The Federal Reserve Board eagle logo links to home page

Remarks by Governor Laurence H. Meyer
At the National Association for Business Economics Seminar on Monetary Policy and the Markets, Washington, D.C.
May 21, 2001

Comparative Central Banking and the Politics of Monetary Policy

Diane gave me a very specific topic: How the Fed balances independence and accountability. The answer, of course, is quite simple: Carefully. I'll have something more to say about this balancing act in a moment, but before I do I thought it would be fun to start out with a broader set of questions, especially given the exciting panel that is going to follow lunch. That panel includes monetary policy committee members from the European Central Bank, the Bank of England, Canada, and the United States. Very nice job, Diane.

So, I would like to get today's program going by posing a question to the first panel, actually a set of three interrelated questions about comparative central banking. First, how similar or different is the practice of monetary policy across the central banks represented on the panel? Second, if practices are different, what accounts for the differences? And, third, what are the consequences of these differences for macroeconomic performance?

Monetary policies may, of course, appear different because the circumstances faced by central banks differ around the world. That would especially be the case if there were important country-specific shocks. But it could also be the case if common shocks affected countries differently because of differences in the structure of the economies, or even if initial conditions differed at the time they were hit by a common shock.

An even more interesting question, and a better lead-in to my assigned topic, is: To what extent do different central banks respond differently to the same set of circumstances? Here's a wonderful game: Assume that the ECB governing council or the Bank of England's Monetary Policy Committee replaced the Federal Open Market Committee and made monetary policy in the United States--or vice versa. How would policy outcomes and, in turn, macroeconomic performance be affected?

Comparative Central Banking
Why would central banks differ in their response to a given set of economic circumstances? Let me offer a few possibilities--some of which I hope will be explored by the next panel.

First, the central banks could have different objectives. How similar, for example, is the U.S. dual mandate--full employment and price stability--to the singular or, more precisely, hierarchical mandates for the ECB and the Bank of England? Under hierarchical mandates, price stability or low inflation is typically the principal objective, and other objectives can be pursued only once the inflation objective is met.

Second, there could be different strategies for achieving a given set of objectives. Of course, the most important factor in determining monetary policy strategy is likely the nature of the objectives themselves, so any differences in strategy may be related to the first question. At any rate, is a Taylor-rule type of approach--well aligned with a dual mandate and therefore a simple representation of the U.S. policy response--different in practice from the flexible inflation-targeting strategy of the Bank of England and the two-pillar strategy of the ECB? If so, what are the consequences of the differences in strategy for how central banks respond to unexpected developments?

Third, what determines how effective a central bank is in meeting its objectives, given its strategy? This takes us into the main subjects that this program will cover. To what extent does the process followed in making monetary policy shape the details and effectiveness of the outcomes? Does the nature of the governance and decision process matter? For example, does it matter whether there is a single governor or a committee; whether there are outsiders as well as insiders on the committee; whether there is individual or collective responsibility for the decisions of the committee; and whether the committee is made up only of centralized members or also includes regional representatives? Does the effectiveness of policy depend on the central bank's skill in communicating the intent behind its policy and the future direction of policy and, therefore, on the transparency of the process itself? Does the effectiveness of the policy depend on the instrument independence of the central banks as well as the overall relationship between the central bank and the rest of government?

It is interesting to try to identify differences in practice among central banks. But is the more meaningful story a convergence in practice around the world? Is there any real difference between a monetary policy executed under dual mandate and under flexible inflation targeting? If central bank practices are converging, is this due to a convergence in the economic environments or to greater acceptance of a common intellectual framework regarding what central banks can and should do as well as how those goals can be achieved?

The Politics of Monetary Policy
Well I have finally reached my assigned topic--the politics of monetary policy. I use the term to describe the relationship between the central bank and the rest of government. In an earlier paper on this subject, I emphasized the way this relationship balances independence and accountability. But the relationship between the central bank and the rest of government typically has four dimensions.

First, the legislature will usually limit central-bank independence by imposing a mandate--a specific set of responsibilities or objectives.

Second, the legislature will often ensure instrument independence for the central bank, thus insulating monetary policy from the short-term political pressures related to the electoral cycle. The mandate and instrument independence usually reflect formal institutional arrangements, typically set by statute (or, in the case of the ECB, by international treaty).

Third, the central bank will typically engage the legislative and executive branches in a variety of informal interactions. Such informal relationships may play a particularly important role in facilitating the coordination of monetary and fiscal policies and, in the case of the legislature, also allow for appropriate oversight.

Fourth, the legislature typically establishes procedures to hold the central bank accountable for its policy decisions. As I noted in my earlier paper, instrument independence can appear inconsistent with democratic ideals. Therefore, there is generally an attempt to balance independence and accountability. This provides incentives for good performance and maintains the connection between the monetary policy process and publicly elected government officials.

It is important to appreciate that steps to encourage accountability typically offer opportunities for political pressure. The history of the Federal Reserve's relationship with the rest of government is marked by efforts by the rest of government both to foster central-bank independence and to exert political pressure on monetary policy.

The Mandate
Let me comment briefly on each of these relationships. The mandate is, of course, central to the conduct of policy, or, at least, it should be.

The Fed's dual mandate--full employment and price stability--is really quite unique. Actually, the specific language of the mandate is that the Fed should promote price stability and maximum employment. We presume that the Congress did not intend to give us contradictory objectives, so we interpret the objectives as price stability and maximum sustainable employment. Maximum sustainable employment is also sometimes referred to as full employment, the maximum level of employment sustainable without upward pressure on inflation.

The key to the dual mandate is that the two objectives are sought at the same time, though the relative weights are left to the discretion of the central bank. Canada, Great Britain, and the ECB, on the other hand, all have hierarchical mandates that single out price stability as the principal objective and allow the central bank to pursue other objectives only after the price stability objective has been achieved. Central banks operating under hierarchical mandates typically set their inflation objective in terms of a forecast or as a medium-term objective, allowing a gradual return to their objective if inflation deviates from it. In addition, the different regimes appear to have converged over time. It is harder to tell the difference between what Lars Svensson refers to as flexible-inflation-targeting and dual-mandate regimes. So the question with respect to the mandate remains: Does it affect the policy outcomes and macroeconomic performance among the countries represented on the first panel?

Assuming there is no long-run trade-off between inflation and unemployment and assuming that the economy gravitates to full employment in the long run, central banks will achieve full employment and price stability in the long run under either mandate. But the nature of the mandate could affect their choice of where their policies might put them on the trade-off between output variability and inflation variability. Does the dual mandate make the Fed more responsive to downward demand shocks? Does the hierarchical mandate permit Canada, Great Britain and the euro area to more tightly control inflation over the shorter run? Or do practices among these central banks differ less than it might appear from a strict reading of their mandates? By the way, I only have time to ask the interesting questions, not to answer them. That will be the panel's job, I hope.

Instrument Independence
Each of the central banks represented on the panel has instrument independence. The motivation for granting such independence is to insulate the central bank from political interference, especially interference motivated by the pressure of elections to deliver short-term gains irrespective of longer-term costs. The purpose of this insulation is not to allow the central bank to pursue whatever policy it prefers--indeed, as I just noted, governments invariably define the broad goals for the central bank--but instead to provide a credible commitment of the government, through its central bank, to achieve those goals, especially price stability.

Central bank independence is in part the result of formal institutional arrangements typically incorporated in the legislation defining and creating the central bank. The most important requirement is that the central bank is the final authority on monetary policy decisions. That is, monetary policy decisions should not be subject to the veto of the executive or legislative branches of government. This is further protected if other institutions of the government--typically the Treasury Department or the Ministry of Finance--are not represented on the monetary policy committee. A lesser protection would be to allow such representation, but only in a non-voting capacity, as is the practice at the Bank of England and the Bank of Japan.

Instrument independence is facilitated by long overlapping membership terms, by limited opportunities for reappointment, and by protecting committee members from removal, except for cause--where cause refers to fraud or other personal misconduct but explicitly excludes differences in judgment about policy. An intangible contributor to independence is the appointment of a capable, respected, politically astute and "independent minded" chairman. Countries have many of the same protections of instrument independence. An interesting question, however, is whether the differences in institutional arrangements across central banks materially affect the degree of instrument independence in practice. Do the shorter terms for members of the Bank of England's Monetary Policy Committee affect their degree of independence relative to the longer terms for Fed governors? Does the presence of non-voting representatives of the executive branch on the committees in Britain and Japan lessen their effective degrees of independence compared to the United States?

Relationships in practice can be different from what they are in principle, so the independence of the central bank depends on the evolution of informal as well as formal relationships between the central bank and the rest of government. The history of the Federal Reserve is replete with examples of attempts by presidents and senior administration officials to influence the conduct of monetary policy, especially in periods leading up to elections. Sometimes this takes the form of appointments--including specific instructions to appointees to vote in a particular way. Sometimes it takes the form of pressure on the chairman and occasionally on other members of the FOMC.

The Fed has--throughout this process--built a reputation for independent policymaking and has had the benefit of strong chairmen to sustain this direction. But independence is not something that should be taken for granted. Rather, it is a work in progress that, now and again, will have to be defended and protected.

Informal Interactions and Policy Coordination
One potential tension is between independence and policy coordination. There is a fascinating and quite famous, or infamous, episode involving the FOMC and the President in 1965. President Johnson did not want the Administration's stimulative fiscal policy to be undermined by restrictive monetary policy. Chairman Martin, on the other hand, supported an increase in the discount rate as an appropriate step to contain the risk of higher inflation. A key vote on a proposed increase in the discount rate occurred at a Board meeting on December 3. Not only did the President try to influence the Chairman's position, but others in the Administration put pressure on other members of the Board. The Board voted 4-3 to support the Chairman. Following the vote the President summoned the Chairman to the President's ranch in Texas. But the vote stood. The independence of the Fed was preserved and indeed used precisely for the purpose it was intended. It is now widely held that raising the discount rate was the correct decision.

The minutes of that Board meeting make interesting reading. The debate was not just about the importance of asserting the independence of the Fed. It was also about the importance of facilitating policy coordination. Some members of the Board were reluctant to go counter to the policy direction of the Administration, and preferred negotiation, given that the Administration also deserved a voice in stabilization policy.

The coordination of monetary and fiscal policies has not been much of an issue since at least the early 1980s. At that time, fiscal policy shifted focus to longer-run issues--such as balancing the budget, or setting a tax policy that supports economic efficiency and long-term growth. This left stabilization policy virtually exclusively to the Federal Reserve. History reveals that this division of labor did not eliminate tensions over monetary policy decisions. On many occasions, an Administration voiced its displeasure--publicly or privately--with monetary policy actions. These communications were, as you would expect, biased--they were all in response to decisions to raise interest rates and most often occurred when those increases came shortly before an election.

The change in views about the role of tax-rate and spending adjustments as part of an activist stabilization policy significantly diminished any need for ongoing coordination between monetary and fiscal policies. The main issues of coordination in recent years have been about how monetary policy should adapt to significant shifts in the direction of fiscal policy, which are motivated in turn by the longer-run considerations. Recent examples include the monetary policy responses to the Clinton Administration's deficit reduction direction and, going forward, to the Bush Administration's longer-run fiscal program. I have described this kind of coordination as sequential decisionmaking. Fiscal policy is slower moving. Once fiscal policy is decided upon, the Federal Reserve takes that policy into account (and/or takes into account its expectations of future fiscal policies) when it sets monetary policy. However, if fiscal policy wanted to play a more important role in active stabilization policy, the potential for tension between the administration and the Congress and the Federal Reserve could increase, especially at times when the fiscal and monetary authorities have different views about the cyclical situation or if they have different objectives.

The relationship between the Federal Reserve and the executive branch was exceptional during the Clinton Administration. Early indications point to a continued excellent relationship with the new Administration. In contrast, the recent relationship between the Bank of Japan and the rest of the Japanese government seems not to have been so good. This undoubtedly reflects in part the fact that the Bank of Japan and the Ministry of Finance are still working out their respective roles following the Bank gaining greater independence. An interesting question, however, is whether the tensions have affected policy outcomes and macroeconomic performance. It sometimes appears that the result has been a stalemate or the outcome of a noncooperative game.

For example, has the Bank of Japan been reluctant to engage in bolder monetization strategy in part because of differences in policy priorities and tensions with the Ministry of Finance? The Bank of Japan seems to believe that operations in longer-term government bonds reduce the incentive of the government to move toward fiscal consolidation. Therefore, monetary policy in Japan might be affected not only by views about how such policies would affect macroeconomic performance, for given fiscal policies, but also by views about how fiscal policy might adjust to monetary policy. Another way to pursue more stimulative policy, even once the nominal policy rate is driven to zero, might be to carry out open market operations in foreign bonds--in effect, unsterilized foreign exchange intervention. But foreign exchange intervention is at the discretion of the Ministry of Finance, not the Bank of Japan, so this direction might at least give the appearance of ceding control of the timing and magnitude of monetary policy actions to the Ministry of Finance. Such an appearance would be a problem even under the best of relationships, but such a direction may be still less likely when there are tensions between the two parties and when the independence of the central bank is so recent. Finally, the Bank of Japan may believe that there are limits to what monetary policy alone can accomplish and, given the uncertainty about the effects of monetization, may resist moving in this direction until the government moves decisively to deal with banking problems and the overhang of corporate debt and more boldly to open markets to domestic and international competition.

Coordination of monetary and fiscal policy in Europe has become much more complex, of course, because the ECB formulates a unified monetary policy, while fiscal policy still is largely determined at the national level.

Procedures Related to Accountability and Transparency
Accountability is facilitated by providing the central bank with a specific, external (usually legislatively imposed) mandate. Two aspects of designing the objectives for monetary policy are important. First, a single objective (typically price stability)--or even a hierarchical mandate--makes the central bank more accountable, because multiple objectives always carry trade-offs, at least in the short run, that are subject to discretion by the central bank. Second, explicit numerical targets make a central bank more accountable than more general targets. Specifically, an explicit numerical inflation target makes central banks more accountable than a more general commitment to price stability.

The assignment of a well-articulated and tight mandate seems to be importantly related to instrument independence. The well-articulated mandate for the Federal Reserve is a surprisingly recent development; it became part of the Federal Reserve Act only in 1977. In this case, the timing of the mandate did not correspond to any change in the degree of independence of the Fed. But, many recent examples exist of central banks being given tighter mandates at the same time they were given formal instrument independence. The tight mandate leaves the most important decision--that of the goals of policy--in the political process, and it also facilitates the accountability of the newly independent central bank. This was the case for both the Bank of England and the Bank of Japan.

Another consideration in the setting of the mandate is a trade-off between accountability and flexibility. Flexibility can be a valuable asset for policymakers, given the variety of shocks that the economy may face, the structural changes that could affect the nature of trade-offs confronting policymakers, and the possibility of trade-offs among multiple targets. Less precise targets and multiple objectives provide such flexibility for the policymaker. The Federal Reserve, with both a dual mandate and a less precise inflation target, therefore appears to have greater flexibility compared to central banks that operate under singular or hierarchical mandates and have explicit, numerical inflation objectives. Does this matter in practice?

There appears to have been some evolution in inflation-targeting regimes toward greater flexibility. It could be that the effort to sharpen accountability through a very tight mandate was viewed as most important immediately after the granting of instrument independence. Such tight mandates might also have been especially important at this juncture, which typically followed a period of poor inflation performance and, hence, diminished credibility with respect to price stability. The tendency may be natural to allow more flexibility when central banks are focused on inflation maintenance than when they seek inflation reduction. More recently, inflation performance has been generally very good and inflation-targeting regimes over time appear to have become somewhat more flexible. Indeed, Svensson has characterized many central banks as now operating under "flexible" inflation-targeting regimes, where the flexibility allows them in practice to take account of smoothing output relative to full employment, perhaps making such regimes observationally equivalent to those operating under an explicitly dual mandate. Is this true? In addition, central banks can also increase their flexibility in an inflation-targeting regime by setting their objective in relation to an inflation forecast, rather than current inflation performance, or by focusing on core, as opposed to headline, inflation--directions in which inflation-targeting regimes appear to have evolved over time.

Once we allow for multiple targets and imprecisely defined objectives, the distinction between goal and instrument independence may become blurred. Multiple targets that give central banks discretion over the relative weights to be assigned to the objectives--and hence, with respect to short-run trade-offs, between the targets--in effect give the central bank some important discretion over objectives. Similarly, the flexibility afforded by imprecisely defined targets also leaves some discretion about goals with the central bank.

A second source of accountability is through the reappointment of committee members. If terms are short and, especially if the chairman and the other voting members can be reappointed for additional terms, more control can be exercised through the reappointment process, and committee members can more easily be held accountable for their policy votes. This is another clear example of the trade-off between independence (facilitated by long terms without the possibility of reappointment) and accountability (facilitated by short terms with opportunities for reappointment).

A third source of accountability is oversight hearings on monetary policy held by legislative committees. In the United States, we have semiannual congressional monetary policy hearings--formerly known as the Humphrey-Hawkins testimony. While many focus on the relationship between the central bank and the executive branch, central banks are typically creatures of legislatures who, in turn, control the degree of the bank's independence. The ECB again is unique as a creature of an international treaty and its independence cannot therefore be eroded except by unanimous agreement of the signatories to that treaty. The ECB participates in oversight hearings by the European Parliament, but that body has no authority over the Bank. This makes the ECB, without question, the central bank with the greatest degree of instrument independence and the least degree of political accountability.

Transparency and disclosure are also essential to accountability. The legislature, for example, needs information about the policy actions and an accounting of the rationale for the policy if it is to hold the central bank accountable. In addition, it is generally agreed that markets work better with more complete information, though some worry that a continuous flow of information about the leanings of members of the policy committee can result in excessive volatility in financial markets.

Transparency can also contribute to policy effectiveness as well as to accountability. Monetary policy works not only by setting the policy rate, but by conveying expectations to the markets about the future course of monetary policy. This may be done intentionally or unintentionally, but markets will make their own attempt to assess the future direction of policy and will try to glean whatever they can from policy statements, speeches of policy committee members, and so forth. To the extent that policymakers expect to ease or tighten in the future, conveying an expectation that policy rates will move in a given direction typically will immediately change long-term rates and asset prices in ways that support the objectives of that policy. That is, longer-term interest rates may move sooner than would otherwise be the case, in effect reducing the lags in the effect of monetary policy on aggregate demand.

Expectations, however, can also constrain monetary policy. If policymakers convey, perhaps unintentionally, expectations of a move at the next meeting, but policy does not in fact deliver that change, longer-term rates may reverse their earlier move. The only way to avoid surprises is, of course, always to deliver to the markets what they expect. And surprises are inevitable and desirable when the market gets it wrong. But concerns about such surprises, especially at times when financial markets are skittish, may complicate the policy decision.

This tension between encouraging more effective policy by guiding private-sector expectations while, at the same time, avoiding policy being unduly affected by the very same private-sector expectations, in my view, was central to the Fed's decision in January 2000 to change the way it communicated to the market. Before January 2000, the FOMC indicated whether policy was more likely to remain unchanged or to move in one direction or the other by its choice of a symmetrical or asymmetrical posture. Thereafter, it communicated only its judgment about the balance of risks to the forecast. While the latter obviously also had implications for future policy, the new approach might be viewed as a way to guide expectations while limiting the market's attaching the same degree of immediacy and certainty to expected future policy actions.

I expect that both panels that follow will discuss the transparency of central banks around the world--especially those represented on the first panel--and perhaps the degree to which this transparency contributes to the effectiveness of policy. I hope they will discuss to what extent central banks can and should guide private-sector expectations about future policy and how they can avoid situations in which they may feel pressure to ratify private expectations to avoid market disruptions. This is part of a broader question about the best amount, frequency, and nature of information to provide the public about the process of setting monetary policy and the details of the debate.

2001 Speeches