May 06, 2015
It is time the central bank comes up with a better framework for measuring inflation expectations
When Reserve Bank of India (RBI) governor Raghuram Rajan made his first monetary policy rate cut on 15 January, this was the rationale he gave: “Households’ inflation expectations have adapted, and both near-term and longer-term inflation expectations have eased to single digits. These developments have provided headroom for a shift in the monetary policy stance.” Rajan has alluded to households’ inflation expectations in every policy decision speech he has made as well as in his opening speech upon donning the mantle as the 23rd governor of RBI in September 2013. In his interaction with analysts on 3 December 2014, Rajan asserted that, “These are real people whose inflation expectations matter. This is your voter, this is your citizen. These people have to feel more satisfied about inflation.”
Evidently, the aam aurat’s expectations of price rise is a cornerstone to Rajan’s regime as governor. And Rajan is on record stating this indicator matters much more for monetary policy decisions than market determined inflation expectations through government bond yields. For all its eloquent significance, the details behind this aam aaurat indicator are shrouded in mystery.
Every quarter, RBI surveys nearly 5,000 people across 16 cities to elicit both qualitative and quantitative responses on inflation expectations over the near term (three months) and long term (one year) periods. These answers are then used to calculate median and mean inflation expectations for the next three-month and one-year periods. The true merit of the survey lies not in the accuracy of the common woman’s inflation expectations but in the trends of their expectations. Trends matter while levels don’t, argues Rajan. Thus ostensibly, an analysis of trends of double-digit versus single-digit inflation expectations in the near term and longer term should be a reasonable framework to glean households’ inflation expectations.
Analysis of 17 quarters from March 2011 to March 2015 of RBI inflation expectations survey provides some interesting insights. For 16 of these 17 quarters, consistently between 60-80% of the respondents expected near term double-digit inflation with no steady declining trend. Inexplicably, in the December 2014 quarter that number dropped sharply to 36% and further to 35% in the March 2015 quarter. Similarly, between 70-80% of respondents expected double-digit longer term inflation. Again, that number fell radically to 39% in the December 2014 quarter and the March 2015 quarter. From the time Rajan assumed office in September 2013, to September 2014 the percentage of respondents expecting double-digit near term inflation has held steady at 64% and those of double-digit longer term inflation at 68%, with no signs of any declining trend. To add, during the same period nearly 50% of all respondents expected longer term inflation to be higher than 15% with no upper limit specified. And 40% expected near term inflation to be greater than 15%. Then abruptly in the December 2014 quarter, those households expecting double-digit near term inflation fell to 36% from 67% in the previous quarter and double-digit longer term inflation expectations fell to 39% in December 2014 from 72% in September 2014. RBI quickly followed suit and dropped rates in January, citing this as one rationale for the decision.
Contrary to RBI’s statements of declining household inflation expectations as measured by the median and average, as measured by percentage of respondents expecting double-digit versus single-digit inflation, there was no evidence of any significant declining trend until September 2014 before the bizarre drop in December 2014. It is apparent that this sharp drop in people’s inflation expectations in one quarter was neither due to some sudden onset of achche din governance nor a case of outsized wage growth.
While measured consumer price index (CPI) inflation has been steadily declining, it cannot be posited as an explanation just for one quarter sharp drop and not as a rationale for other quarters when expectations didn’t fall while CPI inflation did.
Similarly with global crude prices. It is evident that either the inflation expectations survey is insufficient or is moodily reflective of global trends or CPI inflation or is plain whimsical. Whichever way, the household inflation expectations survey in its current form seems more an alibi than any rock solid foundation for monetary policy decisions. To be clear, I am no economist, let alone a monetary one to opine on monetary policy decisions and hence this is not to hold brief against the January rate cut. This column is merely to point out the stated significance of households’ survey on such decisions and the arbitrariness of it.
Inflation expectations evidently play an inordinately large role in the new RBI inflation-targeting regime. Presumably then, measuring inflation expectations reliably will be crucial. And trends in expectations matter more than levels of actual expectations. If these are sacrosanct principles to abide by for an inflation-targeting regime, as governor Rajan has argued for, then it is time RBI comes up with a better framework for measuring inflation expectations, in the absence of a robust market mechanism for doing so.
Praveen Chakravarty is founding trustee of IndiaSpend, India’s first data journalism initiative and a former chief executive officer of an investment bank.