Yes, the Fed (Passively) Tightened in the Fall of 2008

Fed Chair Janet Yellen went before the Joint Economic Committee of the U.S. Congress today. She gave her report on the economy and then took questions from members. Probably the most interesting question came from Senator Ted Cruz:

Thank you, Mr. Chairman. Chair Yellen, welcome. In the summer of 2008, responding to rising consumer prices, the Federal Reserve told markets that it was shifting to a tighter monetary policy. This, in turn, set off a scramble for cash, which caused the dollar to soar, asset prices to collapse, and CPI to fall below zero, which set the stage for the financial crisis.

In his recent memoir, former Fed Chairman Ben Bernanke says that, the decision not to ease monetary policy at the September 2008 FOMC meeting was, quote, "In retrospect certainly a mistake."

Do you agree with Chairman Bernanke that the Fed should have eased in September of 2008 or earlier?

This question is interesting because it presents a more subtle understanding of the Great Recession than is found in the standard narrative. In fact, it may be too subtle since it seemed to have caught Janet Yellen off guard. It also seemed to have tripped up the usually sharp Wall Street Journal reporter Sudeep Reddy in his live blogging of the hearing. Both seemed incredulous that Cruz would imply the Fed actually tightened monetary policy in the fall of 2008. But that is exactly what the Fed did, though to see it one has to understand the notion of a passive tightening of monetary policy.

A passive tightening of monetary policy occurs whenever the Fed allows total current dollar spending to fall, either through a fall in the money supply or through an unchecked decrease in money velocity. Such declines are the result of firms and households expecting a worsening economic outlook and, as a result, cutting back on spending. In such settings, the The Fed could respond to and offset such expectation-driven declines in spending by adjusting the expected path of monetary policy. But the Fed chooses not do so and this leads to a passive tightening of monetary policy.

A passive tightening of monetary policy is no less harmful to the economy than an active tightening. The Fed, therefore, should be no less culpable for passive tightening than it is for active tightening. One way to see this is to imagine the Fed as a school crossing guard. If the Fed failed to prevent a child from crossing into a busy street would we be any less indignant than if it instructed the same child to cross into the busy street? Both mistakes are equally dangerous and the result of choices made by the crossing guard. It is no different with monetary policy. 

So how does this play into Cruz's comments about 2008? The answer is that the Fed began to passively tighten during the second half of 2008. It had actually done a decent job stabilizing aggregate demand for the two years leading up to this point despite a housing recession occurring during this time. But in mid-2008 it allowed a passive tightening to emerge. Arguably, this passive tightening sowed the seeds for the financial panic that erupted later in 2008 and ultimately is what turned an otherwise ordinary recession into the Great Recession. 

Okay, that is story. What evidence do we have for this understanding of the Great Recession?

First, here are two figures that demonstrate the Fed contained the housing recession for roughly two years. They show that employment and personal income outside of housing related sectors actually grew at a stable rate up until about mid-2008. Kudos to the Fed during this time.

But something clearly changes in mid-2008. My argument--echoed by Senator Cruz today--is that the Fed inadvertently allowed a passive tightening of monetary policy. This can be seen in the next few figures.

The first two figures shows the 5-year 'breakeven' or expected inflation rate. It shows that prior to the September 16 FOMC meeting this spread declined from a high of 2.72 percent in early July to 1.23 percent on September 15. This was an unusual sharp decline and was screaming "Trouble ahead!" 

One way to interpret this decline is that the bond market was signalling it expected weaker aggregate demand growth in the future and, as a result, lower inflation. Even if part of this decline was driven by a heightened liquidity premium on TIPs the implication is still the same: it indicates an increased demand for highly liquid and safe assets which, in turn, implies less aggregate nominal spending. Either way, the spread was blaring red alert, red alert!

The FOMC allowed these declining expectations to form by failing to signal an offsetting change in the expected path of monetary policy in both its August and September FOMC meetings. The next figure shows where these two meetings fell chronologically during this sharp decline in expectations.

As noted above, this passive tightening in monetary policy implies there would be a decline in the money supply and money velocity occurring during this time. The Macroeconomic Advisers' monthly nominal GDP data (i.e. NGDP = money supply x velocity) indicates this is the case:

The Fed could have cut it policy rate in both meetings and signaled it was committed to a cycle of easing. The key was to change the expected path of monetary policy. That means far more than just changing the federal funds rate. It means committing to keeping the federal funds rate target low for a considerable time and signalling this change clearly and loudly. With this approach, the Fed would have provided a check against the market pessimism that developed at this time. Instead, the Fed did the opposite: it signaled it was highly worried about inflation and that the expected policy path could tighten. 

On the financial panic side, note that the Fed kept aggregate demand stable during the early stages of the panic in 2007. Again, job well done. Only in late-2008 after the Fed had allowed passive tightening (as seen by the decline in NGDP) does the financial panic spike. This can be seen in the figure below:

So the worst part of the financial crisis took place after the period of passive Fed tightening. This is very similar to the 1930's Great Depression when the Fed allowed the aggregate demand to collapse first and then the banking system followed. 

So it is completely reasonable for Senator Ted Cruz to ask his question today about the Fed's mistake at the September 2008 FOMC meeting. I would only add that this mistake was already in play by that meeting. The September meeting served to confirm the market's worst fear that the Fed was more concerned about inflation than the collapsing economy.

This understanding of the Great Recession is not unique to Senator Ted Cruz, Scott Sumner, or myself. Robert Hetzel of the Richmond Fed has written an entire book that makes this argument (he also has an journal article). So even a Fed insider acknowledges this possibility.

Update: George Selgin shows how the Fed's sterilization in 2008 contributed to the passive tightening of monetary policy.

P.S. This post draws heavily from earlier ones that make the same point. Also see Scott Sumner's arguments for this view: here and here.