Bubbles and potential output

In the boom years

As in many recent strands of the macro literature – think of the debate on balance sheet recessions – the body of work that studies bubbles in the presence of financial frictions emphasizes the importance of distributional concerns for the macroeconomy by studying the kind of transfers that bubbles implement.

In a voxEU contribution, Alberto Martin and Jaume Ventura provide some useful background on the theory of rational bubbles. Samuelson (1958) and Tirole (1985) laid the foundations for such a theory by portraying bubbles as a remedy to the problem of dynamic inefficiency. Their argument is based on the dual role of capital as a productive asset and a store of value. To satisfy the need for a store of value, economies sometimes accumulate so much capital that the investment required to sustain it exceeds the income that it produces. This investment is inefficient and lowers the resources available for consumption. In this situation, bubbles can be both attractive to investors and feasible from a macroeconomic perspective. For instance, a pyramid scheme that absorbs all inefficient investments in each period is feasible and its return exceeds that of the investments it replaces. The Samuelson-Tirole model provides an elegant and powerful framework to think about bubbles. However, the picture that emerges from this theory is hard to reconcile with historical evidence. In the Samuelson-Tirole model bubbles raise consumption by reducing inefficient investments. As a result, bubbles slow down capital accumulation and lower output. In the real world, however, bubbly episodes tend to be associated with both consumption booms and expansions in the capital stock.

In a set of related papers Alberto Martin and Jaume Ventura investigate mechanisms through which bubbles can increase potential output. The starting observation for both papers is quite simple: Bubbles not only reduce inefficient investments, but also increase efficient ones. In the first paper – an OLG model with productive and unproductive agents – bubbles induce two kinds of transfers: an intergenational transfer, which is welfare reducing as it reduces consumption and capital accumulation by the youngs; and an intragenerational transfer, which helps mitigate the effects of financial frictions (to model financial frictions, the authors simply assume that the unproductive guys cannot lend to the productive guys). During bubbly episodes, unproductive investors demand bubbles while productive investors supply them. These transfers of resources improve the efficiency at which the economy operates, expanding consumption, the capital stock and output. In the second paper, bubbles may also transfer resources towards efficient investments through the credit market. This happens when the prospect of a future bubble raises the net worth of efficient investors, therefore allowing them to expand their borrowing and investment.

After the collapse: investigating the Bullard story

David Andolfatto gives some background on the infamous speech. This speech is really about how to interpret the recent performance of the U.S. economy. Is the conventional interpretation, that we are far below “potential” GDP owing to “deficient demand,” the correct view? Or should we instead be thinking in terms of a large negative shock to “potential” GDP, with unemployment returning slowly to its natural rate, according to its normal dynamic?

More interesting than the initial speech – which, because of poorly argued statements gave rise to the initial outcry by bloggers – is Jim Bullard’s “geeky” response to bloggers. In this text, Bullard argues that potential output – as currently measured by the CBO – does not correct for output growth driven by sunspot equilibriums. Because this drove some of the growth in the 2000s, the current output gap even by a production function metric would be smaller since the widespread belief that sustained this sunspot equilibrium – that “house prices never fall” – will not and should not return. 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.

Tim Duy looks more closely at the CBO estimate and sees two recent episodes of output in excess of CBO potential, both of which were associated with what the author believes were asset-price bubbles.  Furthermore, if this was a significant overestimate of potential output in during the housing bubble, we would have expected more severe inflationary pressures. At the core of Bullard’s asset-bubble model of potential GDP shifts it is a demand story with maybe a second-order labor supply aspect. This sounds quantitatively irrelevant and does not explain why no other source of demand can compensate for the lost housing bubble and induce higher labor supply.

Paul Krugman argues that Bullard changed positions following the flood of online commentary. In the first version, Bullard seemed to be arguing that the wealth loss from the burst bubble represented a real destruction of economic capacity. Now he seems to be making a quite different case: that the economy in 2005-2007 was operating at an unsustainably high rate of capacity utilization, driven by the bubble. Krugman likes Bullard’s new version better than the old one, but it’s still wrong. Demand in the mid-naughties was not, in fact, at fever pitch. Yes, we had very residential construction and high consumer spending. But we also had record-high trade deficits, so that overall demand wasn’t that vigorous, after all.

Scott Sumner also notes that although 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. 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 had 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.

Potential output, flexible price output, and trend output

In his post, Jim Bullard notes that the different definitions of potential output make the discussion blurry. Bullard argues that the terms of the debate should be based on the appropriate definition of the output gap according to the available New Keynesian literature: that the output gap is the distance between the actual level of output under sticky prices and the flexible price level of output. For that matter, the flexible price level of output would fluctuate continuously in response to shocks hitting the economy. Tim Duy argues that somewhere in the background there is need for some structural change, toward export and import competing industries.  That said, it is hard to believe that this is the primary story given that the downturn negatively affected employment across almost all industries.  If structural adjustment was the primary issue, I would have anticipated a narrower range of affected industries.

Menzie Chinn had a useful post a few months ago where he presented the difficulty of linking the concept of potential output in DSGE models with the one used for policymaking. DSGEs (and their predecessors, RBCs) are models of the business cycle. As such, they focus on the deviations from trend. However, in order to predict where the economy will be in one year, given current conditions and policies, one needs to know what the trend is. In other words, extracting the cycle from the trend is critically important. Camilo Tovar notes further that the dynamics obtained do not match those required by policy makers, weakening the usefulness of DSGE models as policy tools. For a detailed and technical discussion on some of these issues, see Gorodnichenko and Ng.

What this episode says about the American econ blogosphere

This episode illustrates the importance of the American economic blogosphere in the economic policy debate. And if you believe Paul Krugman, the blogosphere did not only manage to get a proper written response from a policymaker but basically made that person realized that part of his argument was wrong. Scott Sumner also sees this as a victory for the blogosphere, especially for Tim Duy, who was at the center of this controversy. Brad DeLong adds that it is also a victory for Mark Thoma who built “Economist’s View” into a must-visit platform and then shared that platform with Tim Duy.

Brad DeLong further notes that those of us who have wasted our time in this business over the past decade got into it because we were horrified by episodes that showed the conventional press corps engaged in extraordinary efforts to lower the level of the economic policy debate and felt that somebody, somewhere needed to try to do something constructive. The Bullard-Duy exchange is a powerful piece of evidence that we – and those of us whom the fish that the conventional news media have hired in their attempt to turn into birds, and those of our fellow travelers in the conventional press corps who kept the faith and tried to keep the cold altars lit – have to a surprising degree succeeded: our time was not, after all, wasted.