Wednesday, October 6, 2010

Optimistic CBO Economic Assumptions

The other day, I was expressing some skepticism about this next graph, which the Congressional Budget Office (CBO) puts out to show various scenarios for the future federal government revenue/spending balance:

My concern was particularly about the way the revenue and spending curves improve so sharply as we go from actuals to projections.  I commented that the long term budget projection report doesn't specifically document its macroeconomic assumptions.

So I figured that they (the CBO that is) must just be relying on their general economic forecasts.  The most recent of those are here.  They show figures for calendar year 2010, 2011, 2012-2014, and 2015-2020.  So let's have a quick look.

First up, here's the assumptions about about real GDP growth (along with the last 10 years of actual data from the BEA).

As you can see, CBO was assuming a fairly healthy growth rate in 2010, then a little more subdued in 2012 2011 (perhaps as the stimulus effects on growth ended), but then a very strong recovery in 2012-2014.  This strikes me as very unrealistic, and in particular a failure to appreciate the magnitude of the balance sheet explosion we saw in the last decade or two, and how long it's going to take to deleverage back to reasonable levels.  We still have something like a quarter of US households with negative equity.  And the housing market is being propped up by massive government intervention in the form of mortgage guarantees for low down-payment loans (which are at some risk of being withdrawn or reduced if Republicans control congress after the election), tax-credits, etc.  Not only that, but most indications in the last few months show the economy slowing again.  We are certainly not going to be back to normality in 2012-2014, and given that, I find it implausible that growth is going to be above the historical average in that period.

Similarly, I think the CBO assumptions about unemployment are unrealistic.  Here they are, along with the last decade of data from the BLS.  You can see that they are already a little high on the 2010 figure, and the rate of progress actually occurring is lower than what the CBO is hoping for in coming years.  They also show a more rapid fall than we saw following the 2001 recession (which didn't have the nasty balance sheet/financial crisis aspects of the more recent one).

All of this has implications for the deficit.  Higher unemployment means more unemployment benefits to pay out, and less payroll to tax.  Further, much of the fall in the unemployment rate will not be people being reemployed, but rather people leaving the labor force (eg by going on social security earlier than they otherwise would have done, thus requiring more government payments, or going back to school and thus not paying much if any tax).

So I think that indeed, the deficit is going to be higher in the near term than the CBO is allowing for.

Note: An early version of this post (only up for a few minutes) had an initial graph with a slightly non-standard version of the unemployment rate.  It has been corrected.

No comments: