Are jobless claims pointing to structurally high unemployment?
The latest jobless claims data is in, with initial claims registering 530,000 and continuing claims coming in at 6.1 million. These figures are well off the business cycle highs of a few months ago but still quite elevated.
While the as-reported seasonally adjusted initial claims numbers are in the 500s, the actual initial claims have been in the 400s for 6 weeks now. And the 4-week average of just over 442,000 is now within striking distance of the 4-week average at this time last year. I anticipate the gap will close by the end of the year.
On the other hand, continuing claims remain above 6 million and are a full 2.3 or 2.6 million above last year’s levels, depending on whether you use seasonal adjustments or not. Clearly, the initial claims figures are declining faster than the continuing claims figures. And, remember, an awful lot of people have exhausted benefits before finding a job.
So, stepping back for a moment, this picture suggests that unemployment will remain stubbornly high. I think this is a view most economists hold. But, there’s more to the data than this. The data suggest a recession that is ending, but with remaining structurally high unemployment.
Look at it this way: When you think about GDP as a measure of the economy, what you normally hear is the economy grew 1% or 3% or the economy contracted 2% last quarter. This is what is known as a first derivative statistic. That means all we are measuring is the change from one period to the next. That’s it.
Equally, when we think about economic recession and recovery, we are also measuring change from one period to the next. If unemployment is 9% in one period and 9.3% in the next, the difference is minimal enough that a recovery could take hold because of cyclical factors like the inventory cycle and automatic stabilizers. So recovery can come even in the face of high unemployment.
Therefore, when I look at jobless claims, I think not only in terms of absolute numbers but in period-to-period changes because this coincides with the first derivative measures which reported GDP and business cycle measures are. For example, we are fast approaching a point where jobless claims are no higher than they were at this time last year. As a result, one should expect a recovery to take hold.
What are the flies in the ointment?
- Initial claims are still at a level that suggests job losses and rising unemployment. While the trend in jobless claims suggests we will start adding jobs soon (maybe even before the end of the year), it is altogether possible that the move in claims down stops or reverses.
- The salve of stimulus and the boost of cyclical forces like automatic stabilizers are really the only thing keeping us from recession. if these are removed before unemployment stabilizes, the only way to produce economic growth in the short-term would be through the accumulation of debt.
All that said, when jobless claims do slow to a point that allows the economy to add jobs, it will do so at a very high level of unemployment (say 10%). Given the record low capacity utilization rates and the record low hours worked per employee, it is likely that unemployment will remain structurally high for a long period afterwards since employers are not going to add a ton of jobs in that situation.
The upshot of this structurally high unemployment is it puts a debt-laden economy at perpetual stall speed. Any exogenous shock (oil prices, inflation, withdrawal of stimulus, increase in interest rates) would throw us back into recession. And given the already high rates of unemployment and need for deleveraging in the private sector, that recession would likely be very severe.
Therefore, the only way out of this trap is exports as it could grow the economy and reduce slack capacity enough to move us away from stall speed. But, of course everyone is suffering with the same problem of lower growth and excess capacity.
Unemployment Insurance Weekly Claims Report – U.S. Department of Labor