From my friend David Kotok:
September 4, 2012
Bob Eisenbeis is Cumberland’s Chief Monetary Economist. Prior to joining Cumberland Advisors he was the Executive Vice President and Director of Research at the Federal Reserve Bank of Atlanta. Bob is presently a member of the U.S. Shadow Financial Regulatory Committee and the Financial Economist Roundtable. His bio is found at www.cumber.com. He may be reached at Bob.Eisenbeis@cumber.com.
Saturday’s WSJ banner headline trumpeted “Fed Sets Stage for Stimulus,” and this wishful thinking has been widely expressed by other analysts, most of whom are Wall Street economists who were not invited to the Jackson Hole summit. A careful reading of the WSJ article, however, shows it to be inconsistent with the headline. Only the first three paragraphs of the article are devoted to a description of Bernanke’s defense of the use of unconventional monetary policies, and there is no mention that he hinted of a policy move. Most of the article sets out the arguments and beliefs, both within the Fed and outside, that those unconventional policies have sowed the seeds of future inflation, failed to deliver the desired improvement in economic growth, and hurt the value of the dollar. So what did Bernanke actually do?
Bernanke did exactly what any central banker in his position should have done, especially one who is running out of options. He provided a balanced discussion and defense of the policies put in place from late 2007 to the present. The speech provides no hint of future policy moves and simply reiterates the points made in Fed policy-maker speeches and FOMC statements, that is, the Fed is watching incoming data and will act if necessary.
Others have already commented and dissected the speech ad nauseum. Simply put, Bernanke did several things. He described the policies put in place. He detailed the portfolio-balance theory behind the FOMC’s attempt to lower interest rates. There was no discussion of who holds those securities or what their alternative investments might be. Pension funds, for example, are major holders of Treasuries but can’t make loans to small businesses or consumers. Bernanke also cited evidence of how rates did decline.
He defended his belief in the efficacy of the unconventional policies. But most importantly, he not only discussed the risks inherent in those policies, which were discounted, but also pointed out the difficulties in demonstrating the links between those policies and their effects on growth and employment. The only evidence cited here were Fed counterfactual simulations, showing that real growth is 3 percentage points above what it otherwise would have been and that 2 million more people are employed.
So what about that research? The estimates are based on a Fed paper by Chung et. al. (Hess Chung, Jean-Philippe Laforte, David Reifschneider and John C. Williams, “Have We Underestimated the Likelihood and Severity of Zero Lower Bound Events?” Federal Reserve Bank of San Francisco Working Paper 2011-01, January, 2011, http://www.frbsf.org/publications/economics/papers/2011/wp11-01bk.pdf). That very long paper is a complex, academic, econometric exercise that precisely lays out all the simplifying assumptions behind the simulation models employed and details some of the confidence intervals around the forecasts. The caveats read like the lists of possible side effects in the disclosures for over-the-counter and prescription drugs. Whether those assumptions are appropriate or realistic is left to the reader to decide.
For our purposes here, the main model results cited by Bernanke on the effectiveness of policy rely upon the authors’ use of the Board’s FRB US large macro model. The model’s reported confidence intervals don’t give us much comfort in the counterfactual employment and growth results cited by Bernanke. First, as the authors carefully point out, the data and structure of the model do not include events like those we experienced in recent years, and the authors’ out-of-sample forecast comparisons of the predictions of the model with actual performance during the crisis are dismal. Actual results lie outside the 95% confidence intervals for growth and unemployment predicted by the model. And consider those confidence intervals and how wide they are. For example, the 95% long-sample confidence interval for the output gap (the difference between actual economic growth and potential growth, a key variable in the structure of the model) is -4.8 to 4.9, and for unemployment it is 2.3 to 7.8. This means that one can place little comfort in the counterfactual point estimates in the paper, or those cited by Bernanke for growth and employment. The associated confidence intervals for those counterfactuals aren’t reported; but they have to be huge, given the uncertainty in the model’s parameters and its performance.
Putting all these criticisms aside, but recognizing the dilemma any central banker in Bernanke’s position would face, what else could he have done? What follows are three talking points that Bernanke could have used that are equally consistent with the empirical evidence, research, and discussions within the Federal Reserve that have been revealed in the FOMC minutes and in policy makers’ speeches. Of course, Bernanke would never have used them, but the reader may judge how such statements might be received by markets, politicians, or the general public.
Point 1. Bernanke could have said, “We have been exploring other tools available to us and, as detailed in the last set of FOMC minutes, no new options are under consideration at this time other than those we have detailed in previous speeches and that have appeared in the minutes. About the only policy move we haven’t made is to change the interest rate paid on reserves. We could reduce that to zero or even begin charging banks for holding reserves at the Fed. But this would mean that we would have to be prepared to act preemptively to raise that rate once bank lending took off, to avoid future inflation. Unless some of us have some additional ideas, what you see is what we have. Our choices are either to put in place another round of quantitative easing to satisfy the markets or to stand pat. The former risks revealing that our policies haven’t been particularly effective and that future actions are likely to be even less so, especially if some of the FOMC members believe that quantitative easing is resulting in diminishing returns. Standing pat for now, on the other hand, preserves the perception that we can and will do more. Let us just hope we don’t have to do more.”
Point 2. “The Fed’s portfolio has increased by some 1.969 trillion dollars since the end of August 2007. We estimate that 2 million jobs were created as a result of these unconventional policies and the accompanying drop in interest rates. Thus, this puts the cost of each job created at 980 thousand dollars, in terms of our willingness to expand the Fed’s portfolio by that amount. At that rate, to completely regain the 8 million jobs that were lost during the crisis we would have to expand our portfolio by an additional 7.84 trillion dollars. These numbers say that we are way behind the curve, and were we to publicize this, there would be hearings on why we hadn’t done more long ago.”
Point 3. “We are confident that we can execute an effective exit strategy from the current policies. In June of 2011 we detailed our then most current thinking as to how to proceed, which involved a gradualist combination of a change in communications policy, an increase in the target Fed Funds rate from the current 0-0.25 basis points, a cessation of the rolling over of maturing securities, and then possible sales of securities from the portfolio. This should work, provided existing bond holders don’t decide to liquidate their low-yielding bond portfolios to avoid the inevitable capital losses that will occur. If we are wrong about our assumptions concerning investor behavior, we are in for a big interest-rate shock and will find ourselves chasing the market when it comes to setting the Federal Funds rate, and will face dysfunctional markets that will make the 2007-08 period look mild.”
Given the alternatives available to Bernanke and the FOMC, at Cumberland we view the most likely scenario for the near term to be one of watchful waiting rather than one of more aggressive actions called for by many. If the Fed is actually or nearly out of options, then responding to short-term market demands for more quantitative easing seems like a very risky strategy, with the only payoff being requests for more and more.
Bob Eisenbeis, Chief Monetary Economist
To sign-up for Market Commentaries from Cumberland Advisors: http://www.cumber.com/signup.aspx
For Cumberland Advisors Investment Portfolio Styles: http://www.cumber.com/styles.aspx?file=styles_index.asp
For personal correspondence: email@example.com