A LOST DECADE FOR JOBS

Lakshman Achuthan and Anirvan Banerji are co-founders of the Economic Cycle Research Institute in New York City.

Last week’s news of a drop in the unemployment rate to ten percent is a welcome development. It was presaged by earlier strength in reliable leading employment indicators, which suggest that this improving pattern will persist next year. In November employers cut the fewest jobs since the recession began, but how should Americans interpret this news? With unemployment in double digits for the first time since 1983, many still worry about the jobless recovery.

This coming post-recession dip in joblessness is the good news. But, looking ahead to the later phase of the expansion, the post-World War II period shows disturbing cyclical patterns.

The jobless rate usually sees a sizeable drop during the economic recovery – and bigger recessionary spikes in unemployment are typically followed by larger declines during the first year of improving unemployment. So it would be no surprise if, a year after the unemployment rate begins to drop, it falls to the nine percent range.

The real problem is that the rate of decline in joblessness slows during the rest of the economic expansion. The annual postwar pace of decline in unemployment during these periods has been reasonably uniform, the median being 0.5% a year.

If that pattern persists, the U.S. economy needs to keep expanding without interruption until 2020 for unemployment to fall to its pre-recession low. Even to get back to 5%, often considered to be “full employment,” it would take a business cycle upswing lasting about as long as the record-setting 1991-2001 expansion. Should the next recession arrive earlier, as we suspect, it will take much longer. The implications constitute nothing short of a wake-up call for policy makers who promise to get job growth back on track.

Since World War II, there has been a clear easing pattern in the trend rate of economic growth during expansions, culminating in the 2001-07 expansion, which exhibited the slowest trend rate of growth on record – especially in terms of jobs. Ominously, during expansions following the initial year of revival, growth in non-manufacturing employment – now 91% of nonfarm jobs – has been falling in a parabolic fashion since the 1970s. A continuation of this pattern would mean well under a million jobs gained annually during the expansion following the initial jobs revival. If so, official projections that the jobs lost in this recession will be regained in four years are wildly optimistic.

The “great moderation” of business cycles once extolled by many economists, including Chairman Bernanke, is clearly history. Meanwhile, the trend rate of growth is shriveling. In other words, business cycles are back with a vengeance.

In fact, the combination of weak trend growth and the death of the “great moderation” is a recipe for more frequent periods of negative growth, i.e., recessions. Think of Japan over the last two decades, where 1% GDP trend growth made it easy for growth to fall below zero during periods of cyclical volatility, while China – with a 10% GDP growth trend – has not seen a recession in two decades.

In contrast, the conception, popular in some quarters, of an economy now destined to grow smoothly, albeit at a “new normal” anemic pace, is the sort of fantasy typically generated by extrapolative econometric models. In the real world, one never sees such slow and steady growth – especially in the wake of a severe recession featuring massive stimulus that will be difficult to withdraw smoothly.

The real risk is of more frequent recessions repeatedly aborting cyclical downswings in unemployment in coming years. The stark implication is that it will take much longer than a few years – perhaps decades – to get back to the jobless rates we saw only a couple of years ago.
What we have described provides only a historical baseline, rooted in the arithmetic of past cycles, unencumbered by the fallout from the Great Recession. Some consolation comes from the fact that past performance does not dictate destiny, and extrapolation from past patterns is not a reliable forecasting method, especially if the pattern is about to change.

So, it is at least conceivable that either enlightened policy measures, or good luck, or both, will result in a decisive break from these patterns. The silver lining is that even an economy dipping in and out of recessions and keeping joblessness cycling near historical highs is a navigable one for decision makers who keep a closer watch for recessions and recoveries.

~~~

December 7, 2009

Lakshman Achuthan and Anirvan Banerji are co-founders of the Economic Cycle Research Institute in New York City.