We really don’t know what drives inflation

Yesterday, I wrote about a research paper that tested seven of the fundamental assumptions underlying economics, often called stylized facts. While these fundamental relationships often are violated for long times (sometimes even decades), they hold in the very long run. Except for two. Even over 147 years and incorporating the distortions from international relationships between 17 large economies around the globe there is no stable relationship between money supply and inflation, nor between inflation and long-term interest rates.

The chart below shows the ratio of money supply growth to inflation. In essence, the higher this ratio, the more inflation is created by an additional 1% in the money supply in the UK. In the late 19th century 1% growth in money supply led on average to 4% more inflation. Today, a 1% increase in money supply leads to 0.08% more inflation. But this assumption that increases in the money supply should lead to higher inflation is the core assumption of monetarist theories of inflation that have been followed by central banks in the developed world throughout the 1980s and 1990s. Did I mention that monetarism is dead? Pundits who claim that central banks printing money creates inflation should not be taken seriously. The chart below shows that this theory has been dead for decades.

The unstable relationship between inflation and money supply or interest rates

Source: Jorda et al. (2017), Macrohistory database

The second line in the chart above shows the long-term real rate of interest calculated as the long-term interest rate minus inflation. This too is typically assumed to be constant in the long run and while the chart looks like it is reasonably stable, the axis on the right reveals that the swings are enormous, going from -2% real rate to +5% real rate most of the time. Closer statistical analysis shows that the relationship is indeed not stationary over time.

And that is a problem because the central goal of central bank monetary policy since the 1990s was to target the elusive r_star, i.e. the real risk-free rate of interest that keeps inflation at the target inflation rate and unemployment low. Go to any symposium on monetary policy (I recommend drinking many cups of strong coffee beforehand to help you stay awake) and you will hear economists talk about how to estimate r_star and how to adjust interest rates to target the desired level of r_star.

You see the problem with r_star is that it cannot be measured. It can only be estimated based on your assumptions of inflation expectations, and the relationship between inflation, unemployment, and economic growth. The chart above doesn’t show r_star but something that is very close to it. And the new research indicates that we not only can’t measure r_star but that it changes over time and never approaches a stable equilibrium level – not even in the very, very long run.

In other words, central bankers are trying to influence an economic variable that cannot be measured, that changes all the time, is never trending to equilibrium and where we do not know how changes in monetary policy affect the economic variable, to begin with.

Now add to that that other research shows that inflation expectations – the other crucial economic variable for monetary policy these days – do not influence future inflation either, and you come to the following summary:

Central bankers try to manage inflation by targeting r_star which they determine using inflation expectations and influence by adjusting the money supply through changes in policy rates. Unfortunately, there is no relationship between money supply and inflation or inflation expectations and inflation and the level of r_star changes all the time.

We really don’t know what drives inflation and central bankers don’t know how to manage it.