The Federal Reserve Open Market Committee is, yes, a committee. How well does it work? Kevin M. Warsh has had a front-row perspective. Warsh was a member of the Federal Reserve Board of Governors from 2006 to 2011--and thus, right through the heart of the Great Recession. He was asked by the Bank of England to review the deliberations of its own decision-making Monetary Policy Committee, and to study and consider how other central banks work as well. Warsh describes his insights in "Institutional Design: Deliberations, Decisions, and Committee Dynamics," which appears as Chapter 4 in Central Bank Governance And Oversight Reform, edited by John H. Cochrane and John B. Taylor, and just published by the Hoover Institution.
How do you set up a committee to have the highest chance of reaching a wise conclusion? Warsh suggests that a combination of high-quality inputs, genuine deliberation, and optimal committee design all play a role. Otherwise, outcomes can fall short. Warsh writes:
The literature identifies numerous interrelated theories that link internal management inadequacies to organizational failure. These include: • Janis’s canonical Groupthink theory (1972, 1982), which highlights the tendency of small, homogenous management teams to make suboptimal decisions; • Hambrick and Mason’s Upper Echelon theory (1984), which links organizational achievements to the composition and background of an organization’s senior management team; • Staw, Sandelands, and Dutton’s Threat Rigidity Effect theory (1981), which explains the tendency of management groups to stick rigidly to tried and tested techniques at times of threat and challenge, thereby increasing the risk of organizational failure among incumbents at times of secular change.
Does the Fed Open Market Committee operate in a way that seems likely to gain the benefits of committees and minimize the costs? Here's some basic background comparing the Fed to policy decision-making committees at other central banks.
Warsh writes: "The FOMC’s institutional design is not inconsistent with sound practice. But there are certain institutional aspects of the FOMC which differ somewhat from best practice, at least as identified in the literature." Here are some examples of what he has in mind.
Successful committee don't have too many participants. "By statute, the FOMC includes twelve voting members. ... Policy deliberations, however, occur in a much larger institutional setting. Nineteen people convene in the discussion (voters and non-voters alike) and a total of about sixty people are in attendance, including a range of subject-matter experts on key
aspects of the economic and financial landscape."
The members of successful committees have independent information that they can bring to bear. "While the Reserve Bank presidents are supported by large,independent staff s of economists to help inform their forecasts and policy judgments, I would note that the economic models and
forecasting tools are substantially similar across the Federal Reserve System. Th is explains, in part, the remarkable conformity of the so-called dot plots in the projections from FOMC participants."
A lack of disagreement suggests an insufficient breadth of views. W"One simple mechanism for evaluating the breadth of views is to review trends in dissent: that is, the number of FOMC members who voted against the majority policy stance. By both FOMC tradition and practice, the bar for lodging a dissenting vote is high. Neither Chairman Greenspan nor Chairman
Bernanke ever cast a vote in the minority. In contrast, the governor of the Bank of England was outvoted on nine occasions since 1997. And governors of the Federal Reserve, unlike Reserve Bank presidents, only rarely dissented in casting of votes. In the past decade, for example, there has been only one instance of dissent by a sitting governor."
In successful committees, people are willing to express their unvarnished opinions. But since the Fed started publishing transcripts of its meetings in 1993, albeit with a lag, "Meade and Stasavage (2008) find evidence that the Fed’s post-1993 transcript policy led to deterioration in the quality of FOMC
deliberations. In the authors’ formulation, policymakers are motivated to achieve two goals in the policymaking process: making optimal policy decisions and garnering a good reputation in public
(often associated with conformity with the prevailing consensus). The existence of public transcripts, even with a lag, caused FOMC participants to voice less dissent in the meetings themselves and to
be less willing to change policy positions over time. For example, the number of dissenting opinions expressed by voting members fell from forty-eight (between 1989 and 1992) to twenty-seven (between 1994 and 1997)."
In his comments on the Warsh paper, Peter Fisher, who spent a number of years at the New York Fed in the 1980s and 1990s, summarized what he viewed as Warsh's message in this way:
Successful committee don't have too many participants. "By statute, the FOMC includes twelve voting members. ... Policy deliberations, however, occur in a much larger institutional setting. Nineteen people convene in the discussion (voters and non-voters alike) and a total of about sixty people are in attendance, including a range of subject-matter experts on key
aspects of the economic and financial landscape."
The members of successful committees have independent information that they can bring to bear. "While the Reserve Bank presidents are supported by large,independent staff s of economists to help inform their forecasts and policy judgments, I would note that the economic models and
forecasting tools are substantially similar across the Federal Reserve System. Th is explains, in part, the remarkable conformity of the so-called dot plots in the projections from FOMC participants."
A lack of disagreement suggests an insufficient breadth of views. W"One simple mechanism for evaluating the breadth of views is to review trends in dissent: that is, the number of FOMC members who voted against the majority policy stance. By both FOMC tradition and practice, the bar for lodging a dissenting vote is high. Neither Chairman Greenspan nor Chairman
Bernanke ever cast a vote in the minority. In contrast, the governor of the Bank of England was outvoted on nine occasions since 1997. And governors of the Federal Reserve, unlike Reserve Bank presidents, only rarely dissented in casting of votes. In the past decade, for example, there has been only one instance of dissent by a sitting governor."
In successful committees, people are willing to express their unvarnished opinions. But since the Fed started publishing transcripts of its meetings in 1993, albeit with a lag, "Meade and Stasavage (2008) find evidence that the Fed’s post-1993 transcript policy led to deterioration in the quality of FOMC
deliberations. In the authors’ formulation, policymakers are motivated to achieve two goals in the policymaking process: making optimal policy decisions and garnering a good reputation in public
(often associated with conformity with the prevailing consensus). The existence of public transcripts, even with a lag, caused FOMC participants to voice less dissent in the meetings themselves and to
be less willing to change policy positions over time. For example, the number of dissenting opinions expressed by voting members fell from forty-eight (between 1989 and 1992) to twenty-seven (between 1994 and 1997)."
In his comments on the Warsh paper, Peter Fisher, who spent a number of years at the New York Fed in the 1980s and 1990s, summarized what he viewed as Warsh's message in this way:
"I had almost ten years at the FOMC table ... I thought I understood the awkwardness of group accountability when more than once I saw the FOMC gravitate toward no one’s first choice and virtually no one’s second choice, and we ended up with third-best outcomes. But now I’m also worried about individual accountability of a pseudo-nature, which I’m afraid is the regime we now have ..."
Preface
By John H. Cochrane and John B. Taylor
Chapter 1: How Can Central Banks Deliver Credible Commitment and Be “Emergency Institutions”?
By Paul Tucker
Chapter 2: Policy Rule Legislation in Practice
By David H. Papell, Alex Nikolsko-Rzhevskyy and Ruxandra Prodan
Chapter 3: Goals versus Rules as Central Bank Performance Measures
By Carl E. Walsh
Chapter 4: Institutional Design: Deliberations, Decisions, and Committee Dynamics
By Kevin M. Warsh
Chapter 5: Some Historical Reflections on the Governance of the Federal Reserve
By Michael D. Bordo
Chapter 6: Panel on Independence, Accountability, and Transparency in Central Bank Governance
By Charles I. Plosser, George P. Shultz, and John C. Williams