America's potential GDP

Remembering when the future kept getting bigger 

May 24th 2012, 17:17 by G.I. | WASHINGTON 

HOW big can the American economy grow? This week’s Free exchange column tackles the critical question of America’s potential: the maximum output it can sustain given its endowments of capital, labour and technology.

The article notes that economic growth since the recession ended three years ago has averaged 2.5% a year. That is roughly the trend rate of an economy already at full employment. Given that America is still in a deep post-recession hole, such a rate should not be enough to reduce unemployment, and should have left so much spare capacity that inflation ought to have fallen sharply. Instead, unemployment has dropped nearly two percentage points in that time and underlying inflation, after dipping below 1%, is above 2%.

While various idiosyncratic factors can explain this behaviour, it could also be a sign that the crisis has significantly eroded potential GDP, and the output gap is much smaller than generally realised. (This is a topic on which I’ve blogged before, here, here and here.) Since 2005 the Congressional Budget Office has revised down its estimate of potential GDP in the year 2012 by 5%.

Doing this exercise for the late 1990s, a completely different picture emerges. As the accompanying chart shows, in 1997, the CBO estimated potential in early 2001 would be $8.3 trillion (in constant 1996 dollars). By 2001, it had revised that up a whopping 12%, to $9.3 trillion, a figure that looks more reasonable given what we now know GDP actually did.

The CBO’s shifting estimates of potential illustrate two things. One is that potential is almost impossible to pin down in real time since the economy’s equilibrium long-run stock of capital and labour are so difficult to estimate with precision; so we look at what GDP actually did as a hint of what it can do. 

Second, and more important, is that supply (i.e. potential) is itself affected by demand. Potential output is the product of capital, labour and innovation. Since economic booms bring more investment, more risk-taking, and higher labour force participation, they push up measures of potential. The opposite is true of busts. If overall spending is depressed long enough, many workers will experience prolonged unemployment that degrades their skills, making them unemployable; they may eventually quit the labour force altogether. 

Depressed sales also discourage investment in new technology and research, which can degrade productivity and efficiency for years to come. (A counter argument is that depressions may hasten the migration of capital and labour from dying, low-productivity sectors to growing, high-productivity ones. Apparently, scholars are still arguing over whether this happened in the 1930s.) Powerful evidence for this phenomenon comes in a paper that my colleague A.C.S. discussed Monday which found most structural unemployment begins during recessions.

It follows that efforts to preserve demand can also preserve the economy’s supply-side potential. That, too, seems to be one of the lessons of international experience. It is not too late for America to limit most of the long-run damage of its crisis; but it may soon be.

(Note: special thanks to Brent Moulton of the Commerce Department’s Bureau of Economic Analysis for technical advice on how to convert real GDP figures to a common base year.)

Source:  The Economist