Is MacroPru Procyclical?

I know it is heretical to even suggest it, but is it possible, just possible that MacroPru could be procyclical?

The question came up in a conference recently, where I was presenting on the nature of risk and after showing my 42 year MacroPru cycle plot, suggested MacroPru might just be procyclical.

After my talk, a policymaker came up to me and said that could not be possibly true.

So, where are the arguments?

The orthodox view might say that by measuring cycles in risk-taking and implementing MacroPru policies that counteract those cycles, MacroPru is countercyclical by definition.

So what could possibly make it go the other way? There are at least two factors:

  1. Policymakers measure risk with a considerable lag and uncertainty, and it is quite possible that any consequent policy remedies might take place when the cycle is already turning;
  2. At the moment the various parts of the financial system, like insurance companies, asset managers, pension funds and banks are usually regulated by separate authorities, all with their own distinct methodologies. That means they tend to behave differently — not the least and how they react to shocks. If, as many would like, all of these are to be regulated by the same authorities using the same methodologies, things change. These various industries would inevitably behave more similarly — increasing procyclicality.

So, it is entirely conceivable that MacroPru is procyclical, just depends on how it is implemented.