Time for RBI to relook its inflation modelling PDF Print E-mail
Thursday, 05 April 2018 00:00
AddThis Social Bookmark Button

RBI continues to miss all forecasts by big margin – in Feb, it forecast March-quarter CPI at 5.1% against a likely 4.6%


Chances are, later today, the central bank will once again talk of the upside risks to inflation from high oil and commodity prices globally and the government’s plan to offer high MSPs to all farmers; the latter is seen as not just inflationary in a direct manner, to the extent it raises the fiscal deficit, it will further push inflation. The Monetary Policy Committee (MPC), however, needs to be careful in its assertions. It has, in the past, said that a 100 bps rise in the combined government deficit could cause a 50 bps hike in CPI, but the equation is unlikely to hold in a low-demand situation as now. When the combined deficit was high, but fell slightly, from 6.8% in 2013 to 6.3% in 2017, CPI collapsed from 9.4% to 3.7%. Between 2005 and 2010, the combined deficit rose from 7.3% to 9%, while inflation rose dramatically, from 4.2% to 10.5%. And, while deficit levels remained high, but fell slightly, from 9.5% in 1998 to 8.8% in 2004, CPI crashed from 13.3% to 3.9%. So, there is a lot more to inflation than just the deficit. What RBI needs to focus on, a lot more, is the fact that inflation is trending down and, indeed, the central bank has got almost all forecasts wrong in the last few years. As the Economic Survey puts it, in the last 14 quarters, inflation has been overestimated by 180 bps in six quarters, and we are talking of just three-month-ahead forecasts.

In the last policy, in February this year, RBI forecast inflation for the quarter ending March at 5.1%—though the March inflation numbers are yet to come in, they are likely to come in at around 4.1% and, as a result, CPI for the March quarter is more likely to be around 4.6% or a full 50 bps lower than that forecast by RBI. So the first thing RBI needs to do in its monetary policy deliberations is to examine why the model is doing so badly. This is important because, despite the obvious flaws in an inflation-targetting framework, if RBI had got its forecasts right, it would have had scope to cut interest rates and that, in turn, would have ensured bond rates didn’t jump as they have. Indeed, while RBI is not expected to cut rates today, with inflation likely to remain benign in FY19—BoFAML has cut its forecast by 20 bps to an average of 4.7% for FY19 and 4.6% in March 2019—there is an outside chance there could be a rate cut in the August policy if, as expected, the monsoon is normal. In the current policy, of course, RBI needs to spend more time looking at how to bring down bond yields further, perhaps by promising more liquidity. The central bank has, thankfully, just reversed its position by allowing banks four quarters to mark-to-market their bonds portfolio—in January, Deputy Governor Viral Acharya indicated that this would not be allowed and pushed yields up 11 bps in one day; with banks now allowed more time to deal with bond market losses, they will once again turn buyers in the bond market and that will also help cool yields. Similarly, bond yields rose 18 bps the day the government announced, in late December, that it was raising its borrowings by Rs 50,000 crore; on March 26, when the government announced a sharp curtailment in its first-half borrowings target, yields fell. So, if, between the government and RBI, bond yields are kept under check, this could help monetary transmission.


You are here  : Home Miscellaneous Time for RBI to relook its inflation modelling