Vice Chairman Donald L. Kohn
At the American Economic Association Session, Allied Social Science Associations Annual Meeting, New Orleans, Louisiana
January 4, 2008
Expertise and Macroeconomic Policy, Comments on "Insiders versus Outsiders in Monetary Policy-Making," by Timothy Besley, Neil Meads, and Paolo Surico
This panel is dealing with an important topic--in effect, the design and functioning of monetary policy committees. Monetary policy decisions are judgment calls, informed by forecasts and discussions about how the economy is likely to evolve, alternative possibilities, and potential responses to policy actions. The question is how to bring the best possible judgments to bear on these questions.
Besley, Meads and Surico's contribution is to ask whether certain characteristics of the policymakers matter for how they vote on the monetary policy committee. For the Monetary Policy Committee (MPC) at the Bank of England, they find heterogeneous views among voters, but no systematic differences based on whether the MPC member is an insider--that is part of the hierarchy at the Bank--or an outsider appointed for a short term just to the MPC; whether the person comes from an academic background; or whether that person has worked at the Treasury. I found the piece well done and interesting, and I have only a few comments before moving on to some questions and observations about the Federal Open Market Committee (FOMC) sparked by reading this research.1
My first comment is that I'm not surprised by the results. The MPC has a well-defined objective given by the government. To be sure, members may differ about the precise time frame in which to meet that objective, but these distinctions should be small relative to the importance of the overall objective. In addition, at every policy meeting, each member of the MPC is operating with the same information from the staff, and before the quarterly meetings that precede an Inflation Report, the members have sat through a number of discussions covering all aspects of the forecast. Everyone should be equally familiar with the key issues and analysis. Finally, dissent is accepted, indeed encouraged in some respects, by insiders as well as outsiders, reducing any differences in incentives facing the two groups.
Nonetheless, some aspects of the paper raised questions in my mind about the robustness of the results. For one, the authors, in testing for differences in policy preferences, assume everyone has the same forecast, thus eliminating one channel through which different backgrounds may manifest themselves. The MPC does go through an extensive internal process when producing its forecast, which as I just noted, may mitigate the differences among members. Still, some of the differences in policy choices probably reflect underlying differences in forecasts, not just differences in Taylor-rule parameters. Certainly, the minutes of the MPC meetings suggest divergence in how members view the outlook, and the Inflation Report routinely notes differences in views about the central tendency forecasts. Small differences, especially if sustained over time, could well have a substantial effect on voting patterns.
In practice, policy decisions are likely to depend on more than the modal forecast used by the authors, such as the degree of uncertainty, the risks around the central tendency, and a weighting of the costs to public welfare from missing the forecast on one side or another. In this regard, the Inflation Reports also note that members may have different views about the balance of risks. I don't know how the authors could go about testing whether the backgrounds of MPC members are systematically related to differences in forecasts or in risk assessments, but surely these are important possibilities.
I was also struck by the finding that the output gap did not play a significant role in policy decisions. As the paper notes, the MPC takes account of the gap in its inflation forecast, but the results suggest members give no independent weight to deviations of output from potential. On the surface, this may suggest that either the MPC has faced only demand shocks--unlikely given the rise in energy prices--or it is not really a flexible inflation targeter. Indeed, figure 1, column 3, indicates that many MPC members have significantly negative coefficients on the output gap. I suspect the reality is that members rely on much more than the projected output gap to judge the future balance of demand and potential supply, but the econometric results on the output gap suggest the need for further explanation.
Finally, the authors find that the change in the policy rate at the last meeting matters for each person's vote at the current meeting. It is not clear why the decision to change rates at the last meeting should affect judgments on the appropriate policy at the current meeting, independent of the outlook. Perhaps the dissenters believe they've made their point and unless they feel especially strongly they don't need to dissent again, but the rationale for this variable should probably be explored further.
The purpose of bringing outsiders onto the policy committee of the Bank of England is to assure that a variety of views will be considered in the policy debate. I'll spend the rest of my time addressing how this works at the Federal Reserve. I'll do this in the form of questions about the personnel and functioning of the FOMC, with my tentative answers.
1. Does it matter who is on the FOMC other than the Chairman?
My answer is yes, and I think I have reasons other than simple self-esteem. To be sure, the Committee looks to the Chairman to provide strong leadership. The Chairman's special role is partly legal and structural--as the chair of the Board and the Committee and as the Committee's representative in front of Congress where the FOMC is held accountable for its actions. In addition, the Committee is so large--in effect nineteen people--that strong leadership is required to construct a policy position that can command a substantial majority and to present a coherent rationale for that policy to the public. But in my experience, the other members of the Committee are also important for good policy. Effective Chairmen cannot operate independently of the sentiment on the Committee; ultimately, the authority of the Chairman rests on his ability to persuade the other members of the Committee that the choices they are making under his leadership will accomplish their objectives. The better the collection of individuals he needs to persuade, the better the policy is likely to be. Individual members can exert considerable influence over time through the cogency of their arguments, and having a variety of perspectives on the Committee will strengthen this process. In addition, all FOMC members play important roles in explaining policy decisions to the public, and how they present their views can influence public attitudes, perceptions, and expectations.
2. Are there insider/outsider types of distinctions on the FOMC?
Although the process by which people come onto the Committee leads naturally to considerable diversity, the insider/outsider distinction does not apply cleanly to the FOMC. Board members of course are appointed by the President with the approval of the Senate. The members of the Board are almost always outsiders--the appointment of a staff member, like myself, is very rare. And Board members typically come from a variety of backgrounds, including top-notch academic economists, but also bankers and others with experience in the financial arena, and occasionally business people. We are appointed for fixed terms, encouraging independent views.
Reserve Bank presidents owe their appointments to both the board of directors of their regional Reserve Bank and the Board of Governors in Washington. Presidents can be either insiders--long-time staff members--or outsiders with relevant experience in academia, public policy, or the financial sector. Once appointed, a president can serve until retirement at age 65 or for a minimum of ten years, affording scope for maintaining their individual views.
Nonetheless, FOMC members try to find consensus under the leadership of the Chairman, and neither Governors nor presidents dissent very often. Presidents have tended to dissent a little more than Governors. The greater number of dissents by presidents might reflect a number of factors. For one, presidents have their own staffs, which can help support alternative views in preparing for a meeting. Board members share a common staff with the Chairman, and, being in the same building, perhaps have a greater opportunity to influence and be influenced by the Chairman. In this regard, Bank presidents may act like "outsiders" more readily than Board members. At the same time, Bank presidents tend to be longer tenured than Board members, and can contribute the institutional memory one might expect from "insiders" when Board membership turns over rapidly.
3. Why so few dissents?
FOMC members recognize the degree of uncertainty around their judgments. Unless they perceive that a serious misjudgment is being made, they expect, given common objectives, that when they prefer another policy and their analysis later proves correct, the Committee is likely to move in their direction in time to forestall problems developing. And they may see their influence on policy as greater over time if they are usually part of the consensus. Finally, they could perceive that a series of close votes could create uncertainty about policy and the direction of the institution.
The absence of dissents should not be taken as indicating that alternative views are not expressed at FOMC meetings. Often these are manifest in participants' forecasts. Because we are such a large and geographically separated Committee, the FOMC, unlike the MPC, does not attempt to arrive at a common forecast, and the dispersion of individual forecasts illustrates the range of views. For example, participants' projections in October of gross domestic product (GDP) growth for 2008 differed by one full percentage point and differences were still a half a percentage point for 2010. Participants had alternative views about current developments affecting the economy and about its long-run growth properties, and they argued their points in the meeting. Nonetheless, all but one of the voting members was able to agree that the action taken that day was consistent with moving the economy toward meeting the Committee's objectives.
4. Do you need to be a monetary policy expert to contribute?
The Committee is strengthened by having people with a wide variety of backgrounds, including those whose principal area of expertise of lies outside of monetary economics. To some extent, the range of backgrounds on the FOMC reflects the need for expertise to carry out the non-monetary policy responsibilities of the Federal Reserve. Any monetary policy committee needs to have a number of people on it that fully understand and can deal readily with the model behind policy deliberations, and who have an awareness of the lessons of monetary policy history. But other types of expertise also are valuable. Policy works through financial markets, and those with that background contribute to understanding how policy impulses are being transmitted to savers and spenders. Experience in private industry is valuable for gathering and interpreting information from many business sources that we bring to FOMC deliberations.
Any member needs to understand the basic framework to have influence on the debate. But you don't need to be a monetary policy expert. Those who come on the Committee with policy credentials can be very effective immediately. But judgment and experience count. Noneconomists can become very effective as they learn the lingo and the framework. The large size of the Committee and long tenure of members allows time and space for this expertise to develop. One of the most difficult aspects of policymaking is recognizing what you don't know, perhaps because the world is no longer conforming to your models. A high level of monetary policy expertise is not necessarily a good guide to who will possess that skill.
The proof of the pudding is in the eating. Since the late 1970s in the United States and since the establishment of the MPC in the United Kingdom, economic performance in both countries has, on balance, been good. A number of factors have contributed to this result, but the respective monetary policy committees must have played some role.
To be sure, in the United States, the leadership of Chairmen Volcker, Greenspan, and Bernanke has been critical to the results, but I believe a diverse and independent minded group of policymakers has contributed as well, and it is important to maintain this mix and quality. Could the functioning of the FOMC be improved? Certainly, and Chairman Bernanke has moved to further utilize the diverse views of the FOMC members, while maintaining the leadership necessary to keep this diverse group headed in the appropriate direction.