Marvin Goodfriend at the Richmond Fed: Recollections

Marvin Goodfriend joined the Richmond Fed Research Department as an economist in 1978, in the middle of the Great Inflation, and worked there until his retirement in 2005, when he became professor of economics at Carnegie Mellon University. He viewed the 20th century monetary policy experience as an "odyssey" from the gold standard to today's inconvertible paper standard supported by a credible Fed commitment to price level stability.¹ While in Richmond, Marvin participated meaningfully in the latter stages of that odyssey and the substantial progress it represented. I was privileged to work closely with him for most of the time he was in Richmond. I'm happy to have this opportunity to share a few memories of what our Reserve Bank achieved in those years with Marvin's extraordinary intellectual and personal leadership along with a few details of how we achieved it.

I think I can recall the exact time and place I first met Marvin. It was Wednesday, December 28, 1977. Bob Hetzel and I were in a room in the NY Hilton interviewing candidates to join us in our Richmond Fed Research Department. Bob stepped out for a minute. When he returned, he looked at me and said, "G. William Miller." A big question mark must have appeared in a bubble over my head, so Bob added that President Carter had just nominated Miller to succeed Arthur Burns as Fed chair. We returned to interviewing candidates. Little did I know that one of the most important and fortunate events of my life and in the history of the Richmond Fed as a policy unit was about to occur.

When the final candidate of the day knocked on the door, it was getting dark in Manhattan, and I remember being weary. Marvin was the candidate. I don't recall for sure, but I'd bet he was wearing a beige and brown argyle patterned sweater vest. I do recall clearly that my first impression of him was altogether positive. There was none of the awkwardness or resume padding that often characterizes these interviews. I remember thinking "with this guy, what you see is what you'll get." I liked what I saw, and Bob did as well. At the time, we were hiring with the objective of strengthening our department's ability to contribute meaningfully to Fed monetary policy; we wanted the significant resources we were devoting to policy research to be justified by increased influence in the broader policymaking process. We were therefore talking to a number of very well-qualified people that we knew would be hard to attract to Richmond. In the interview, Marvin expressed some interest in what we were doing, but as he left, I recall thinking our chances of attracting him were slim.

What a pleasure to learn not long after that Marvin would accept our offer. I would have been even happier had I known then that he would stay almost 30 years and that my Richmond colleagues and I would enjoy the extraordinary stimulation he would bring to the department's intellectual environment throughout those years.

The central theme that motivates virtually all of Marvin's work is the overarching importance of credibility in conducting monetary policy successfully. To achieve credibility, Marvin thought it essential to keep the Fed independent within the government so that its policy decisions and actions were well separated from fiscal policy, Treasury actions outside the Fed's purview, and partisan politics. As I see it, this theme — which was firmly aligned with Marvin's core personal values — was not only a guiding principle but an enabler in practice of most of his policy positions and proposals. Credibility was not a soft concept to Marvin but a critical precondition of effective monetary policy that should be built into the expectations components of policy models. These views are conventional now, but Marvin embraced them well before they became standard and argued for them relentlessly throughout his career. More compactly, in his own words,

Fed (and other central bank) policies only have lasting effectiveness if the policies are credible to the public, i.e., the public is confident that the Fed's actions are free of political influence or manipulation and seek consistently to advance attainment of the Fed's central mandates of maintaining price stability and promoting maximum sustainable economic growth.²

This theme was a natural and powerful elaboration of the policy positions advocated by Robert Black, Richmond Fed president when Marvin arrived in 1978. It came to encapsulate the Bank's permanent overall position on how the Fed should pursue its mandates

Gearing up

Marvin's arrival required some adjustments by all of his colleagues in the department — essentially a higher bar for our policy research and advice to Black. Actually — and fortunately — an initial round of such adjustments was occasioned by the arrival of Bob Hetzel about four years earlier, fresh from writing a PhD thesis under Milton Friedman. I will be forever grateful to Bob for helping get me ready for Marvin.

In the years following their respective arrivals, both Marvin and Bob argued persuasively that influencing broader Fed monetary policy positions meaningfully would require articulating our Bank's views more forcefully and visibly to economists — inside and outside the Fed — focused on monetary policy. The department initiated a concerted, continuing effort to attract recent PhD graduates from university economics departments recognized for their effective research and influence on monetary policy issues. We set an expectation that all economists would produce high-quality and relevant policy research, seek its publication in top professional journals, and present it at influential meetings and conferences inside and outside the Fed. We also began to hire leading monetary and banking economists as consultants and gathered them in Richmond for several weeks during the summer. Through seminars, lunch roundtables, and office visits with our economists, the consultants strengthened our research and broadened our contacts in the profession.³ (They also helped Marvin grow and thrive in Richmond, I believe, despite the absence of the richer economic research environments enjoyed by Reserve Banks in larger cities such as Boston, New York, and Chicago.)

It took some time to build, but over the course of the late 1970s and 1980s, a critical mass of research bearing the Richmond Fed imprimatur was produced and the Bank's influence and recognition in the policy arena increased. This reflected in significant part Marvin's own research since his arrival, and it provided a suitable platform for what was to come.

Research and Preparation for FOMC Meetings

While the department's policy team recognized the need to broaden the Bank's influence, it also understood that the central channel through which any Reserve Bank influences monetary policy is the Bank president's participation in meetings of the Federal Open Market Committee (FOMC), the Fed's principal monetary policymaking body. Even before Marvin's arrival, the department had developed an effective procedure for preparing the president for these meetings including a "pre-FOMC" briefing attended by the president and the full policy team late in the week before a meeting and a subsequent final briefing on Sunday afternoon. The Sunday briefing was attended by the president and the research director — traditionally the president's principal advisor, who typically attended FOMC meetings with him — and two or three other senior members of the policy team.

Over the years, as he gained greater experience with the FOMC, Marvin raised the level of these preparatory meetings substantially, especially after he became my principal advisor when I was appointed Bank president in 1993. Team members presented high-quality memos on various topics relevant to the upcoming meeting. Some of these memos addressed matters expected to be the principal focus of the meeting. Some of the most valuable, however, provided broader relevant background on issues like the inflation process, operating procedures used in conducting monetary policy, and labor and financial market conditions.⁴ I attended these pre-FOMC briefings throughout my term as president and benefited greatly from them not only for their content, but also because the interaction reinforced my relationships with individual members of the Research Department and the policy team. These initial stages of our FOMC preparations were central to our ability to participate effectively in FOMC meetings and constructively influence their outcomes.

The heart of our Bank's policymaking process occurred after the pre-FOMC briefing, culminating in the Sunday afternoon session. During my tenure, Marvin led these meetings. Before the meeting, he would prepare a draft proposal for the statement I would make at the upcoming FOMC meeting. This statement summarized our Bank's view of the economic outlook and our recommendations for the policy that would be implemented following the meeting.⁵ The team members and I would then discuss Marvin's draft in detail and hone it to a "final" product.⁶ The discussions were lively and expert. By the latter stage of my tenure, Marvin had established himself as a widely respected macroeconomist and constructive critic of Fed policy. He spent substantial time drafting the statement and did not readily agree to modifications. I, in turn, pushed for modifications I thought were necessary for me to present our positions comfortably and effectively to the FOMC. A typical exchange went like this:

Broaddus: Marvin, I think we'd be more persuasive if we said such and such rather than what you have.

Goodfriend: You can make that change if you want to, Al, but it will gut the whole point I'm trying to make.

After some good-natured but serious back and forth, we'd reach an agreement Marvin could live with. These brief but sometimes intense fine-tunings played to our respective strengths and, in my view, contributed greatly to our effectiveness in the FOMC meeting "go-around" discussions. They were reinforced by innumerable one-on-one discussions with Marvin, during walks in Richmond or Washington or late at night in one of our offices, where we hammered out our joint positions on core monetary policy issues.

Following the FOMC meeting in Washington, Marvin would grade my performance on the drive back to Richmond — always fairly, but as anyone who knows Marvin would expect, no punches Marvin felt were needed were pulled. There's no question in my mind, though, that these critiques served me well and elevated my ability to represent the Bank and present its positions effectively.

Making the case to the FOMC — some examples

Against this background, the following sections describe several experiences over about a quarter-century that I recall especially well, where Marvin and our Richmond policy team sought to convey some of Marvin's core policy themes persuasively to the FOMC and the broader Fed's policymaking staff.⁷

a. Policy Transparency and Marvin's Secrecy Paper

The first of these experiences was the writing and eventual publication of Marvin's seminal paper on transparency, "Monetary Mystique: Secrecy and Central Banking."⁸ If memory serves, Marvin began thinking about this paper not long after arriving in Richmond. As many readers are probably aware, early drafts of this paper made some people in the Fed uncomfortable, not least in Richmond. At the time, the cultural consensus inside the Fed was that it was inherently and continuously exposed to often politically motivated external scrutiny and attack. Many felt the wagons needed to be circled pretty much all the time. Moreover, it was generally recognized that responsibility for protecting the Fed and the part of the public interest the Fed served resided primarily with the Fed Board of Governors in Washington.

Marvin's paper challenged important aspects of this consensus when an early version was sent to the Board for review. In brief, the 1979 Merrill lawsuit forced the Fed — more precisely the FOMC — to provide an explicit defense of its routine delay (i.e., secrecy) in releasing policy directives after FOMC meetings.⁹ Marvin scrutinized this defense rigorously using relevant tools of economic analysis, including rational expectations. His paper caused discomfort, I think, because the FOMC's defense of its secrecy took the form of affidavits submitted during the litigation by one of the Board members and senior staff at the Board. Therefore, Marvin was directly (and potentially publicly) challenging these statements and their authors, albeit in a balanced and professional manner. Board positions had certainly been challenged before, by the St. Louis Fed in particular, but not as directly and forcefully by us.

Over time — a fair amount of it, actually — things worked out. Marvin never wavered, and the department's and the Bank's leaders consistently supported making the paper available to the public. It was published in 1986 in the Journal of Monetary Economics and played a significant role in the advancement of transparency in Fed monetary policy. It also permanently set a higher standard for our Bank's effort to contribute meaningfully to monetary policy.

b. Preemptive Policy in 1994

A second experience Marvin and our policy team shared was our especially active participation in FOMC meetings in 1994. This was an eventful year for the FOMC and for us. It was my first year as a voting FOMC member and Marvin's first accompanying me to Washington as my official advisor. That last point might suggest to some that Marvin reported to me. But most readers of this article are probably aware that on matters of monetary policy, in most respects effectively, I reported to Marvin. In any case, I was now positioned to present and advocate Marvin's views, often edited by me to soften their hardest edges to make them more palatable to my colleagues at the FOMC table. My statements typically included Marvin's thoughts, along with my own, on immediately current policy issues, but also Marvin's longer-term core principles summarized above. He sat directly behind me, and I felt a strong need to convey these principles accurately, frequently, and convincingly.

Several issues arose during the year. The first was whether the Fed would act promptly and with sufficient force to preempt any material increase in inflation or, equally importantly, emerging inflation expectations as the economy completed its recovery from the 1990-91 recession. Marvin had been greatly impressed by the preemption of inflation that Chair Volcker had overseen in 1983 and 1984, when for the first time the Fed had increased its policy federal funds rate materially without a sustained prior increase in inflation. With long-term Treasury bond rates rising since October 1993, signaling a rise in inflation expectations, Marvin wanted a preemptive encore in 1994. To that end, we argued for relatively aggressive tightening at each meeting throughout the year, which occurred, although not as aggressively as we wanted at the September meeting, when I dissented for the first time.¹⁰

Marvin believed that the FOMC's preemptive policy rate increases in 1994 anchored inflation expectations in the US and prevented the increase in actual inflation that appeared possible, even likely, at the beginning of the year. This was a substantial accomplishment for the FOMC, and I believe that Marvin's advocacy of this strategy, through my FOMC statements and other channels, played an important role in making it happen.

The year 1994 was also the year in which, at Marvin's instigation, we argued strongly against Fed involvement in Treasury initiatives to lend money to Mexico to assist the country in dealing with its peso crisis and prevent Mexico's problems from destabilizing broader international financial markets.¹¹ While persuasive arguments were made supporting such actions, Congress had explicitly declined to authorize them, and Marvin was appropriately concerned that Fed involvement would threaten the Fed's independence. Later in the year Marvin pushed me to dissent against renewal of the Fed's foreign exchange swap lines because they facilitated foreign exchange market intervention, which he felt undermined monetary policy credibility. He also believed they didn't work well, as illustrated especially clearly by an unsuccessful joint intervention with several countries in June of that year to support the dollar. Marvin insisted that I repeat the language of that dissent verbatim, annually, for the remainder of my time at the Fed. I would object, annually, that the FOMC had already heard it more than once. With a hint of annoyance, he would tell me, in effect, to "play it again, Sam."

c. What Assets Should the Fed Buy if Treasury Bonds are in Short Supply?

In 2001 and 2002, not long before both Marvin and I left the Fed, tax revenues arising from the late 1990s technology boom produced federal budget surpluses. Remarkable as it seems now, Treasury debt outstanding was declining and the FOMC began to worry about what it would do in the event there were insufficient Treasury securities available for the Fed's routine open market purchases. An FOMC subcommittee led by Don Kohn and Peter Fisher suggested several potential ways to address the problem.

As he made clear in his proposal later for an "Accord" on Federal Reserve credit policy, Marvin believed passionately that the Fed should avoid putting private assets on its books.¹² Consequently, faced with the prospect of a shortage of bonds, at the January 2001 FOMC meeting I summarized in detail Marvin's proposal for having the Treasury issue additional bonds to keep the Fed's operations "Treasuries Only." Chair Greenspan, true to his jazz band accounting experience, wanted to know what the Treasury would do with the proceeds of such bond issuance.¹³ Wouldn't the Treasury then have to buy private assets? I mumbled something; Greenspan repeated the question. Marvin, clearly alarmed that I was about to blow the opportunity, raised his hand and asked if he could respond. I said "sure," but I wasn't sure what the chair would say. In almost 30 years of attending FOMC meetings, this was the only time I ever saw a Reserve Bank advisor intervene in an FOMC meeting without an invitation to do so. I suppose I should have been embarrassed since it revealed who was driving policy in Richmond, at this meeting certainly. But I was proud of Marvin, and the point that needed to be made was made — that the Treasury would either need to buy private assets or the proceeds would have to be eliminated by increasing government spending or reducing taxes. I was also impressed that Greenspan showed no displeasure at the deviation from protocol and had a brief but natural exchange with Marvin.

d. The Initial Inflation Targeting Debate in the FOMC, 1995-98¹⁴

As indicated above, the progress toward bringing inflation down began with Chair Volcker's decisive actions in the early 1980s. While it took well over a decade, by the late 1990s there was at least an implicit consensus within the Fed that we were in the neighborhood of "price stability."

At this point, naturally enough, questions arose. First, what should the ultimate numerical objective for the desired steady-state inflation rate be? Second, and related, with inflation now persistently at historically low levels with market interest rates trending downward, was there a level below which a further decline in inflation might harm the economy?

Attention thus turned to the idea that the Fed should consider setting an explicit numerical inflation target.¹⁵ Not surprisingly, Marvin viewed inflation targeting favorably as a way of reinforcing a central bank's credibility for low inflation, and he pushed me to indicate our Bank's support for the concept when I became an FOMC member/participant in 1993. When the idea gained traction among some members of Congress, Chair Greenspan asked then Fed Governor Janet Yellen and me to lead a discussion of the pros and cons of inflation targeting at the FOMC meeting in January 1995.

In the week preceding this meeting, Marvin enthusiastically drafted the case for targeting that I delivered at the meeting. There, I summarized the benefits that a credible Fed precommitment to low inflation via a target would foster, most notably higher economic growth and employment. In particular, I argued that the greater credibility resulting from a target would reduce the sacrifice ratio:

What it would do – and this is probably the most important thing I'll say today – is discipline us to justify our short-term actions designed to stabilize output and employment against our commitment to protect the purchasing power of our currency.¹⁶

Governor Yellen made the case against a target, which she believed would downgrade the Fed's high employment goal.¹⁷ She also doubted that a target would increase the Fed's credibility and thereby reduce the sacrifice ratio, i.e., the loss in jobs and output required to resist short-run increases in inflation.

In the discussion that followed, the committee was about evenly divided on the desirability of a target. A subsequent discussion at the July 1996 meeting, however, briefly encouraged Marvin and me at one point when it appeared that the committee was approaching a consensus on "holding the line" at the then current 2 percent inflation rate (as measured by the Fed's preferred personal consumption expenditures (PCE) price index), which would have locked in the significant reduction in inflation already achieved. Our hopes were dashed when the discussion ended without an explicit recognition of the progress just achieved toward consensus on the issue. Marvin was especially disappointed.

With Marvin's encouragement, I again proposed an inflation target at the February 1998 FOMC meeting. At this point, the core PCE inflation rate had declined to below 2 percent and concerns about "unwelcome disinflation" and the risk of deflation had begun to arise. With this in mind, we argued for an explicit lower bound for any inflation target that might be put in place. This was a precursor of Marvin's subsequent, influential work on conducting monetary policy at the zero lower bound (ZLB).¹⁸

e. Monetary Policy at the ZLB

As just discussed, the almost 20-year effort to achieve credibility for low inflation appeared to bear fruit in the late 1990s. Attention began to turn — slowly at first — to the challenges posed for monetary policy by persistently low inflation and short-term interest rates approaching the ZLB. Marvin was at the forefront of path-breaking research about these challenges and ways to deal with them.¹⁹

In October 1999, at a Fed conference in Woodstock, Vermont, Marvin presented the results of his initial research in this arena, "Overcoming the Zero Bound on Interest Rate Policy."²⁰ In the paper, he described three approaches to retaining the ability of monetary policy to cushion a deflationary downturn in the price level and economic activity: a "carry tax" on cash balances and two forms of "quantitative" policy actions, either large-scale purchases of longer-term government securities (or similary illiquid private assets), or direct transfers of money to the public. At the January 2002 FOMC meeting, Marvin was invited to summarize his two quantitative policy alternatives. Both alternatives derive from the idea that long-term government securities offer what Marvin calls "broad liquidity" services in that they can be converted to liquid assets or used as collateral to borrow liquidity. Consequently, Marvin argued, the Fed (or any central bank) can affect the economy by purchasing broad liquidity assets and thereby affecting their yield.

While I did not participate directly in Marvin's preparation for this meeting, I was present and listened attentively to his presentation and the discussion that followed. Participants generally indicated interest in his results but seemed to view them as preliminary and academic rather than of immediate relevance to the policy issues of the day. Their relevance soon became apparent, however, during the financial crisis of 2008-09 and its aftermath, as Marvin's ZLB research provided a starting point for the Fed and other central banks as they confronted the radically different policy environment that emerged following the crisis.

Against this background, it is interesting and satisfying to consider Marvin's views regarding the role of Fed credibility in conducting pol-icy at the ZLB, thus returning full circle to his core credibility principle discussed at the beginning of this article.²¹ As Marvin indicates, the quantitative policy approaches described above are likely to increase, significantly, the public debt and the monetary base. Therefore, in resisting excessive disinflation or a deflation, the Fed may create a risk of a rapid reemergence of inflation. Its willingness to mount a successful defense against deflation, then, requires that it sustain, permanently, its credibility against inflation. As Marvin put it to me (and to many others), full credibility against deflation requires continuing full credibility against inflation. It was at this moment, late in my career at the Fed, that I grasped fully the comprehensive power of Marvin's credibility principle for monetary policy.

Concluding comments

Hopefully the preceding discussion has conveyed adequately the range and depth of Marvin's contributions to Federal Reserve monetary policy and central banking more broadly. All of us who worked with Marvin at the Richmond Fed are proud that our Reserve Bank was the setting for much of his most important research. We are also grateful for what he did to raise the standing of our Bank and enrich our individual careers. By the same token, I know also from conversations with Marvin that he greatly appreciated the supportive research environment he enjoyed in Richmond just as he appreciated Chair Green-span's willingness to foster a collaborative culture across the Federal Reserve, understanding that each strengthened his research and made it available to a broad international audience. Most importantly, I hope that what I've summarized here will confirm again what I and many others, inside and outside the Fed, have long believed: that among those who have labored to improve the conduct of American and global monetary policy, Marvin Goodfriend is a giant.

References

Goodfriend, Marvin. 1986. "Monetary Mystique: Secrecy and Central Banking." Journal of Monetary Economics 17, no. 1 (January): 63-92.

Goodfriend, Marvin. 1997. "Monetary Policy Comes of Age: A 20th Century Odyssey." Federal Reserve Bank of Richmond Economic Quarterly 83, no. 1 (Winter).

Goodfriend, Marvin. 2000. "Overcoming the Zero Bound on Interest Rate Policy." Journal of Money, Credit and Banking 32, no. 4, Part 2 (November): 1007-35.

Goodfriend, Marvin. 2008. "We Need an Accord for Federal Reserve Credit Policy." Paper prepared for a Shadow Open Market Committee Symposium, Cato Institute, April 24.

Goodfriend, Marvin. 2010. "Policy Debates at the FOMC: 1993-2002." Paper prepared for the Federal Reserve Bank of Atlanta-Rutgers University Conference "A Return to Jekyll Island: The Origins, History and Future of the Federal Reserve," November 5-6: 1-2.

Goodfriend, Marvin. 2016. "The Case for Unencumbering Interest Rate Policy at the Zero Bound." Paper presented at the Federal Reserve Bank of Kansas City Symposium on "Designing Resilient Monetary Policy Frameworks for the Future," Jackson Hole, Wyoming, August 25: 127-160.

Goodfriend, Marvin, and J. Alfred Broaddus Jr. 1996. "Foreign Exchange Operations and the Federal Reserve." Federal Reserve Bank of Richmond Economic Quarterly (Winter): 1-19.

Federal Open Market Committee v. Merrill, 443 U.S. 340 (1979).

Federal Open Market Committee. 1995. "Meeting of the Federal Open Market Committee, January 31-February 1." pp. 39-41.

Williams, John C. 2020. "Research, Policy and the Zero Lower Bound." Remarks at the Shadow Open Market Committee Spring Meeting, New York City, March 5.

Williams, John C. 2021. "The Theory of Average Inflation Targeting." Remarks at Bank of Israel/CEPR Conference on "Inflation: Dynamics, Expectations, and Targeting," July 12.

Cite as: Broaddus, J. Alfred Jr. 2022. "Marvin Goodfriend at the Richmond Fed: Recollections." In Essays in Honor of Marvin Goodfriend: Economist and Central Banker, edited by Robert G. King and Alexander L. Wolman. Richmond, Va.: Federal Reserve Bank of Richmond.