Policy: Lack of tools and a drought of data

The Reserve Bank of India has been contemplating adopting Inflation Targeting (IT) in India and replacing the time-tested multiple indicator approach (MIA) used successfully since 1998.

As the name indicates, MIA covers different aspects of the economy like exchange rates, growth, liquidity conditions in the market, financial stability of institutions, employment and also inflation, while the focus of IT is only inflation. It is for this reason that since the onset of the global financial crisis, the major economies are shifting towards MIA and shunning away from IT.

An important consideration in IT is the ability to apply the Taylor Rule (TR), named after Prof John Taylor of Stanford University. In fact, interest rate adjustments are attempted by the central bank based implicitly on TR which implies calculation and forecast of output gap and deviations of inflation from the stipulated target. The estimation and forecast of output gap, and interest rate path involve use of many sophisticated econometric tools which India lacks. These tools are like the telescope which gives the policy maker a peep into the future.

The right models

Among the important features of IT, in most cases the target horizon for operation is six to eight quarters and in some cases, even three years. And, in most cases, sophisticated models like dynamic stochastic general equilibrium (DSGE) models, autoregressive time series models and semi-structural models are used taking into account quarterly data.

In almost every case, forecasts are regularly disseminated along with the assumptions. In specific and successful cases like Australia, many other models along with DGSE are used. Similarly, in New Zealand, a host of models like calibrated gap-form models, and factor-augmented autoregressive models are used.

In Sweden, a suite of models like time series models, structural models and many small models, in addition to DSGE are used for forecasting. The Bank of England uses both statistical and theoretical models, and a new DSGE model called ‘central organising model for projections analysis and scenario simulation’ has recently been introduced in 2011.

The important issue in India is lack of expertise in the above mentioned models and non-availability of reliable data for obtaining these forecasts on series like gross domestic product, inflation rates and employment. In advanced countries, one of the ways to compute GDP is through the rate of unemployment which is reliably computed every month, given that the authorities provide unemployment allowance.

In India, such data, not even unreliable data, exists. Therefore, making policy decisions which impact long-term decisions based on scantily available data could have perilous consequences. In a country like India, where food prices account for nearly half of the newly adopted consumer price index, not only the vagaries of the monsoon, but also supply side problems could play havoc with decision making. International oil prices could be another unpredictable factor. In fact, the need has arisen for a robust and representative price measure — which is still unavailable in India.

The RBI has proposed flexible IT (FIT), but FIT loses its significance as IT per se is expected to guide the general business and industry about its policy path in the next few quarters to facilitate investment decisions. Also, FIT is operationally non-feasible as the Parliament cannot be expected to fix or revise the target on a monthly and quarterly basis and therefore discretion will continue to play a role. The change in the interest rate on a regular basis will also not be useful in making long-term investment decisions.

The transparency factor

IT also requires and implies high levels of transparency and though RBI has been advocating IT, and despite a tradition of transparency in articulation of the policy measures and analysis, the trend has reversed in recent months. In RBI’s annual monetary policy announcement (AMPA) of April 1, 2014, in sharp contrast to last year, the policy documents were very brief. To illustrate, the Macroeconomic and Monetary Developments (MMD) issued on May 2, 2013 along with AMPS for 2013-14 had 49 pages as compared with the recent similar document with just 16 pages.
 
The crucial issue is lack of substantive information contained in the documents that deprives researches and market analysts of valuable data. If there is a change in the policy of data dissemination, RBI could consider issuing a press release as to where rare historical data can be located. Important data, for example, that could help a researcher is related to variations in the housing indices, non-performing assets and their sectoral distribution, and external sector vulnerabilities. In fact, as expected, AMPA should have contained the largest dataset in detail, as it shines the light on the recent past and immediate future; for, non-release of data only fuels suspicion.

Besides, according to the global financial stability report released by IMF last week, sovereign wealth funds are progressively expanding their investments in India. If IMF is correct, it would have been interesting to know from the policy documents as to who these sovereign wealth funds are and in what segments of the economy they are investing.

Historically, IT has been generally adopted by countries recording hyperinflation. In India, that has never been the case. Therefore, inflicting such high interest rates on the Indian public in the name of combating inflation has only resulted in lower investment and growth, and higher non-performing assets. In view of the young population of India, our priorities should probably be on ensuring higher employment and growth and not just low inflation.

(The writer is RBI Chair Professor of Economics at the Indian Institute of Management, Bangalore)