Scott Sumner  

Should the Fed raise rates? Will the Fed raise rates?

What I Didn't Get to Say to Mi... What I Didn't Get to Say to th...

These two questions are much more similar that we tend to assume, despite the normative/positive distinction emphasized in EC101 textbooks.

They are not identical questions in the short run---say the October or December Fed meetings---but over a long period of time they are pretty much the same question.

To see why, start with the thought experiment that the Fed had a 2% inflation target, or a 4% NGDP growth target, and used a futures market to implement its policies. If they pegged the prices of CPI futures along a 2% inflation rate, then the money supply and interest rates would be completely endogenous---controlled by the market, not the Fed. Interest rates both would and should adjust along the path required to keep inflation expectations at 2%.

Even under the current policy regime, where there is a bit more discretion, there is actually far less discretion than most people assume. The Fed had originally planned to raise interest rates in mid-2015. They also wanted to raise interest rates in mid-2015. Why didn't they? Because the markets began telling them that a rate rise was inconsistent with their macro goals, especially 2% inflation.

That's not to say that the Fed has no discretion. They are free to make mistakes in the very short run. (That's what discretion means, freedom to make mistakes.) They are free to set rates at a level that will cause inflation to diverge from the target. But if they want to avoid a macroeconomic disaster (and they do, even the ECB does) then they need to quickly "make-up" any error with future rate settings.

Here's another way of making my point. I argued against a rate increase in September, and will probably argue against one in December. If it turns out that 2 or 3 years from now the Fed is setting rates where the Fed currently expects to set rates, then I will have to concede that my current views are wrong, and that a rate increase in December would have been called for.

In contrast, if 2 or 3 years out in the future the Fed is setting rates where the market currently thinks they will be setting rates, then the Fed hawks (and even some doves) that now favor a rate increase in December will have clearly been wrong.

What discourages me most is not that people are right or wrong (there's a good chance my views will turn out to be wrong.) Rather what discourages me most is that people don't seem to even understand that after the fact it is possible to ascertain who was right and who was wrong about monetary policy. For instance, in retrospect Richard Fisher was clearly wrong in advocating a rise in interest rates in July 2008. In retrospect the Fed was wrong to not cut rates in September 2008 (as Bernanke acknowledges in his book.) But we actually know far more than that. We know that the Fed set rates too high throughout late 2008. We know that the Fed hawks have been consistently wrong since 2008, every single time they disagreed with Fed policy. And I mean consistently wrong, not right once during a single debate in the past 7 years. Not once. But I don't see the hawks acknowledging that fact.

There seems to be this weird view that even after the fact, "Who can say?" who was right and who was wrong about policy. But that's not how the world works. If you want to be even close to 2% inflation you need to set rates at a level that leads to that outcome. The market determines the path over any extended period of time, even if the Fed can deviate for a few meetings. The market eventually tells us who was right and who was wrong, it's not "debatable".

Of course the policy target itself is debatable. I prefer NGDP targeting to inflation targeting. But as long as the Fed has an official agreed upon 2% inflation target, they have to eventually set interest rates where the market tells them. They wanted to raise rates last summer, and the markets said no. They currently want to raise rates to 3.5% over the next few years, and I suspect the markets will again say no. If I'm wrong and rates do rise that much, then the markets and I will have been wrong. I will have no right to say, "Yes they raised rates to 3.5%, but they should not have." They have some discretion, but not that much. If they are even close to that level then I will have been wrong.

PS. I hope after reading this post you understand how idiotic it is to complain that the Fed's been running a "low interest rate policy". Nonsense. They have a 2% inflation target, and the markets decided that this target requires low interest rates. When the ECB tried to fight the market in 2011, and raise rates, the were brutally rebuffed, and now eurozone savers face years or even decades of near zero rates, as punishment for the ECB daring to defy the Credit Market Gods.

Screen Shot 2015-10-26 at 3.47.59 PM.png

Comments and Sharing

COMMENTS (15 to date)
TravisV writes:

Prof. Sumner,

I wish more people appreciated that your macro model is the one that is perhaps the most consistent with basic finance theory (CAPM, Damodaran, etc.).

When the Fed provides a positive surprise, stock prices instantly increase, and later that day, thousands of analysts update their valuation spreadsheets by plugging in slightly higher estimates for future sales.

When the Fed provides a negative surprise, stock prices instantly decrease, and later that day, thousands of analysts update their valuation spreadsheets by plugging in slightly lower estimates for future sales.

Your model is actually quite simple and fairly easy to understand for the thousands of people who have performed standard DCF valuation analysis.

P.S.: At least that's what analysts SHOULD be doing. If they're changing the discount rate instead and don't touch future sales estimates then I think they're certainly doing it wrong.

P.P.S.: In general, I sense that most valuation models use a discount rate that's too high but that's offset with future sales estimates that are also too high.

P.P.P.S.: In my first finance class, one of the first things we learned is that the interest rate used should reflect the opportunity cost of capital (i.e., rate of return of alternative potential investments in the real economy).

John Hamilton writes:

I have two unrelated points that I write here, since it seems easier than writing you an email:

(1) Check out this podcast from Vox (, where Ezra Klein pretty much endorses a progressive consumption tax in the minute and a half after ~11:15.

(2) Why don't you talk/blurb about your book coming out in December more? I've heard a lot more about Garrett Jones's Hive Mind, for example. Where's the hype? (Is it on the way?) The book description on Amazon makes it seem like a pretty big deal...

SG writes:

Brilliant post. I've always had the vague sense that finance pundits going on and on about "ZIRP" were clueless, but it wasn't until I read this post that the error became crystal clear.

Njnnja writes:

Isn't the difference really just the difference between agreeing on a policy (say, 2% inflation) and disagreeing about what policy levers to pull to implement that policy, versus disagreeing about what policy should be followed?

I don't think that the Fed hawks want a 2% inflation target. Perhaps more precisely, they would define a 2% inflation to mean lower inflation than what others would mean, such as by including food and energy when inflation in those are high but excluding them when they are low, or talking about threats of long-term inflation when short term inflation is low, and talking about the immediate need to deal with short term inflation when short term inflation is high.

So they weren't wrong, because they got exactly the desired policy that they wanted.

Scott Sumner writes:

TravisV, Yes, if more people thought about monetary policy with the EMH in the back of their mind, there would be much less confusion on these issues.

Thanks SG.

Njnnja, You seem to be claiming that they were corrupt, not wrong. But it's even worse to be corrupt. They are part of a committee that voted democratically for 2%. If they want to remain on the Fed board they are duty-bound to try to hit that target. Their competence will be judged on that basis.

Cyril Morong writes:

I don't know what the Fed should do but people should consider buying this book

The Midas Paradox: Financial Markets, Government Policy Shocks, and the Great Depression

Dan W. writes:


I was going to make the same point that Njnna did. You are being presumptuous that the FED has a hard target. For if they did then they would be obligated to hit that target. But they are not acting that way. The rational interpretation is the FED's inflation target is a desire but not a commitment. They have reasons for declaring the target and they have reasons for not trying hard to make it happen. What might those reasons be?

Njnnja writes:

I don't mean to imply anything nefarious, I just mean to say that all the FOMC votes on (AFAIK) is a FF rate target of x%, or buying $yMM of mortgages each month, etc. I don't believe that it is like a SCOTUS opinion, where everyone who signs their name to a document agrees not only with the result, but also the logic behind it. If they agree with the result, but differ in opinion, they write a separate opinion.

But there is never a concurring statement coming out of the FOMC, and members who vote against the policy just say "I would vote for a FF rate of z% instead of x%." So why couldn't someone who wants to keep inflation very low vote for the same policy as a dove who wants to get higher inflation via the same policy? And if you in fact get low inflation, wouldn't it then be the *dove* who was wrong, not the hawk? If the statement says that "the Fed" wants 2% inflation, and a hawk agrees with the settings of all of the policy instruments that are agreed to but doesn't agree with the same 2% inflation (or has a different definition of inflation than, say, medium term PCE or whatever), do they vote yes or no? I think historically concurring opinions just vote yes to the policy instruments and stay silent about the rationale behind it.

I guess a reasonable question is how is the statement prepared? Who writes it and how do they edit it? Does everyone who agrees with the policy rates have to agree to the wording of the statement? I honestly don't know but given how important a change in any particular adjective can be I guess I should.

ThomasH writes:

The Fed not only waited too long to reduce rates but it waited too long to begin QE and then stopped too soon, before the price level was back on its pre-crisis trend. Not only should the Fed not reais rates in December, it should start back with more QE until inflation has brought the price level back to trend. The Fed's fundamental problem is that is has NOT had an inflation target, it has had an inflation rate ceiling (and maybe a floor, too). This leaves great uncertainty about future price levels that a price level trend target does not.

I agree that an NGDP target would be better than a price level target, but the big improvement would be if the Fed actually had a price level target.

And the inflation hawks have not been "wrong" because they never cared about keeping the price level on a steady trend. They wanted higher rates, period. Any prediction made that X would or would not happen if rates were not raised was purely coincidental. It's like politicians who predict cutting taxes on the wealthy will raise revenue. It's not an argument; it's a pose.

ThomasH writes:

@ Travis

The politicians who gave us austerity flunked both finance and economics classes; the current deficit is NOT an argument (much less with a negative sign) of a proper investment function.

CMA writes:

Previous comment was a typo.

"If they pegged the prices of CPI futures along a 2% inflation rate, then the money supply and interest rates would be completely endogenous"

Endogenous money expansions are debt expansions which have a limit due to willingness to borrow and debt to gdp making borrowers less creditworthy. Too much endogenous debt means we hit the ZLB because as endogenous money expands so does the debt to GDP level.

Exogenous money can be simply created as a good and placed into circulation without expanding on debt. Therefore stimulus can be performed without the same limitations.

[previous comment, corrected by this one, has been removed--Econlib Ed.]

Jason writes:

Scott, will your book be available for Kindle (e-book)?

Jose Romeu Robazzi writes:

Some commentators have pointed out that the data supports the idea that the FOMC targets an inflation band 1-2%. But if that is true, their communication is very bad. If I recall correctly, one of the last statements from chair Bernanke said that the FOMC was willing to accept 2.5% inflation in order to achieve faster job creation. But that never happened (above 2% inflation), despite declining unemployment ...

Larry writes:

I see no possibility of inflation liftoff before 2016. Why not end the suspense with a "see you next year". That would calm everybody down.

Scott Sumner writes:

Njnnja, Ask them. I very much doubt any of them would say "The Fed has set an official inflation target of 2%, but I'm voting for interest rate settings in the hope that this will cause the Fed to miss their official 2% inflation target." If anyone is willing to say that, then I will admit that I am wrong.

Look at the minutes of the meetings, I doubt they'd ever say anything like that.

Jason, I'm told it will be on Kindle at some point in December.

Comments for this entry have been closed
Return to top