Via Mark Thoma, James Bullard has a reply to Tim Duy that sheds further light on his argument. In this piece he concentrates his criticism on the concept of “potential output.” Here I actually agree with Bullard; potential output is hard to measure, may move around at times, and can easily lead monetary policy astray. But I still have major problems with his overall argument:
So, what Irwin’s picture is doing is taking all of the upside of the bubble and saying, in effect, “this is where the economy should be.” But that peak was based on the widespread belief that “house prices never fall.” We will not return to that situation unless the widespread belief returns. I am saying that the belief is not likely to return–house prices have fallen dramatically and people have been badly burned by the crash. So I am interpreting your admonitions on policy as saying, in effect, please reinflate the bubble. First, I am not sure it is possible, and second, that sounds like an awfully volatile future for the U.S., as future bubbles will burst once again.
Now we can see the damage to monetary policy by the widespread (but false) view that this recession was caused by the bursting of the housing bubble, and the resulting financial crisis. Interestingly, this is exactly that same argument that the Fed used in the 1930s; “Do you mean you want us to go back to the false prosperity of 1929?” Most economists would now say; “Yes. The 1927-29 expansion saw no inflation at all. There is no evidence the economy was overheated in 1929.”
Now I’m willing to concede that this perception may be wrong, about both 1929 and 2007. Both years might have seen output slightly above the mythical “natural rate.” There’s really no way to know for sure. But it’s important to recognize the nature of Bullard’s argument—in some respects it’s a return to 1930s macro after a 70 year hiatus. Why did the Fed ignore bubble theory for so long? Two reasons. Modern macro models really don’t have much of a place for bubbles, except to the extent they throw monetary policy off course and lead to excessive NGDP growth. But NGDP growth followed a fairly stable path along a 5% trend line after 1990. It might have been slightly too high in 2007, but nothing of the sort that would cause a major crisis. The second reason is that between 1929 and 2000 we didn’t seem to have bubbles that influenced the business cycle. The huge 1987 stock market crash didn’t even dent the economy. So macroeconomists naturally (and correctly) ignored bubbles. Bullard continues:
This is admittedly a qualitative story, but much of the rhetoric about the U.S. economy during this period fits this description. So it is not that the bubble destroyed potential, instead it is that actual output was higher than properly-defined potential during the mid-2000s, and then it crashed back as the bubble burst.
But where is the evidence for this? I think everyone agrees that housing construction was too high in 2004-06. But that doesn’t mean RGDP was too high. After all, the resources that went into housing could have come at the expense of other sectors. Housing is only about 5% of GDP. Is there evidence that AD was too high? Maybe a tad too high, but nothing out of the ordinary compared to other business cycle expansions. How about SRAS? Bullard mentions a rise in labor supply caused by the housing bubble, but I don’t see the logic. What would be evidence for a rise in labor supply? Presumably you’d see unusual patterns in the employment to population ratio during the 2000s, but I just don’t see it. His best argument would be that the desire to work has been dropping sharply for many years, and that this increased preference for leisure was temporarily covered up by a housing bubble that mysterious caused people to want to work more. Then when they found they could no longer build houses, they decided they didn’t want to work at all. They didn’t like other jobs being offered. I’m probably being unfair here, but I just don’t see the argument, or the data to support such an argument.
Of course you could make a “recalculation” argument, but that wouldn’t fit Bullard’s claim of a semi-permanent downshift in trend output. And it doesn’t even fit the data, as the big housing construction crash was mostly during the period of January 2006 to April 2008, and yet the unemployment rate was unusually low during that period of supposed “recalculation.”
Here’s the Bullard argument I find most troublesome:
As I noted earlier, the Irwin description is the dominant view of the U.S. economy. But, as you and many others have stressed, we have not seen the bounce back toward potential that would be suggested by that picture. That is giving me pause, and frankly I think it is giving everyone who follows the U.S. economy pause.
At a superficial level Bullard may be right. We may not ever get back to “potential output” as described in many conventional macro models. But I think at a deeper level Bullard is confusing two completely unrelated issues, the link between monetary policy and NGDP, and the link between NGDP and RGDP. Let’s start with the data. Bullard’s right that potential output is very misleading. In my view employment is a better indicator of cyclical patterns. For example, over the past 14 months the level of employment has risen by somewhere between 2.3 million and 2.9 million. Both of those numbers are above trend growth in jobs. So we are having an (admittedly weak) recovery in jobs. But at the same time RGDP is growing well below the 3% long run trend. Let’s assume Bullard’s right that the potential output estimates are too high, perhaps due to reasons outlined in Tyler Cowen’s The Great Stagnation. In that case the jobs numbers show us slowly recovering toward a reduced trend. The next question is; why isn’t the recovery faster? That’s where I suspect Bullard and I would part company.
Now let’s look at NGDP growth. During this period of recovery NGDP has been growing at about 4%, which I consider tight money. So the explanation for the slow recovery is quite easy, money has been too tight. Bullard would presumably reject that argument, as the Fed insists that money has been very easy—even though they’ve followed the same low interest rate/high monetary base policy that the Hoover Fed adopted. But even if money has been easy, it doesn’t help Bullard’s case. Now the mystery would be why the “easy money” hasn’t boosted NGDP, not why slow NGDP growth hasn’t triggered fast RGDP growth. There’s no direct effect of easy money on RGDP. It works, if it works at all, by boosting NGDP growth. Then if we assume wages and prices are sticky in the short run, the higher NGDP growth will (partly) translate into higher RGDP. But only if you get the desired NGDP growth. And that hasn’t happened. In contrast, during the first 6 quarters of the 1983-84 recovery, NGDP grew at an 11% rate, and RGDP grew at a 7.7% rate. The proximate cause of the slow recovery is obviously slow NGDP growth. The only debate is (or should be) whether the Fed or the fiscal policymakers are to blame for the slow NGDP growth.
[BTW: Ben Bernanke has called for fiscal stimulus at various times, tacit admission that the Fed sees an AD shortfall.]
The Fed’s current position is very similar to the Fed’s stance in 1932—they’ve done their job, provided low rates and a greatly enlarged monetary base, now why isn’t the economy recovering? I think we now know why the economy wasn’t recovering in 1932, money was way too tight. As soon as FDR adopted a (Woodfordian) price level target, the economy turned around on a dime. Even Vincent Reinhart, the guy who (according to Lawrence Ball) turned Bernanke from a bold advocate of monetary stimulus at the zero bound to a timid Fed-clone, now says the Fed isn’t moving fast enough in a Woodfordian direction.
If we get adequate NGDP growth, growth high enough where Bernanke doesn’t have to worry about Congress suddenly getting religion on the deficit, and if the economy still doesn’t recover, then we can entertain real theories of the recession.
Mark Thoma also links to Barkley Rosser:
The basis of this argument is that the fall in output following the collapse of the bubble has reduced the rate of real capital investment. Of course, the sharpest decline was in the real estate construction sector, the part of the economy that was most severely distorted by the housing bubble, and we should expect that part of investment to remain reduced for some time.
If Rosser means “reduced from 2006 levels” I obviously agree. But housing construction is also greatly reduced from normal levels, and hence there’s no reason why some recovery isn’t possible. What about the huge “overhang” of housing caused by “overbuilding?” Actually housing construction over the past 10 years has been well below normal. That’s why many young people are living with their parents. They have no job, because the Fed hasn’t provided enough AD (aggregate demand).
I do like Bullard’s new argument better than the old one. This is what I strongly objected to earlier:
A better interpretation of the behavior of U.S. real GDP over the last five years may be that the economy was disrupted by a permanent, one-time shock to wealth. In particular, the perceived value of U.S. real estate fell substantially with the 30 percent decline in housing prices after 2006. This shaved trillions of dollars off of the wealth of the nation. Since housing prices are not expected to rebound to the previous peak anytime soon, that wealth is simply gone for now. This has lowered consumption and output, and lower levels of production have caused a significant disruption in U.S. labor markets.
This is doubly wrong. There is no reason to assume that building too much investment goods (housing) would cause people to want to build fewer consumer goods. Indeed in a well functioning economy overbuilding in one sector should cause resources to re-allocate into other sectors. Output of consumer goods would probably rise (although it’s theoretically possible that this could be offset by an increased preference for leisure that was even bigger than the fall in labor utilized in housing construction.)
And even if I’m wrong about the effect of less housing wealth on consumer goods production, the “consumption and output” connection is completely unjustified. You can’t simply assume that lower consumption translates into lower output; it could lead to higher investment or exports. Indeed that’s the sort of adjustment predicted by most equilibrium models. Now you might make a Keynesian argument that less wealth depressed AD, but Bullard is claiming the problem is not a lack of AD.
I notice that Bullard did not repeat this argument in his second piece, so the widespread criticism by economics bloggers may have nudged him in the right direction. This represents a sort of victory for the economics blogosphere, and especially Tim Duy.