|Remarks by Governor Roger W. Ferguson, Jr.
Before the The National Economists Club, Washington, D.C.
September 9, 1999
Transparency and Responsibility in Monetary Policy
It is a pleasure to address the National Economists Club. I understand that the Club is an educational organization with the goal of encouraging and sponsoring the public discussion of significant economic issues. This is a worthy goal, and I am pleased to support it.
Today I would like to spend a few minutes discussing the issue of transparency and responsibility in monetary policy. I am sure you will all agree with me that the first task of the monetary authority is to get the right policy setting to achieve national economic objectives. However, that task is complemented by and intertwined with another -- telling the public, both the general public and active market participants -- what the central bank is trying to accomplish, what forces it sees impinging on meeting its goals, and how it might cope with those forces.
The latter job of communication has received increasing attention in recent years as central banks worldwide have moved toward greater "transparency" in their operations, objectives, and economic assessments. In my view, this trend is healthy and should be extended when it would be useful. But we must be mindful that this new openness can have an impact on how effectively central banks pursue their primary mission -- making the best possible monetary policy. I want to take a few minutes to review the actions taken at home and abroad toward greater transparency, and assess some of the issues that arise as central banks contemplate further steps in this regard.
Of course, the views expressed here are my own and should not be interpreted as the position of the FOMC or the Board of Governors.
The Recent Trend Towards Openness and Transparency
Let me start this discussion by focusing on the United States. The Federal Open Market Committee in recent years has increased in several ways the transparency with which policy decisions are made and implemented. In 1994, the FOMC began formally announcing, immediately after any meeting in which a policy action had taken place, the change in the targeted federal funds rate and a brief rationale for the decision. Until that time, changes were "signaled" through open market operations. It sometimes took several days before market participants and the broader public were certain that a new policy setting was in effect. In that earlier regime, the public rarely received a prompt explanation for the change.
This year, the FOMC went a step further and began a policy of communicating major shifts in its views about future policy even when the current policy setting has not changed. The idea is that providing more information about the Committee's views of the economic outlook may allow financial market prices to reflect more accurately the likely future stance of monetary policy. The Committee made its first such announcement after the May meeting. Both the markets and the Committee are learning to live with the new disclosure policy, and our experience will help to guide any subsequent modification of these practices.
As technology continues to reduce the costs of disseminating information, and as Americans participate in financial markets in greater numbers, the Federal Reserve has endeavored more broadly to improve communication with the public about its views and actions. A major component of that effort is our website. The texts of policy announcements, testimony, speeches, and minutes of meetings are available through this medium simultaneously with their initial release or presentation. As a consequence, the public and market participants can form their own views on the implications of these texts, and do not rely solely on press reports for interpretation.
These changes have been part of a broader trend among central banks in the industrialized world. For example, in 1997 the Bank of England took a number of steps to increase the openness and transparency of its policy making. It now announces monetary policy changes immediately after meetings, provides press releases or conferences at pre-announced times following monetary policy meetings, and has shortened considerably the lag time in releasing its meeting minutes. The press conference announcing its quarterly inflation report is also now televised.
Others have become more transparent as well. In 1998, the Bank of Japan's governing law was revised. The Bank now announces policy changes immediately after meetings, publishes minutes of the meetings, and produces two formal reports each year on monetary policy, which it explains to the Diet.
The new European Central Bank incorporates a number of these aspects of openness. It too announces policy changes immediately after meetings, along with a detailed rationale in a press release or conference. Furthermore, it intends to sometimes provide occasional post-meeting announcements that give a sense of the direction of future policy when the current policy setting has been left unchanged. The ECB also publishes a Monthly Bulletin, which provides an assessment of monetary, financial, and economic developments, and explains monetary policy decisions of the Bank.
Openness and the Effectiveness of Policy
I believe that there are two main forces driving this move towards greater openness and accountability. The first is that openness may improve the effectiveness of monetary policy. The Federal Reserve controls a very short-term interest rate -- the federal funds rate -- but it is the longer-term interest rates at which businesses and households borrow to finance spending on capital goods, homes, and durable goods which matter most for the economy. Those longer-term rates reflect expectations of future short-term rates, as well as a premium for uncertainty. If the monetary authority can be clearer about what it is doing now and plans to do -- not in the sense of setting future moves in stone, but rather in terms of explaining risks that might influence future policy -- then market participants can improve their expectations of future short rates, and possibly reduce the premium for uncertainty. Both of these changes ought to bring the rates that matter most for the macroeconomy into closer alignment with the intentions of monetary policymakers.
Openness on the part of the central bank does not guarantee that markets will accurately forecast interest rates, however. Markets can err because the economy changes or because, despite greater openness, they don't fully understand the central bank's intentions. One aspect of this seems to be a tendency to extrapolate the recent trend of rates into the future. For example, after a series of tightenings in 1994, the federal funds rate topped out at 6 percent, and short-term Treasuries remained below that level for virtually the entire subsequent five-year period. During this period of tightening five-year Treasury yields soared to nearly 8 percent, a rate never fully justified by the extent of firming in short-term rates, suggesting that greater openness by policymakers does not preclude the natural uncertainty involved in interest rate forecasting.
I might note that the decline in the usefulness of the monetary aggregates may have contributed to the type of openness and transparency in use today. The regime targeting money growth was, in a sense, a transparent one -- the targets were contained in the biannual Humphrey Hawkins report and widely known, and money numbers were published weekly. As the monetary aggregates became less reliable and consequently receded in importance, maintaining openness called for increased communication about the rationales for discretionary interest rate policy.
I do not wish to leave the impression that market participants accurately anticipating policy actions is sufficient for central bankers to play their assigned role. While a tighter link between monetary policy and long-term interest rates may lessen the need for the central bank to move its policy instrument as much or as quickly, it should be clear that markets are not a substitute for monetary policy action. Market movements in response to anticipated central bank actions must be validated when they are, in the view of the policymakers, correct, in order for the market to continue to reflect the intentions of policymakers.
Openness, Independence, and Democracy
A second force behind the trend towards openness and transparency is the increased importance attached to, and increased prevalence of, independent central banks. The notion that central banks need to be insulated from political pressures, especially regarding the financing of national budgets, has been embraced and implemented throughout much of the developed world. And, in many cases, openness and transparency on the part of the central bank have been something of a quid-pro-quo for that independence. Here I am referring to independence of instrument rather than independence of goal. It is appropriate that in a democracy voters' representatives should decide the central bank's ultimate goals, as they have in the U.S., for example, by providing the Federal Reserve with a mandate to pursue stable prices, maximum employment, and moderate long-term interest rates.
The main reason that central banks need to be free of political pressure in using their instruments is that effective monetary policy requires a distant time horizon. Because changes in monetary policy generally affect output well before inflation, policymakers with nearer time horizons could be tempted to favor output in excess of capacity, at the expense of higher inflation in the future, and the well-known and inevitable disruptions that accompany it.
Most countries have come to believe that the solution to this problem lies in providing the central bank with independence in choosing settings for its policy instruments, and with an institutional structure protective of that independence as well. For example, Federal Reserve governors' terms are long and staggered, the Federal Reserve's budgets are not subject to the congressional appropriations process, and it has no obligation to purchase or support the prices of Treasury debt.
Along with their independence, central banks are required to account for their decisions in various ways. This too is appropriate in a democracy: the public has a right to know what an unelected body as important as the central bank is doing, and why. To some extent, accountability is fostered by openness and transparency because letting people know what you are doing gives them the tools to hold you accountable. The Federal Reserve's accountability is collective, and it is often centered in the office of the Chairman. For example, while the minutes of FOMC meetings generally refer to views of "members of the Committee," the Chairman represents the Committee in testimony and answers questions about the semi-annual Monetary Policy Report to Congress under the Humphrey-Hawkins law. When a committee makes policy, a process involving consensus building and compromise, collective rather than individual accountability seems preferable because policies are not simply weighted averages of members' wishes.
Having outlined some of the advantages of transparency and accountability in central banking, let me note that there are limits to how open central banks can practically be. The Federal Reserve, like many but not all central banks, releases minutes after its next regularly scheduled meeting. The ECB, representing a potentially more diverse set of national constituencies, has decided neither to release minutes nor to publish the voting behavior of its individual members. In this way, it hopes to avoid pressures to act in what may be perceived to be a national interest rather than in the interest of the euro area as a whole. Similarly, most central banks omit committee members' names in the discussion portion of the meeting minutes, for fear that attributing remarks during the discussion would undermine the free ranging nature of meetings and encourage members to arrive with prepared positions. In fact, many members of the FOMC now do rely on prepared remarks for certain segments of the meetings, in part due to the greater transparency that they face, given that in the mid-1990s transcripts of meetings started to be released after five years. To the degree that policy decisions benefit from the give and take of opinions and evidence at meetings, something less than full openness to public scrutiny is required to assure the quality of monetary policymaking.
Price Stability, Inflation Targeting, and Credibility
While central banks' mandated goals often include many things, price stability has emerged as the preeminent one, either formally -- as in inflation targeting countries or the Maastrict Treaty establishing the European Central Bank -- or informally, through a greater understanding of how economies work. That is, policies that foster low inflation are more likely to deliver high employment and maximum sustainable output growth as a by-product, while policies which aim directly to keep employment and output high are more likely to result in poor inflation performance. To paraphrase Chairman Greenspan, the economy works best when inflation is so low that businesses and households do not have to take it into account when making everyday decisions. In the short run, however, difficult choices may be required about the timing and magnitude of policy actions consistent with achieving price stability.
One approach to such difficult choices, as I have mentioned, is to mandate a quantitative inflation target for the central bank, or to have the central bank select an inflation target that it must endeavor to achieve. A number of other countries, like Britain, have adopted inflation targeting monetary policy regimes. In many cases, that shift has been part of a move toward greater independence of central banks. Such regimes might lend themselves to both openness and accountability. A single quantitative goal might focus explanation of policy actions and provide a simple yardstick for assessing performance. However, this approach faces at least two sets of important issues.
The first is what to do about supply shocks, like large increases in oil prices, which tend to increase both inflation and unemployment. Many of the countries that have adopted inflation targeting avoid having to respond to the first-round effects of these disturbances by targeting a "core" price index that excludes goods like food and energy that are often subject to shocks. However, increases in raw materials prices will still tend to pass through to higher prices for other goods, and thus raise core inflation, perhaps above the target rate or range. In such cases, bringing inflation back down rapidly may entail high costs in terms of unusually elevated output gaps and unemployment rates.
A related issue is which other price changes, beyond those induced by supply shocks, the central bank should ignore in inflation targeting. Some central banks use indexes that exclude certain prices, such as mortgage rates, in order to avoid the perversity of tightening policy feeding into higher measured inflation for the targeted index. Occasionally, central banks effectively make special adjustments for other price changes, such as excise tax hikes, that clearly have an impact on "core" price indexes but do not signal an imbalance between demand and potential supply that might give rise to broader inflationary impulses.
Inflation-targeting regimes may allow some consideration of real-side costs either by specifying relatively long adjustment periods, to allow a high probability that the central bank can bring inflation down to the target within the allotted time, or by including "escape clauses" that grant temporary exemptions for large supply shocks. The question then becomes whether the use of either of these elements of flexibility maintains the credibility of the central bank better than a system of multiple objectives, like that of the Federal Reserve. The longer the policy timeframe, the less content there is to an inflation-targeting regime, and presumably the lower is its credibility. Similarly, the more often a central bank has to declare emergencies, use escape clauses, or otherwise allow price increases to go unchecked, particularly when the rationale for such decisions is difficult to communicate to the public, the less credibility it will have.
Why is credibility so important? There is a large theoretical literature on credibility in central banking, arguing for the most part that central banks need independence to be credible inflation fighters, to be able to disinflate at a lower short-run output, and social, cost. However, the evidence does not support a particular empirical relationship between credibility and costs of disinflation, nor one between independence and costs of disinflation. In the real world, there are two reasons why central bankers still prize credibility, even if it cannot be shown to reduce the costs of disinflation. First, central banks need the latitude to change operating policies when circumstances warrant -- such as de-emphasizing growth rates of monetary aggregates when their velocities become unstable -- without the markets fearing that a central bank's commitment to the goal of price stability has been compromised. Second, credibility is an asset during a financial crisis, when the central bank may need temporarily to take extraordinary measures. For example, it was helpful in the fall of 1998 for the markets to understand that the FOMC's policy easings represented a response to a financial crisis, rather than a reduced concern about inflation.
The second issue with inflation targeting is what rate or range of inflation to choose. A rate above zero (in terms of published inflation indexes) may be appropriate, for a couple of reasons. One is that it may be difficult to conduct monetary policy at levels of inflation near zero because nominal interest rates -- including the central bank's policy rate -- cannot be below zero. Many observers believe that monetary policy in Japan is constrained in this way, with its current policy setting an essentially zero short-term interest rate. A second reason is the statistical biases in inflation indexes, resulting from slow inclusion of new goods, improvements in quality, substitution from more expensive to cheaper goods, and other factors. In the United States, the 1996 Boskin Commission report, estimated that these biases may have overstated true inflation in the consumer price index at that time by three-quarters to one-and-one-half percentage points per year. In such an environment, a central bank mandated to pursue price stability can be flexible according to the circumstances, while one with a numerical target may need to obtain formal modification of its objectives.
The movement towards more central bank transparency, independence, and accountability that has taken place over recent years constitutes an exciting and welcome development. In striking the appropriate balance going forward, the advantages of further steps must be carefully weighed against the risks of impairing the deliberative process or indeed destabilizing financial markets. By proceeding cautiously, but keeping the goal of optimal transparency and accountability in mind, I am confident that central banks will continue to contribute to the national welfare in their respective countries.