What Does the Fed Know, and When Does it Know It?

Marvin Goodfriend was my student at Brown University; he was awarded his PhD in 1980. I am pleased to contribute an essay to honor his memory.

When I was appointed to the Council of Economic Advisers (CEA) in 1982, I immediately reached out to Marvin and persuaded him to spend a year on the CEA staff in 1984. During my time as president of the Federal Reserve Bank of St. Louis (1998-2008), Marvin and I crossed paths regularly. We often chatted briefly at Federal Open Market Committee (FOMC) meetings before he left the Richmond Fed in 2005.

Shortly after returning to the Richmond Fed from the CEA, Goodfriend began working on a paper evaluating the Fed's case for secrecy.¹ He picked apart the arguments the Fed presented as it attempted to fend off the FOIA suit brought by David R. Merrill in 1975. That litigation found its way to the Supreme Court and back to Federal District Court before being finally concluded in 1981. That paper of Marvin's is my motivation for this essay.

As St. Louis Fed president I attended every meeting of the FOMC between March 1998 and my final one in January 2008. The committee often grappled with the wording of policy statements, as anyone can see from reading transcripts of the meetings. For example, at the end of my first meeting in March 1998, Ned Gramlich, a Fed governor appointed in 1997, interrupted the roll call vote on the policy directive. As reported in the transcript of the meeting:

MR. GRAMLICH. Do we have to have the "slightly lower" phrase? Am I out of order? [Laughter]

CHAIRMAN GREENSPAN. That is the conventional rhetoric.

MR. GRAMLICH. Yes, but—

CHAIRMAN GREENSPAN. We have been butchering the English language in this directive for years, but let's not change it just yet. Why don't you bring that up at a later meeting? [Laughter]

MR. GRAMLICH. On that advice, I vote "yes." [Laughter]

During my entire tenure we struggled with how best to communicate with the markets and the general public. I gave several speeches on the subject (see Poole [2003] for an example). By 1998, when I arrived in St. Louis, the case for secrecy in the abstract was dead but saying that does not per se make a case for disclosure. Disclosure of what, and when? We agreed that putting the FOMC meeting on C-SPAN made no sense because doing so would inhibit a full discussion of the policy issues the committee faced and would simply move the real discussion into the hallways.

Most of the literature on Fed communication is normative — what should the Fed disclose, when, and why? As illustrated by Ned Gramlich's question, the issue was how we could best communicate the policy issues we faced and why we made the decisions we did. Goodfriend examined the Fed's case for secrecy; my purpose in this essay is to approach the policy communication issues by first concentrating on what the Fed knows and when. When does the Fed have an information advantage, or disadvantage, relative to the market? What is the nature of the advantage, or disadvantage, and why does it matter? This essay is more along the lines of a positive than a normative analysis of the role of information in the Fed's monetary policy.

Proposition 1: Absence of a systematic Fed information advantage. The Fed has a minimal advantage, if any, in terms of knowledge of facts. The Fed chair gets the unemployment report through the chair of the CEA late Thursday afternoon, just a few hours before the Bureau of Labor Statistics releases the data publicly on Friday at 8:30 a.m. This same "advantage" of a few hours is true of all the data released by federal statistical agencies. Today, the Fed, governments at all levels, and private firms are wrestling with COVID-19. It is hard to believe that the Fed has a COVID-19 information advantage over other entities; the Fed has made this point clear on many occasions as it has discussed the pandemic.

Federal statistical data are the raw material of macroeconomic and financial analysis. While it is true that Fed economists can tap directly into employees at the statistical agencies for additional insight, it is also generally true that private economists and analysts can reach the same experts. As a close approximation, the Fed has no advantage whatsoever over the private market in access to data. It is laughable to believe that the Fed has an inside track on future fiscal policy when the federal government itself seems so obviously dysfunctional in planning just about everything. All too much of what the federal government does is a consequence of last-minute deals brokered among many competing interests. Relative to political experts, the Fed is probably at a disadvantage in predicting what will happen.

It is also true that certain private parties sometimes have access to data that the Fed does not. An example is the story spun by Michael Lewis in The Big Short.² Lewis explains how a few hedge fund managers developed a firm conviction that many housing-related securities would crater in a financial crisis. It is not infrequently the case that industry experts understand developments that will make an import-ant difference to the macroeconomic environment in the months and quarters ahead. Another example might be the expansion of oil and gas production through fracking. That expansion affected the outlook for energy prices and, through those prices, the direction of the international economy. Yet another example might be the development of COVID-19 vaccines. The companies involved understood the process and prospects much better than the Fed. Because of insider trading restrictions, pharmaceutical firms had to be very careful about the release of information.

I have long held the view that the best way to think about the arrival of new information is to assume that the market and Fed get the information at the same time. Since the rational expectations revolution in macroeconomics in the early 1970s, it has been standard to model new information as "innovations" that push the state of knowledge one way or the other from previous expectations for every variable in the economy.

This view of the information process motivated my 1999 speech, "Synching, not Sinking, the Markets." In that paper, I argued that the Fed ought to think of policy in the context of how to change the federal funds rate in response to arriving information. With Fed clarity and transparency, one of its goals should be leadership and, through disclosure, alignment with private sector reactions to incoming information.

Proposition 2: Absence of a systematic Fed processing advantage. Analysts must process the flow of raw data to develop useful ideas as to what new data might mean for the direction of the financial markets and the economy. The Fed has a very large and expert staff. Does the Fed have a processing advantage over the private market? Can the Fed's experts distill knowledge out of raw data more quickly and more accurately than the private market?

My instinct is that the Fed does not, overall, have a processing advantage. Fed economists and private economists read the same professional papers and attended the same graduate schools. They go to the same conferences. Fed and private sector economists move back and forth in their employment. The literature on the accuracy of economic forecasts does not suggest that Fed forecasts are clearly more accurate than private forecasts.³

Obviously, there are occasions when the Fed is ahead of the private market and some when the Fed is behind. In thinking about the normative aspects of information sharing and secrecy, both the average situation and the special cases are relevant. My general proposition is that the Fed and the markets have the same information base and the same ability to process that information.

Obviously, not everyone in the market has the same information and the same skills in processing information. And that is also true of Federal Reserve officials. What makes sense to me is that it is best to assume, absent evidence to the contrary, that market prices accurately reflect implications of available information for all the reasons discussed at length in the rational expectations and efficient markets literature.

The Fed does have one power the market does not — the power to print money. In the short run, the Fed can set almost any interest rate on government securities at almost any level it wants, and the same is true for foreign exchange rates. "At almost any level" is obviously an exaggeration, but it is not far off if we think in terms of basis points for a few weeks rather than percentage points for a few years. Basis points for a few weeks are of critical importance to market speculators.

However, the Fed does not have the power to set real variables, except real interest rates temporarily, at any level it wants. This proposition goes far back in the monetary theory literature. Its corollary is that an effort by the Fed to set real variables, such as the unemployment rate, will fail. Moreover, a determined Fed effort to set real variables at levels materially different from market equilibrium levels will create large problems.

Illustrative examples

The terrorist attacks on Tuesday, September 11 illustrate these themes. Neither the Fed nor the private market anticipated the attacks. Here was an information innovation writ large. Because the government shut down air travel, the Fed was forced to shut down the clearing of checks. At that time, the Fed clearing process involved shipping physical checks by air from Fed processing facilities in Reserve Banks and their branches across the country to Reserve Banks and then by van to banks on which checks were written. Commercial banks make payments from accounts that have received funds. With no funds coming in, banks could not honor checks written on accounts on which firms were making payments. The Fed made clear to banks that they could borrow from the Fed whatever amounts they needed to be able to honor checks on September 12 and later.

The Fed's power to print money prevented the terrorist attacks from creating a financial disaster. No private entity could have done what the Fed did. A similar and more familiar story arose after the failure of Lehman Brothers in September 2008. The Fed established several special facilities, and expanded others, to provide funds in vast amounts to the markets. The Fed does have powers that permit it to respond to information in ways the market cannot. That said, the Fed's powers are not indefinitely large.

There is a flip side to these examples. When the Korean War broke out in June 1950, the Fed found itself stuck with the World War II policy of pegging yields on long-term Treasury bonds. That policy was unsustainable as the market began to dump bonds on the Fed in massive quantities. Looking at the Federal Reserve Bulletins for June 1950 and January 1951, we see that the Fed's holdings of government securities at the end of 1950 were 28 percent above the level at the end of May 1950. In March 1951, the Fed was able to negotiate the Treasury-Fed Accord that discontinued the Fed's obligation to peg Treasury bond rates. Clearly, the Fed must be very careful about the commitments it makes or implies. As I write, the Fed needs to find a way to extract itself from its announced plans to buy bonds and hold the fed funds rate near zero.

My personal view at this time is that waiting until spring 2022 to begin the process of raising the federal funds rate increases the risk, dramatically, that 2021's surge in inflation will continue. More generally, though, isn't this situation of exactly the sort that led many economists to favor an announced inflation target? I will return to this question below.

A somewhat similar process to the one in 1950 occurred in the late 1970s, as the Fed attempted to hold down interest rates as inflationary pressures blossomed. The inflation accumulated to the point of creating a variety of ills in the economy. The Fed's power to print money was then the problem and not the solution. The Fed had no information advantage or processing advantage over the private market. It took Paul Volcker's leadership advantage to deal with the growing inflation turmoil. No private entity could fix the inflation problem.

A more recent Fed processing failure is worthy of comment, especially since it illustrates my own failure to understand what was happening with the house price bubble that ran from 2000 to 2006. I was a member of the FOMC during those years and still have a vivid memory of the special FOMC study of the housing situation. It is a sobering exercise to review the FOMC transcript of the meeting of June 29-30, 2005. By coincidence, this was also the last FOMC meeting that Marvin attended, as Chairman Greenspan noted at the beginning of the meeting.

The staff analysis of the housing situation, and my own contributions to the FOMC debate, demonstrate how wrong the FOMC can be. I had given several speeches, as had Greenspan, worrying about the potential for financial chaos should Fannie Mae and/or Freddie Mac find themselves approaching insolvency. They did become insolvent in 2008, but the problem was easy to fix. All the government had to do was to take them into conservatorship, making the implicit guarantee explicit.

A much more serious problem was the accumulation of weak mortgage paper on the books of investment and commercial banks. There was not a word of that development in the FOMC record until 2008, and even then the severity of the problem was not understood, at least from my reading of the FOMC transcripts of this period. The private market understood the developing problem before the Fed did. The Fed and the Treasury did eventually deal with that problem by recapitalizing major banks and taking other steps familiar to students of the 2008 financial crisis. However, there was no satisfactory way of dealing with the insolvency of several million households that defaulted on mortgage and other debt. The FOMC and its staff seemed not to have a clue about these risks until they arose.

Market speculation on what the Fed knows

I am often amused by press accounts guessing as to what the Fed knows that the private market does not know. During most of my tenure in St. Louis, I was fortunate to have Robert Rasche as research director. Bob and I had known each other for many years, and he was a full coauthor on most of my speeches, many of which were mini research projects.

Particularly after a Fed surprise policy adjustment, Bob and I would joke, before release of the news, about the likely effects on the markets. I wish I had kept a record of what we guessed would be the impact on the stock market. We generally got the direction right but neither of us had confidence as to the magnitude of the effect. I doubt that any member of the FOMC or its staff then or since has a better track record than we did. The Fed's lack of ability to forecast the effects of its policy changes was one of the reasons I was always opposed to the Fed creating policy surprises.

Given that private participants in financial markets are deeply knowledgeable and highly motived to understand the significance of Fed policy adjustments, the observation just made should be a warning. If market experts cannot figure out the likely effects of Fed policy adjustments, why should economists and other outside observers believe that the Fed itself has a good idea about the effects of its policy adjustments?

One of the complications Fed policymakers face is that some market participants interpret policy adjustments as evidence that the Fed knows something the market in general does not know. As already emphasized, that view seems to me to be mistaken most of the time.

Market participants sometimes interpret Fed policy adjustments as "sending a message" of some sort, or "priming the market." That interpretation is probably correct on occasion, but it is an example of the Fed's failure to explain its policy clearly. Fed words and policy actions should be consistent with one another to avoid confusing the market.

Application to current policy situation

Marvin's 1986 paper concerns interactions with the markets of Fed statements, policy actions, and stance on disclosure. The Fed's adoption in 2012 of a formal inflation target was a new phase in its communication with the markets.⁴

In 2012, the FOMC settled on a target of 2 percent annual increase measured by the annual change in the price index for personal consumption expenditures. In August 2020, the FOMC changed the target to be an undefined average over time: "The Committee seeks to achieve inflation that averages 2 percent over time, and therefore judges that, following periods when inflation has been running persistently below 2 percent, appropriate monetary policy will likely aim to achieve inflation moderately above 2 percent for some time." How moderate and how long? The problem is not that the Fed won't provide an answer, but that it does not know the answer.

Why the change in the inflation target? I'll offer a speculation. The Fed wanted to pull out all the stops to make clear that it was doing everything possible to fight the COVID-19 pandemic. Easy money was the only tool it had. However, the Fed badly miscalculated the inflation risk. As the months passed, how could the Fed backtrack without appearing to give up its battle against the pandemic or admitting that it had miscalculated the inflation risk?

As many have noted, by changing to an ill-defined average over time the FOMC has debased the clarity of the original target. Also, the committee's turn to vague language has walked away from one of the important original arguments for an inflation target — providing discipline to the FOMC itself. As I write, with an Excel spreadsheet in front of me, I see that (using continuous compounding) the average annual percentage change over the past 36 months of the Fed's favorite inflation gauge is 2.5 percent for data through October 2021. The three-year average breached 2 percent in June 2021, and yet the FOMC continues to buy assets and continues to hold the federal funds rate near zero. The Fed has achieved its objective — real interest rates are now substantially negative.

The Fed, by taking variance and uncertainty out of the financial markets with its policy of asset purchases and near-zero fed funds rate, has increased variance and uncertainty in both quantities and prices in the goods and labor markets. Fifty years ago, I published a paper⁵ on this exact topic. FOMC members who are especially concerned about the real economy have the story upside down — Fed policy has increased, not reduced, variance and uncertainty in the goods and labor markets. Why should they be so terribly solicitous of the financial markets while ignoring what is going on in the goods and labor markets?

The June 2021 FOMC meeting was June 15-16. By that time the committee knew that its previous projections had gone seriously wrong. On June 10, the BLS had released data for the consumer price index for May, which showed a year-over-year increase of 5.0 percent and ex food and energy of 3.8 percent. The index for used cars was up a remarkable 29.7 percent. On May 28, the BEA had released data on the personal consumption expenditures price index for April, showing a year-over-year increase of 3.6 percent (3.1 percent ex food and energy).

The median of the FOMC's inflation projections (PCE price index) for 2021, released December 16, 2020, were already obsolete. The FOMC had projected for all of 2021 an increase of 1.8 percent. In June 2021, the projection for the full year was 3.4 percent. By December 2021, the projection was 5.3 percent. Yet, at that time, the FOMC continued to purchase billions of dollars of bonds and continued to maintain a near-zero fed funds rate.

Without question, COVID-19 has created a severe disturbance. That said, we should not forget that the start of the great inflation in the mid-1960s also saw severe disturbances. The Vietnam War and President Johnson's efforts to prevent disputes over the war from disrupting his plans for expansion of Great Society programs were active Federal Reserve concerns. 

Anyone who doubts that statement should spend time studying the detailed economic history of this era. I was an economist on the Fed's Washington staff from May 1969 through June 1973. I used to commute to the Fed on Rock Creek Parkway and remember antiwar demonstrators throwing park benches off overpasses onto the parkway.

In 1965 and thereafter, the Fed was also concerned with the solvency of the savings and loan industry. Pushing up interest rates would lead to disintermediation. Then, in 1971, Nixon imposed wage-price controls. Then, in October 1973, OPEC's oil embargo sent petroleum prices surging. I was among those who had problems finding an open gas station. And then there was the Watergate affair. Pumping money into the economy did not help to resolve any of these nonfinancial disturbances.

I do not mean to downplay the importance of COVID-19, which has killed an estimated 800,000 Americans as of this writing. That number may be compared with about 50,000 Americans dead from the Vietnam War. The point is that the Fed cannot affect a real variable — the number of vaccinated Americans — by monetary expansion. Nor will monetary expansion alleviate supply chain problems. The economy is very liquid. Bank credit is readily available for any trucking company that wants to buy more trucks to get goods off Pacific Coast docks. Eighteen wheelers going down the highway are traveling billboards for trucking companies trying to hire more drivers.

A policy of drift, month after month, for one reason after another, created the Great Inflation of 1965-80. The same policy today is likely to yield the same outcome.

Continuing to pump cash into the economy will not encourage more vaccinations and will not alleviate supply chain disruptions. As I write this just before Christmas 2021, the Federal Reserve is at least six months behind in responding to the flow of formal data and the many anecdotal observations familiar to anyone who is looking. Help wanted signs are everywhere, and employee turnover is the highest in the history of the JOLTS data. Residential property prices month after month have been rising at a rate higher than the peak rate during the house price bubble before the 2008 financial crisis.

So far, inflation expectations data suggest that the market has not lost confidence in the Fed. The risk is there, however; Goodfriend explained the process in his 1993 and 2007 papers, and elsewhere.

Knowledge of this history, in the United States and abroad, led many economists (including me) to advocate a Federal Reserve commitment to an announced inflation target. The announced target would communicate more clearly with markets and — very importantly — constrain the Fed to act as it had promised. The literature on inflation targeting is voluminous; I will refer to just one paper, by David Archer (2000).

Archer — assistant governor at the Reserve Bank of New Zealand at the time — begins his paper with four observations about the New Zealand inflation targeting regime.

1. A nominal variable (such as the price level or the inflation rate) is recognized as the sole achievable medium-term objective for monetary policy.
2. An attempt to drive policy directly at the medium-term objective via a tightly specified inflation target, rather than indirectly via an intermediate target.
3. An institutional structure that clearly articulates the respective roles and responsibilities of the key actors (the central bank and the government).
4. Heavy reliance on transparency to support the arrangement and cover the "weak points" in the institutional structure.

In the context of Goodfriend's work and this essay, points (3) and (4) are especially relevant. Just following the above points, Archer refers to, "the interactions of 'public choice' incentives and expectations." The US experience of 1965-80 displays these interactions in spades. The Federal Reserve today faces the same sorts of issues. The public debate about how best to deal with COVID-19 is confused. The Fed, of course, has expressed its commitment to do "whatever it takes" — to use all the "tools" at its disposal. The public and the Fed itself seem baffled as to what these phrases mean. Has the Fed published a list of the tools it can use to tackle COVID-19?

If the Fed had not confused matters by fuzzing up its inflation target in August 2020 — walking away from transparency — it would have been in a position in mid-2021 to end asset purchases immediately and to begin to raise its fed funds rate target. The Fed could have said that the situation was extremely difficult with the new delta variant and the best monetary policy was not clear. The Fed could have said that given the inflation data and its commitment to an inflation target, it was time to begin raising rates. If the increase in rates turned out to be premature, the policy adjustment could be reversed in the future.

Some will be reminded of the quote attributed to Keynes: "When the facts change, I change my mind. What do you do, sir?" As a bit of research will reveal, there is controversy over whether the quote is accurate, but it fits here anyway. By June 2021, the facts had changed dramatically.

I am also reminded of the legend of Odysseus. "The technique is called Odyssean self-control, and it is more effective than the strenuous exertion of willpower, which is easily overmatched in the moment by temptation."⁶ In August 2020, the Fed cut the ropes by which it had lashed itself to the mast, and now all of us will pay for that mistake.

An economist who discusses his outlook is always risking ridicule in retrospect. That understood, I have put my cards on the table. I am well aware of the first law of forecasting: if you name a number, do not name a date; if you name a date, do not name a number. I say "here goes" because the only true test of any empirical proposition intended to be taken seriously is an out-of-sample forecast. My outlook is dependent on the data I have observed and on Federal Reserve statements about its policy. I believe that the inflation rate in 2022 will be at least as high as the 2021 rate.

My sense of the policy environment as I write in the fourth quarter of 2021 is that the Fed has created a generalized asset price inflation. At the time of the FOMC's December 2021 meeting, stock prices were high, as measured by P/E ratios. Bond prices were high — interest rates low — as compared with data over past decades. Residential property prices were rising at a more rapid rate than at any time during the house price bubble of 2000-06. The second derivative of house prices — the speed with which house price inflation year over year has increased — is higher than at any time in the history of the broad repeat-sale house price indexes. Farmland prices are increasing. However, gold prices — a traditional measure of inflation concerns — have not moved by much.

Here I am writing in late December 2021, and at best the Fed has indicated that it might begin to increase the federal funds rate in the middle of 2022. I am well aware that I am offering pointed criticism of the Fed in a volume to be published by the Federal Reserve Bank of Richmond. Marvin Goodfriend spoke his mind, as evidenced especially in his paper on the Merrill case, which was highly critical of Fed leadership. It was gutsy for him to write this paper while he was in the Research Department at the Richmond Fed.

I would like to think that a small fraction of his approach to policy came from his old professor with the initials WP. I am honored that I was invited to contribute a paper to this volume.

References

Archer, David J. 2000. "Inflation Targeting in New Zealand." IMF seminar on inflation targeting, March 20-21.

Board of Governors of the Federal Reserve System. 1950. Federal Reserve Bulletin. June.

Board of Governors of the Federal Reserve System. 1951. Federal Reserve Bulletin. January.

Chang, Andrew C., and Tyler J. Hanson. 2015. "The Accuracy of Forecasts Prepared for the Federal Open Market Committee." Board of Governors of the Federal Reserve System Finance and Economics Discussion Series 2015-062.

Federal Open Market Committee. 2005. "Meeting of the Federal Open Market Committee, June 29-30."

Federal Open Market Committee. 2012. "Statement on Longer-Run Goals and Monetary Policy Strategy, Adopted effective January 24, 2012."

Federal Open Market Committee. 2020. "Statement on Longer-Run Goals and Monetary Policy Strategy, Adopted effective January 24, 2012; as amended effective August 27, 2020."

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

Goodfriend, Marvin. 1993. "Interest Rate Policy and the Inflation Scare Problem: 1979–1992." Federal Reserve Bank of Richmond Economic Quarterly 79, no. 1 (Winter): 1-23.

Goodfriend, Marvin. 2007. "How the World Achieved Consensus on Monetary Policy." Journal of Economic Perspectives 21, no. 4 (Fall): 47-68.

Lewis, Michael. 2011. The Big Short: Inside the Doomsday Machine. New York: W.W. Norton & Company. Kindle Edition.

Pinker, Steven. 2021. Rationality: What It Is, Why It Seems Scarce, Why It Matters. New York: Penguin Random House. Kindle Edition.

Poole, William. 1970. "Optimal Choice of Monetary Policy Instruments in a Simple Stochastic Macro Model." Quarterly Journal of Economics 84, no. 2 (May): 197-216.

Poole, William. 1999. "Synching, Not Sinking, the Markets." Speech at the meeting of the Philadelphia Council for Business Economics, Federal Reserve Bank of Philadelphia, August 6.

Shapiro, Adam, and Daniel J. Wilson. 2019. "The Evolution of the FOMC's Explicit Inflation Target," Federal Reserve Bank of San Francisco FRBSF Economic Letter, April 15.

Cite as: Poole, William. 2022. "What Does the Fed Know, and When Does It Know It?" 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.