Scott Sumner  

What does it mean to say "The Fed did too much"?

The Flynn Effect vs. the Great... The Unbearable Arbitrariness o...

MRU has a video entitled "When the Fed Does Too Much". That led me to wonder, "too much what"? Too much discretion? Too much regulation? Too expansionary a policy? So I decided to watch the video.

Screen Shot 2017-12-05 at 2.41.12 PM.png
By conventional standards the video is perfectly fine. But then I'm not a conventional economist, and I disagree with several parts of the video.

It begins by discussing the theory that the Fed helped to inflate a housing bubble with a relatively low interest rate policy during the period from 2003-05. Later in the video it suggests that monetary policy is a blunt instrument, and that it might have been better to address the housing issue with regulation rather than with tight money--which affects the entire economy. That was my favorite part of the video. Even later, there is discussion of cases where the Fed did too little, as during the Great Depression. It concludes with a look at whether monetary rules such as NGDP targeting could have prevented the Great Recession.

Here are some reservations that I have with the video:

1. No persuasive evidence is presented in favor of the claim that there was a housing bubble. There is no discussion of the fact that Australia, New Zealand, Canada and the UK all had similar home price run-ups during the housing bubble, or that in those countries prices remained high after 2006. No discussion of the recent sharp recovery in housing prices, which have erased much of the post-2006 decline.

2. There is no discussion of what it would mean to claim that the Fed created the housing bubble. Does that mean there is some other Fed policy that would have prevented a bubble? Or that the Fed adopted a policy that was too expansionary for its own dual mandate, and the bubble was a side effect of that overly expansionary policy? Those are two vastly different claims. For instance, in their discussion of the Great Depression, Alex and Tyler omit any mention of the fact that the Depression was triggered by a tight money policy expressly aimed at restraining a stock market "bubble". Surely that case is relevant to the issue of whether the Fed should have adopted a more contractionary monetary policy to prevent the 2006 housing bubble!

In the case of 1928-29, the Fed discovered that a somewhat tighter policy was unable to restrain stock prices, and that only a highly contractionary policy that tanked the entire economy was able to end the stock boom. In fairness, this dilemma is alluded to in the part of the video that views the Fed as a blunt instrument, but unfortunately this problem leaves the opening section somewhat incoherent. What would it mean for the Fed to have caused the housing bubble? Would it mean they failed to create a depression? (Presumably not.)
Perhaps that there was some slightly more contractionary policy that would have prevented a bubble, while still keeping us within shouting distance of the dual mandate? (I doubt that would have been possible.)

If the video had been entitled "Was Fed policy too expansionary?", then the meaning would have been clearer. Then evidence could have been presenting showing inflation and employment data during the bubble period, and comparing the actual performance to the Fed's dual mandate. Without that data, we have no way of evaluating whether Fed policy helped fuel the bubble.

3. Part of the problem with the opening section is that "monetary policy" is implicitly equated with "interest rate policy". The authors don't say this explicitly, but the viewer is led to believe that a more contractionary policy path during 2003-05 would have been a path of higher interest rates. As we saw in the eurozone after 2011, however, that is not necessarily the case. In my view, the adoption of a tighter monetary policy in 2003 would have created a double dip recession (just as in the eurozone) yielding lower interest rates by 2005.

4. The video suggests that a doubling of the money supply in late 2008 was not enough to keep NGDP from falling, and hence that (under NGDP targeting) even more money would have had to be created to prevent a deep recession. Needless to say, this is the part of the video with which I disagree most strongly. Under a regime of 5% NGDP level targeting, the Fed would have been able to keep NGDP growing at an adequate rate with a much smaller increase in the money supply. Interest rates would not have fallen to zero in 2009, and hence the demand for bank reserves would not have risen sharply.

Even worse, there is no mention of the Fed's creation of a program of interest on reserves during late 2008, which was adopted for the express purpose of sterilizing the injection of new money. The video leaves the impression that the Fed tried to adopt an expansionary policy during late 2008 (with heroic injections of money), but it was not enough. This is simply false. The policy of IOR was adopted because the Fed was worried about high inflation, and hence it is not correct to say the Fed was trying to stimulate the economy when they injected lots of money in the fall of 2008. Indeed the Fed's justification for IOR cited its contractionary impact. The monetary injections were aimed at helping the banking system, not boosting the economy.

This video presents the events of 2002-08 the way most economists remember things, not the way they actually occurred. I am currently nearing completion of a book manuscript that tells the story of what actually happened in 2008, not what people (wrongly) remember as having happened.

Despite all of these reservations, the MRU video is an excellent teaching tool for giving students an idea of how most economists understand the Great Recession. Beginning students should be taught the standard view, not the heterodox view. (Perhaps after I read Eliezer Yudkowsky's new book I'll change my mind on this issue, and become less modest.)

However I hope instructors in more advanced undergraduate courses will supplement this standard view of the crisis with something I have written, once my book comes out. Until then I have some articles available online, including a new paper that just came out in the Journal of Macroeconomics.

HT: David Siegel

PS. I noticed a typo in my Journal of Macroeconomics article:

"Indeed, in March of 2009 the dollar rose 6 cents against the euro on the day that the Fed announced QE1."

I meant the dollar price of euros rose by 6 cents on the day that QE1 was announced. That means the dollar depreciated. Not sure how I didn't spot that embarrassing mistake.

Comments and Sharing

CATEGORIES: Macroeconomics , Money

COMMENTS (11 to date)
bill writes:

I don't think the Fed made any mistakes during 2003 to 2006. It wasn't until some time in 2007 or maybe even 2008.

John Hall writes:

Let us know when we can pre-order!

Miltie writes:

I'm sorry I must have missed an earlier post. Are you working on another book?

Jacob Egner writes:

Scott: you said...

Perhaps after I read Eliezer Yudkowsky's new book I'll change my mind on this issue

I'd love to hear your thoughts on the book once you read it. I hope we can look forward to a blog post or two on the matter.

Also, I'll echo Miltie some; what's this new book of yours that you mention?


Cloud writes:

Your new book would be a textbook style kind of book? I can't wait!!

Cloud writes:

BTW, if you don't like that video, this one might make you happier. (Esp around the 4:50 mark)

Mark Bahner writes:
Despite all of these reservations, the MRU video is an excellent teaching tool, to give students an idea of how most economists understand the Great Recession. Beginning students should be taught the standard view, not the heterodox view.

I agree with this only to a certain extent. I'll take the example of the heterodox view that the earth is flat:

1) There would be no need to mention that view because it's 99.999999999999999+ percent certain to be wrong.

2) Let's say it instead had a ~10 percent chance of being right. Then I think ~10 percent of the class time should be spent on explaining it. Perhaps that class time should include a brief explanation of why the heterodox view is likely to be wrong, and the orthodox view right, such as, "When sailing ships disappear over the horizon, the hull disappears first. This shows the earth is curved, not flat."

The reason I think it's important to teach a heterodox view, at least briefly, is because presenting a heterodox view is more interesting then just a recitation of "facts." Also, presenting heterodox views can get students actually assessing what's going on, rather than simply trying to remember "facts."

Scott Sumner writes:

John, Unfortunately the publishing industry moves slowly.

Miltie and Cloud, I am working on two books, a principles text (mainstream) and a book based on my blog posts---The Money Illusion

Jacob, Thanks.

Scott Sumner writes:

Mark, Yes, 10% of the time on heterodox views seems about right.

Mark Bahner writes:
Mark, Yes, 10% of the time on heterodox views seems about right.

Hi Scott,

Sorry, I probably didn't make my point very clearly. What I was trying to say is that the amount of time devoted in class to the heterodox view should be approximately equal to the heterodox view's likelihood of being correct. So since a flat earth is, to put it mildly, highly unlikely to be correct, no time should be devoted to that particular heterodox view.

But if the heterodox view has a 10 or 20 percent chance of being correct, then 10 or 20 percent (respectively) of the class time should be devoted to the heterodox view.

Presumably, you think your particular heterodox view regarding the 2008-2009 recession has much more than a 10 or 20 percent chance of being correct. (And your views have always seemed persuasive to me...but that might just be because you write well and persuasively. I don't think I've seen anyone trying to rebut your view, and promote the orthodox view instead.)

If the heterodox view is almost as likely as the orthodox view to be correct, then I think the two views should be given almost equal time, even in an Intro class.

Scott Sumner writes:

Mark, I'm in no position to offer an unbiased estimate of the probability of my own view being correct. I'm not sure anyone is.

Students should mostly be taught the consensus view.

Comments for this entry have been closed
Return to top