Scott Sumner  

Everyone needs to be accountable

Overselling Online Education 2... Paul Samuelson's Take on the G...

All human institutions are fallible, and hence need to be accountable. That should be obvious, but when it comes to the Fed some people seem to think we just need to trust them. Allan Meltzer knows otherwise, and did a nice job explaining this to Congress (this quotation is from John Taylor's blog):

Meltzer spent a lot of time explaining in simple terms the merits of the policy rules bill, especially in comparison with other reforms, including the Audit the Fed bill. It's worth listening to. Here are some highlights:

In his opening, Meltzer said "We need change to improve the oversight that this Committee and the House Committee exercises over the Fed. You have the responsibility. Article I, Section 8 gives that to you. But you do not have the ability to exercise authority. You are busy people. You are involved in many issues. The Chairperson of the Fed is a person who has devoted his life to monetary policy. There is not any series of questions that you can ask on the fly that they are not going to be able to brush aside. That is why you need a rule. I agree with John Taylor about some of the reasons for the rule, but I believe one of the most important is that Congress has to fulfill its obligation to monitor the Fed, and it cannot do that now because the Chairman of the Fed can come in here, as Alan Greenspan has said on occasion, Paul Volcker has said on occasion, and they can tell you whatever it is they wish, and it is very hard for you to contradict them. So you need a rule which says, look, you said you were going to do this, and you have not done it. That requires an answer, and that I think is one of the most important reasons why we need some kind of a rule."

Later in the hearing Senator Brown raised the issue of the audit the Fed bill, saying "Dr. Meltzer, be specific, if you would, about your thoughts about the Audit the Fed proposals," and Meltzer answered: "I think you do not get what you want. Suppose you knew everything. Suppose you found out that the Fed chooses its policy using a ouija board. What would you be able to do with that? What you want to do is get something which permits you to see that the policies that are carried out, are carried for the benefit of the public....The information you want has to come from having something very deliberate that you know they are going to do and that they tell you they are going to do, and you are able to say, 'You did not do it,' or, 'You did.'"

Meltzer later elaborated: "The problem with many of these proposals [release transcripts earlier, audit the Fed, etc.] is they don't look at what would be the circumvention....What I think the Congress needs to do, it needs to face up to its responsibilities. Its responsibility is to be able to say to the Fed: 'You told us you were going to do this, and you didn't do it. Why?' That's what the rule gives you. That's more important....You have to get a discipline in the Fed to tell you what it is going to do and then do it." At this point Senator Shelby interrupted: "That is more important, isn't it?" and Meltzer answered "That is more important than any other single thing you can do. You do not have the ability now to monitor them."

Speaking of John Taylor, back in 1993 he explained that a rule did not have to be completely rigid, for instance you could instruct the Fed to target nominal GDP:
Moreover, in my view, policy rule need not be a mechanical formula, but here there is more disagreement among economists. A policy rule can be implemented and operated more informally by policymakers recognize the general instrument responses that underlie the policy rule, but who also recognize that operating the rule requires judgment and cannot be done by a computer. This broadens the definition of a policy rule significantly and permits the consideration of issues that would be excluded under the narrower definition. By this definition, a policy rule would include a nominal income rule in which the central bank takes actions to keep nominal income on target, but it would not include your discretionary policy.

These two quotes get to the heart of the problem. Right now Congress has no ability to hold the Fed accountable. Most Congressmen don't even know that the Fed is trying to keep the growth path of aggregate demand fairly stable. Most don't know that a steep fall in AD is a monetary policy failure. Most don't know that 2008-09 saw the steepest fall in AD since the 1930s. So they have no way to hold the Fed accountable.

Congressmen need to be able to say, "You guys at the Fed are supposed to deliver this price level path, or this NGDP path, and you are not doing so. Why not?" That's how you hold institutions accountable. That doesn't mean Congress must pick the target. One option is for Congress to set a broad mandate, such as stable prices and high employment, and instruct the Fed to set a clearly identifiable path of some sort of macroeconomic aggregate that is consistent with that mandate, against which their policy could be judged.

Right now it doesn't seem like the Fed is even holding itself accountable. In the minutes for 2009 I'm not seeing Fed officials saying, "we should have done X, and we failed." There's no sense that they are actually responsible for the path of AD. And yet when things go well (say in the period of stable and low inflation prior to the Great Recession) the Fed takes credit for that success. They can't have it both ways, but right now they do have it both ways. Opinion leaders like Robert Samuelson say that we should not be bashing the Fed. Here George Selgin pushes back, he understands that even the Fed must be held accountable:

Mr. Samuelson worries that Fed "bashing"--by which he seems to mean any criticism of the Fed that seeks to justify a reduction of its considerable power--"adds to uncertainty and subtracts from confidence" upon which economic growth depends. In truth, the Fed's actions are themselves often unpredictable, and especially so when it comes to their influence on the long-run course of prices and spending. Were the Fed really a sort of Ambrose Light of financial markets, as Mr. Samuelson imagines it to be, Fed watching, instead of being a growth industry, would be about as useful--and as boring--as watching paint dry.

But the Fed needs more than mere watching. It needs scrutiny. It needs criticism. Above all, it needs to be reigned in--not for conservatives' sake, but for everyone's. Mr. Samuelson may not like it. But I, for one, intend to keep bashing away.

Surely the Fed must be trying to do SOMETHING. In that case the Fed needs to tell us what that something is, and how we can judge what sort of success they've had. It's that simple.

HT: Frank McCormick, Evan Soltas

Comments and Sharing


COMMENTS (16 to date)
Brian Donohue writes:

Excellent post.

Jose Romeu Robazzi writes:

La guerre! C’est une chose trop grave pour la confier à des militaires.

Michael writes:

You assume that politicians know anything at all or care about finance or economics. I do not know if that assumption is warranted: Occupational backgrounds. Even in b-school, most people did not care at all about econ. Most are just looking for any heuristic that doesn't make them really think about how hard econ can be given all the moving variables.

David S writes:

I'm not sure I agree - the only thing worse than an unaccountable Fed would be a Fed controlled by Congress. The Fed, as it is now, seems to be at least somewhat of a meritocracy - they had to prove to those around them that they knew what they were doing. Under Congress control, the only qualification for the job will be your political connections.

I think there is enough oversight via 2 mechanisms: 1) the existing appointment process, and 2) because whenever something "goes wrong" Congress can and does threaten to put them on a leash. Since they don't want to be on a leash for many reasons, the threat alone is sufficient.

Jon writes:

I'm not saying this because I believe the audit critique but the audit the fed movement had nothing to do with judging policy and acting as if does is a misdirection.

The conspiracy theory behind audit the fed stemmed from two scenarios, one of which is plausible/classic corruption.

The first one is that the fed was manipulating the gold market In the 1990s by repo transactions which sold the same gold many times over. Mind you the gold has stayed at Fort Knox the whole time so this is not an obvious fraud to detect.

The second is that the set of firms designated as broker-dealers has an opaque selection process and there is a revolving door between employment at the fed and their primary dealers. Further there is question whether the OMO have been done at market rates or if the fed would deliberately offer the broker dealers favorable terms.

Neither of these points has to do with the Policy decisions by the fomc.

Scott Sumner writes:

Michael, You said:

"You assume that politicians know anything at all or care about finance or economics."

Just the opposite. This post tries to address the fact that they know nothing about economics, and hence as Meltzer pointed out are not able to evaluate the Fed's statements when they testify.

David, You said:

"I'm not sure I agree - the only thing worse than an unaccountable Fed would be a Fed controlled by Congress. The Fed, as it is now, seems to be at least somewhat of a meritocracy - they had to prove to those around them that they knew what they were doing. Under Congress control, the only qualification for the job will be your political connections."

I must have done a poor job explaining my proposal as I certainly don't want Congress controlling monetary policy---that would be a disaster. I want the Fed to be accountable. Right now they are not, as there is no clear metric by which to hold them accountable.

Jon, You said:

"I'm not saying this because I believe the audit critique but the audit the fed movement had nothing to do with judging policy and acting as if does is a misdirection."

I'm not sure if this comment is directed at me, but this post has no bearing on the "audit the fed" debate. Meltzer's right that the audit the Fed argument is a sideshow, the real problem is the lack of a clear target.

Gordon writes:

Scott, regular readers of the MM blogs probably have seen a number of posts on those blogs related to the importance of clearly defined policy targets because expectations matter so very much. I'm wondering if some of the misunderstanding about this post is because the readers have not been acquainted with this.

A writes:

What good comes out congressional accountability unless congress holds economic opinions? The Fed can just say that they changed the rule because their economic expertise suggested it be so. "Audit The Fed" looks rather like an institutional validation of the macroeconomic thinking of congress people.

If the goal is to force the Fed to clarify their intentions, perceived duties, and metrics, then incentivizing them to perfect congress speak seems counterproductive.

A writes:


It seems like the controversy is more about the nature of "accountability". Is it accurate to believe that Congress is a group of philosophers who will force the Fed to be logically coherent and consistent? Or does it create an institutional imperative for congress people to present macroeconomic theorizing, as well as for the Fed to improve congress people manipulation?

Gordon writes:


Scott wasn't saying that the Fed needs to be accountable to Congress. I've seen enough of his writings as well as some of the other MM economists to know that he wants the Fed to have a clear cut target for the market. He referenced the Congressional testimony because it's a symptom of the fact that no one knows what the hell the Fed's target is.

Scott, Marcus Nunes, and other MMists have compared having a clearly defined target to the destination of a passenger ship. If the ship is drifting off course, the ship's steering needs to be adjusted until it is back on course. But as things stand now with the Fed, no one knows if the destination of the ship is Charleston, New York, or Boston. And none of the passengers know if the ship is off course. Also, if you were to question the different ship officers on the destination or the course corrections that were needed, you'd get different answers. Congress questioning the Fed is akin to a passenger committee trying to get answers out of the ship's officers. But that committee should not have control over the course or the destination.

Jimo writes:

It is true that the Fed is not accountable to Congress. How would Congress hold the Fed accountable? How often would Congress scrutinize the Fed? What actions would be taken if Congress decided that the Fed was not meeting the policy criteria? How would Congress enforce these actions?

From my reading of various blogs, it would not be easy to choose agree on a policy because of a lack consensus between various economic and political factions.

In some ways a fixed policy statement could be implemented by a computer program. Of course, economic data is imperfect and the possibility of making specific decisions on imperfect data can cause bad outcomes. The data problem would cause Congress to weigh various datum items based on beliefs about their soundness.

Daublin writes:

This would be obvious anywhere but Washington.

Somehow Washington has some sort of magic spell around it that affects even otherwise level-headed people. They just have to believe that somewhere in the world there is a group of people who are honest and good and just trying to make everything better for us. They then attach that belief to Washington.

I really hope the Fed doesn't matter that much. The fact that it's so hard for the world's best economists to pin this down suggests that maybe that is the case.

Scott Sumner writes:

I'm surprised there are people who don't think institutions should be held accountable. Would you be OK with Zimbabwe-style inflation? If not, does that mean you think the Fed should be held accountable?

Bernanke blamed the Fed for the Great Depression. Was he wrong to hold the Fed accountable? He blamed the BOJ for the Japanese deflation. Was he wrong to hold the BOJ accountable?

George Selgin writes:

Although Scott didn't intend to be understood to be endorsing proposals for further Fed audits, I think those proposals are being too readily dismissed by many on the grounds that they would undermine the Fed's independence. In fact, the claim that removing present restrictions on GAO audits of the Fed (especially by allowing GAO "audits," meaning investigations, of various open-market transactions and programs) would have the effect of letting Congress influence the stance of monetary policy is about as ludicrous as some of the more preposterous conspiracy theories offered by Fed critics. Moreover, it would be a simple matter to include language in a bill relaxing current GAO Fed auditing "exemptions" that would explicitly rule out any GAO "second guessing" of the direction of basic monetary policy. The GAO has itself endorsed such an approach--and has done so since 1978, when the current law was established.

In short, I'm with Scott of course on the general matter of the Fed having to be held accountable. But I also think that a greater scope for GAO "auditing" of Fed activities is a perfectly reasonable means for achieving this goal.

Roger McKinney writes:

Congress will never impose a rule upon the Fed because most mainstream macro economists don't want one and Congress has no idea what rule to impose.

Congress needs to do better at picking Fed chairs who promote a policy of rule following. But there is the rub, isn't it? There are only a handful of economists who want the Fed to follow any rules so Congress has very few people to choose from.

The big banks have bought Congress with campaign contributions and determine Fed policy.

Jimo writes:

I firmly believe the Fed should be held accountable for its actions and the results derived from its actions. I think the Fed should largely be held accountable for the Great Recession and the limp recovery. If the new policy accountability had been in place. What would have changed and how would the changes been implemented?

Comments for this entry have been closed
Return to top