Food Prices and Distribution Margins in India

 
Feb 3rd 2011, Jayati Ghosh
The dramatic increase in food inflation over the past two years has been associated with several surprises. One major surprise has been how the top economic policy makers in the country have responded to it. The initial response was one of apparent disbelief, followed very quickly by the frequently repeated but thus far unsubstantiated conviction that prices would come down very soon.

Then this massive increase in the price of essential commodities was welcomed, even by those who should know better, as being a sign of greater material prosperity in the country and the success of ''pro-poor'' schemes of the government, reflected in increased demand for food. Could it be that the economists who are running the country apparently believed that food demand should not increase much even in periods of significant aggregate income growth, and among a population that has the some of the worst nutrition indicators in the world? Is that why they did not see any need to work towards increased supply of food and have been so surprised by even a slight increase in demand?

As it happens, in fact demand for food has been growing much more slowly than could be anticipated by both income and population growth. Much of that has to do with the distribution of that growth, which has disproportionately denied benefits to the poor who would naturally consume more food. But even so, the fact that it is really the conditions of supply – reflecting the continuing policy neglect of agriculture as well as the nature of distribution and the pressures on the market from speculative activity – that have driven food prices up.

This recognition may be why the official arguments have changed somewhat recently. Most recently the officially stated position has been to blame inadequate existing distribution chains – focusing on their inefficiency, rather than any speculative pressures that could also affect supply. This has become the most popular interpretation of the ongoing food crisis in the corridors of power and their stenographers in the financial press. This has consequently led to the demand that modern corporate retail chains (ideally with FDI) be brought in to manage food distribution.

As a result, there are now those who have argued that the only solution to the problem of high food prices is to bring in FDI in retail! It is argued that this will reduce wastage in storage and costs of transport of food items, cut out intermediaries in distribution and provide food more effectively to consumers at lower prices.

Of course this argument is rather foolish, at several levels. First of all, if the traditional supply chains in food items are so faulty and deficient, why did they not create such massive food price spikes earlier? Why was food inflation relatively low in the period until 2006, despite equally rapid GDP growth and the same system of distribution that is being faulted?

Secondly, if the problem is inadequate infrastructure, including cold storage facilities and quicker distribution networks from farm to market, what stopped the government from more proactive intervention to ensure better cold storage and other facilities through various incentives and promotion of more farmers' co-operatives? To announce such measures only now, as a weak response to a period of raging food inflation is futile, because obviously such measures operate only with a significant time lag. This is all the more so because such proposals are explicitly mentioned in the Farmers' Commission Report, which has been lying with the government for half a decade now. The idea that cold storage and other facilities can only be developed by large corporates once they get directly involved in retail food distribution is ridiculous at best.

Thirdly, this entire argument completely ignores the critical role that can be played by a public distribution system in moderating such food price spikes and dampening inflationary expectations and tendencies of hoarding. Instead of accepting the failure of the government to use this system effectively so far, the tendency is to throw up hands and declare that only the large private sector can save us, even though international evidence indicates that corporate monopoly in food trade typically increases distribution margins rather than reducing them.

Unfortunately, though, we are forced to take such arguments seriously because they are being repeated ad nauseum by the media and pushed into government policies by corporate lobbies. So let us consider what the recent evidence on distribution margins indicates.

In fact, there is significant reason to believe that the margins between wholesale and retail prices of many important food items have increased in the recent period. (See MacroScan, Businessline, 23 February 2010) The point is that this has been happening in a period of increased corporate involvement in food distribution and food retail. The share of corporate retail in food distribution in the country as a whole is estimated to have tripled in the past four years, and has grown even faster in major metros and other large cities. And this is also the period when retail food prices have shown the greatest increase!

The other point that emerges from a comparison of retails margins across major towns and cities is that such margins are lowest in the states (like Tamil Nadu and Kerala) where there is an extensive, well-developed and reasonably efficient system of public distribution that provides a range of food items on a near universal basis to the population. In regions where such a public distribution system is weak or non-existent (such as Utter Pradesh and Bihar) the margins tend to be much higher and growing faster, even though corporate food retailing in such regions has been expanding.

So to look at corporate retail as the solution to the current food price increase is more than irresponsible. There is no question that the current system of food procurement from farmers is inadequate, faulty and often quite anti-farmer. There is much that needs to be reformed in the way that market yards are organised and in the options available to farmers to get their produce to market. There is a range of necessary and possible interventions for this, most of which have been stated many times to the government by various Commissions of its own. Yet thus far the UPA government has done little about any of these, even in terms of working with state governments to improve the situation, and instead seems to think that simply allowing more corporate (and FDI) activity in retail will allow it to wash its own hands of the matter.

In this context, consider how retail margins have been behaving in the very recent past, in just one location, the city of Delhi. Charts 1, 2 and 3 describe the price behaviour of three significant but relatively less perishable food items: rice, sugar and tur dal.

Chart 1  >> Click to Enlarge

Chart 2  >> Click to Enlarge  

Chart 3  >> Click to Enlarge

It is evident that the retail prices have generally been tracking the wholesale prices in terms of direction of movement, but still there are some noteworthy variations. On average, retail margins have increased for all these commodities, and quite sharply for tur dal. This may be the result of a number of features, and obviously requires more investigation. But even so it is worth noting that Delhi is a city that has witnssed a signfiicant increase in corporate food retail. And the role of inflationary expectations in being able to influence retail price behaviour is obviously much greater for larger players.

The food prices that have been most talked about of course are those of onions. Onion prices are widely perceived to have great political significance, especially in North India. Because onions like other vegetables are highly perishable, supply conditions should play a major role in their price. Charts 4 and 5 describe the wholesale and retail prices, and the total market arrivals of onions and tomatoes in the city of Delhi.

The evidence is somewhat surprising. For much of the period of falling market arrivals over the past year, onion prices were rather stable and the retail margin actually shrank. Prices started rising sharply only in October – and this is the period after which supply was actually increasing quite sharply! In November and December, market arrivals increased but prices continued to shoot up. Surely inflationary expectations and hoarding must have played roles, along with speculative pressure, and this was not sufficiently counteracted by public intervention through its own food distribution network.

Chart 4  >> Click to Enlarge

Chart 5  >> Click to Enlarge

The case of tomato prices is similarly interesting. It is evident that neither wholesale nor retail prices had much relation to market arrivals, even for this extremely perishable commodity. But what the period of higher prices has been associated with is a significant increase in retail margins in October and November.

Dealing systematically with the problem of high food prices in a context of a largely hungry population should normally be a priority issue for any government. There are certainly crucial medium term policies that reverse the longer run neglect of agriculture, that must be implemented. The issue of rapidly rising cultivation costs that are making farming unviable once again, needs to be addressed in a holistic way. The concerns of storage, distribution and post-harvest technology also need to dealt with. But in the short run, the problem cannot be avoided by talking of astrologers and the inability of mere humans to predict the future. Instead, creating a viable and effective public distribution system that will counteract tendencies to price spikes in essential commodities is an immediate requirement.
 

Print this Page

 

Site optimised for 800 x 600 and above for Internet Explorer 5 and above
© MACROSCAN 2011