The constant revision of estimates of potential output and thus of the output gap also underscore one of the difficulties policymakers have in trying to use gaps as a guide to policymaking in real time. Indeed, Athanasios Orphanides and Simon van Norden have argued that a major problem that gave rise to the great inflation of the 1970s was the mismeasurement of the perceived output gap. They explained how the Fed consistently relied on estimated output gaps that were subsequently revised and ended up being much smaller than initially thought. They argue that policymakers' reliance on estimated gaps led to overly expansionary monetary policy and the resulting high rates of inflation.

A different conceptual approach to defining a gap is implied by a class of economic models in wide use today the new Keynesian dynamic stochastic general equilibrium, or DSGE, models. DSGE models explicitly posit that firms have some pricing power; that is, there is imperfect competition so that a firm can choose to sell more of its output by lowering its price or to sell less of its output by raising its price, and the firm will set its price at a markup over marginal cost to maximize its profits. DSGE models also assume that firms are able to only adjust prices infrequently. This form of sticky prices, together with imperfect competition, allow monetary policy to have real effects in the short run, while remaining neutral for the real side of the economy in the long run. The sticky prices generate distortions that mean allocations and output can be inefficient in the face of shocks. In these models, the efficient level of output is the level of output that would prevail in the absence of the sticky prices and other market imperfections that allow deviations from perfect competition. In this framework, the relevant output gap to be addressed is the difference between the efficient level and that level generated by the distortion introduced by the sticky prices and market imperfections. The behavior of the efficient level of output is unlikely to be a smooth or a slowly evolving series like the CBO concept. In fact, it could be quite volatile and may bear little or no resemblance to the traditional concept of potential used by the CBO and others. Efficient output would be altered by changes in technology that affect productivity or changes in agents' preferences. The role of monetary policy in these models is to react to economic conditions in a way that minimizes the potential for distortions arising from the price stickiness or other market imperfections. The general policy prescription is to minimize the gap between output and the efficient level of output. In the absence of unexpected events that lead firms to change their desired markups over marginal cost, or other real rigidities like real wage rigidities, this would be equivalent to stabilizing inflation.

The accumulating evidence does appear consistent with the financial crisis and the associated recession having reduced the potential growth rate of our economy somewhat during the past few years. In particular, slower growth of potential output would help explain why the unemployment rate has declined in the face of the relatively modest output gains we have seen during the recovery...

Notably, because of ongoing weakness in labor demand over the course of the recovery, nominal wage increases have been roughly offset by productivity gains, leaving the level of unit labor costs close to where it had stood at the onset of the recession. Given the large share of labor costs in the production costs of most firms, subdued unit labor costs should be an important restraining influence on inflation.

WSJ:  How will you judge when it will be the right time to begin tightening policy?

LACKER: Tough, tough question. It is really hard to write down a recipe or a check list about this. The growth rate matters, more than the overall magnitude of slack in the economy. I would again emphasize that we need to be sure we act before inflation expectations shift. We need to act in a way to prevent inflation expectations from shifting.

See Chairman Bernanke's comments on the same topic earlier in the week.

I have been critical of gap-based analyses of inflation dynamics in the past.  Those criticisms still apply:

• Theoretical issues are unresolved. 
• Gap measurement issues are acute. 
• Empirical relationships between gaps and inflation are shaky.

Still, the idea of “global output gaps” is one way to frame the recent criticism of the Fed and promote fruitful debate.

We are barely cruising above what we at the Dallas Fed call “stall speed.” Annual real gross domestic product (GDP) growth below 2 percent has predicated every recession since 1970. If we continue to barely clear the 2 percent hurdle, the pace of economic recovery will be insufficient to create the number of jobs the United States needs to bring down unemployment significantly in the foreseeable future. If we cannot generate enough new jobs to sufficiently absorb the labor force over the intermediate future, we cannot expect to grow the final demand needed to achieve more rapid economic growth.

Let me do a little math for you. The San Francisco Fed estimates that the potential level of output is increasing roughly 2½ percent a year due to growth in the labor force and increases in productivity. Hence, over the next two years, potential output will increase by about 5 percent. My forecast is that real GDP will increase about 8 percent during that period, or 3 percentage points more than potential output. This implies that the output gap will shrink from its current level of negative 6 percent to around negative 3 percent by the end of 2011. In fact, I don’t expect the output gap to completely vanish until sometime in 2013.

Third, central banks need to be very careful in using information extracted from market data when conducting monetary policy. The anchoring of inflation expectations is a very important benchmark for assessing whether the stance of monetary policy has become too expansionary and whether there are risks of falling behind the curve. However, the pre-crisis period and the crisis itself have shown that expectations are not always formed in a rational way and might themselves be influenced by the behaviour of the monetary authorities. Quite often markets participants form their expectations by looking at central banks’ behaviour, on the assumption that the latter have better and more information about underlying inflationary pressures. Under these circumstances, market-based inflation forecasts tend to be biased and may react in a perverse way following a tightening of monetary policy. In particular, inflation expectations may be revised up, rather than down, after an interest rate increase, especially at the beginning of a tightening cycle. If the monetary authorities’ assumption that inflation expectations are well anchored proves to be wrong, the whole yield curve might move upwards and the decision to increase the rate of interest may result in a much sharper tightening than expected. ^[4]

Another problem with inflation expectations concerns their measurement, especially in periods of financial turbulence, when the liquidity of the underlying instruments can affect their signalling content. A further distortion may arise when the central bank is itself a major buyer and holder of indexed-linked assets, which may distort prices and thus the information content. Measurement problems arise also with other components of the analytical framework, such as the output gap. Experience shows that measurement of the output gap varies over time. For instance, when measured ex post, with the data available until 2008, the US output gap over the period 2002-2004 turned out to be very small, even non-existent, while ex ante it was estimated at around 2% of GDP. ^[5]

Editor's note:  We don't ordinarily include comments from central bankers in other countries in this database, but some comments are too insightful to pass up.

I’m on the record not putting as much weight on the slack argument as others might. It is a factor but it is not as big a factor as many people make it out to be. You always have to look at slack in combination with inflation expectations. And right now, inflation expectations are about as uncertain as they’ve been since the 1980s. There are widely varying opinions about what might happen over the next three or four years. Some people think you might get a lot of inflation in this circumstance with a bloated balance sheet and fiscal deficits. Other people think that the economy will recover slowly and that will put downward pressure on core. I’d be in the former camp on this issue.

Even as the economy begins to recover, substantial slack in resource utilization is likely to continue to damp cost pressures and maintain a competitive pricing environment. I expect that the persistence of economic slack, accompanied by stable longer-term inflation expectations, will keep inflation subdued for some time. Indeed, if inflation expectations were to begin to ratchet down toward the actual inflation rates that we have experienced recently, inflation could move appreciably lower.

We can never directly observe the level of economic potential--it is largely inferred from the behavior of related variables, like output, productivity, costs and prices. In that regard, a widely discussed upside inflation risk is the possibility that as a result of the financial turmoil and deep recession, the extent of economic slack in the economy is not as great as is commonly estimated. One possibility is that the steep drop in investment has caused a decline in capital services that could damp the rise in productivity. Another possibility is that the needed reallocation of resources away from a number of sectors--including finance, construction, and motor vehicles--will have a restraining influence on potential output for a time. In addition, prolonged periods of unemployment could have adverse effects on the skills of workers and their attachment to the labor force.

The financial crisis may also have affected potential output by reducing the ability of financial markets to effectively lubricate the flow of credit throughout the economy--and to allocate capital resources to their most productive uses. The deterioration in the health of the financial system conceivably may have disrupted the credit allocation system enough to seriously impair the efficiency of business operations, and this impaired efficiency could show up at some point in more meager gains in productivity. And, some have argued, as governments seek to build more stable financial and economic systems, they may impede innovation and efficiency.

Each of these arguments contains a grain of truth, and they are worthy of further research. But in my view, the cumulative reduction in aggregate demand has been much greater than any possible cutback in potential supply. The unemployment rate has risen by 5 percentage points in a very short period, and capacity utilization in industry hovers just above its lowest level in the history of the series, dating back to 1948. The downward pressures on both prices and labor compensation reinforce my impression that our economy is operating well below its productive potential. And, if anything, productivity has been surprisingly strong, not weak, in recent quarters.

Recent studies done at the Chicago and San Francisco Feds find little evidence that sectoral reallocation or other factors are increasing the unemployment rate or reducing measured output gaps on a very large scale.¹⁶ So I believe that resource gaps remain substantial today. That's a significant mitigating factor against inflation pressures.

Many observers see substantial slack in the economy that could persist for some years. Economists' more formal term for slack is “output gap.” We at the Atlanta Fed see a meaningful output gap developing, but in our view it is smaller than would normally be associated with the weak pace of growth we expect over the next couple of years because all the obstacles to the natural pace of growth already mentioned have brought down the economy's potential for the medium term.

I do think that when output gaps reach the level that we are currently seeing that it's no longer the case that we can really debate that the output gap exists. I think there clearly is an output gap and the experiences that in previous recessions that inflation has tended to fall as after the recession.

I think that's the reliable empirical regularity. And the size of the current output gap will be a drag on inflation. You fall into the output gap camp. Mr. Plosser does as well. He's simply saying we shouldn't put too much weight because it's very difficult to measure them.

In response to a question about Charles Plosser's output-gap critique.