Technology-Enabled Agent Choice and Uptake of Social Assistance Programs: Evidence from India’s Food Security Program

Beneficiaries of social assistance programs that distribute undifferentiated commodities are typically assigned to a designated agent for collecting their entitlements. This arrangement grants monopoly power to agents, who often face little incentive to follow operational guidelines – especially in the absence of strong monitoring – leading to reduced uptake by beneficiaries. In response, some governments have introduced agent choice, allowing beneficiaries to select from multiple agents. However, the effectiveness of this policy may be constrained by the limited availability of alternative agents, the lack of meaningful competition in undifferentiated commodities, and the risk of collusion among agents.

The paper examines the impact of technology-enabled agent choice on entitlement uptake in India’s large-scale public food distribution program. The program, which accounts for nearly 1% of India’s GDP, provides subsidized food grains to around 160 million households through licensed agents operating Fair Price Shops (FPSs). Despite the significant budget allocation, beneficiaries often struggle to access their entitlements due to poor service, agent monopoly and weak monitoring.

The study focuses on the Indian states of Andhra Pradesh and Telangana, which implemented the agent choice policy at different times. The researchers employ a difference-in-differences framework, comparing areas that implemented the choice policy to those that did not, before and after its rollout. They also leverage variation in agent density to analyze how the intensity of competition affects outcomes.

The findings are striking: introducing agent choice led to a 6.6% increase in the quantity of entitlements collected by beneficiary households. The effect was significantly larger in areas with higher FPS density, highlighting that the presence of alternative agents is critical for the policy to be effective. More interestingly, nearly all of the increased uptake came from new beneficiaries collecting entitlements from their preassigned agent. This indicates that the mere presence of choice acts as a credible threat – agents improved adherence to operational norms, including better shop opening times, to avoid losing beneficiaries to neighbouring agents. The results indicate that even limited forms of choice – where location is flexible, but product and price are fixed – can meaningfully improve outcomes.

The study contributes to the growing literature on service delivery in public welfare programs, showing how simple policy design changes, coupled with digital infrastructure, can greatly enhance program effectiveness. The authors highlight the broader applicability of the model to other large-scale welfare programs, particularly in low- and middle-income countries, where agent monopolies and weak accountability mechanisms are common. The paper has important implications for public policy design, particularly in using technology to foster choice, reduce inefficiencies and empower beneficiaries in accessing critical social services.