Thursday, August 22, 2013

What the FOMC is Thinking: Parsing the FOMC Minutes

Minutes for the July 30-31 FOMC meeting were released this week. There are a few interesting tidbits in them that are worth discussing. You'll recall that, up to and around the June 18-19 meeting, the FOMC was having trouble getting its message across. Apparently driven by the interpretation of statements by Fed officials - primarily Ben Bernanke - the real yields on TIPS (see the chart for the 10-year yield) increased substantially from early May, and took a big jump around the time of the June FOMC meeting. The chart also shows the breakeven rate on ten-year Treasury bonds (nominal yield minus TIPS yield), which fell during the same period, though it has increased somewhat since June.

In the June meeting, the FOMC took an unusual step in authorizing Bernanke to expand on the FOMC statement in a press conference. The key extra information that Bernanke added concerned how "tapering" might take place. Basically, Bernanke said that the rate of long-maturity asset purchases by the Fed could be reduced toward the end of the year, and might be reduced to zero at about the time the unemployment rate crossed the 7% threshold. Perhaps surprisingly, none of that information actually made it into the FOMC statement following the July 30-31 meeting. Clearly, given what you see in the chart, the FOMC appears to have thought that communication had not gone so well. What financial market participants seem to have read in the tea leaves was a tightening message, but that wasn't the message that the Fed wanted to transmit.

Here's the spot in the July 30-31 meeting where the FOMC tries to figure out what happened:
In discussing the increases in U.S. longer-term interest rates that occurred in the wake of the June FOMC meeting and the associated press conference, meeting participants pointed to heightened financial market uncertainty about the path of monetary policy and a shift of market expectations toward less policy accommodation. A few participants suggested that this shift occurred in part because Committee participants' economic projections, released following the June meeting, generally showed a somewhat more favorable outlook than those of private forecasters, or because the June policy statement and press conference were seen as indicating relatively little concern about inflation readings, which had been low and declining. Moreover, investors may have perceived that Committee communications about the possibility of slowing the pace of asset purchases also implied a higher probability of an earlier firming of the federal funds rate. Subsequent Federal Reserve communications, which emphasized that decisions about the two policy tools were distinct and underscored that a highly accommodative stance of monetary policy would remain appropriate for a considerable period after purchases are completed, were seen as having helped clarify the Committee's policy strategy.
First, the bit about "little concern about inflation" reflects Bullard's dissent at the June meeting. Presumably Bullard thought that the important news at the June meeting was that inflation was lower than expected. So the message the markets should have received was "easing" rather than "tightening." Second, the part about "decisions about the two policy tools" being distinct doesn't make any sense. If decisions about the two policy tools currently in play - quantitative easing (QE) and forward guidance - are not linked, then they certainly should be. Maybe that's why people are confused.

There follows a discussion of what Bernanke said in the press conference following the June meeting, and the committee seems to have "reaffirmed" this - they don't have quarrels with the general idea. But the FOMC appeared gun-shy about putting any of that stuff in the FOMC statement:
participants considered whether it would be desirable to include in the Committee's policy statement additional information regarding the Committee's contingent outlook for asset purchases. Most participants saw the provision of such information, which would reaffirm the contingent outlook presented following the June meeting, as potentially useful; however, many also saw possible difficulties, such as the challenge of conveying the desired information succinctly and with adequate nuance, and the associated risk of again raising uncertainty about the Committee's policy intentions. A few participants saw other forms of communication as better suited for this purpose. Several participants favored including such additional information in the policy statement to be released following the current meeting; several others indicated that providing such information would be most useful when the time came for the Committee to begin reducing the pace of its securities purchases, reasoning that earlier inclusion might trigger an unintended tightening of financial conditions.
So, they are learning. Too much information can be a bad thing.

Following that, there is a discussion in the minutes about tweaking forward guidance. Should the 6.5% unemployment rate threshold be lowered? Should there be more information about what happens after the threshold is crossed? Thankfully, those ideas appeared to have been killed. In this case, I think simpler is better.

This gives you an idea about the fuss that goes into some words in the FOMC statement:
The Committee decided to indicate in the statement that it "reaffirmed its view"--rather than simply "expects"--that a highly accommodative stance of monetary policy will remain appropriate for a considerable time after the asset purchase program ends and the economic recovery strengthens.
You might think there's not much difference between reaffirming your view about what you expect as opposed to just expecting - e.g. "Jane reaffirmed she is pregnant." But, it might actually matter to people whether the Fed just decided today that they expected something rather than deciding, say, six months ago.

So, what to expect now? Here's my prediction: Tapering will be announced at the October meeting, and the Fed will reduce asset purchases by $10 billion per month until the purchases end in mid-2014. Then, the 6.5% unemployment rate threshold will be crossed in mid-2015. Then, who knows? Of course, some dramatic event - financial crisis in China, foreign exchange problems in other countries, whatever - could blow that forecast out of the water.


  1. Two questions:

    1. Why October? Don't you think the FOMC needs the press conference to explain the changes?

    2. Why do you think that tapering will be date based (i.e. $10 billion a month) rather than based on economic conditions?

  2. 1. Forgot about that. Though they may be considering having a press conference after every meeting, so that no meetings look "special."
    2. That's what they seem to have been doing. They say that economic conditions will matter for the rate of purchases. But so far the only way economic conditions seem to matter is for when they end the program. They seem to want to end the program gradually, so they'll say something like "the rate of asset purchases will be reduced at a rate of $10 billion per month unless economic conditions warrant a change" or some such. You can tell I'm not practiced in constructing these statements.

    1. Yes, press conferences come with the September and December meetings, but not with October. October seems like about the right timing, given how they think about things. But we know now that the FOMC might be inclined to let Bernanke take the heat in the press conference on major decisions. On the other hand, Bernanke's on the way out, so maybe he doesn't want to "own" that part of the exit strategy. And look what happens when he talks too much in public.

  3. Why does it not make sense to just stop buying bonds and go "all in" for MBS?
    MBS are obviously useful,and last days' indicators prove that a rise in the mortgage rates might do a lot for a major slowdown on a still fragile housing market.
    But now focusing on Treasuries. We have a short-term problem. Should we not be looking for a short-term solution before bothering with anything else?
    What we need is demand now, no matter how we get it. The problem is within retail sales and all those things. We need consumption, right?
    A lot of it, we have a big gap to fill, and probably some more extra as what I am going to suggest could do some damage to investment (which I think could in this case be offset by some government spending in infrastructure... but I do not want to buy a war with austerians, so let me keep quiet about this for now). Would it not be wise to keep short-term rates low but let long-term rates rise as much as "we could"?
    It is (as far as I know)well accepted that high inflation "calls for" high interest rates. Given this, perhaps we could consider that keeping long-term rates low gives limited room for inflation to grow, and creates some kind of "resistance", financially speaking. I think that allowing for higher long-term rates would definitly boost, or at least would do some good, to expectations of inflation, something that is so widely regarded as our "ultimate hope".
    As everybody is coming to the conclusion that rates are not much of a solution, I do think that they would not pose much of a threat if they were moderatelly (or not so moderatelly, that could be another thing to rethink) higher, and would certainly do something for inflation, which would probably (or at least possibly, and that is not bad in the current situation) give some good results.

    1. Is it me or this is plain nonsense?

    2. Well, it isn't plain, but it does appear to be nonsense.

  4. higher spreads between short and long-term would lead to a higher supply of loans, as it is currently a supply-side problem in this case.