stephen roach and the output gap

  1. 13,013 Posts.
    lightbulb Created with Sketch. 99
    Global: Long Live the Output Gap

    Stephen Roach (New York)

    Macro certainly has its moments in captivating financial markets. I suspect another one of those moments is now at hand. The debate over deflation has been given a new lease on life by Federal Reserve Governor Ben Bernanke. He has now set the risks of deflation squarely in the context of an “output gap” framework -- long a central tenet of macroeconomic analysis (see his July 23, 2003, speech, “An Unwelcome Fall in Inflation?” available on the Fed’s website). Using this macro construct, Bernanke has concluded that even if the US economy now enters a period of solid recovery, the risks of deflation are going to be with us for some time to come. I couldn’t agree more.

    Like most concepts in economics, the output gap is a complex restatement of a very simple premise -- the inflationary consequences of disparities between aggregate supply and demand. Alas, what always sounds simple in macro rarely is. Economists attempt to get at the notion of aggregate supply by assessing the growth of “potential” output (GDP) -- defined broadly as the sum of labor force growth and trend productivity. In essence, the output gap is then calculated as the difference between an economy’s growth potential and its actual level of aggregate activity. When output gaps are at “zero,” it’s the best of all worlds -- supply and demand are in perfect balance and, at least theoretically, the economy is at full employment and able to enjoy the luxury of a stable inflation rate. When demand exceeds potential -- a positive output gap -- inflation can be expected to accelerate. Conversely, shortfalls from potential -- a negative output gap -- are invariably associated with falling inflation; they reflect excess slack that gives rise to a phenomenon referred to as “disinflation.” As such, recessions are generally depicted as disinflationary macro events, while recoveries are thought to be inflationary.

    As presented in this fashion, the output gap is all about levels of aggregate activity. This stands in contrast to the growth rates that color most of our impressions about the performance of economies. This is a critical distinction. An economy can be growing at, or even above, its potential growth rate and still have ample margins of slack capacity in labor and product markets; disinflationary pressures would prevail in such instances. Conversely, a fully employed economy growing slower than its potential growth rate would still be biased toward an inflationary outcome. In other words, initial conditions matter. An economy’s growth speed is not enough, in and of itself, to determine the ups and downs of the inflation cycle. The verdict is critically sensitive to the state of resource utilization.

    Which takes us to the case in point. In his latest speech, Fed Governor Bernanke has used this framework to make some important inferences about the economic and policy outlook for the United States. His most salient conclusion, in my view, is the premise that America’s output gap is likely to remain wide even in the face of a fairly vigorous recovery in the US economy. Bernanke comes to that conclusion by inserting a few key numbers into the output gap framework. Operating under the premise that America’s potential growth rate is around 3%, he points out that even a 4% growth outcome in 2004 will not close the output gap for an underemployed US economy. In that context, a further deceleration in inflation can be expected. Inasmuch as inflation is already quite low -- averaging 0.9% for the core CPIU in the first six months of 2003 -- it’s that next leg of disinflation that becomes so problematic. While Bernanke celebrates America’s achievement of what he calls “the de facto equivalent of price stability,” that may not be a reason to jump for joy. In fact, it doesn’t take much of an imagination to envision what lurks on the downside of this threshold. On that basis alone, the Fed has good reason to remain vigilant in the fight against deflation -- even if the US economy now moves into a solid recovery mode, as the central bank and most other forecasters expect. And that, of course, is exactly the key conclusion that points to a protracted period of monetary accommodation.

    Ben Bernanke is hardly alone in reaching these conclusions. Analysts at the OECD have come to the same realization. In their June 2003 assessment of the global economic outlook, OECD economists estimate that America’s output gap will hit 2.1% of potential GDP in 2003 -- the widest such shortfall in the US economy since 1991, when it rose to an estimated 2.5%. Moreover, over the four-year period, 2001-04, the OECD estimates that the United States will record a cumulative output gap of 5.9% of potential GDP. That’s two full percentage points larger than the 3.9% widening of slack expected in the euro area and only fractionally below the 6.0% cumulative output gap expected in Japan, the land of deflation. Nor can the OECD be accused of sounding the “output-gap alarm” on the basis of an overly pessimistic growth forecast for the US economy. Their current prognosis calls for a 4.0% increase in real GDP in 2004 -- not unlike the outcome Bernanke has built into his stylized depiction of deflation risks. In other words, among the major economies of the industrial world, America is expected to be right at the top of the charts in feeling the full force of disinflationary pressures through the end of 2004. That’s obviously a new role for the unquestioned engine of the global economy.

    Having said all that, it pays to take a deep breath and remember that this is macro -- not nuclear physics. The output gap analysis hardly provides an ironclad guarantee of deflation risk and the policies required to cope with such risks. As has long been noted, it is based on a number of heroic assumptions -- including, but not limited to, assessments of trend productivity, the inflation-stable unemployment rate, and implied rates of full-employment capacity utilization. Moreover, the output gap construct is largely a “closed” macro model -- driven mainly by domestic considerations. In this era of globalization, macro models must be more “open” -- allowing for the possibility that aggregate supply curves are now global in scope. Not only is that true for tradable goods, in the form of Chinese-based outsourcing platforms, but it is now increasingly true in once non-tradable services, as exemplified by Indian-based IT-enabled service exports. In his latest speech, Bernanke concedes that the output gap could be considerably wider if aggregate supply curves were underestimated; in that case, disinflationary pressures would be even more intense as a result. The ever-increasing pace of globalization of goods and services suggests that is hardly idle conjecture.

    Notwithstanding these important caveats, I think it pays to take the output gap seriously in assessing the risks of deflation. While the framework is hardly perfect, it provides a reasonably good assessment of the balance between aggregate supply and demand in the US economy. If anything, globalization tells us that the risks to this macro construct are tilted more toward an understatement of slack rather than the opposite. That points to a potential underestimation of the intensity of disinflation, suggesting that it will take a lot more than a year or so of vigorous growth in the 4% vicinity in 2004 to end the current deflation scare for a low-inflation US economy. Conversely, it follows that any setback in the pace of recovery from the desired 4% trajectory would only widen the output gap even further -- intensifying the disinflationary pressures already bearing down on the US. Either way, the output gap is not about to disappear into thin air. And the longer it persists, there’s no escaping the bottom line for a low-inflation US economy -- the greater risk of deflation.

arrow-down-2 Created with Sketch. arrow-down-2 Created with Sketch.