A good central banker has to think like a good football player. A footballer usually tries to pass the ball to the position where his colleague will reach a few seconds later rather than where he is right now. Anticipation is important. A central banker has to similarly fix interest rates depending on where inflation will be a year down the line rather than where it is right now. Why? Because monetary policy works with a lag of around three or four quarters. A rate cut or a rate hike today takes time to work through an economy.

That is why getting the inflation forecast is so important in modern central banking. The actual inflation rate in the future cannot be known today. So the inflation forecast becomes the intermediate target of monetary policy even though the formal target may be some specific measure of actual inflation. To switch metaphors: Central bankers have to look ahead through the windshield rather than look back through the rearview mirror. Too many Indian monetary policy debates are based on recent inflation rather than future inflation, through the rearview mirror rather than the windshield.

The problem is that central banks across the world have struggled to get their inflation forecasts right in recent years. The Reserve Bank of India (RBI) has been no exception. In an earlier column in this newspaper, I had used data from Monetary Policy Reports that the central bank releases every six months to show that RBI has been overestimating potential price pressures in its inflation forecasts.

A new research report released by RBI earlier this month under its Mint Street Memo series shows that “episodes of large inflation forecasts errors for India were associated with large and unanticipated shocks emanating from prices of food items, especially perishables such as vegetables". The more general observation in the report is that countries where food items account for a large part of consumer price inflation tend to have larger inflation forecast errors. In other words, the problem is with forecasting food prices rather than the rest of the consumption basket. Food accounts for a little less than half of the Indian consumer price index.

Every central bank has a statistical model to forecast inflation. The RBI is no exception. The inflation estimates it releases are not plucked out of thin air. However, the RBI inflation forecasting model—the Forecasting and Policy Analysis System—has parted ways with standard central bank forecasting models by taking a disaggregated look at food, fuel and core inflation. In other words, there are separate forecasts for food inflation within the overall inflation forecasting framework.

A lot of the immediate estimates of food price inflation are based on trends in wholesale prices in mandis across the country as well as surveys in consumer markets. Some have begun to scrape data from e-commerce sites as well. Economists in RBI as well as the private sector depend on such data to understand the direction of food prices in the next few weeks. Food price estimates for longer periods of time require more rigorous forecasts that feed into the inflation forecasting system.

This is where RBI seems to be struggling. The dominance of food prices in the Indian consumer price index as well as the food price forecast errors of RBI in recent years has become a problem for monetary policy makers. Here are two radical suggestions. First, the Indian central bank needs to bring Big Data experts in to help keep a closer eye on factors that affect food prices—be it understanding soil moisture levels using satellite data or getting online data from various e-commerce sites to pick early signs of big moves in consumer prices. Second, RBI should consider getting more agricultural economists into its policy process, be it as staff economists, outside advisors or even as members of the Monetary Policy Committee. In fact, the central bank has a big stake in promoting more agricultural economics in the country.

Credibility also has a big role to play. Sudden moves in food prices need not unleash big changes in the inflation expectations of households as long as people have confidence that monetary policy will maintain inflation close to the promised target. There are thankfully early signs that such credibility is being reestablished—going by the recent decline in inflation expectations.

The RBI is not alone. Other central banks have also struggled with their inflation forecasts. It is important to communicate the fragility of forecasts. Something that former Federal Reserve vice-chairman Daniel Kohn said recently is worth repeating: “My recommendation for the presentation of economic projections and their use in policy explanations is to align them better with the rhetoric of uncertainty and data dependency, and the true state of economic knowledge."

Niranjan Rajadhyaksha is research director and senior fellow at IDFC Institute. Read Niranjan’s previous Mint columns here

Close