Testimony to the House Financial Services Committee, Subcommittee on Trade and Monetary Policy Titled “A Further Examination of Federal Reserve Reform Proposals”

January 10, 2018

PDF pdf_small

I want to thank Chairman Barr and Ranking Member Moore for inviting me to testify before the subcommittee. The issue of reforming the Federal Reserve Board has been debated for the whole time I have been in Washington, and I appreciate the opportunity to share my views on the topic.

Before I address the specific proposals now being considered, I will give my view of the proper relationship of the Federal Reserve Board to the president and the Congress. I will then assess these proposals against that background.

The Federal Reserve System has an unusual status as being a mix of public and private entities. The governors are of course explicitly part of the public sector, as presidential appointees subject to congressional approval. However, the 12 regional banks are private, being owned by the member banks in the district, which have substantial control over the district bank’s conduct.

This structure was put in place more than a century ago to fit the politics and the economy of the time. It is inconceivable that anyone constructing a central bank today would use the same framework. The archaic nature of the Fed’s design is perhaps best demonstrated by the distribution of the regional banks. Two are located in the state of Missouri. Meanwhile, the San Francisco region not only includes the whole state of California, but the rest of the west coast, and the states of Alaska, Hawaii, Nevada, Utah, Arizona, and Idaho, in all accounting for more than 20 percent of the nation’s economy.

While there were reasons that a mixed public–private central bank and regulatory system may have made sense at the start of the last century, this is no longer the case today. The United States is the only major economy with this sort of mixed approach. The Bank of England, the Bank of Canada, the Bank of Japan, and the European Central Bank are all purely public entities. It is recognized that the conduct of monetary policy, along with the lender of last resort and regulatory functions of the central bank, are necessarily responsibilities of the government.

When the central bank is fully public, there is an appropriate concern about political influence. Few would want the party in power in either the White House or Congress to be using the Fed’s power for narrow political advantage. For this reason, it is necessary that the Fed be to some extent insulated from political control. Other central banks in developed countries have been quite successful in bringing about this degree of insulation. While finance ministers and other government representatives necessarily have contact with central bank presidents and other bank governing officials, as well as professional staff, there is little evidence that in recent decades they have managed to have the central bank alter its monetary policy for narrow political ends.

The structure of the Fed similarly provides substantial insulation from political influence. The long terms of the governors mean that they need not be concerned that their actions will anger an incumbent president or powerful members of Congress. They cannot be removed except for malfeasance. Looking back over the history of the Fed, most economists and economic historians would agree that the instances where the Fed may have acted to advance narrow political ends are extremely rare. Even in the cases that are most frequently cited, such as the decision by the Fed under Chair Arthur Burns to have accommodative policy prior to the 1972 election, are very much subject to debate over motives.

While there does not seem to be much basis for concerns that the Fed will act to support the political party in power, there is a real concern about a structure that gives the financial industry a direct voice in the conduct of monetary and regulatory policy through their control of the regional banks. This is really an extraordinary structure without any obvious parallels in our governmental system.

Both aspects of this relationship make little obvious sense. The financial industry certainly has useful insights on the conduct of monetary policy, but it makes no more sense to give them seats at the table than the manufacturing or tech industry. Monetary policy has an enormous impact on the national economy and affects every sector in it; there is no reason to believe that the perspectives gained from working in the financial industry are uniquely valuable.

Similarly, the idea that an industry would be able to pick its own regulator is truly extraordinary. It is understandable that industry groups will try to lobby and in other ways influence the decisions of regulatory bodies. The pharmaceutical industry places pressure on the Food and Drug Administration (FDA) to approve drugs more quickly, the telecommunications industry lobbies the Federal Communications Commission (FCC) for looser standards on universal service, but in neither case are they given a direct role in appointing their regulators. No one would suggest that Pfizer or Merck should be able to appoint a commissioner on the FDA or that Verizon and Comcast should select one of the members of the FCC. The Federal Reserve Board is unique in this way, as the member banks within a district largely have the ability to control the selection of the bank president who plays a direct role in both determining monetary policy and regulation of the banks within the region.[1]

Most of the efforts at reform of the Fed over the last four decades have been in the direction of making it more of a public institution answerable to Congress. For example, the Humphrey–Hawkins Full Employment Act of 1978 requires the Fed give semi-annual testimony to Congress reporting on its progress in meeting the employment and inflation targets set in the law. As a result of a 1993 agreement between then Chair Alan Greenspan and Representative Henry Gonzalez, who was chair of House Banking Committee, the Fed now releases full transcripts of the Open Market Committee’s meetings with a five-year lag. And, the 2010 Dodd–Frank Wall Street Reform and Consumer Protection Act sought to reduce the power of the banking sector in the regional Fed banks by taking away the votes of Class A directors in selecting the bank president.

It now appears that Congress is interested in going the opposite direction with this latest set of proposals. Instead of increasing the public accountability of the Fed and following the example of every other major central bank, several of these proposals seem designed to strengthen the power of private banks within the Fed, thereby reducing its accountability to democratically elected officials.

The move away from accountability to democratically elected officials would be understandable if there had been a track record of failure. The standard concern is that a central bank that is controlled by elected officials will likely be too tolerant of inflation, as it is pressured to have an accommodative monetary policy in the period just before elections.

This clearly has not been a problem with the Fed in recent decades. Inflation has been at relatively steady and low levels for most of the last three decades. In fact, since the Fed officially adopted the 2.0 percent average inflation target in 2012, the core inflation rate has consistently been below this pace. In other words, if we view the 2.0 percent inflation target as a proper goal of monetary policy, the Fed has failed by having too little inflation, not too much. This raises the question, what is the problem that these new proposals are intended to fix?

Also, we do have a track record showing how bank presidents have voted on FOMC decisions. In most cases, most bank presidents do vote with the majority, since dissents are relatively rare. However, the dissents by bank presidents have been overwhelmingly in the direction of tighter monetary policy, in pursuit of lower inflation at the expense of higher unemployment.

An analysis of every dissent by a bank president since 1993, when the minutes first started giving reasons for dissents, showed that in 64 of 72 dissents bank presidents were pushing for more contractionary monetary policy.[2] This meant they were either arguing for an interest rate hike when the majority wanted to hold rates steady or that they were arguing for holding rates steady when the majority supported a cut in rates.

Since inflation has been low and mostly below target for most of this period, while unemployment has been higher than levels consistent with full employment, the implication is that if bank presidents had more authority in determining the Fed’s monetary policy over this period, they would have needlessly curtailed growth and cost workers’ jobs. It is difficult to understand why Congress might view this as a positive outcome.

Specific Proposals Being Considered by the Subcommittee

With this as background, I will briefly address the seven proposals currently being considered by the subcommittee. My ordering follows the order used in my invitation letter rather than my assessment of their relative importance.

1) Requiring salary information and financial disclosures for Fed officials whose salary exceeds that of a GS-15 federal employee and assigning at least two staff positions to each governor.

This disclosure portion of this proposal seems largely unobjectionable, although the cutoff may be somewhat low. The Federal Reserve System’s Board of governors is unquestionably a part of the federal government. For this reason, the public should have the right to know both what high-level officials at the Fed are being paid and also be assured that they do not have any clear conflicts of interest that could affect their actions.

My concern is that the pay level is not especially high for an experienced economist working in Washington, DC. I assume that the intention is to require salary and disclosure information from the most senior staff who have an important role in advising the governors, not to harass mid-level staff whose work is largely directed by more senior staff and the governors themselves. For this reason, I would suggest a higher cutoff so that only the top level staff is affected by these requirements. I do not know enough about salaries at the Fed to recommend a specific cutoff, but I’m sure that the subcommittee could get this information.

Ensuring that each governor has at least two designated staffers to provide information seems like a reasonable use of resources. Since there are only seven governors and the chair already has designated staff, any additional commitment of resources would be minimal. From my understanding of the Fed’s operations, the governors already have substantial access to information/advice from Fed staff, but I can’t see any harm in requiring a minimum amount of designated personnel. It certainly would not be a major expense.

2) Establishing a blackout period of one week for FOMC members for one week prior to a meeting and extending to midnight the day after a meeting takes place.

The principle of having a blackout period before meetings and continuing after the meeting is a good one. This is already the current practice with the period beginning on the weekend before meetings, which are held on Tuesdays and Wednesdays, and continues through the following Thursday. The logic is that we do not want FOMC members to be dropping hints just before a meeting which could provide information that market actors can trade on.

While there is little reason to question the wisdom this practice, it is not clear why Congress would feel the need to extend it and to enshrine it in law. At least, based on what is publicly known, the Fed has been very responsible in not leaking items that could allow for profitable trading by connected individuals. The one notable exception was a leak by former Richmond Bank President Jeffrey Lacker, which led to his resignation last year.

Here also, the question is what problem is this proposal meant to solve? This is an area in which the Fed has been acting responsibly. It’s difficult to see why Congress would feel the need to micromanage its operations with this sort of rule. Furthermore, while it is not clear (at least to me) what the optimal length of a blackout period should be, it is important to recognize that there is a real cost to making it too long.

When unexpected events happen, the public, and certainly financial markets, welcome the opinions of the Fed chair and other members of the FOMC. With eight meetings a year and a blackout period that runs for a full 10 days surrounding each meeting, this law effectively is requiring the Fed chair and other members of the FOMC to remain silent on key issues for 80 days of the year, or more than 20 percent of the time. That seems excessive. If there were evidence that the current blackout period is insufficient and has allowed for improper trading, then perhaps the requirement in this proposed legislation would make sense, but absent such evidence, it is difficult to see why Congress would feel the need to require this longer blackout period.

3) Amending the Federal Reserve Act to bring the non-monetary policy related functions of the Fed under the general appropriations process.

The Fed, like other banking regulators, has the authority to set its own budget with its income from services provided to the financial industry, and more importantly, from the interest earned on assets purchased with Federal Reserve notes.

This authority does seem inconsistent with a public agency accountable to Congress. There is logic to the idea that we would not want the Fed budget to be subject to congressional whims, where budget cuts could be used to punish it for pursuing monetary or regulatory policy against the wishes of the majority in Congress. But this goal can be met by having multi-year appropriations from which Congress could not easily deviate. Establishing a formula for the Fed’s budget, similar to what Congress did for the budget of the Consumer Financial Protection Bureau, could be a useful way to meet the twin goals of having spending set by Congress, while still keeping it insulated from political influence.

In this respect, it is worth noting that the Supreme Court relies on Congress for appropriating funding for its operations. To my knowledge, this funding has never been used as a tool for influencing court decisions. There is little reason to believe that subjecting the Fed’s funding to congressional approval would interfere with its ability to pursue an independent monetary policy.

4) Modifying the Federal Reserve Act to Allow Class A directors to vote for district bank presidents

This proposal would reverse the provision of the Dodd–Frank Wall Street Reform and Consumer Protection Act which takes away the vote from Class A directors on selecting district bank presidents. This provision of Dodd–Frank was put in place to reduce the power of the banks in selecting both their own supervisor and also a member of the FOMC. (The Class A directors are appointed by the member banks.)

As noted in my earlier discussion, it is very difficult to understand the motivation for reversing this Dodd–Frank provision as opposed to going further in taking away the power of the financial industry over the Fed. I could not imagine members of Congress suggesting that industry groups directly appoint their own regulators in any other sector of the economy. It is difficult to understand why they would somehow view it as appropriate in the case of the financial industry and the Fed.

It is also difficult to understand why they would think it appropriate to delegate a fundamental public responsibility — control of monetary policy — in part, to the financial industry. As noted before, it makes no more sense to give the financial industry a direct role in setting monetary policy than the tech industry or the telecommunications industry. Monetary policy affects the whole economy; there is no obvious reason we should want to give the financial industry an outsized role in setting its course.

5) Modifying the Federal Reserve Act to allow all district bank presidents to vote at every meeting

If I understand the proposal by Representative Williams correctly, it calls for having all bank presidents vote at every meeting. (All the bank presidents are already present for the discussions that precede a vote, so the issue is not having the opportunity to benefit from their input.) This proposal again seems to go in the opposite direction of other central banks, and recent policy on the Fed, by taking a big step away from a democratically controlled central bank.

It is difficult to understand the motivation for a measure that would assign the bank presidents a majority voice in determining monetary policy. The problem is compounded if it is coupled with the proposal to restore the vote of the Class A directors in selecting bank presidents.

6) Amend the Federal Reserve Act to require the FOMC to determine interest rates on balances held on deposit at the Federal Reserve System by member banks

This proposal would take away the power of the governors to have control over the interest rate paid by on reserves and instead have it determined by the FOMC. This also seems to be an effort to move away from a Fed controlled by officials appointed through the democratic process to one in which the member banks have more voice. I have given reasons before on why I consider that to be a move in the wrong direction. However, I will also point out that it is not clear what this provision is intended to accomplish.

The FOMC sets the target rate, which is the federal funds rate in overnight markets. The interest rate paid on reserves is simply the tool that the Fed uses to reach this target. So the FOMC is already fully involved in the key policy decision.

It may make sense to insist that the bank presidents have a voice in the mix of tools used by the Fed to reach this goal if there was some reason to believe that they had expertise in this area that the chair and the other governors lacked; however, this hardly seems plausible. Furthermore, I am not aware of any evidence that the federal funds rate has diverged to any substantial extent from the policy rate set by the FOMC. Given that reality, it is not clear what the motivation would be for this proposal.

7) Amend the Federal Reserve Act to require the Vice Chair for Supervision of the Board of Governors to testify twice a year before Congress

This proposal is a useful supplement to the Humphrey–Hawkins Act provision requiring the Fed Chair to testify biannually on the Fed’s progress in meeting its goal of full employment. In effect, this would require the Vice Chair for supervision to provide comparable testimony.

The Fed has often neglected its regulatory responsibilities, a fact that became painfully clear during the run-up of the housing bubble and the subsequent collapse and the resulting financial crisis. This provision will help to give the Fed’s regulatory responsibilities more visibility. It will also encourage Congress to focus more attention on the stability of the financial system and to take note of potential risks that have come to the Fed’s attention.

While both Ben Bernanke and Janet Yellen have taken more of an interest in regulation as chair than did Alan Greenspan, and have mentioned risks on occasion in their testimonies, it would be helpful to have the Fed official most directly responsible for oversight give regular testimony. This is a very useful proposal.


This subcommittee is considering a wide range of proposals that would alter the structure of the Fed. Several are quite useful in increasing openness and accountability. However, the ones which aim to give more control of the Fed in the hands of the banking industry, rather than officials appointed through the democratic process seem at odds with recent trends both in the United States and the rest of the world. It is difficult to understand the effort to privatize the conduct of monetary policy and to turn over control of financial regulation to the industry that is being regulated.

[1] The Dodd–Frank financial reform bill did weaken the banks control by taking away the votes of the Class A directors, who are appointed directly by the member banks, in selecting a president. Nonetheless, they still are likely to control the process since the Class B directors, who have half the votes, are appointed by the Class A directors. This is of course one of the issues the subcommittee is now considering.

[2] Baker, Dean, Sarah Rawlins, and David Stein, 2017. “The Full Employment Mandate of the Federal Reserve: Its Origins and Importance,” Washington, DC: Center for Economic and Policy Research and Center for Popular Democracy, available at http://cepr.net/images/stories/reports/full-employment-mandate-2017-07.pdf.