Why Didn’t Inflation Collapse in the Great Recession?
Marco Del Negro, Marc Giannoni, Raiden Hasegawa, and Frank Schorfheide
Liberty Street Economics

 

 

GDP contracted 4 percent from 2008:Q2 to 2009:Q2, and the unemployment rate peaked at 10 percent in October 2010. Traditional backward-looking Phillips curve models of inflation, which relate inflation to measures of “slack” in activity and past measures of inflation, would have predicted a substantial drop in inflation. However, core inflation declined by only one percentage point, from 2.2 percent in 2007 to 1.2 percent in 2009, giving rise to the “missing deflation” puzzle. Based on this evidence, some authors have argued that slack must have been smaller than suggested by indicators such as the unemployment rate or deviations of GDP from its long-run trend. On the contrary, in Monday’s post, we showed that a New Keynesian DSGE model can explain the behavior of inflation in the aftermath of the Great Recession, despite large and persistent output gaps. An implication of this model is that information about the future stance of monetary policy is very important in determining current inflation, in contrast to backward-looking Phillips curve models where all that matters is the current and past stance of policy.

In the New Keynesian DSGE model, inflation depends on a measure of slack and on expected inflation. In turn, expected inflation is determined by expected future marginal costs. Putting the two things together, today’s inflation depends on current and expected future marginal costs. According to the model, inflation did not fall much during the recession because expectations of future marginal costs, and therefore inflation expectations, remained anchored. In other words, marginal costs were expected to revert back to their normal level even though current marginal costs were low.

This raises the question of what determines the expected reversion of marginal costs. In our paper “Inflation in the Great Recession and New Keynesian Models,” we show that if the prices of individual goods are sufficiently sticky, then monetary policy can have substantial effects on future marginal costs and therefore on inflation. Specifically, if the central bank is committed to stabilize inflation around its target, then it will lower its policy rate and indicate that it will maintain it at a low level for an extended period of time when output is below potential. This announcement tends to lower longer-term rates, which stimulates consumption and investment demand, and in turn raises expectations of future marginal costs.

To see why the degree of price stickiness matters, the left panel of the chart below shows the actual path of marginal costs that are not directly observed but are inferred by the model, along with the forecasts by the agents in the model at two different points in time. The solid red lines are forecasts using our estimated value of the degree of price stickiness, and the dashed lines are forecasts using a lower estimate of price stickiness from the well-known paper by Smets and Wouters. The chart shows that if prices are relatively flexible, marginal costs revert quickly to their steady state. In contrast, marginal costs revert slowly to their steady state when prices are sticky. Since in New Keynesian models inflation is determined by the present value of future marginal costs, this finding implies that with flexible prices, current marginal costs largely drive inflation (the intuition being that firms quickly lower prices in line with current marginal costs). Conversely, with sticky prices, the entire future path of marginal costs becomes relevant for inflation determination as firms take into account future marginal costs when setting current prices.

Chart 1 shows Marginal Costs and Marginal Cost Forecasts

If prices are sufficiently sticky, monetary policy has a considerable impact on the dynamics of marginal costs. To substantiate this claim, the right panel of the above chart shows marginal cost forecasts under two different assumptions about the policy response to deviations of inflation from its target. The red lines show the marginal cost forecasts under the baseline behavior of policy. The blue dash-and-dotted lines instead assume that the central bank responds less to inflation deviations from target. This weaker policy response to inflation makes little difference to the expected path of marginal costs with flexible prices since marginal costs revert back to their steady state regardless of the conduct of monetary policy. However, with price stickiness, the marginal cost forecasts become very sensitive to the central bank’s reaction to inflation movements. Thus, if prices are sticky, the path of expected future marginal costs can be more effectively controlled by a monetary policy that reacts strongly to inflation. As a result, monetary policy is still able to anchor inflation expectations and therefore current inflation.

It has been argued that it is harder for central banks to affect inflation when inflation is unresponsive to current slack, that is, when the Phillips curve is flat. For instance the 2013 IMF World Economic Outlook asks, “Does Inflation Targeting Still Make Sense with a Flatter Phillips Curve?” Seen from the perspective of our model, a flatter Phillips curve does not imply that monetary policy control is diminished. On the contrary, by affecting future marginal costs, monetary policy can effectively anchor inflation expectations, which is why, in our story, inflation did not fall in the Great Recession.

Disclaimer
The views expressed in this post are those of the authors and do not necessarily reflect the position of the Federal Reserve Bank of New York or the Federal Reserve System. Any errors or omissions are the responsibility of the authors.


Del_negro_marco
Marco Del Negro is an assistant vice president in the Federal Reserve Bank of New York’s Research and Statistics Group.

Giannoni_marc
Marc Giannoni is an assistant vice president in the Research and Statistics Group.

At the time this post was written, Raiden Hasegawa was a senior research analyst in the Research and Statistics Group.

Frank Schorfheide is a professor of economics at University of Pennsylvania.

Category: Federal Reserve, Inflation, Think Tank

Please use the comments to demonstrate your own ignorance, unfamiliarity with empirical data and lack of respect for scientific knowledge. Be sure to create straw men and argue against things I have neither said nor implied. If you could repeat previously discredited memes or steer the conversation into irrelevant, off topic discussions, it would be appreciated. Lastly, kindly forgo all civility in your discourse . . . you are, after all, anonymous.

2 Responses to “Why Didn’t Inflation Collapse in the Great Recession?”

  1. DeltaV says:

    Actually, according to the Federal Reserve, GDP and CPI tracked closely from 2007Q4. The only period of significant divergence was from early to mid-2009, when GDP change (QoQ) averaged about -0.25% while CPI change (MoM) averaged about +0.25%.
    Source: http://research.stlouisfed.org/fred2/graph/?g=HMS

    If looked at YoY (and data should always be normalized thus) then CPI followed GDP with a predictable lag.
    Source: http://research.stlouisfed.org/fred2/graph/?g=HMT

    In fact, the only real divergence has been since 2010, during which GDP has averaged about 3.5% YoY, while CPI has only averaged about 2% YoY.