As a regular contributor to the Philadelphia Fed’s Survey of Professional Forecasters (SPF) I am always curious about the forecasts my fellow contributors are making. The latest set of forecasts came out on Friday and this time I was particularly focused on the unemployment rate forecasts. There has been much talk about the importance of the unemployment rate at the time of the presidential election, and the SPF has been shown to produce pretty good forecasts of the US unemployment rate. The unemployment rate for last month was 8.3%, a surprisingly good number. The unemployment rate has fallen faster over the last few months than forecasters predicted in the SPF just last quarter, although some of the decline was due to people leaving the labor force rather than job creation. Following the surprising performance of the unemployment rate generally, the SPF forecasters indeed changed their forecasts substantially this round. Here’s a graph:
The key forecast I was looking at was for 2012Q4 – right around election time. As you can see, the forecasters now see a much better labor market outlook for the end of 2012 than they did just three months ago. The median forecast has dropped from 8.7% to 8.1%. It has not, however, dropped below the threshold of 8% (my personal forecast is currently right at 8% for November 2012). I think there is something to round numbers in human psychology, but I also recognize the importance of trends. A declining unemployment rate, particularly one that is declining faster than previously predicted, looks good for the incumbent. One question that remains for me, however, is has this decline come too soon? If the improvement slows before the election, will voters forget that not so long ago we were wondering if the unemployment rate might go back up above 10% or at least not drop below 9% before the 2012 election?