In a depressed economy, disinflation ensues, to a point.

The important lessons of PLOGs for Macroeconomics

While a PLOG doesn’t sound very helpful, they have actually proved quite important for our understanding of macroeconomics. A PLOG refers to a period of a ‘persistent large output gap’, which means an extended period of time where an economy is performing well below its potential. These periods are significant because they provide a perfect opportunity to examine the dynamics of inflation under these circumstances.

A PLOG episode will be one where unemployment is higher than it normally is. With unemployment relatively high, there will be less pressure on wages as workers see a pool of people waiting for work and will be less likely to push for large pay increases. Similarly, with an economy below potential and aggregate demand lower than usual, those who set prices in the economy will also be less likely to push them up than when the economy is at full potential and demand is strong. In economic theory, this relationship between unemployment and inflation is captured in the Phillips Curve, as shown below.


This is an example of a New Keynesian Phillips Curve. The coefficients on the inflation terms refer to the influence of the previous period’s inflation and the expectations of next periods inflation. The last term on the right hand side refers to the output gap and the estimate of theta measures the impact of this gap on current inflation.

A lot of theoretical work has been done on the Philips Curve and the nature of this macro relationship. While this is necessary of course, in 2010 the IMF provided insightful analysis that sidestepped the theoretical and empirical issues. Specifically, they used an event analysis style of research where they examined 25 PLOG episodes across 14 countries and 40 years.  

How they did it

The IMF used OECD data of high income countries for which quarterly output gap estimates are available. The OECD measures potential output with a production approach based on total factor productivity, capital services and potential employment. They define a PLOG as an episode where an output gap exists for at least 8 quarters and it exceeds 1.5% in absolute terms.  

What they found

Looking at the equation cited above, it posits a linear relationship between the output gap and inflation. What the IMF study showed, however, was that this relationship shows evidence of non-linearity. That is, at low levels of inflation the relationship between the output gap and inflation slows down.

 IMF Plog chart

This non-linearity can be seen in the chart above, taken from the IMF study. It shows the final 9 quarters of the PLOG episode, clearly displaying the disinflation of the cohort but also the flattening out at lower levels of inflation. The next chart shows this evidence across countries. The 45 degree line represents inflation rates in countries that would be the same at the start and the end of the PLOG episode. The vast majority of countries are below that 45 degree line, clearly displaying the disinflationary pressure during a PLOG episode.

IMF chart cross countries 

Why does disinflation not easily turn into deflation?

The IMF paper discussed different reasons why disinflation might not turn into deflation. Two of these reasons, in particular, seem very important and the most likely candidates to explain this phenomenon.  

Reason 1: Resistance to nominal wage cuts

Paul Krugman, on more than one occasion, has highlighted the importance of downward nominal rigidity as playing a role here. This is the notion that employers and employees are resistant to outright wage cuts. If this is the case, Krugman suggests that these ‘sticky’ wages should leave a signature in the data, which is exactly what is shown in the figure below. The large spike at zero represents wage changes that are not turning negative.

Wage distribution 

Reason 2: Central Banks & Inflation expectations

At the Free Exchange blog of The Economist, Ryan Avent wrote a post titled ‘When the Fed fights, it wins’. Avent highlights the importance of inflation expectations and in the case of the US, the ability of the monetary policy of the Fed to maintain these positive expectations.  

Avent posts the following chart of inflation expectations with the timing of Federal Reserve policy actions. Highlighting a critical difference between the Great Depression and the Great Recession, Avent notes that the Fed has acted forcefully at appropriate times to encourage and support positive inflation expectations. He believes that the Fed’s actions have been necessary and sufficient to arrest the disinflation process.  

Infaltion expecations

It is hard to say what is more relevant for the ‘stickiness’ of inflation at lower levels, the resistance of nominal wage cuts or monetary policy and its impact on inflation expectations. But the important point is, however, that both are playing a role in preventing episodes of PLOGs turning into outright deflation. For a country like Japan, who has experienced deflation, the IMF notes that it might take a series of negative shocks to move a country into outright deflation. This is probably true, but perhaps there is also something to be said for well-entrenched inflation expectations. Japan has only recently formalised an inflation target, meaning that when these shocks hit the economy, there wasn’t a history of a central bank encouraging a level of inflation to hold inflation expectations in positive territory. Regardless of the Japan example, the evidence of disinflation during PLOG episodes seems overwhelming and an important lesson for macroeconomists.


1 Comment

Filed under Uncategorized

One response to “In a depressed economy, disinflation ensues, to a point.

  1. Pingback: Inflation in the Great Recession: The importance of a flatter Phillips Curve | globalmacromatters

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s