Grasping Reality on TypePad, by Brad DeLong

Jon Hilsenrath: >Fed's Computer Models Pose Problems:

When the Federal Reserve said in December that it would keep short-term interest rates near zero until the unemployment rate falls to 6.5%, it was backed by… [the] computer-modeling programs the central bank uses…. In December…. Fed officials decided to announce they would keep interest rates near zero until the unemployment rate drops to 6.5%. The models told them such a commitment would help nudge unemployment lower—it was 7.7% in November—and wouldn't risk a big burst of inflation.

But here is the problem: The models are deeply flawed. They failed to foresee the financial crisis in 2008 and have tended to overestimate the strength of the economy for several years. Could they fail the Fed again?… Despite all their complexity and sophistication, they have long been plagued by gaps in how they read and project the economy. One of the biggest is that they have ignored the nuances of the financial system—one of the primary channels through which Fed policy works….

The financial gaps in the models worry Nathan Sheets, a Citigroup C -0.62% economist who was Mr. Bernanke's top international adviser at the Fed from 2007 to 2011. Mr. Sheets points to 1994 as an example of why. Back then, small increases in short-term interest rates by the Fed led to a surge in long-term interest rates. The models didn't foresee the financial chaos that would ensue, including the collapse of investment bank Kidder Peabody & Co., the bankruptcy of Orange County, Calif., and the Mexican peso financial crisis…

I was just last week going through my notes from 1994. Yes, the Federal Reserve's model (and other models) "failed": they all forecast that the 1994 raise in the Federal Funds rate from 3%/year to 6%/year would carry with it an increase in long-term nominal Treasuries from 6%/year to 6.75%/year; instead, long-term Treasuries peaked at 8%/year because of (a) an absence of forward interest rate guidance by the Fed and (b) unexpectedly-large endogenous moves in the duration of MBS.

But the unexpectedly-adverse reaction of the bond market in 1994 did not cause the Mexican peso crisis: an extra 1% rise in U.S. bond interest rates is not big enough to cause the Mexican peso to lose 75% of its value. Neither Orange County nor Kidder Peabody had any business holding a portfolio that would be bankrupted by a 10% fall in the value of the long-term Treasury bond. And it is not the business of the Federal Reserve to make sure that nobody holding an over leveraged portfolio goes bankrupt. The "financial chaos" of 1994 had no effect on the real economy in the United States: there was no recession in 1995, but rather the beginnings of the great dot-com boom.

The point of Hilsenrath's article appears to be that something bad may

happen when Fed officials raise interest rates…. The models make predictions about unemployment and inflation, but Mr. Sheets worries they might not foresee new financial shoes poised to drop. "They've really internalized these models and that's very risky," Mr. Sheets said…. in a lower interest rate, and Ferbus spits out predictions of how much growth and inflation should follow. Punch in some other event—like a sudden contraction in government spending or a jump in tax rates—and it spits out predictions of how the economy should respond…. The fingerprints of Ferbus and its friends are all over the Fed's latest interest-rate decisions…. [G]iven the unknowns about unconventional Fed policies, Mr. Sheets's warning warrants attention.

Mr. Sheets's warning certainly warrants attention if you are a hedge fund or an investment bank thinking of replicating Kidder Peabody's or Orange County's portfolio--but you should have already known not to do that.

But should Mr. Sheets's warnings have led the Federal Reserve to do something different than it has done? In retrospect the Federal Reserve certainly wishes that it had given more guidance at the start of 1994 as to the likely shape of its interest-rate tightening cycle and that it had at the start of 1994 incorporated the endogenous duration of MBS into its models in the way that it had incorporated them by the end of 1994. However, if the employment and output recovery evolves going forward as the recovery of the 1990s evolved in 1994 and thereafter, the Federal Reserve will be not dismayed but ecstatic.