The Fed Has Painted Itself in a Corner

Bernanke managed to shoot global markets in the head the day before yesterday, and then, as has become typical when investors throw brickbats at what the Fed has said, various mouthpieces go about talking the markets back up (this case, in the form of San Francisco Fed president John Williams pointing out that the central bank could ease off or reverse its QE exit if the economy faltered).

I welcome readers telling me I’ve missed something, but looking at the Fed’s problem from 50,000 feet, it appears that the the monetary authority appears to have set boundary conditions for its QE exit that it can’t meet.

1. The Fed is committed to communicating interest rate increases well in advance so as to give investors time to adjust exposures. This policy dates from 1994, when an unexpected mere 25 basis point rise kicked off a derivatives meltdown, producing more losses than the 1987 stock market crash. Punters had been almost universally of the view that any rate changes would be reductions.

2. The Fed designed QE by setting fixed amounts of purchases, and not by seeking to achieve certain levels for particular interest rates, or for MBS, spreads. This gives the Fed much less control in how it exits QE. As Marshall Auerback pointed out when QE was implemented, the central bank can’t control both the rate it achieves and the amount of its purchases. Choosing the latter meant it lost the opportunity to target the former. Thus, it can’t assure investors of a measured exit, since it can’t control rates when investors start dumping bonds.

In other words, the Fed wants a gradual exit,6 but the interaction of 1 and 2 means it can’t achieve that. And on top of that, we have:

3. By moving to influence longer-term interest rates and mortgage spreads, the central bank has come to have a bigger impact on more asset classes than it did in the past. So it has managed to set up a badly designed program with far greater consequences than anything it has done in the past. For instance, Nick Timiraos of the Wall Street Journal pointed out on Twitter yesterday, mortgage rates rose considerably after the Bernanke remarks. That’s not exactly positive for the housing recovery, such as it is.

So despite the Fed starting to try to prepare Mr. Market for an eventual end of QE, the factors above strongly suggest, as critics fear, that the central bank will tighten late. But that won’t necessarily be because they’ve read the data wrong. In fact, Fedwatcher Tim Duy’s latest article if anything suggests that the Fed is suffering from a bad case of confirmation bias. Since it needs to believe its policies are working, it is motivated to see a recovery in the data. Plus its need to promote the confidence fairy also biases its perception (studies have shown that lawyers who defend a client that they suspect is guilty often come to believe in his innocence by virtue of having to sell his case). Duy believes that the central back will back out the effects of “fiscal drag,” aka deficit cutting, from its assessment of growth, yet appears not to be allowing for how much worse unemployment is than the headline statistics thanks to the exclusion of discouraged workers (and that’s before we get to underemployment). The Fed thus seems to have a predisposition to read data releases in a self-flattering manner. That would tend to favor an early exit from QE.

However, the market upset it triggered should give some pause. It may believe it has successfully contained any damage; the Japanese market, which had a wild ride, with the Nikkei down 3.3.% in the afternoon, still ended the day modestly higher (although that had more to do with Kuroda than the Fed’s ministrations). But stock valuations in the US are looking frothy in the face of declining earnings, which increases their vulnerability to shock. So if mere talk of ending QE produces this sort of reaction, the Fed is likely to hesitate to end it, even if it does have a bias to optimism in its data readings.

The one reason this might not be as terrible as it sounds is that with labor markets slack, it’s harder to generate real economy inflation than you think. But the longer the Fed carries on with its poorly-designed QE, the more fear-inducing its end will be.