Scott Sumner and Arnold Kling have been having a very interesting blogosphere debate about the causes behind the recent recession. Scott believes that the Fed made a mistake by letting GDP fall 8% below trend and that when the monetary authorities allow such a large fall in nominal GDP a severe recession is unavoidable. Arnold thinks that the recession is a symptom of a great recalculation in which the actors in the economy realized their previous investments were actually unprofitable malinvestments, but aren’t sure where to allocate resources towards next. This post is mainly focused on one of Scott Sumner’s arguments, while next week I will take a deeper look into the recalculation story.
One of Scott Sumner’s ways of demonstrating the excessive tightness of the Fed has been to focus on the TIPs yield, which showed deflationary expectations in the end of 2009. The 5 year inflation rate that TIPs priced in was at times as low as negative 0.8% a year, and the 5 year 5 year forward rate was as low as 0.42%
The blue line is the time series of 5 year forward 5 year inflation expectations, from The TIPS Yield Curve and Inflation Compensation by Refet S. Gürkaynak, Brian Sack, and Jonathan H. Wright. The green line is the 5 year forward 5 year inflation rate calculated from the real and nominal 5 year yields. The bottom orange line is the five year breakeven inflation calculated as the spread between the nominal and real rates.
One of the reasons that breakevens got to be so extreme is that with the decline in leverage it became a lot more expensive to hold onto TIPs with borrowed money. Furthermore, the flight to liquidity meant that for many funds TIPs were among their most liquid assets and as such had to be sold while for others TIPs were no longer liquid enough to safely hold in such a volatile environment. In this environment, market inflation expectations were probably not as volatile as they appeared and the methodology of Gürkaynak, Sack and Wright created in 2007 a forward inflation measure that never dropped below 2% during the crisis. This may be partially because when they modeled the TIPs curve the left out the particularly illiquid front 18 months.
That the Fed wasn’t forecasting deflation is also shown in the Taylor rule chart that accompanied Bernanke’s speech earlier this month. The Fed saw the signs in the market to cut rates before their forward looking Taylor rule told them they should. Of course, in the Taylor rule chart below the only forward looking component is inflation, not the output gap. The Fed cut rates in advance of a declining output gap but after the financial panic. Furthermore, once rates approached zero they started paying interest on reserves in order to keep their attempts from saving the financial sector from spilling over as inflation into the rest of the economy. In effect this tightened policy just as the forward inflation looking Taylor rule was hitting the zero bound and possibly going negative.
Of course, the chart from Bernanke’s presentation doesn’t go negative (Perhaps to avoid questions from angry senators who were wondering why he wasn’t doing everything possible to promote full employment). However, when this chart is compared with other Taylor rule charts it is obvious that the Taylor rule is suggesting something below zero even now, which implies that the Fed may be too early in withdrawing quantitative easing. For example of these rules, take the work done by Glenn D. Rudebusch at the SF Fed in May of 2009 in which he forecasts a negative rate far into the future, which was updated by Krugman here.
Source: FRBSF Economic Letter, 2009-17; May 22, 2009
As a side note, Menzie Chinn at Econbrowser has an interesting post about how the fed funds futures are pricing in a policy rate much higher than anything the Taylor rule is currently predicting under a wide range of assumptions.