Markets were disappointed by today’s unemployment data because they reveal how difficult it has been for the employment situation to improve in a meaningful way. The employment market is improving, just not as much as one would hope. Below are four charts derived from the household survey data that tell the story, but also point to the secular nature of this problem.
Secular trends in unemployment?
This first chart demonstrates that the secular trend in the unemployment rate has mirrored the secular trend in the stock market over the last 40 years. The data reveal a secular increase in unemployment during the 1970s which I believe is related to economic ‘recalculation’ as the economy made a secular adjustment out of manufacturing and dealt with rising inflation (see my post Looking at structurally high unemployment as recalculation). In the 1980s and 1990s these trends were reversed. It seems we are entering another secular period of increasing unemployment which began at the beginning of the last decade. However, notice the slope of the line is enormous. As I have indicated in the past, I believe we are only in the middle of a depression and secular bear market. This would suggest that unemployment would rise even higher in the next downturn and points to the recent rally as a cyclical bull market aka bear market rally.
This time is different?
Is this recession different? In some ways, yes. The change in the unemployment rate has been unusually high, peaking at 4.0% in April and June 2009. I would expect that June date to be right around the official dating for a technical recovery when the determination is eventually made. However, recovery or not, you have to back to 1949 to find a steeper one-year rise in the unemployment rate. This is a truly depressionary indicator that, combined with the private sector debt load, suggests a future of secularly weak consumer demand. While I am far from convinced the balance sheet recession means deleveraging will be large during a recovery, it does suggest demand retrenchment will be unusually severe when the next downturn hits, one reason, I expect the unemployment rate to shoot up even higher.
Not a garden variety recession but the beginning of a secular trend
This chart only goes back twenty years as there have been increases in the labor force which make longer comparisons deceptive. But, you can clearly see the trend in the number of unemployed workers of the last twenty years has been broken.
If stretched this period out to the 40 years of the previous charts you would see a repeat of the secular trends from the unemployment rate.
Is the fall in labor participation unusual?
Finally, there is the labor participation rate. This usually falls when the jobs market is poor as people become discouraged and drop out of the labor force. You can see this in the data. The worrying thing is that this number is still dropping. However, I would point out that this is not at all unusual in our services-based economy. For example, in the last recovery, which began in November 2001, the participation rate didn’t bottom until January 2005. The 1990-91 recession ended in March 1991, but labor participation fell until December 1991.
I don’t have a feel as to which specific structural issues are behind this jobless recovery syndrome and comparisons to the past are poor because the data only go back to 1975 when women were entering the workforce en masse. But the chart does show a dramatic and unprecedented change in participation rates from this last downturn.
I would say that this fall in labor participation rate, combined with stagnant or falling house prices, is likely to limit household formation, creating another pro-cyclical effect that will be a drag on the economy and on the housing sector going forward.
