Friday, April 1, 2011

Fed Presidents Talking

There has been a flurry of news reports in the last few days concerning public remarks by regional Fed Presidents, and what that may or may not signal about future policy. There was a presentation by Jim Bullard (St. Louis Fed, a speech by Charles Plosser (Philadelphia Fed) an interview with Jeff Lacker (Richmond Fed), another interview with Narayana Kocherlakota (Minneapolis Fed), and speech by William Dudley (New York Fed).

Some people seem to be complaining about all this public discussion, for example Tim Duy (blogger) objects to it. I think Plosser has a good retort to this, which is:
Because we find ourselves in unfamiliar territory, it is understandable that there is less of a consensus among economists about the right actions to take to promote sustainable growth and price stability. As a result, debates about policy have been robust, with bright and talented people on every side. And it should not be surprising — indeed, it should be reassuring — that debates within the FOMC are similar to many that are carried out in more public forums.

A few months ago, I came across a quotation by the not-so-well-known French essayist Joseph Joubert from two centuries ago. It captured my belief about the importance of this honest debate so well that I have begun to cite it — even if Joubert is not a household name. He wrote: “It is better to debate a question without settling it than to settle a question without debating it.” You may have also heard me quote the American journalist Water Lippmann, who said, “Where all men think alike, no one thinks very much.”

Healthy debate is necessary for better-informed decisions. These debates also serve to enhance the Fed’s credibility and transparency as an institution. We owe it to the public to communicate the thoroughness of those discussions
A strength of US central banking is the decentralization in the system. Semi-independent regional Federal Reserve banks promote competition in ideas, and healthy debate about policy. It would certainly give a false impression for the Fed to give the public the idea, especially now, that it has all the answers. The regional Fed Presidents can play an important role in introducing ideas to the public discussion, and in stimulating active discourse about policy. Further, in all of the public remarks I link to above, these people are for the most part standing behind previous policy decisions, and making mild speculative comments about potential future policy decisions and the issues involved, though maybe there is dissent brewing (see below).

What do the Fed Presidents have to say? Here's Kocherlakota:
If underlying inflation rises to about 1.3 percent by the end of the year, the so-called Taylor rule that describes how policy-makers should calibrate interest rates in responses to changes in employment and employment, would call for a three-quarters percentage point increase in rates, he said.

"That means you should be raising the target rate by more than 50 basis points," he said.
Kocherlakota is thinking very much like a New Keynesian here. The basic approach is: (i) treat historical behavior by the Fed as being optimal; (ii) fit a Taylor rule to the data; (iii) point out that this Taylor rule predicts a substantially negative nominal rate under current conditions; (iv) argue that, since we are at the zero lower bound, we have to do something else; (v) we did the something else, which is QE2; (vi) therefore, the Fed is currently behaving optimally; (v) resume policy as before the financial crisis, but now treating the interest rate on reserves as the policy instrument.

To me, this reasoning seems faulty, as: (a) I have no good reason to believe that what the Fed did pre-financial crisis was anywhere close to optimal; (ii) I'm not sure what the effects of QE2 are relative to, say, moving the fed funds rate by 1/4 point in normal times or, indeed, whether QE2 is having any effects at all; (iii) I have no good reason to believe that we are anywhere close to optimal policy right now.

Now, Plosser and Lacker seem to be roughly on the same page as Kocherlakota, but for different reasons. Plosser is worried about repeating the 1970s:
Some fear that the strong rise in commodity and energy prices will lead to a more general sustained inflation. Yet, at the end of the day, such price shocks don’t create sustained inflation, monetary policy does. If we look back to the lessons of the 1970s, we see that it is not the price of oil that caused the Great Inflation, but a monetary policy stance that was too accommodative. In an attempt to cushion the economy from the effects of higher oil prices, accommodative policy allowed the large increase in oil prices to be passed along in the form of general increases in prices, or greater inflation. As people and firms lost confidence that the central bank would keep inflation low, they began to expect higher inflation and those expectations influenced their decisions, making it that much harder to reverse the rise. Thus, it was accommodative monetary policy in response to high oil prices that caused the rise in general inflation, not the high oil prices per se. As much as we may wish it to be so, easing monetary policy cannot eliminate the real adjustments that businesses and households must make in the face of rising oil or commodity prices. These are lessons that we cannot forget.
The report on Lacker's interview says:
Richmond Federal Reserve President Jeffrey Lacker told CNBC Friday that he "wouldn't be surprised" if the central bank raised interest rates before the end of the year. In an interview at a banking meeting hosted by the Richmond Fed, Lacker said ending the Fed's bond-buying stimulus program also "deserves consideration."
Bullard's presentation discusses some of the exit strategy issues, in particular the "normalization" of the Fed's balance sheet through asset sales, the return of the policy rate (now the interest rate on reserves) to more normal levels, and the timing of those two actions.

Some of this discussion seems to run counter to Bernanke's views and the last FOMC statement that suggest to me a commitment to leave the interest rate on reserves at 0.25% for an "extended period" up to two years. Maybe there is a fight in the making, but who knows?

Suppose we assume that the Fed ultimately intends to return to a state where the quantity of excess reserves held overnight is essentially zero, and the fed funds rate target determines the structure of short-term interest rates, with the interest rate on reserves less than the fed funds rate. Achieving that state will require that the Fed: (i) raise short-term interest rates, initially by increasing the interest rate on reserves; (ii) sell assets. How should the Fed go about this process? This is essentially a question of how the Fed wants to distribute wealth. The Fed could choose to sell its assets first, and then raise the policy rate. This would have the effect of forcing capital losses on those holding long-maturity debt instruments - banks for example. If the Fed chooses to raise the policy rate before selling assets, then the Fed bears the capital losses. Presumably the Fed then makes up for the losses by issuing more liabilities, and we get more inflation than we would have otherwise. Either way, there will be political heat for the Fed to bear.

9 comments:

  1. "How should the Fed go about this process? This is essentially a question of how the Fed wants to distribute wealth. The Fed could choose to sell its assets first, and then raise the policy rate. This would have the effect of forcing capital losses on those holding long-maturity debt instruments - banks for example. If the Fed chooses to raise the policy rate before selling assets, then the Fed bears the capital losses."

    How is this consistent with rational expectations? If the Fed sells assets first but the market has unbiased expectations of its future interest rate hikes, the prices of long-maturity debt instruments will change before the Fed has the chance to sell them.

    ReplyDelete
  2. Even if fully anticipated, there's a different path for the short rate in the two cases. The long bond yields can't be the same in the two cases even though, as you're saying, the value of the assets will depreciate in both cases if the private sector moves closer the date at which it expects the tightening to occur.

    ReplyDelete
  3. To me, this reasoning seems faulty

    You're right, it is faulty. Specious, even. One could just as easily invoke a Taylor rule argument to justify keeping rates low (I suspect Bernanke is thinking along such lines).

    It seems to me that Plosser is talking a lot of sense, and Kocherlakota has come up with the the right conclusion despite flaky reasoning. Having said that, it's far from obvious to me what the path of US inflation is going to be. There's a risk of the 1970s but there's also a risk of 1990s Japan. I guess the Fed should sit tight for the moment.

    My preference would be to unwind the QE sooner rather than later. It's not obvious to me why selling assets first should cause significant capital losses. Wouldn't the Fed just be selling at roughly the same price they paid in the first place?

    ReplyDelete
  4. 1. Yes, I quite liked Plosser's speech as well.
    2. I was thinking the following. Suppose the Fed held the interest rate on reserves at 0.25%, and sold the long-maturity assets in its portfolio, say over a period of one to two years. It gets excess reserves down to zero, then raises the policy rate at the end of the one-to-two-year period. The private sector is then holding the long-maturity assets, and it is the private sector asset-holders that suffer the capital losses. Now, alternatively, suppose the Fed raises the interest rate on reserves now. Then the Fed suffers the capital loss now as it is holding the long-maturity assets when rates go up. The Fed then sells the long-maturity assets over a one-to-two-year period at a loss.

    ReplyDelete
  5. Suppose the Fed held the interest rate on reserves at 0.25%, and sold the long-maturity assets in its portfolio, say over a period of one to two years. It gets excess reserves down to zero, then raises the policy rate at the end of the one-to-two-year period. The private sector is then holding the long-maturity assets, and it is the private sector asset-holders that suffer the capital losses.

    Well, yes, but it's only the same loss they would take if the Fed had never bought the assets in the first place. The same loss all asset holders would take when interest rates go up.

    I don't think the Fed should worry about such losses, they are a normal part of monetary policy. On the other hand, it has to worry about selling the assets after raising rates. I can't see how they can make up the loss without undermining monetary policy, so I guess they would need to be bailed out by the taxpayer.

    ReplyDelete
  6. Kocherlakota's speech was quite interesting. He argues that, in practice, a central bank is not independent of a reckless fiscal authority. That it because allowing a default jeopardizes price stability, so a bail-out (via monetization) winds up being optimal. I would file this under, "things we know but shoulnd't say."

    Exiting by selling assets might well drive up term Treasury rates, increasing the fiscal deficit. As Bernanke recently wrote, a "doom loop" occurs when higher deficits beget higher interest costs beget higher deficits. Kocherlakota is telling us such a loop will likely end in monetization and, presumably, high inflation. Is this relevant or merely academic? Bond prices are saying its the latter, for now.

    ReplyDelete
  7. One last thing: the significance of Excess Reserves lies not in their inflationary potential. Yes, they can be converted to currency, but the banking system can borrow and convert reserves at will at any given FF rate, so what's the difference? Instead, ER's are just the after-effect of an attempt to depress term Treasury rates. The question is, what happens to term rates when those ER's stop accumulating, or worse, decline? The Japanese answer to that question is, "not much". Is this the right analogy?

    ReplyDelete
  8. anonymous,

    Yes, it's the same loss. Some people think that the financial sector is fragile and needs to be pampered. Others seem to think not.

    anon1,

    Yes, I think the fiscal policy issues are important. I'm not sure what you mean when you say "the banking system can borrow and convert reserves..." The banks can't make the reserves go away, unless they are withdrawn as currency. On the question of what happens to Treasury rates, that all depends on what, if anything, happened due to the QE2 accumulation. Is that moving long rates or not?

    ReplyDelete
  9. Stephen,

    What I meant was that, in a FFR targeting regime, the supply of reserves to meet credit (or currency) demand is perfectly elastic. This renders the pre-existence of Excess Reserves irrelevant from the standpoint of the potential money supply at any given FFR.

    I imagine QE has moved rates significantly lower than they would otherwise be, simply because the volume of QE2 has equaled net Treasury issuance since November. What happens to yields post-QE2 will depend on whether markets adopt deflationary expectations (forcing private savings into risk-less assets, as in Japan, and causing nominal yields to fall); or whether, instead, nominal and real yields rise as Treasury is forced to compete for private savings. In either case, the Fed is likely to feel the need to launch yet another round of QE to counter-act the resulting drag on employment growth. There is a "best case" where QE2 actually produces sustainable, organic growth, which allows the economy to withstand higher real yields. I would not bet on it...

    ReplyDelete