Scott Sumner  

The Fed is caught in a vise

PRINT
Is TOMS Different?... Casey Mulligan Visits Cuba...

I often read the business press, which has a lot of articles on Fed policy. One recurrent theme is that the Fed is sort of kowtowing to the markets. Here's a typical Bloomberg article:


It's the Greenspan put gone wild. Or so the Federal Reserve's critics would have it.

No longer is the Fed just waiting for financial markets to be hit by a bout of turbulence and then lowering interest rates in response -- as former Chairman Alan Greenspan did. Instead, the critics contend, it's become so sensitive to the risk of sharp market moves in the future that it's pulling its policy punches now by repeatedly holding off on raising rates.

"They used to respond, I think excessively, to what financial markets did," said Willem Buiter, chief economist at Citigroup Inc. and a former policy maker at the Bank of England. "Now they don't act in response to the fear that they have of how the market might respond to their actions."

New York Fed President William Dudley suggested earlier this week that the central bank is paying close attention to how its possible policy moves might play out in the currency and other financial markets. The risk in such a strategy is that the Federal Open Market Committee loses credibility and finds itself increasingly hamstrung in carrying out policy, said investment analyst Jim Bianco.

Running Policy

"The market now thinks it runs monetary policy," said Bianco, president of
Bianco Research LLC in Chicago.

In a Bloomberg Television interview on July 29, Buiter put it another way: "The Fed more than any serious central bank unfortunately appears to be the captive of the financial markets."


I happen to be a market monetarist, and thus I think the markets should be running monetary policy. My biggest fear is that the Fed will stop letting markets run monetary policy the moment the economy tips into the next recession.

The Fed's critics overestimate the extent to which the Fed controls interest rates. I thought about this issue recently after reading a Jeffrey Hummel paper entitled "The Myth of Central Bank Control Over Interest Rates", presented at the recent APEE meetings in Las Vegas. Hummel does not deny that the Fed can nudge overnight bank rates a bit higher or lower, in the very short run. Nor does he deny the Fed's influence on inflation, and hence longer-term nominal interest rates. Rather, he showed that most people vastly overestimate the extent to which the Fed controls rates via the liquidity effect.

Right now, the Fed is basically caught in a vise-like grip of two markets---TIPS spreads and fed funds futures. The fed funds futures market shows short-term interest rates remaining below 0.75% for the next three years. Meanwhile the bond market shows TIPS spread of 1.68% on the 30-year bond. That's consistent with roughly 1.4% PCE inflation, which is the variable that the Fed targets at 2%.

Admittedly, TIPS spreads can be distorted by liquidity factors. However, the absolute level of rates on the 30-year bond---2.31%---is also unusually low. It's difficult to argue that the bond market expects high inflation; given those extraordinarily low bond yields.

Here's the problem faced by the Fed. They currently expect to raise rates by several hundred basis points over the next few years. The bond market thinks they will not do so. In addition, the bond market thinks the Fed will fall short of hitting its 2% inflation target even if they adopt the much lower interest rate track predicted by the fed futures markets. Just think about that for a moment. The markets aren't just saying, "We think you are wrong about inflation". Nor are they just saying, "We don't think you'll raise rates as much as you anticipate." They are saying that the Fed policy will be much more "expansionary" (in the conventional sense of the term) than the Fed expects, and even with this highly "expansionary" policy the Fed will still fail to hit its inflation target.

For this year, it's already clear that the bond market was right and the Fed was wrong. Going forward, the Fed really needs to see strong evidence that the bond market is wrong, in order to raise rates as much as they currently contemplate. If they don't, they'll have to continue holding off on rate increases.

See if this analogy works. I say our family will vacation in Brazil as soon as the Zika virus problem ends. I predict it to end in three months, and thus plan a vacation in November. After three months the Zika problem is still plaguing Brazil. I delay the trip but predict that the problem will go away in another three months, and we'll do the vacation at that time. And this cycle repeats a few more times. Have I changed my plans? Yes and no. I said we'd visit Brazil when it was Zika free, and I have continued to adhere to that rule. I also said that I expect Brazil to be Zika free in three months, and for our trip to occur at that time. That prediction was inaccurate.

The Fed targets both inflation and interest rates, but in very different senses of the term 'target'. The Fed says it will adjust short-term interest rates as appropriate to keep inflation close to 2%. They also make forecasts of what sort of rate increases will be needed, going forward. At various times, the markets make it quite clear that the Fed can't adjust rates as they anticipated, and still hit their 2% inflation target. At those points in time, the inflation target takes precedent, and the Fed backs off on its previously announced intention to raise rates.

All of the Fed's plans going forward are built on the assumption that the markets are wildly inaccurate in their forecasts. Each month that goes by, we find evidence supporting either the Fed or the market's view of reality. During this particular year, the market view has been more accurate, so the Fed appropriately adjusts its policy as needed to hit its 2% inflation target. On occasion, the Fed's view will be more accurate than the market view. In those cases the market will adjust.

The press likes to mock the Fed, portraying them like a big dumb ox with a nose ring, being led to and fro by the markets. But that's exactly how things should be---the Fed should take its marching orders from the markets. Instead, we should fear the day when this no longer occurs---as in September 2008.

PS. Here is an earlier version of Jeffrey Hummel's paper.


Comments and Sharing






COMMENTS (8 to date)
Scott Sumner writes:

Ha! No one saw my original title was misspelled. I guess I spend a lot more time thinking about vice than vise.

B Cole writes:

In the vise-grip of vice.
Good post.
Lately I have been pondering IOER.
Banks generally lend out at 300 basis point spread, that is cost of funds vs. return on loans. But they also have administrative costs and overhead.

Interest on excess reserves is banks getting paid to do nothing. I think it is a USDA program in drag.

The point is, if interest on excess reserves goes to 1% does it make any sense for banks to extend loans or just to collect interest?

The bulk of bank business is extending loans on real estate. There is of course a endogenous vs. exogenous argument on the creation of money.

AntiSchiff writes:

Scott,

I would make a stronger case than your EMH-based one for why the Fed should target the market forecast. Even EMH aside, it's the markets that need to be convinced there's enough monetary expansion anyway, since monetary policy works almost entirely through expectations.

This would be an improvement over the typical stupid arguments favoring this or that pet non-monetary policy to boost market confidence.

James writes:

"The press likes to mock the Fed, portraying them like a big dumb ox with a nose ring, being led to and fro by the markets."

None of these Fed watchers in the press want the Fed to to ignore markets. Their complaint is that the Fed is paying too much attention to financial markets instead of labor and goods markets.

That's an interesting debate for those who believe successful central banking is mostly a matter of choosing what variable to target. I suspect it makes about as much difference as whether a state run nail factory has a production target expressed in tons of nails or inches of nails.

"But that's exactly how things should be---the Fed should take its marching orders from the markets."

That has never happened and it is not going to happen in an organization not subject to market discipline. The Fed is a government sponsored monopoly which doesn't face the risk of bankruptcy if its clients become unsatisfied and take their business to rival entrants.

bill writes:

Check out this link to a post by Ben Bernanke. The Fed would sure simplify its life if it would just focus on an NGDPLT target. How in the world anyone would ever think they could figure out the NAIRU in real time is beyond me.

https://www.brookings.edu/2016/08/08/the-feds-shifting-perspective-on-the-economy-and-its-implications-for-monetary-policy/

bill writes:

PS - in the absence of an NGDP futures market, the Fed could glean some information from a 10 year Treasury paying less than 3% (3% being its own projected terminal r). I hope the problem is simply that nobody tells them the interest rate on the 10 year T, because that problem would be fixable. If they actually knew that the 10 year paid 2.25% in December and are still talking about raising the Fed funds rate (and raising IOR) since then with a 10 year paying 1.37% to 1.65%, then I don't think the Fed should be trusted to handle monetary policy any longer.

Scott Sumner writes:

Anti-schiff, That's right. One way of making that point is that investment today depends heavily on expected NGDP one and two years out. If those are on target, then investment projects are less likely to be cancelled during a period of temporary market turmoil, such as September/October 2008.

James, Financial markets are the best indicators we have of the future condition of the markets for goods and services.

Thanks Bill, I'll take a look.

James writes:

Scott,

"Financial markets are the best indicators we have of the future condition of the markets for goods and services."

That is incorrect. In "best" means minimizing forecast error, the best indicators of inflation, unemployment and output are models based on lagged inflation, unemployment and output. Financial markets can make these models incrementally better but the lagged dependent variables are the most important.

Even if you were correct, there is no reason to expect an organization exempt from market discipline to "take its marching orders" from markets. That has never happened before and there is no reason to expect it to happen.

Comments for this entry have been closed
Return to top