

The “inconsistency” noted by McCallum is not an unusual one in the annals of central bank legislation (see Siklos, supra note 5) and can easily be addressed by ensuring that the respective responsibilities between the government and the central bank are clearly laid out. As noted above, that is the fundamental problem with the BOJ legislation.

An excellent ongoing source of information documenting the tensions created by the ECB, as envisaged in the Maastricht Treaty, is the “Monetary Dialogue” between the Committee on Economic and Monetary Affairs of the European Parliament and the President of the ECB, together with academic views commissioned by the Committee. The relevant documents are available at

See C. Snyder, “The Problem of Monetary and Economic Stability,” Quarterly Journal of Economics (February 1935), at 173-205.

It is conceivable that the failure to provide for the possibility of dismissal reduces the likelihood of conflict since there is, in effect, no end game. On the other hand, the failure to provide for the possibility of conflict can exacerbate the severity of the outcome of disputes between a government and a central bank, assuming these are inevitable.

Accountability may be defined as consisting of the following criteria: the clarity and precision with which the objectives of monetary policy are stated and/or communicated; the degree and force of communication of policy decisions and strategies to accomplish stated objectives; and, if required, the extent to which the central bank must answer for past decisions, the form such responses must take, and the nature of the persons or bodies to whom the chief executive of the central bank is responsible. See Siklos, supra note 5.

The concept of disclosure puts the emphasis on the quality, as opposed to the quantity alone, of information provided to the public by the central bank. Hence, while “disclosure” and the more commonly used term “transparency” can be used interchangeably for the purposes of the present discussion, they do not have the same meaning. Id.

Greenspan, supra note 2.

See Siklos and Bohl, supra note 22; Siklos, supra note 5; and A. Blinder, C.A.E. Goodhart, P. Hildebrand, D. Lipton, and C. Wyplosz, How do Central Banks Talk: Geneva Report on the World Economy (Geneva: International Center for Monetary and Banking Studies, 2001).

Greenspan, supra note 2.

    Other Resources Citing This Publication