Eight months into an informal adoption of inflation targeting with headline CPI as nominal anchor, there are signs of discomfort in RBI about the evolution of inflation expectations in India. As per the latest survey of households (HH), inflationary expectations persisted at higher levels in spite of headline CPI declining quite sharply. RBI seemed a little puzzled as median expectations for current, 3-month and 1-year ahead inflation in April-June 2014 came in at 13.3%, 14% and 15%, respectively, as against 13.3%, 12.9% and 15.3% in the previous (January-March) quarter. At the post policy conference call on August 6, RBI Governor admitted that disinflation has not shown up as yet in inflationary expectations and he would not take the level of these surveys as Gospel truth. Not surprisingly, he pointed out that policy would be more driven by actual inflation than by these survey inflationary expectations.
Barely weeks down the line, RBIs Annual Report (Box II.3) went a step further in diluting the significance of the outcomes of the household surveys and instead, displayed a marked preference in favour of professional forecasters (SPF) survey, which shows moderating medium- and long-term inflation expectations. As per the latest SPF, 5-year and 10-year ahead median expectations for CPI inflation declined to 7% and 6% respectively in April-June, from 7.75% and 7% the previous quarter; in line with the medium-term CPI inflation target of 6% by January 2016. The report emphasised that as the professional forecasters consider all the available information, past and future, their expectations could be considered more rational than that of households.
This was bit of a surprise given the fact that the Urjit Patel committee report on the new monetary policy framework (UPC) presented clinching evidence of high correlation of CPI-combined inflation with 3-months-ahead household inflation expectations, with persistence up to eight quarters (annex 2 of the report). RBIs panel data analysis shows both 3-month and 1-year ahead expectations are significantly influenced by food as well as fuel inflation shocks, which have a much larger and more persistent impact on inflation expectations than shocks to core (non-food, non-fuel) inflation; elevated inflation expectations then spill over into inflation of other items, especially services, with a lag.
So while survey-based measures may not provide a fully clean read on inflation expectations at this point, and financial market measures are hard to construct given Indias underdeveloped, long-term bond markets, the preference for a rational measure (SPF) over the HH-measure, which could be considered adaptivehouseholds place a large weight on recent inflation data, especially food and fuel, a highly variable componentis hard to understand. But food shocks and their role in shaping HH expectations are the bedrock of the new monetary policy framework with headline CPI as nominal anchor. In this light, the doubt expressed on the HH-measure of inflation expectations in such a short time span is a surprise. More so as it isnt evident at this stage that the SPF-measure of inflation expectations reflects beliefs that headline CPI, dominated by large-sized disturbances, will no longer transmit into higher core inflation with a lag. The point, argued by the UPC, is that Stabilising and anchoring inflation expectationswhether they are rational or adaptiveis critical for ensuring price stability on an enduring basis, so that monetary policy re-establishes credibility, visibly and transparently, that deviations from desirable levels of inflation on a persistent basis will not be tolerated.
International evidence on survey-based measures of inflation expectations shows that in general, their rationality is well regarded. In fact, many research findings show that survey measures of expected inflation provide better inflation forecasts than any other alternative, including variants of the formal Phillips curve and term structure models. Most evidence however, exists for advanced countries; very few studies deal with emerging economies who are relatively recent inflation targeters. But recent work by the Board of Governors of the Federal Reserve System (International Finance Discussion Papers, No.1098, March 2014) considers if long-term inflation expectations are well anchored in Brazil, Chile and Mexico, three countries that adopted inflation targeting at the end of nineties. The study finds that though inflation expectations have become much better anchored over the past decade in these three countries, both survey-based and market-based readings on the long-term inflation outlook have been consistently above the inflation target in Brazil and Mexico, but more so in Brazil.
Given the evolution of experience over one and a half decades of inflation targeting in comparator EMEs, it is perhaps bit too early to expect public attitudes to inflation to change so soon in India. In fact, looking at the large gap in households current inflation perceptions13.3% regionand the official 8% rate of inflation (more than 5 percentage points, compared to an average 1.5 ppt differential in households perceived inflation rate versus the official rate in countries like UK, say) an issue to be contended with is the influence of monetary policy upon the behaviour of households, or their saving-spending decisions.
Why is this a problem Because if households perceive current inflation as 13.3%, they are equally likely to consider financial instruments like bank deposits as yielding negative real returns; interest rates on these are benchmarked to a much lower, official inflation rate. If this be the case, a shift from gold and real estate towards financial savings may not follow as result. The problem for monetary policy, now anchored upon headline CPIthis is what households use to deflate nominal returns and make portfolio choices, as UPC arguesis of successful stabilisation, viz. influencing inflation expectations of all economic agents via real interest rates; their behaviour, in turn, and thereby, aggregate demand. From these perspectives, it would seem that not only anchoring HH-expectations around the inflation target may take far longer than a few months, but whether real interest rates are at all regarded positive by households is itself an issue to be grappled with.
The author is a New Delhi-based macroeconomist.