Dynamic concept Monetary policy

Transparent versus fuzzy

The U.S. added just 38,000 jobs in May, the Labor Department reported on last Friday. That was far fewer than the 158,000 jobs economists had expected. The employment figures for the previous two months were revised down by 59,000 jobs which are significant. However the unemployment rate fell to 4.7% from 5% due to the fact that number of people looking for jobs fell sharply.

The uncertainties in the labor measurement create challenges for the central bank which has repeatedly stated that changes in the future monetary policies would be “data dependent”. If the data available have unusual large amount of uncertainties, the correctness of monetary policies can be in question.

In process control, when the feedback measurement have uncertainties, typical design will ask for a synthesis of a number of measurement, rather than only one; or if only one measurement is available, then the measurement will go through a “filter” which can reduce noises. The benefit of using filter comes with a price to delay the feedback signals.

The Fed has adopted both methods – by using a wide range of measurements and by averaging data over several months. The Fed under Bernanke and Yellen has promoted transparency. Monetary policies changes were explained in detail and the criteria for future policy changes are also explicit in each FOMC statement. This is sharply different from Greenspan’s Fed where rationales behind policy changes are often vague and fuzzy.

I think transparency has the benefit when the monetary policy is saturated and the policies more rely on time dimension, i.e. so called “communication tool”. In this case, momentary policy does not need much feedback information as control is in saturated zone.

However, in normal time when the monetary policy needs to be adjusted or regulated, feedback becomes important. Unfortunately uncertainties are not avoidable in any measurement. In order to properly making control decision, feedback has to be processed by using other measurement or by filtering out noises. In both case the control will sacrifice by delaying control decision.

That might be a reason why Greenspan chose a fuzzy control strategy. Fuzzy control does not depend on precise measurement, and control decision is also fuzzy.

I am not sure if Yellen’s Fed will adopt fuzzy control, but when the economic and financial environment become normal, conducting monetary policies will be more of art and less mechanical. “Data dependent” can be technically very confusing.

Leave a Reply

Your email address will not be published. Required fields are marked *