Jobless Claims Could Spike by 40% in Coming Weeks—Even if Claims Fall.

Text size

Is the number of Americans filing initial claims for unemployment insurance going up or down? It may seem like a simple question, but the best answer we have right now is “both”—and the reading is only going to get blurrier over the next few months thanks to the vagaries of the Labor Department’s seasonal-adjustment algorithm.

This is not a conspiracy, but another consequence of the unprecedented nature of the coronavirus contraction. Layoffs normally follow predictable seasonal patterns, with big spikes at the turn of each calendar year and smaller spikes around the end of June and beginning of July. The Labor Department adjusts these patterns in the raw data when it presents the headline numbers, which makes it easier for journalists and traders to track the underlying state of the economy.

The problem is that the recent downturn is so massive that the normal seasonal patterns probably shouldn’t be applied. So far, that hasn’t mattered too much: the seasonal-adjustment algorithm has inflated the total number of initial claims for unemployment insurance filed since the beginning of March by less than 8%.

Since the end of May, however, the seasonal adjustment algorithm has implied a much sharper decline in the number of Americans filing new claims for benefits than the raw data.

On a seasonally adjusted basis, 1.3 million Americans filed initial claims in the week ended July 11, down from 1.9 million in the week ended May 30. By contrast, the unadjusted numbers were 1.5 million in July and 1.6 million in May. That’s a much smaller improvement.

Moreover, the unadjusted number of initial claims actually began rising in the first half of July. That could be a seasonal quirk—but it could also be consistent with an economy starting to roll over thanks to the worsening viral outbreak.

Regardless of how observers interpret the recent past, the seasonal algorithm is about to make the incoming data much tougher to interpret. As Ed Harrison recently explained in his Credit Writedowns newsletter, the end of the traditional midyear layoff cycle is going to cause a sharp increase in the multiplier the Labor…