Monday, June 8, 2015

Interest rates: natural or artificial?

The debate about who is responsible for the low level of interest rates that has prevailed in most economies over the last years heated up when Ben Bernanke wrote a series of blog posts on what determines interest rates. He argued, once again, that it is the global dynamics of saving and investment the one that created a downward trend in interest rates starting in the mid 90s and that it accelerated as a result of the crisis. In his story, central banks are simply reacting to economic conditions rather than driving the interest rate (always refreshing to see a former central banker explaining how powerless central banks are). What Bernanke described can be interpreted as a decrease in what economists called the natural real interest rate.

There are, however, those who have a very different interpretation of the persistent low levels of interest rates. They see central banks as the main drivers of this trend and they think about current levels of interest rates as being artificially low and forced on us by central banks. The popular press is full of references to artificially low interest rates causing bubbles, imbalances, hurting savers and being the seed of the future crisis (about 1 million results if you do a Google search).

From the academic world, John Taylor has been very vocal about the negative effects of artificially low interest rates. He stresses the fact that interest rates have been below what a Taylor rule indicates, a sign that there is a mispricing created by central banks. In a recent blog post he refers to the the results of a paper by Fitwi, Hein and Mercer that tests whether Bernanke or Taylor are right when it comes to explaining interest rates. The paper shows that both theories are possibly right. That low interest rates are the result of both a saving glut (Bernanke's explanation) and central banks pushing rates below the Taylor rule level. I find the evidence that the paper presents very weak but my main issue is much more on the interpretation of the hypothesis of artificially low interest rates.

The first question is how can central banks be seen as so powerful as to control and distort a market price for such a long period of time? Typically, the models where central banks are powerful enough to do this are those with nominal rigidities in prices and wages. But these rigidities are assumed to be temporary as prices and contracts adjust. How can it be that central bank have managed to affect a real price (the real interest rate) for more than a decade? I cannot think of an accepted model that would support this. What is more paradoxical is that those who tend to support this view are in some cases those who are critical of models with price rigidities. So on hand they dislike models where central banks are powerful and on the other hand they argue that central banks have been super powerful over the last 10 or 15 years. This is very inconsistent.

The second question is how can it be that such a low level of artificially low interest rates has not had any effect on inflation. The original interpretation of the Taylor rule was always about the level of interest rates that was consistent with a stable inflation rate. How can we explain a deviation from the Taylor rule that lasts for many years and that instead of causing an increase in inflation it is producing a low level of inflation everywhere where interest rates are low? Once again, not sure what model can explain this.

Finally, the view of those who talk about artificially low interest rates tends to be driven by an analysis of the US economy in isolation. Interest rates are low at a global level, they are at low record levels everywhere in the world. What type of coordination exists between all central banks in the world to keep artificially low interest rates everywhere without generating inflation anywhere? The paper by Fitwi, Hein and Mercer tries to look into this issue by analyzing capital inflows to the US and its potential influence in interest rates (as a test of the Bernanke hypothesis) but this is not a good test. If you take the world, there are no capital inflows from other planets but a shift towards higher saving will still cause lower interest rates. 

In summary, there are two very simple facts that provide strong support to the Bernanke hypothesis on why interest rates are (naturally) low:

1. Interest rates are low everywhere in the world.
2. Inflation remains low everywhere in the world.

These two facts are very difficult to square with a world where the US federal reserve is keeping interest rates artificially low for many years. 

Antonio Fatás