Bernanke Keeps Door Open for More Easing

• In a speech before the National Association of Business Economists, Federal Reserve chairman Ben Bernanke reviewed two important labor-market issues: why the labor market has been outperforming growth, and why long-term unemployment remains so high. In both cases, his findings pointed to the need for continued, highly accommodative Fed policies.
• He argued that the recent strong performance of the labor market may reflect the "flip-side" of fear-driven layoffs during the worst part of the recession. If so, further sharp drops in unemployment will require faster economic growth.
• He also argued that the high level of long-term unemployment remains mostly due to cyclical rather than structural factors.
• The implications? The Fed should keep monetary policy extremely accommodative as it waits to see whether labor-market improvements are followed by faster growth. In addition, we can conclude that QEIII (or a sterilized version of QE III) remains on the table as a possibility for later this year should growth fail to pick up.

Chairman Bernanke reviewed two important labor-market issues before a meeting of the National Association of Business Economists today—issues with important implications for the conduct of monetary policy.

Unemployment and Economic Growth

Unemployment has been falling more rapidly than economists would normally have expected given the tepid pace of economic growth. In fact, over the course of 2011, it is surprising that unemployment fell at all. From the fourth quarter of 2010 to the fourth quarter of 2011, the unemployment rate fell by nine-tenths of a percentage point, from 9.6% to 8.7%, while the rate of GDP growth was 1.6%. Since a growth rate of 1.6% is below even pessimistic estimates of the economy's long-run growth potential, normally one would have expected the unemployment rate to have risen, not fallen.

The long-established "Okun's Law" (really a statistical regularity, not a law) says that the unemployment rate falls by one-half a percentage point for every one-percentage-point excess of actual growth over potential growth. So, if we take a cautious view and peg the economy's potential growth rate at 2.0%, a drop in unemployment of nine-tenths would normally be associated with a 3.8% growth rate, not 1.6%.

In 2012, the labor-market improvement has if anything accelerated, but growth remains modest, most likely around 2% for GDP in the first quarter. So the puzzle remains—and it is one that is greatly exercising minds at the Federal Reserve as they ponder their monetary policy options.

The Fed chairman considered a number of possible explanations for the apparent conflict between unemployment and growth data:

• It might reflect statistical noise (i.e., growth has really been better than estimated). He said that this is "certainly possible," but noted that there's no specific evidence for this yet. In particular, the evidence on gross domestic income (calculated by adding up incomes, in contrast to GDP, which is calculated by adding up spending) currently shows less growth than GDP in 2011, rather than more.
• It might reflect potential workers giving up and leaving the labor force, so that the unemployment rate is overstating the improvement in the labor market. He pointed out that a broader measure of labor underutilization (including people only marginally attached to the labor force) has come down roughly in line with the unemployment rate since late 2010, suggesting that the fall in the unemployment rate is not due to people leaving the labor force and becoming "marginally attached." We would argue that declining participation has played an important role in lowering the unemployment rate over the whole period since its peak in late 2009, but that it has been far less important over the past year, i.e., the recent improvement reflects a genuine acceleration in employment growth.
• The final explanation considered by Bernanke is that the employment improvement reflects a "catch-up" from the huge job losses during and just after the recession. Okun's Law broke down during and after the recession in the exact opposite way to today, i.e., unemployment rose much more than one would have expected based on economic growth. This may have reflected "fear-driven" layoffs that took employment down more than justified based on the loss of demand—a shortfall that is now being reversed. We would add that during the downturn firms may have made severe cutbacks in functions not immediately important for short-term production – such as in marketing departments, planning departments, R&D departments – producing what looked like huge gains in productivity. Now that the "fear factor" is less important, and firms are planning for long-term growth again, they may be restoring these functions—important for long-term growth, but not for immediate production—resulting in what look like weak gains or even declines in productivity.

Bernanke did not wholeheartedly endorse the "catch-up" explanation—but he gave it more favorable treatment than the other two. The implication? To the extent that the catch-up has run its course (and we do not know how much farther it has to run), further rapid labor-market improvement will hinge on faster growth in demand and GDP. So the Fed will need to watch carefully to see if that acceleration in growth begins to materialize.

Long-Term Unemployment

The chairman also looked at the high rate of long-term unemployment. More than 40% of the unemployed have been out of work for more than six months—a level without precedent in recent history. Even during the 1982 recession, when the unemployment rate rose even higher than during this recession, the proportion of long-term unemployed peaked at around 25%.

The key question is whether this high rate of long-term unemployment means that unemployment is now primarily structural in nature (reflecting a lack of skills or a mismatch of skills supplied and demanded). Bernanke's view is that only a modest portion of the increase in long-term unemployment is due to structural factors, for several reasons:

• The job-finding rates (how many of the unemployed find jobs over the next month) of both the recently unemployed and the long-term unemployed all fell over the recession in roughly the same proportion. So the labor market worsened for all the unemployed to roughly the same extent—suggestive of cyclical rather than structural unemployment.
• Labor demand is weak in most industries and locations. This also suggests an aggregate demand shortfall.
• The rate of vacancies has moved up along with the rate of unemployment (suggestive of a mismatch)—but this is not unusual after a deep recession (employers may be more selective when hiring is not urgent; individuals may take longer to search when extra unemployment insurance is available).

Bernanke concluded that the substantial increase in long-term unemployment is primarily cyclical rather than structural. We suspect that the structural component of unemployment may be more important than Bernanke suggests—for example, "job-finding rates" have started to improve for the short-term unemployed, but have not yet done so for the long-term unemployed—although we would still agree that the high level of unemployment remains mostly a cyclical phenomenon.

Implications

If the labor market improvement reflected an economy "taking off," and if the remaining unemployment were mainly structural, not cyclical, there would be no case for any further Federal Reserve easing. Indeed, the Fed would have to start thinking about when to tighten.

But if the labor market improvement reflects a "catch-up," and cannot be maintained at the same pace without an acceleration in growth, and if unemployment remains mainly cyclical, then the Fed needs to stay highly accommodative, and to watch whether growth remains tepid or starts to accelerate. It also needs to keep further quantitative easing on the table as a possibility—whether unsterilized (balance-sheet expansion) or sterilized (in effect, an extension of "Operation Twist").

We do not think that it would take much—in terms of growth disappointment—for the Federal Reserve to come in with more easing. But if the Fed does act, a "sterilized" intervention seems more likely than a full quantitative easing. This could take the form of purchases of mortgage-backed securities, to try to hold long-term mortgage rates down, with the liquidity injection neutralized by borrowing back the proceeds at short maturities.

by Nigel Gault