September Employment

Should be an interesting month.  On the inflation side of things, I'm getting nervous (that housing will have an inflationary supply shock while the rest of the economy has a deflationary demand shock).  But, employment continues to look strong.  Here is a comparison of the forecast, based on continued unemployment claims and a steady decline in very long term unemployment, and actual unemployment.  Continued claims continue to surprise to the downside, so that the forecasted unemployment rate has accelerated downward and is now below the lower linear range.  For the unemployment rate to fail to decline this month, there would need to have been a breakdown in the relationship between insured unemployment and total unemployment.  I don't see any evidence to expect that kind of breakdown in recent labor market behavior or in the historical data.


This graph shows the bullish indications from insured unemployment even more starkly.  This is the graph I have been using, comparing insured and total unemployment rates.  Now, I have added the unemployment forecast indicator.  We can see how reasonably this indicator has predicted unemployment over the previous 18 months.  With the continued decline in insured unemployment so far in September, this indicator forecasts September unemployment at 5.6%.  That would be quite a decline.

This forecast includes a linear decline in very long term unemployed.  The measured number of VLTUE had declined linearly along with the forecast until the last couple of months.  There is always some noise in this data.  But, even in an unrealistic worst case, where VLTUE completely stalls at its current level, all but about 0.2% of the forecasted decline has already happened, so, at most, the forecast would suggest an unemployment rate of about 5.8%.

I think it is very likely that we will get our gap down this month.  The question is how far below 6% will we get.