The increase in prices has been lower for many food items without MSP as compared to those that have MSPs
Everyone agrees that food inflation is not just a serious problem now, and many feel that it is raging out of control. All measures of consumer prices have been mostly running over 10% for over a year. Within that, food prices, which comprise about half in some of these measures, have increased at a faster rate.
One major factor that affects food prices is the procurement price system. Many economists are of the view that the minimum support price (MSP) at which the government buys foodgrains and some other food items from farmers has been instrumental in keeping food inflation high. The most vocal and prolific proponent of this view is Surjit Bhalla, who has written several columns on this matter, over the last two years, a noteworthy one being ?Price of Paddy Populism? (FE, May 10, 2012, http://goo.gl/XgeVH).
These columns draw upon a longer paper of Bhalla?s titled ?Indian Inflation, Populism, Politics and Procurement Prices? (Oxus Research, July 2011). In that paper, he presents a regression result indicating that MSP inflation is affected by lagged CPI inflation, and that it rises in election years. Specifically, a 1 percentage point rise in CPI Inflation raises MSP inflation by 0.53 percentage points the next year. In turn, the change in average procurement prices (derived from the above equation) predicts CPI inflation for that year: a 10% rise in the average procurement price leads to a 3% increase in the CPI.
Predicting out of sample, he concludes that procurement price inflation has pushed up inflation by about 4.5 percentage points higher than the base value of 5.1% predicted by his model. Adding the two values, the model predicts 9.6%, almost equal to the actual inflation of 9.5% over the period 2007-11. This result holds without bringing in the usual variables?money growth, fiscal deficit, the output gap etc used to explain inflation. (The reader is invited to check out his results ?a forecast of 5-6% CPI inflation by the end of the year does not appear to be outlandish? made in July 2011 in this paper.)
In a very recent article he proclaims, ?RBI should state clearly that Indian inflation has been primarily food inflation?and that a necessary and sufficient condition for food inflation to decline is that procurement prices are contained? (?Over to RBI?, FE, March 10, 2013, http://goo.gl/dDF8O). While his position is extreme, the annual Economic Surveys from the finance ministry have also stressed the role of MSP hikes in high inflation, and so has RBI in its various publications in recent years.
Let us take his results as valid, noting that Oxus Research has developed and used a proprietary index of procurement prices. Ideally, one should first be able to replicate his regression results before analysing them. Even when all data is in the public domain, such replication often cannot be done.
However, his conclusion about MSP is not justified. The cost-push inflation view of inflation is flawed. As explained by Milton Friedman in his 1966 debate with Robert Solow on wage price controls, special factors?such as union mandated wage increases, monopoly pricing power, supply shocks (in India?s case procurement price hikes)?can only raise the relative price of that item. They cannot raise the inflation rate. If the central bank keeps overall spending (nominal GNP) at its prevailing level, higher spending on the item whose relative price has gone up will be offset by less spending on others, and inflation for other items would tend to fall.
Drawing upon this argument, a comparison of the relative price of food to non-food items in India with that in Asian countries reveals a robust fact. Over 2006 to 2011, the relative price of food to non-food in India, compared to major Asean countries, was among the lowest, and was much lower than in China. However, India?s inflation rate was very much higher than these countries indicating lax monetary policy and not MSP hikes are to blame. (Shrikant Kolhar, IIMB unpublished thesis on Inflation Measurement, March 2013).
The observed statistical links between MSP and overall inflation are, most likely, a manifestation of the inflation adjusted (or expectations augmented) Phillips curve process. In any given year what is or appears to be cost-push reflects a delayed response to demand overheating. Surjit Bhalla seems to recognise this by stating, ?the price of food goes up which sets an increase in wages, which results in costs going up, which means output prices go up, which means an increase in CPI and WPI (July 2011).? From a classical standpoint, the MSP hike may potentially trigger the Phillips curve process, but a continuing ?wage-price spiral? depends on the central bank?s interest rate decisions to allow demand to increase.
Going beyond this macroeconomic logic, there is no solid evidence to indicate that MSP hike, rather than prolonged demand overheating, is the trigger. There is a simple way to evaluate the MSP hike versus excess demand explanations of inflation. Not all commodities have MSP. In fact, most of them do not have MSP. For these latter items, price is clearly ?demand determined? not ?cost determined?. (Quotes are used since this distinction, and RBI?s approach of trying to decompose inflation into cost versus demand factors, is not very meaningful. It would take another article to explain why.)
Let us look at the price rise over the last five years for five items with MSP and those without. Robust statistical studies use such control groups, instead of trying to establish causality from time series data. The accompanying table is self-explanatory. For many food items in the latter group, without MSP, the increase in prices has been larger than for those with MSP. These items were chosen randomly.
These findings are not definitive. The topic of agriculture requires vast domain knowledge, and one could have tripped up somewhere. But, more recently, Commission for Agricultural Costs and Prices chairman Ashok Gulati stated that food prices have risen more for some commodities without MSP than those with MSPs.
MSP hikes and price support systems are not harmful. They distort production and wreak grave damage on the economy, and push up the fiscal deficit. Surjit Bhalla should be commended for tirelessly denouncing the socialism of Joseph Stalin and Madam Sonia Gandhi. Nevertheless, contrary to his repeated assertions, one can conclude that the MSP system does not have a major impact on inflation.
The author is a research associate at IIM Bangalore
