In the past, I’ve argued that tight money caused the Great Recession. But what caused the tight money? One answer is too much weight placed on inflation targeting, combined with a backward-looking approach to inflation. Consumer price inflation was quite high during the first half of 2008, due to rising oil prices and a weak dollar.

An excellent paper presented by Robert Hetzel at the recent Mercatus/Cato Monetary Rules Conference suggests another possible reason why the Fed erred in 2008:

The FOMC had started lowering the funds rate from its cyclical peak of 5.25 percent at its September 2007 meeting out of concern that a disruption to the flow of credit to mortgage markets would weaken growth. Following its lean-against-the-wind procedures, it lowered the funds rate to 2 percent at its April 2008 meeting. However, at that meeting, the FOMC signaled an end to the easing cycle.

One reason for this signaling was that economy appeared to stabilize in 2008Q2. Temporary factors made it appear that the economy was reviving. A decline in net exports boosted GDP. The enormous boost to disposable income provided by the Bush tax cuts temporarily halted the decline in real personal consumption expenditures that had begun in December 2007 (Hetzel 2012, 213).13 Most important, the FOMC became concerned about the persistent overshoot in its inflation target and also about depreciation of the dollar. Its communication delivered the message that the easing cycle was over so that the next move in the funds rate was likely to be upward.1

In footnote 13, Hetzel explains the tax cut in more detail:

The following figures are for annualized growth rates of monthly real PCE:

12/2007 – 2/2008: -1.9%
3/2008 – 5/2008: 1.7%
6/2008 – 9/2008: -3.8%
10/2008 – 12/2008: -4.5%

There is a pause to these negative growth rates in the months of March, April, and May. That pause came from the boost to income from the Bush tax cut, which President Bush signed into law on February 12, 2008. The actual rebates arrived in the month of May, but households anticipated their arrival.

The following graph shows how the tax cut impacted disposable income:

Screen Shot 2016-12-19 at 12.15.25 PM.png

It’s worth noting that Bush’s 2008 tax cut was an almost ideal example of fiscal stimulus. It did not involve wasteful bridges to nowhere; it was a pure lump sum transfer to households. It was far better timed than usual, occurring before mainstream economists or the Fed even forecast a recession. (Notably, it was far better timed that the 2009 fiscal stimulus.) Kudos to Bush!! And yet it failed to accomplish its goal–preventing a recession. Indeed the recession became very severe just a few months later. Why?

One possible culprit is the financial crisis that occurred in the fall of 2008. But on closer examination, the economy worsened significantly during the summer of 2008. Indeed I’ve argued that the economic decline that occurred in the second half of 2008 had the effect of dramatically lowering asset prices, and pushing highly leveraged investment banks like Lehman into bankruptcy. But why was the economy so weak in the summer of 2008? Perhaps the answer is tight money.

Europe and the US had pretty similar monetary policies during 2008, and pretty similar recessions. So let’s compare the two cases:

Screen Shot 2016-12-19 at 12.11.29 PM.png

The blue line is the eurozone. Notice that NGDP growth remained robust up through the first quarter of 2008. Then the eurozone went into recession. (This is nominal GDP, not real, so even a flat line is recessionary.) The US (red line) went into recession in the first quarter of 2008, but then had a reprieve in the second quarter—perhaps due to fiscal stimulus. After that, both regions had a deep recession in late 2008 and early 2009.

Hetzel considers the failure to cut interest rates between April and October of 2008 to be an effective tightening of monetary policy, and I agree. As noted earlier, one reason for the tightening might have been high inflation. Another might have been that the fiscal stimulus gave the Fed a sort of “head fake”. At a time of high inflation, the strong consumption numbers during the spring of 2008 convinced the Fed that it had done enough to prevent recession. If you read Bernanke’s memoir you sense a sort of “eye of the hurricane” moment in the spring, when it seemed to the Fed that they had weathered the subprime mortgage crisis, and that the economy was holding up OK due to the Fed’s bold action when Bear Stearns failed, as well as the aggressive rate cuts of early 2008.

But of course this was a false dawn. The Fed’s refusal to ease monetary policy during the late spring and summer of 2008 caused NGDP to start falling during the summer months. This is because the housing/banking distress had caused the Wicksellian equilibrium rate to fall sharply during 2008. Here are some estimates of the equilibrium rate from Vasco Curdia:

Screen Shot 2016-11-07 at 1.28.33 PM.png

BTW, the slight increase in Q3 NGDP is misleading, as the level in September was considerably lower than in July, based on monthly estimates of Macroeconomics Advisers. This graph shows that monthly nominal consumption started falling sharply in September, but even the June to August period is weaker than it looks, as that slight decline occurred against a backdrop of rapid inflation (Hetzel used real consumption data):

Screen Shot 2016-12-19 at 12.13.20 PM.png

Nominal investment spending did even worse, falling at a 9.5% annual rate between Q2 and Q3.

I still think the high inflation rates were the main reason why the Fed erred in mid-2008, but the fiscal stimulus almost certainly was a contributing factor. The fiscal boost lasted only a few months, but it left us with a monetary policy that was far too tight for the needs of the economy. It’s quite possible that the Fed’s overreaction to the spring 2008 fiscal stimulus actually made the recession worse than otherwise.

Monetary offset can be less than 100% or more than 100%. I still think the assumption of 100% monetary offset of fiscal policy is the baseline assumption we should work with, until proven otherwise.