Sunday, September 27, 2015

Central bank independence: India does better than Turkey and Brazil?

Well, that's what an article in the FT claims. Both Turkey and Brazil have inflation targeting as the objective of their central banks, as India does now. But both countries have missed their targets- thanks to political pressures, according to the article:
Recep Tayyip Erdogan, Turkey’s president, has repeatedly tried publicly to bully the central bank into cutting interest rates, calling its governor a “traitor” and evoking the peculiarly paranoid notion of an “interest rate lobby” that was damaging Turkey by demanding high borrowing costs.

In Brazil, the central bank is still part of the finance ministry, and its head is answerable to the president. Its governors are also appointed by the president and have no fixed terms. In last October’s election, Marina Silva, the Brazilian Socialist party candidate, argued strongly that the central bank had done too little to control inflation in an election year because of political pressure from Dilma Rousseff, the president.
In India, in contrast, the author contends, the RBI has managed to get on with its job of containing inflation, thanks to a determined person at the top. So it's best to leave inflation targeting to technocrats free from any political interference.

Well, matters are not that simple. Inflation targeting is not the norm amongst central banks. The Fed does not follow inflation targeting nor does the Bank of England. Secondly, if we even believe in inflation targeting, leaving matters entirely to the RBI is not a great idea. There is always need for an external input. And government nominees are not necessarily handmaidens of the government. Economists nominated by the government have their own reputations to protect. Indeed, one could argue that they are, perhaps, better placed to act independently that some of those inside RBI, as the latter would have their own career ambitions to worry about and have every incentive to stay on the right side of the government.

Lastly, what constitutes 'optimal' or 'threshold' inflation is not easily determined. The RBI has set itself a target of 6% for January 2016. A recent paper in EPW points out that there is a wide range for inflation thresholds, as estimated by various studies. One study, which covered 127 countries in the period 1960-92, found that the inflation threshold is as high as 20% if outliers (those with hyperinflation of over 40%) are excluded from the sample. The EPW paper itself places the threshold at 11% for Asia, 23.5% for Latin America and the Caribbean, and 23.6% for sub-Saharan Africa.

There could be costs - in terms of lost output- to keeping inflation below a threshold if the the threshold is incorrect. Given the difficulties in estimation, there is a good chance that the threshold is indeed incorrect. That's why being fixated on inflation targeting may not be a good idea. Politicians, including those in Turkey and Brazil, may have a point when they complain about the monetary polices of their central banks.

No comments: