My model for how the Fed thinks about withdrawing QE

Bernanke believes QE works, but having been caught off-guard once before, in 2007-2008, he doesn’t fully trust his own judgment.  He fears some risk of bubbles, or excess private disintermediation, in either case resulting from low interest rates.  He lets Tarullo and Stein carry related messages to the markets to signal possible fears without having to endorse them.

Let’s say he assigns these risky scenarios a fairly small p = 0.05.  Still, another financial collapse would be a disaster, all the more for political economy reasons.  Bernanke has spent down his own political capital and these days Republicans are more likely to be obstructionists.  Fear of that disaster leads him to withdraw QE sooner than his “most likely to be true” opinion thinks prudent.  Economists respond by defending the “most likely to be true” opinion, and by arguing moralistically rather than probabilistically.  That doesn’t convince Bernanke, because said economists can only convince him that they are likely right, not that he should obliterate his p = 0.05 fear of being wrong.  The current policy course continues, early withdrawal from QE is contemplated, and economists complain all the more.  Outside observers find it hard to understand the disconnect.

Here are some related probabilistic considerations from Brad DeLong.


"Outside observers find it hard to understand the disconnect."

Not sure you are helping with your reading of the Fed tea leaves. I know very little about any of this, but I am dismayed at outside economists who act like they know it all about monetary policy. (We don't even 'know' what the initial conditions are ... how can we know the right path?) An economist who "doesn’t fully trust his own judgment" is a good economist ... nothing radical there.

One tidbit that only applies to staff forecast (and not necessarily anyone you mention in the post): the staff prepares a modal forecast, that is, the most likely outcome for the economy ... not an average forecast which falls out of most forecasting models. And then we do risk scenarios around the baseline, always asking are they balanced, skewed, etc. You can go read old Greenbooks to see this. The FOMC, of course, can use whatever information or framework they see fit, but I do not see why much in this post (even if may not be entirely correct) should be baffling.

Great post. I love reading about the inside of Ben Bernanke's mind. I'm writing an upcoming Benny B fan-fic called "Ben There; Munn That." It's a crossover fan-fic, where B-B meets Olivia Munn, and he has to combat inflation and she has to get to a photo shoot! Wacky stuff.

Tyler ought to make this a weekly feature.

Monday morning column: "What I think Bernanke is thinking this week"

Monthly ought to do.

Still, another financial collapse would be a disaster, all the more for political economy reasons. Bernanke has spent down his own political capital and these days Republicans are more likely to be obstructionists. Fear of that disaster leads him to withdraw QE sooner than his “most likely to be true” opinion thinks prudent.

This doesn't make sense at all. If I wanted to cause a financial collapse, I'd go about engineering an unexpected and sudden reduction in the growth of the money supply. Even if I made it an expected reduction in the money supply growth rate, it would put pressure on the financial sector, and who knows what the result might be.

On the other hand, if I wanted to ensure their was no financial collapse, I'd go about reinforcing over and over that the money supply will grow with a predictable floor rate, and at a higher rate if a 2007-2008 level systemic collapse looks likely.

You realize that that's a theory, right? Begging the question, etc.?

You realize everything's a theory, right?

As theories go, I think "Chaotic changes in the growth of rate of money create problems" is better supported than "Systematic changes in the growth rate of money create problems".

True, except the second one isn't really on the table. What we actually have is "chaotically high rates of growth of reserves and money relative to a depressed economy we hope will restore trend growth without any wacky overshoots or distortions on a fragile financial system."

In the 'got Bin Laden' movie the Jessica Chastain character sits in the meeting to decide whether to carry out the attack. The chair asks for likelihoods and several moderate statements ensue. When asked she says, "100 percent! ! Oh, all right, none of you believe in one hundred percent of anything. Ninety-five percent." Got to play by the rules of the game. Let's get on with naming Janet Yellen and let The Benank return to peaceful Princeton. Promote him to the Inst. of Advanced Study for a well-earned rest.

According to Delong, two things have changed. First is the ability of the financial system to produce assets that people trust. Second is the ability of the US economy to produce employment gains.

As for the first, the policy reaction has been to buy a large portion of the trash that the financial system produced. The Federal Reserve is saying in effect "these fools and liars who made up all this garbage, sold it to you and screwed you are now sound because we bought the garbage".

As for the second, the policy reaction has been what? Low interest rates that hopefully create another bubble somewhere. All the while the costs of doing business in the US when it comes up results in temporary measures or measures that look great but in fact increase costs.

Bernanke will probably keep doing what he is doing because he is stuck in a rut of his own creation. Maybe there is no exiting the zero interest rate policy. This policy tells us two things: first, money is of no value because I can rent it at no cost, second, there is no yield, no return on risk, so play it very very safe. And third, which is really ugly, is that a rise in interest rates, or any change will cause a catastrophe in any environment which has acted as if money is free. We will watch this lesson in Japan over the next while.

A while ago I was talking to a mechanical engineer about systems, specifically ones that didn't work, and he said that there is a learning curve in these things. What that meant was that some sucker spends a pile of cash building and installing an expensive system that probably won't work very well, and the engineer walks away saying, 'hmm, I won't do it that way next time". It was a remarkable admission by a professional. I think we are in that situation with macro.

One of the ignored risks of QE is that the Fed is buying a lot of the trusted paper (Treasuries). Yes, they have to stoop to buying junk paper at these volumes, but the financial system is running low on quality assets.

@8 - "the financial system is running low on quality assets" - it costs the treasury practically nothing to buy an asset (what's the cost of electronically printing money?). So even if the assets are "worth", say, only half as much as the amount paid, it is still a great deal for the treasury. But of course there is actually no lack at all of quality assets - if necessary the treasury can start buying shares, land and houses and so on, in fact all the world. There is is really no practical limit to a Treasury determined to print money.

I suspect that what you mean by a non-high quality asset is a highly leveraged one, if the leverage is related to an underlying asset, then if the Treasury succeeds in raising real asset values there should be no problem.

What I mean is the financial system needs "risk-free" assets for collateral. You can pledge other assets as collateral, but they're often pro-cyclical. This doesn't bear on the U.S. financial system, but imagine lending money to a Chinese homebuilder who uses copper as collateral. If the Chinese housing market goes bust, there is a very good chance that copper collateral goes bust as well. The Fed is sucking the good collateral out of the financial system and that is creating the potential for greater volatility during the next downturn.

Could it be that the main benefit of the QE has been to raise the stock market and make rich people richer and when it ends the stock market will crash and make everybody poorer? I don't see this QE generating any new real assets.

Take out the Occupy Wall Street tone, not that it's wrong. Maybe monetary manipulation never does exactly what you want it to.

Yes, but the side-effects need not be random. There could be an assymetricity favoring the rich. Doesn't Inflation hurt at the bottom most?

That's what she said.

Maybe, but I don't think so. Inflation's first order effects are to hurt creditors and help debtors. Same as in William Jennings Bryan's "cross of gold" days.

I was reading this post, and I thought it sounded just like de Long's paper and speach.

The other part of the story was that de Long ascribed the 5% probability scenario -- of withdrawing too early -- to a disaster.

the paper is well worth reading.

Unemployment is still sky high. Does he have forecasts saying that it will be down soon?

Here's a cynical model. They launched the expanded QE late last year and they're not going to end it in less than a year. However, they see Japan and the U.S. stock markets going up and up. I doubt they keep up on the financial scuttlebutt, but it reaches them that people are starting to talk of a 1987 style crash. (Japan crashed 20 percent quickly, with one day falling 9% intraday from top to bottom, so now they have confimation if they're thinking at all along these lines). They can see the weaker data out of China and use a ruler to extend the stock market rally into September/October. They know there are German elections coming in September. Bernanke gets out in January. If the wheels fall off, he wants to make sure it's the next guy's fault.

The problem for this model is that Ben has to simultaneously believe that he is seriously wrong in his understanding/predictions but that he is also the guy to steer the ship through that 0.05%. If his model crashes he shouldn't have any fear of having lost his "political capital" to the obstructionists since their position would then be the favorite for the best action to salvage the economy in the second crisis. His main fear should be doing to little while being labeled as a guy doing to much which would then lead to doing less than his already inadequate job.

I heard an economist say last week that Bernanke's exit model was "he's going to go back to Princeton and let Janet Yellen figure it out."

So, this all depends on there not being another financial crisis. That makes me feel better.

Comments for this post are closed