Saturday, March 29, 2008

Lid Blown off Low Cost

One of my very first posts here was What is the Low Cost Alternative which I put up on November 20th, 2004. In it I made the observation that since 1997 Low Cost, purportedly the upper end of a possible range of economic outcomes, in fact always gave the same result: fully funded Social Security through the 75 year actuarial window with a steady reserve. It consistently returned what I call Baby Bear results, the porridge never being too hot or too cold. I discussed the implications of this in my 2006 Post Goldilocks and the Three Social Security Bears.

Well this held up excellently through the 2005, 2006, and 2007 Reports each of which gave the same basic Baby Bear outcome. Which made 2008 kind of a shock. For the first time the Trustees present a Low Cost model that would officially have Social Security over funded after mid century. Low Cost is outcome I in the following figure Figure II.D6.—Long-Range OASDI Trust Fund Ratios Under Alternative Assumptions [Assets as a percentage of annual cost] A constantly rising Trust Fund is a long range menace to Social Security, there is a point when having too much in reserve becomes a bad thing, something I expand on some in Interest on Interest: a Threat.
(I finally figured out how to upload images. II.D6 now displays at the top of this post.)

There is only one way to flatten the tail after 2041 to get Social Security back to long term Baby Bear status. And that is some combination of cutting revenues or increasing future benefits. Which is to say exactly 180 degrees reversed from pretty much universal policy positions. To say the least not at all the kind of crisis people are envisioning.

Monday, March 17, 2008

Reality vs Intermediate Cost vs Low Cost

This table (which looks a lot nicer in edit mode-oh well) shows projected Real GDP figures for both Intermediate Cost and Low Cost from the last eleven Social Security Reports along with actual Real GDP numbers. The 1997 numbers make some sense, in the face of a 1996 2.5% number Intermediate Cost projected a repeat at 2.5% with a more optimistic Low Cost number of 3.2%. But oddly each model predicted a sharp downturn in the second year. In the event the economy returned 3.8%. Now one would think this result would tug at the model and at a minimum establish a new ceiling since obviously 3.8% was a possible outcome. Instead the 1998 Report stuck with 2.5% as a median IC number and set LC at 3.1%. That is they suggested that the very best the economy could be expected to do was a 20% slowdown in the rate of growth of GDP. Well the real economy returned 3.9%. What was the response? The Trustees stayed with their slowdown narrative. Why did the Trustees not concede that growth above 3% was the new median? Well the answer appears in the post above this one. Or will when I write it.

Report Year Prev Year IC Current Year IC Second Year LC Current LC Second Year Actual Result
1997 2.5 % 2.5% 2.0% 3.2% 2.7% 3.8%
1998 3.8% 2.5% 2.0% 3.1% 2.5% 3.9%
1999 3.9% 2.6% 2.0% 3.4% 2.5% 4.0%
2000 4.0% 3.5% 2.7% 3.9% 3.1% 5.1%
2001 5.1% 3.1% 3.1% 3.5% 3.5% 1.0%
2002 1.0% .7% 3.8% 1.6% 4.6% 2.4%
2003 2.4% 2.9% 3.6% 3.8% 4.1% 3.1%
2004 3.1% 4.4% 3.6% 4.9% 3.9% 4.4%
2005 4.4% 3.6% 3.5% 3.9% 3.8% 3.6%
2006 3.6% 3.4% 3.3% 3.8% 3.5% 4.7%
2007 4.7% 2.6% 3.0% 3.4% 3.4% 2.2% (prelim BEA)