Scott Sumner  

Does a measly quarter point matter?

Robert P. Murphy: We've "Been ... Confirmation bias?...

Paul Krugman and Larry Summers have recently argued that the Fed should not raise rates later this year. I agree, mostly because I believe it will prevent them from hitting their announced inflation target, but also because it slightly increases the risk of another recession. If anything Summers and Krugman seem even more concerned about recession risk than I am. Here's Krugman:

Larry Summers argues that a Fed rate hike would be a big mistake; I completely agree. Yet he also suggests that the Fed "seems set" to do this foolish thing. . . . I'm with Larry here: this attitude has the makings of a big mistake. Think Japan 2000; think ECB 2011; think Sweden. Don't do it.
Rather than focus on the risk of recession, I'd like to use this example to illustrate a point that causes endless confusion. (Partly because it really is confusing.)

Tyler Cowen mentioned to me that Krugman's worry that a 1/4% rate increase might push us into recession seems at odds with his frequent claim that the 1/4% interest on reserves can't explain very much. In those earlier posts, Krugman seems to suggest that just a quarter point isn't all that important. And yet the 2000 increase in Japanese interest rates was just a quarter point, and the 2011 ECB rate increase was just 1/2%. Krugman cites both examples, and I think he's right to do so. In both cases a small interest rate increase seemed to tip weak economies back into recession.

To make things even more confusing, I often argue that interest rates tell us nothing about the stance of monetary policy. So how can I argue that a quarter point increase risks tipping us back into recession?

I'll try to explain this with an example. Suppose what really matters is the difference between the Fed target rate (fed funds rate) and the Wicksellian equilibrium rate, which is the interest rate setting that would produce macroeconomic stability (perhaps defined as 4% NGDP growth.) Now here's the tricky part. When the Fed tightens, it slows the economy and reduces inflation. And this reduces the Wicksellian equilibrium interest rate. If the Fed doesn't later cut its target rate, the gap begins to widen, and monetary policy becomes tighter and tighter.

Now in practice the Fed generally does realize its mistake, and begins cutting rates. But not fast enough to close the gap, at least initially. And this leads to a strange paradox, in most cases monetary policy is tight when the Fed is cutting rates, and it's usually expansionary when the Fed is raising rates. That's why I always say that interest rates are not a reliable indicator of the stance of monetary policy. And yet on any given day, money is tighter if the Fed raises rates than if they don't. In the very short run the conventional view is right, but we are mostly living in the long run, when peoples' intuition is backwards.

Because the Wicksellian equilibrium rate is unobservable (at least without an NGDP future market), it's difficult to be certain how much impact any quarter point change in rates has on the economy. All we know is that is could be highly consequential, or it might not. For instance:

1. The increase in reserve requirements in 1937 effectively pushed up short-term rates by a quarter point. We now know that policy was far too contractionary, but don't know the extent to which that would have been true without the reserve requirement increases.

2. The institution of IOR in late 2008 initially raised rates by more than 1/4%, but soon after settled into a 1/4% higher IOR (relative to pre-October 2008.) We also know that in retrospect policy was far too contractionary, but don't know the extent to which policy would have been far too contractionary without the higher IOR.

3. We know that, in retrospect, the 2000 decision of the BOJ to tighten policy was a mistake, as was the 2011 decision of the ECB to tighten policy. But we don't know the extent to which the small interest rate increases caused that tightening, and to what extent it was errors of omission.

If the Fed raises rates later this year, and we then go into recession, critics will rightly point out that policy was too tight. But we won't know the extent to which the interest rate increase itself caused that tightness, we'll just know that it was a mistake--the Fed shouldn't have been tightening.

My only real complaint with Summers and Krugman is that they need to be consistent. If a 1/4% increase in rates could set in motion cumulative forces with vast consequences, and it's certainly possible that it could, then one should not discount the possibility that the October 8, 2008 institution of IOR (which the Fed did for reasons that even it admits had a contractionary intent) might have been a very consequential error that dramatically worsened the recession. Perhaps it even created the liquidity trap. We simply don't know.

Interest rate changes may or may not have big effects; it depends on what's happening to the Wicksellian equilibrium rate. But one thing is clear; one should never discount the importance of an interest rate change by a central bank merely because it looks small.

Some pretty big avalanches have started from a small pebble being dislodged.

PS. In fairness to Krugman, in some of those earlier posts he was also making a separate point, that a liquidity trap could form even without IOR. That's true. But what is not true is the implicit suggestion that a mere 1/4% IOR couldn't plausibly have big effects. For instance, here's a typical quote by Krugman:

Incidentally, small nerdy note. Some people argue that the concept of the monetary base has lost its relevance now that the Fed pays (trivial) interest on reserves. I disagree. Reserves and currency are fungible: banks can turn one into the other at will. But the total of reserves and currency is fixed by the Fed -- nobody else can create either. That, as I see it, makes them a relevant aggregate -- and anyone who believes that all those reserves are sitting idle because of that 25 basis point reward is (a) silly (b) ignorant of Japan's experience, where the BOJ sharply increased the monetary base without paying interest on reserves, and what happened looked exactly like our own later experience.

This post certainly leaves the impression that if the number were larger, than the argument would no longer be "silly." And that's also Steve Waldman's reading, commenting on this quote:

Perhaps there are people in the world who think that paying 25 basis points of interest on reserves means that base money doesn't matter, but I have not met any of them. I certainly agree with Krugman that those 25 basis points have a pretty negligible macroeconomic effect now.

Again, the 1/4% might be irrelevant, as we saw Japan fall into a zero rate trap without IOR, or it might be really, really important. But the fact that it's only a measly 1/4% is not the deciding factor.

Comments and Sharing

COMMENTS (14 to date)
E. Harding writes:

"But one thing is clear; one should never discount the importance of an interest rate change by a central bank merely because it looks small."

-I don't see how that follows from the rest of your post, which I found perfectly understandable and acceptable.

Maybe what Krugman and Summers are really fearing is the risk of policy inertia were the Fed to even negligibly start tightening by raising rates.

Dan W. writes:

In the past 35 years every cycle of increased rates has been followed by a recession. Each recession has ended with rates lower than they were in the previous cycle. In the past 5 years simply holding rates at "zero" as resulted in economic slowdowns that have required direct monetary infusions of liquidity. I will go out on a limb and prognosticate:

(a) an increase of rates will result in a recession

(b) absent additional infusions of liquidity the US economy will enter a recession.

foosion writes:

My simplistic analysis, which I thought you also believed, is that the Fed raising rates is a signal that it wants the economy to slow. The exact mechanics are less important than that the Fed has a tremendous ability to get what it wants (more so on the contractionary side than the expansionary side, at least around zero).

Even 1/4% is enough to signal desire by an entity that has the ability to make it happen.

bill writes:

When you write: "here's atypical quote by Krugman"
I think you're missing a space? Or did you mean "an atypical"? I think the former.
Great post.

ThomasH writes:

Isn't the importance of any increase mainly that is confirms the Fed's abandonment of its average inflation (PL) target and increases uncertainty about what the real target really is?

Vaidas Urba writes:


Krugman's claim was that there is no big difference between a permanent liquidity trap at 25 bps compared to a liquidity trap at zero percent. On the other hand, you are considering a transition between 0 bps and 25 bps which is not what Krugman discussed.

As it happens last week I was reading October and December 2008 Fed transcripts. Please take a look at chart 28 "Fed Funds Rate Trades Below the Target" here:
The chart clearly shows that IOR has lowered interest rates which were too high before October 9. Based on that chart, I would say that mismanagement of fed funds market in the second half of September and early October raised rates by more than 1/4% and it was one of the triggers of the crisis. The policy was too tight after since October 9 too, but it was a more usual case of Fed cutting rates too slowly when money is too tight, it was not a case of Fed increasing them.

Scott Sumner writes:

E. Harding, My guess is that you didn't understand the rest of my post. The statement you question is a direct implication of everything I wrote. The entire post was devoted to showing that even small changes in a policy rate can start a cumulative process that has vast implications.

Take another look and see if it make sense.

Dan, I'm afraid I don't follow that analysis.

Foosion, What the Fed wants is of no importance if they lack the ability to make it happen. Here I was trying to explain the mechanism, using the Wicksellian framework. Of course I could have used other frameworks such as the supply and demand for money approach, and the result would be the same.

Bill, Thanks, I'll correct it.

Thomas, That's certainly possible, but it would be another way of looking at the same story.

Vaidas, I strongly disagree. Krugman is clearly implying that even if the Fed had not raised IOR by a quarter point, we'd still be in the low growth zero rate trap environment. And that's simply not necessarily true. If it were true then the next 1/4% rate increase could not matter very much. In fact, that 1/4% could have been the difference between a hyperinflationary boom and a depression.

Waldman read it exactly the same way as I did (as did Tyler Cowen), and Krugman is certainly not a poor communicator. He knows exactly the impression he leaves. Under your interpretation Krugman's argument would make no sense, as the other side was claiming that without the IOR there would be no liquidity trap. So he was clearly considering the counterfactual.

A higher IOR does not cause problems only because it raises market interest rates, it is contractionary because it increases the demand for the medium of account.

Larry writes:

If the Fed wants normalcy, why not get rid of IOER?

Philip George writes:

Is it of any interest that real interest rates are at the level they were in October 2006, a few months before the financial crisis began to unfold?

See the graph on Real Interest Rates

J.V. Dubois writes:

Excellent article. What I find incredibly fascinating in these "monetary policy is manipulating interest rates" discussions is total ignorance of japanese example (and latest western examples).

Since 1995 Japan keeps interest rate in 0% - 0.5% range. So if small changes do not matter does this mean that Japan basically stopped doing monetary policy back in 1995? Then how can you explain that the CPI was at 101.1 points in 1995 and it was 102.78 points in 2014 - that is 1.6% increase in total over almost 20 years? Is this pure coincidence that absence of "conventional" monetary policy brings such an unprecedented "price stability"?

Or could it be that they were using some kind of "unconventional" monetary policy all the time? It seems incredibly plausible given the circumstances. So what was it?

PS: also how much time will it take for something unconventional to become conventional? Is 20 years in japan and now 7 years in the west not enough time? So how long to change it?

Vaidas Urba writes:

"even small changes in a policy rate can start a cumulative process that has vast implications."

I completely agree.

"Krugman's argument would make no sense, as the other side was claiming that without the IOR there would be no liquidity trap"

Yes I think he was arguing with such people. See

"higher IOR does not cause problems only because it raises market interest rates, it is contractionary because it increases the demand for the medium of account."
The Fed has sterilized the contractionary effect of IOR by expanding the quantity of base money, just like Bank of England did when it started paying 4 or 5 percent IOR in 2006. Fed's policy was contractionary in second half of 2008 because the Fed was cutting rates way too slowly.

Scott Sumner writes:

Larry, That's what I keep asking.

Philip, Perhaps it's just a coincidence.

JV, Excellent.

Vaidas, Yes, and Krugman was wrong when he argued with that group. There is a possibility that the IOR might really have been important.

You have causality backwards on IOR. The Fed did the QE first (to save the banks), and became upset that it was having the expansionary effect of reducing the fed funds rate below target. To prevent that expansionary effect, they adopted IOR to sterilize the impact. Thus IOR had a contractionary effect.

Vaidas Urba writes:


I disagree. The Fed was always sterilizing QE. Prior to IOR, the expansion of Fed's balance sheet was facilitated by expanding the amount of Treasury deposits. IOR announcement simply changed the mode of sterilization. How could the change of the mode have contractionary effect? In fact, the new system was more expansionary as fed funds rate was trading above the target prior to IOR, and it started trading below the target only after they started paying IOR.

Kyle writes:

[Comment removed for supplying false email address. Email the to request restoring this comment and your comment privileges. A valid email address is required to post comments on EconLog and EconTalk.--Econlib Ed.]

Comments for this entry have been closed
Return to top