The right to choose your food
Over the past six months we have been working with the government of Bihar on the details of a cash transfer pilot and evaluation. One component of this work was an April 2011 survey of 500 BPL and Antyodaya card-holding households in Patna district — 300 rural and 200 urban. Our survey aimed to understand households' experiences with the PDS and the extent of their interest in cash transfers as an alternative.
The results confirm a number of commonly cited problems with the PDS: fair price shops (FPS) are open infrequently (10 days/month in urban areas and 6 days/month in rural areas on average), have long lines even when they are open (52 per cent of rural respondents waited more than 45 minutes), and frequently adulterate food grains (80 per cent of respondents) and/or under-weigh them (75 per cent of respondents). In light of these issues, it is not surprising that 94 per cent of respondents in our survey said that they would like to participate in a pilot program that provides a cash equivalent subsidy in lieu of their current PDS entitlements.
This large number might seem a compelling argument for switching to a system of cash transfers. There are several reasons to exercise caution, however. First, respondents may have unrealistic expectations about how well cash transfers could be implemented. Second, in a recent survey by Jean Dreze and his colleagues, only 54 per cent of (approximately 136) households in Bihar expressed a preference for cash. This could reflect sampling and site variation, but also suggests that responses to hypothetical questions can be quite sensitive to the manner in which they are posed. Third, even if a majority of households prefer cash, we should also be concerned about protecting the interests of the minority who do not, especially if they are among the most vulnerable. Finally, households may prefer cash ex ante but end up using it irresponsibly ex post.
How then should we proceed?
The approach we propose is both simple and innovative: we recommend conducting pilots that give households a choice between cash and in-kind transfers. Several states are planning to modernise the PDS using biometric authentication technology such as the one developed under Bihar's e-Shakti initiative. While this technology may reduce outright leakage, it does not allow flexibility in consumption, or address issues of adulteration and under-weighing of goods. To test the added value of cash, a selection of households (chosen by lottery) can be offered an option to receive cash transfers in lieu of their rations. The technological back-end can track whether the form of entitlement opted for is cash or kind, and disburse the corresponding amount of cash (through business correspondents) or grains (through the FPS). The system can also accommodate beneficiaries changing their choices at quarterly intervals.
Introducing choice will help clarify the issues raised above. Households will have the opportunity to learn from experience as to which option suits them best — for example, to learn whether or not cash transfers are implemented well and adequately indexed to variation in food prices. Households that prefer the status quo will be protected. Comparing nutrition and health outcomes across households offered the choice and those in the status quo will allow us to rigorously quantify the extent to which cash transfers are used responsibly and ensure food security.
Maintaining both the PDS and a cash transfer system in parallel over the longer term may involve higher overhead costs, but would allow different households to choose the benefit that best matches their needs. But, regardless of whether the long-term structure of food security involves PDS, cash, or a choice, our key argument is that juxtaposing the two approaches on a pilot basis and letting beneficiaries choose between them can generate critical insights on a key policy question that ought to be decided with empirical evidence, not just theoretical arguments.
This approach mirrors a key theme of the draft of the food security bill recently approved by the eGoM, which recognises the diversity of situations across states and avoids a "one size fits all" straitjacket. While it states unambiguously that the PDS remains the default, it leaves open the option for states to consider alternative approaches. Similarly, even within states, it is possible to preserve the PDS as the default but leave open the option for households to try an alternative. Aggregating individual preferences through actual choices can also serve as a powerful feedback mechanism to policymakers about beneficiary preferences. Finally, providing such a choice puts beneficiary voices at the heart of anti-poverty policy — a powerful precedent for the world's largest democracy to set. As Dreze says, "if people are best judges of their own interest, why not ask them whether they prefer food or cash?"
Muralidharan is at UC San Diego and J-PAL. Niehaus is at UC San Diego, and Sukhtankar is at Dartmouth. They have been working with the Bihar government on a cash transfer pilot scheme