Social Assistance in Developing Countries Database Version 5.0

In this new version of the database we have included pilot social assistance programmes. A number of pilot cash transfer programmes have been introduced in Latin America, Asia and Africa in the last year or so, and a few more are in the design stage. Their scale and rationale suggest there is a good chance they will be scaled up in the near future. In theory, pilot social protection programmes should imply experimentation in the face of uncertainty regarding the way forward, but several of the pilots covered in the database, and many of those in the pipeline, represent instead a specific route to the extension of social protection, and as such they merit discussion. The main purpose of this brief note is to provide such discussion, and illuminate on this specific mode of development of social protection in developing countries. In Sub-Saharan Africa, there are pilot cash transfers schemes in place in Kenya, Malawi, Ghana and Zambia; and in the implementation stage in Nigeria, Liberia, Uganda, and Tanzania. In Latin America, pilot programmes have been rolled out in Paraguay, Honduras, Nicaragua, Panama, Argentina, and the Dominican Republic. In South Asia, ’s Challenging the Frontiers of Poverty Reduction - Targeting the Ultra Poor programme is in fact a pilot programme, as as is Pakistan’s Child Support programme. Why the high number of pilots? In the context of technocratic models of policy making, pilot programmes would make a great deal of sense if policy makers are uncertain of the feasibility and likely impact effectiveness of interventions. Before introducing innovative, complex, and costly interventions, sensible policy makers would recommend testing the interventions in a small scale experiment. Knowledge from the delivery and impact of the interventions could then inform the desirability and design of a scaled up programme. There is a sense in which the social protection pilot programmes referred to above, and described in the database, do not fit fully into this description. We have accumulated a large body of evidence and knowledge about the design, delivery, and impact of cash transfer schemes in Latin America to be reasonably confident that, adequately designed, they can achieve their short term objectives. Why is further testing necessary? The strongest available evidence on cash transfer programmes comes from middle income countries in Latin America, Mexico’s Progresa/Oportunidades, and to a lesser extent Brazil’s Bolsa Escola/Familia. Naturally, questions remain over whether similar programmes can work in other environments. Would cash transfer schemes work in Africa? Would they work in low income countries in Latin America? Low income countries have higher incidence of poverty; lower capacity in terms of designing, delivering, and evaluating transfers schemes; and less developed administrative and financial systems. It makes sense to check whether cash transfers are appropriate and effective in these, more adverse, environments. Even then, fewer pilots would still deliver answers to our questions. We know from the Zambia Kalomo Social Transfer Pilot Scheme that cash transfers are feasible and effective in low income countries, providing that technical support is available and community selection of beneficiaries is feasible. The spread of pilot social assistance schemes is also explained by domestic policy processes and funding modalities. In countries where policy makers, and perhaps civil society, are reluctant to innovate, pilots provide an opportunity to enable learning from new approaches to poverty and vulnerability. It also provides a well defined time frame in which donors could use existing funding modalities to support the extension of social protection. DFID, for example, is committed to shifting focus from emergency aid to regular forms of support in Africa. In Latin America, IADB support for social protection initiatives normally extends for periods of up to five years. Given the time frame of available international aid , the expectations are that pilot schemes could be instrumental in building learning and support for social protection among domestic policy makers, that they would have strong ‘demonstration effects’. Risks and opportunities There are significant risks with this strategy, and even more significant opportunities. The risks are to do with pilots failing to generate the expected ‘demonstration effects’, and with changes in international economic conditions that shift attention to other problems. The opportunities could potentially be very significant, successful pilot transfer schemes could mark the beginnings of a process leading to the implementation of effective anti-poverty programmes at a scale capable of making a large dent on global poverty. Paying attention to the design of pilots and to associated policy processes could help minimise these risks and maximise opportunities. Designing pilot social assistance programmes as if they are a first phase of a fully scaled up programme is essential. This involves avoiding short cuts in the pilot stage, and making the necessary investment in information systems, delivery institutions, and beneficiary selection. These set up costs can be substantial. Process considerations are important in ensuring the pilots are part of national social protection strategies, and involve a wide range of stakeholders. It is vitally important that pilots achieve a good balance of design and process considerations. As much else in development policy, pilot social transfers are as much about politics as they are about the economic and technical issues of poverty reduction.