“Growth and Welfare Consequences of a Rise in Minimum Support Prices: An Applied General Equilibrium Analysis for India”

 

Citation: Parikh, K. S., A. Ganesh-Kumar and G. Darbha. 2002. “Growth and Welfare Consequences of a Rise in Minimum Support Prices: An Applied General Equilibrium Analysis for India”. Report of the project “Minimum Support Price for Food Grains and its Implications for Food Security of the Poor”, submitted to the Department of Economic Affairs, Ministry of Finance, Government of India, New Delhi.

 

Executive Summary:

1. Introduction

1.    The mounting food stocks amidst substantial poverty and under nutrition raise questions on our agricultural policy. Government food stocks were 59.1 million tonnes as of 30th November 2001. With the expected bumper crop in the agricultural season July 2001 to June 2002, stocks are bound to increase. The existence of such large stocks in a country with 200 million persons below poverty line is an absurdity. The question is why are the stocks mounting? What is wrong with our policy?

2.    The main reason for increase in food stocks is related to the government procurement price policy. The problem is the level at which the government sets procurement prices, which have become minimum support prices (MSP) over the years. These prices for many years now have been higher than what free market prices would have been. The government, through Food Corporation of India (FCI) buys up whatever is offered at procurement prices ensuring that the harvest price for a particular commodity does not fall below its procurement price. Under the pressure of the farm lobby the prices are set high. At these prices, what farmers produce consumers do not demand for want of adequate purchasing power.

3.    The procurement prices are set by the government on the recommendation of committee on agricultural costs and prices (CACP). The CACP’s recommendations are based on costs of cultivation and adequate return to farmers. For a number of recent years, the government has set prices particularly of wheat, at higher levels than recommended by the CACP.

4.    What happens when government increases procurement prices? High procurement price gives farmers incentive to produce more. They will use more fertilizers and increase yield. Higher price would reduce demand. To support price FCI would have to procure more. Stocks would rise further. Government will have to finance the addition to stock. This is done by cutting some other expenditure. The easiest to cut is investment. Less would be invested in agriculture. Irrigation capacity would not grow as much. In a year or two the cumulative impact of lower irrigation would reduce growth rate of agricultural output despite higher procurement price. Farmers themselves could be worse off compared to what they could have been had investment in irrigation not reduced.

5.    Apart from that consumers and particularly the poor consumers may also be hurt. The poor consisting of landless labour, small and marginal farmers are net purchasers of food. They are able to buy less food even when one accounts for increase in wage rate that may follow higher procurement price.

6.    In order to appreciate better the consequences of a procurement price hike, it would be instructive to analyze the exact quantitative impact of a price hike on the growth and welfare of producers and consumers. To undertake such an exercise in a credible way, we need a numerical simulation model that accounts for the inter-sectoral interactions that one finds in a modern economy, price determination and income distribution mechanisms. This study explores in an integrated way the consequences of increasing MSP of wheat and rice by 10 per cent.

2. An applied general equilibrium model of India

7.    For this purpose we built a dynamic Applied General Equilibrium model for India. The model parameters are estimated and the model itself is calibrated for the base year to match the Social Accounting Matrix (SAM) for 1996-97. The SAM, besides accounting for all the inter-sectoral and agent transactions for a given year, is made consistent with a demand system estimated over various consumption and commodity classes. Other notable features of the model are: endogenous determination of level of stocks within the model without imposing any arbitrary stock limits, self or home grown consumption possibilities and consumption out of ration shops for some household classes. The later two features are of crucial importance for the purpose on hand as the impact of a price hike on different population classes would differ depending upon whether they are net producers or consumers of the food commodities, which is determined by the proportion of their total consumption they buy from ‘home’ and ration shops. If a large proportion of their consumption is ‘home’ grown and comes from ‘ration’ shops, an increase in market prices cannot affect their real consumptions and welfare significantly. However, the levels of home grown and ration consumption would themselves be affected by the equilibrium prices prevailing in the economy and hence a model that attempts to capture the welfare impacts of a price hike must explicitly model for the endogenous determination of different components of household consumption. The proposed model accounts for these as we describe in detail below.

8.    This model used in this study is characterised by 65 commodity producing sectors 10 types of households classified first by their location of residence into two broad groups – RURAL and URBAN and within each of them into 5 classes based on their annual per capita real expenditure. The model is sequentially dynamic in the sense that it is solved for every year in a sequential manner with the previous year’s solution providing a starting value for the next year. The policy shock that we analyze is a one time 10% hike in procurement price of Wheat and Rice.

3. The scenarios

9.    Policy analysis using a model requires specification of the prevailing and changed policy regimes. To facilitate comparison a base / reference scenario is first generated and the outcomes of policy scenarios are compared with that of reference scenario.

10. The reference scenario, labelled “REF”, against which policy scenarios are compared is a business-as-usual scenario in which the policy environment specified in the model corresponds to the actual policies observed during the base period, 1996-97. The policy scenario is designed as a one time increase by 10% in the support price of these two commodities in the second year, with all other parameters kept at their base scenario levels. That is, the government assures the producers of rice and wheat a 10% higher price by committing to buy whatever amounts of these two commodities that they may offer for sale. Given fixed rates of trade and transport margins and indirect tax / subsidies on these two commodities, this hike in procurement price and hence producer price will translate into a commensurate rise in both their market price and the ration price.

4. Results

4.1 Macro impacts

11. A hike in procurement prices of rice and wheat results in a marginal decline in total GDP (at constant prices) in the first year of the price hike, because of a fall in GDP non-agriculture. Agricultural GDP is marginally higher. This increase in agricultural GDP is, however, insufficient to compensate for the fall in non-agricultural GDP and consequently, total GDP is lower in in all the years.

12. The fall in non-agricultural GDP is attributable to the decline in aggregate demand due to rise in the overall price indices and decline in total fixed investments in the economy. The rise in overall price indices to the tune of 1 to 1.5 percent leads to a general reduction in real incomes and hence demand. Non-agricultural output being demand determined consequently declines. The decline in fixed investments is due to the additional expenditure that the government incurs to support the hike in procurement prices. This additional expenditure arises due to both an increase in procurement prices as well as the quantity procured. Given the overall resource constraint, the additional expenditure on building up stocks comes at the cost of a decline in fixed investments. While the additional expenditure on stocks favours only rice and wheat, the decline in fixed investments adversely affects the demand for many non-agricultural sectors, and hence the decline in non-agricultural GDP.

13. The results in decline in overall investments decline in agricultural investment also despite the higher relative price of agriculture and in the later years is felt in terms of virtually insignificant growth in gross irrigated area and gross cropped area, and hence agricultural GDP, despite a favourable shift in relative prices towards agriculture.

14. Thus from a macro perspective, the results indicate that the policy to hike procurement prices of rice and wheat has an adverse impact on overall growth, raises the level of inflation, reduces investment, and has a miniscule impact on agricultural GDP.

4.2 Impact on Agricultural Production, Consumption and Stocks

15. With the hike in prices of rice and wheat, the output of these sectors increase by 1.6 and 2.6 percent, respectively, only in the year immediately following the price hike. This increase in output is not sustained over time because of the fall in agricultural investments and irrigated area seen earlier. The hike in procurement prices results in increase in all prices, producer, ration as well as consumer prices, as explained earlier. This in turn, leads to a significant decline in total private consumption of these two commodities by around 3 to 3.5 percent. Indeed self-consumption of both these commodities is higher the ration and the market consumption are lower in response to increase in prices. The increase in output along with a decline in consumption results in a build up of stocks of these two commodities. For rice, stocks are larger by 12.9, 28.6 and 43.4 percent for the years 1998, 1999 and 2000 respectively compared to the REF. For wheat these are larger by 11.0, 23.3 and 25.8 percent, respectively. The combined stocks of rice and wheat at the end of three years come to around 66.6 million tonnes compared to 50.1 million tonnes in the REF. The effects of an increase in relative prices of rice and wheat on output, consumption and stocks of other agricultural commodities are quantitatively marginal.

4.3 Welfare impacts

16. The cross-sectional impact of the price hike in terms of distribution of population, income and welfare are instructive. There is a significant increase in the population in the poorer classes, both in rural and urban areas. Comparison of the two income distributions under assumptions of Pareto principle, anonymity and aversion to regressive transfers, clearly indicate a significant worsening of welfare of 80 percent of population in rural areas and all persons in urban areas with higher MSPs. Further, the welfare loss is in general larger for urban population than for rural population across classes and over time. The welfare loss is in general regressive in the sense that the lower classes seem to lose more than the upper classes. Only the top 20 percent of the rural population experience welfare increases in the third year. It should be emphasized that rural population get part of their income from non-agricultural activities. Thus the decline in non-agricultural output affects adversely the income of rural population also, which to some extent neutralizes the gains from increase in agricultural prices. The top quantile of rural population has a larger marketable surplus of wheat and rice and they also get a relatively larger share of non-agricultural income.

17. While the immediate welfare loss due to a general price hike (procurement, ration and market price) is obvious, the welfare loss over time can be explained in terms of the decline in overall GDP attributable to a decline in non-agricultural GDP and mild growth in agriculture output as the decline in fixed investments almost completely offsets the incentive effects of higher MSPs.

5. Conclusions

18. We have seen that a 10 percent increase in minimum support prices of wheat and rice leads to decline in overall GDP by 0.33 percent, increase in aggregate price index by 1.5 percent, reduction in investments by 1.9 percent and a miniscule impact on Agricultural GDP by the third year. What is more important to note is that in terms of welfare the bottom 80 percent of the rural and all of urban population is worse off.

19. The results are easy to understand. Higher minimum support prices reduce demand and lead to larger stocks. Thus the stocks of wheat and rice are larger by 16.5 million tonnes after 3 years and reach 66.5 million tonnes compared to a reference scenario. We have assumed that a government under hard budget constraints can finance the expenditure on additional stocks only by reducing its investment expenditure, including investment in irrigation. We consider this to be a realistic assumption. This leads to lower agricultural output overtime and even the farmers get hurt.

20. Thus the minimum support prices at levels much above market clearing prices are not desirable. One should emphasize here that the adverse impact of increases in MSP is related to the level of MSP to begin with. Only when the increases push up prices above the market clearing prices that the problems arise.  Thus we conclude that government should announce minimum support prices that remain below the market clearing prices in a normal or modal year. In an exceptionally bumper year the market clearing prices would be lower than the announced MSPs. This, however, is fine as in such years, the farmers need support which should be provided.