Delivering public services effectively is essential for development. Yet governments often have good policies in place—promising to provide universal primary school, health care, and social assistance for the poor—but fall short on implementation, especially for the most vulnerable. Overcoming “last-mile” problems in public service delivery—such as absenteeism among front-line service providers and rent-seeking—is one of the most intractable development challenges.
Traditionally, governments have a monopoly over local service delivery, and improvements are difficult in part because central governments have a hard time motivating and monitoring their local employees. One solution is contracting these functions out to private service providers, who may be able to deliver services more effectively and at a lower cost.
However, this approach poses its own set of obstacles. Without sufficient competition, any efficiency gains may accrue to the contractor rather than to the public. Vested interests may block or subvert the selection process to protect rents in areas where reforms are needed the most. Private contractors may compromise on quality in ways the government can’t anticipate or easily measure in the contracting process. Private sector actors could end up being even more difficult to monitor and sanction than local government employees. In a developing country context, there may not be sufficient private sector capacity to take on service delivery.
A new study in the Journal of Political Economy by Abhijit Banerjee, Rema Hanna, Jordan Kyle, Benjamin Olken, and Sudarno Sumarto tackles these questions in the context of Indonesia’s Raskin program (“Rice for the Poor”). Raskin is designed to provide 15 kg of heavily-subsidized rice to 17.5 million of Indonesia’s poorest households, making it the country’s largest targeted transfer program. In practice, local rent-seeking is substantial: Large amounts of rice disappear and households pay 41 percent more than the official subsidized price.
Outsourcing for better public services
Working with the government of Indonesia, we designed a field experiment to test whether outsourcing public service delivery could deliver public benefits. In 191 villages (randomly selected from among 572 villages across 3 provinces), the government introduced a process for competitive bidding for the right to distribute Raskin locally.
Bids specified the price that households would pay for the rice and key aspects of the distribution process. Members of the Lembaga Pemberdayaan Masyarakat (Agency for Community Empowerment, LPM)—established small local committees comprised of informal community leaders and charged with overseeing local development projects—developed a set of criteria to help select the winner (before seeing any of the bids). For households, these typically included Raskin prices, pick-up locations, and payment methods; for distributors, assets, experience level, and overall character.
After hearing the bids, the committee members ranked them using the LPM’s pre-set criteria and picked a winner. This meant, for example, that the committee did not always choose the low-price bidder as the winner, as committees took bidders’ character and experience level into account as well. After six months of service, the committee could decide to re-contract with the winner, to implement a new bidding process, or to revert to the status quo.
One possibility is that the bidding process leads to improved service delivery through an information rather than an efficiency channel. Given that the government had always held a local monopoly over service delivery, making information available on its logistics (as the bidding process requires) could enable local accountability and facilitate service improvement on its own. To control for these bidding process information effects, we also randomly selected 96 villages for an information-only treatment, where the current Raskin distributor made a presentation on delivery logistics, and the LPM (and any interested community members) gathered to hear the presentation, then monitored Raskin delivery. This mimicked the first part of the bidding process.
Another possibility is that the bidding process improves outcomes—but only if there is sufficient competition. If there is only one bidder, for example, the private contractor can pocket any efficiency gains made during service delivery, without any value for the public. To probe the role of competition, in 96 of the 191 villages assigned to the bidding process, we instituted an “extended bidding” treatment, in which if at least 3 bids had not been received by the end of the tendering process, the process was extended by 10 days. This treatment increased the number of bidders by about 30 percent.
We conducted over 10,000 household surveys across six months to probe how outsourcing affected the benefits that citizens received from the Raskin program.
Overall, offering villages the opportunity to outsource public service delivery reduced the costs of delivering the rice (improving efficiency), with no detectable declines in service quality. However, service providers only lowered Raskin prices in villages where the extended bidding window had been implemented. Where there was outsourcing but no additional time to develop more competition, Raskin distributors reported lower distribution costs—more efficiency—but households still paid the same high price for Raskin as before. By contrast, in villages with the extra competition, price mark-ups for households fell by 11 percent. The information-only treatment did not affect program outcomes on its own; however, it enabled us to net out the effects of allowing bidding from the effects of increased transparency.
Addressing policy concerns
One serious concern about introducing an outsourcing option was that vested interests would block outsourcing efforts, and that communities would be ill-equipped to monitor and sanction bad actors.
To address this, we tested whether areas where baseline levels of rents were highest were more likely to see elites coordinating to block or thwart the bidding process. While there is some evidence that elites acted to prevent winners from assuming their distribution duties, areas where rents were highest also attracted the most bidders, increasing competition. The positive effects of additional entrants far overwhelmed any negative effects from blocking.
Governments around the world face challenges in ensuring that social assistance programs reach their intended beneficiaries. Providing an option to outsource public service delivery can yield benefits for citizens—but only where there is a sufficient supply of potential local service providers, where governments maximize competition, and where rents in the system are particularly high.
Our work was supported by the Australian Government’s Poverty Reduction Support Facility and the collaboration of the Abdul Latif Jameel Poverty Action Lab (J-PAL) and the National Team for the Acceleration of Poverty Reduction (TNP2K). The analysis was made possible thanks to many enumerators and survey coordinators at SurveyMeter and many excellent analysts at J-PAL.
Jordan Kyle is a Non-Resident Fellow in IFPRI’s Development Strategy and Governance Division and a Senior Research Fellow at the Tony Blair Institute for Global Change. She was an Associate Research Fellow at IFPRI during part of the time that she was working on this project.
