Sweet incentives and unintended consequences

India is the World’s second largest producer of sugarcane (after Brazil). Approximately 45 million Indian farmers are involved in cane production, and cane processing is the second largest agro-processing sector in the Country. In her dissertation research on this industry (supervised by Goodhue and Sexton) Sandhya Patlolla encountered an unusual marketing practice that is unique to the Andhra Pradesh (AP) State. Private sugar processors issue ‘permits’ to selected cane growers a few weeks before harvest. These permits allow growers to deliver a specified amount of cane during a specified period of time. Farmers without a permit must sell their cane at a reduced price for manufacture into gur, a traditional Indian sweetener. The price difference averaged 43% over the six years of data acquired for our study. We wondered why sugar processors in AP would create uncertainty among farmers by using ex post permits instead of offering ex ante production contracts?

Our research, published recently in the World Bank Economic Review, found the answer in the complex maze of government regulations that impact the Indian cane sector. The Indian government imposes a factory-specific price floor for cane used to produce white sugar, but no floor exists for cane used to manufacture gur or khandsari, another traditional sweetener. The government also limits competition among sugar processors, which are either private firms, public firms, or grower cooperatives, by assigning a reserve area to each processor within which the processor has a right of first refusal, i.e., a grower must sell cane to the processor at any price meeting or exceeding the floor.

Cane production in AP normally exceeds the combined capacity of the factories, and on average annually about a fourth of AP cane farmers do not obtain a permit and must sell in the discounted gur market. Permits are issued (or not) by fieldmen, who visit the fields prior to harvest and measure the brix content of the cane. Yield of sugar is increasing in the brix level of the cane.

Under this arrangement cane producers in AP must make decisions about planting and allocation of inputs without knowing whether they will obtain a permit and, hence, the price they will receive. Almost surely these small farmers are risk averse, so the uncertainty created by processors’ procurement policies is a departure from the predictions of a traditional principal-agent model, wherein the presumably less risk averse processor should remove (rather than create) risk for farmers.

Our theory is that this marketing mechanism works to create competition among farmers to increase their cane quality as a way to improve their odds of obtaining a permit. Although the Indian price floor system nominally specifies price premiums for higher cane yields, in practice such premiums aren’t paid. Thus, by incentivizing farmers to compete for permits through improving quality, processors reduce their input acquisition costs per unit of refined sugar and increase profits.

The obvious alternative strategy of issuing ex ante contracts in an amount equivalent to processing capacity doesn’t work due to moral hazard concerns. Farmers who had a contract would have no incentive to improve quality above the minimum required to receive the floor price. Processor actions to enforce quality-enhancing production practices would be prohibitively expensive, given the large number of small cane farmers that a typical processor needs to engage to meet its capacity requirements.

We found empirical support for this hypothesis using cane farmers in AP who are members of cooperatives as a control group. No private processors in AP use ex ante contracts, but the arrangement between a co-op and its members functions much the same as an ex ante contract, making these farmers an effective control group. We develop detailed arguments in the WBER paper that there is no selection problem in comparing farmers selling to private factories vs. selling to cooperatives. We found that farmers selling to private factories had statistically significantly higher cultivation costs than farmers selling to cooperatives, with most of those extra costs being for labor, which comports to the notion that cultivation practices associated with higher quality are very labor intensive. In further support of the hypothesis, we also found that private processors obtained higher quality cane on average than cooperative processors, with the difference being statistically significant.

The permit system is not efficient, however, as the cost savings to processors from higher sucrose yield of cane was less than the additional cultivation costs incurred by farmers in producing it. In short, although the mix of government regulations regarding the sugarcane market is intended to protect the interests of both processors and farmers, the market conditions in AP enable processors to capture additional rents through the ex post permit system at the farmers’ expense. This system thus illustrates a classic unintended consequence of government regulation.

Published in collaboration with The World Bank

Author: Richard J. Sexton is a professor and chair in the Department of Agricultural and Resource Economics, University of California, Davis.

Image: Farmers walk through a paddy field towards a sugarcane field at Moynaguri village. REUTERS/Rupak De Chowdhuri