Asset Transfers or Cash Transfers: The Design of Anti-Poverty Transfers to the Ultra-Poor
Principal Investigator: Professor Imran Rasul, University College London
More than 40 million people live below the national poverty line in Pakistan. How can poverty alleviation programmes be designed to target and permanently raise such households out of poverty? Recent evidence suggests that the “Targeting the Ultrapoor” (TUP) program pioneered by BRAC in Bangladesh has proven very effective and portable in achieving such aims across a diverse set of low-income settings [Banerjee et al. 2015, Bandiera et al. 2016]. As a consequence, interest in the program has been growing and up to 30 governments had been piloting variants of TUP by the end of 2015 (CGAP 2015).
The TUP program consists of an in-kind social protection programme: it provides a one-off large-scale asset transfer (typically livestock) combined with complementary training. The key open question is whether beneficiaries could do better with an equivalent-valued cash transfer. Namely, whether a social protection programme designed around cash transfers could lead to better outcomes than transfers in-kind. This is the research question at the heart of this project, and helps to inform the design of social protection programmes across the developing world.
Standard economic theory suggests under a certain set of conditions, households can do no worse with cash transfers than in-kind transfers. We aim to test whether this is so, and if there is a divergence in the returns to transfers in-kind versus cash transfers, what is the origin of this divergence? Does it relate to market imperfections, or does it relate to particular behavioural biases or psychological traits that might lead households to treat cash and in-kind transfers differently?
Our evaluation answers these questions using a large-scale randomized control trial to compare the classic TUP design to a modified design where beneficiaries have the choice of cash instead of an asset-skills bundle. The Asset Transfer Program is a regional poverty alleviation development intervention, being implemented in four of the poor districts of Southern Punjab namely; Bahawalpur, Muzaffargarh, Lodhran and Bahawalnagar.
Based on our preliminary findings two years post intervention, we have observed that transferred assets have been largely retained and household’s labor market activities now begin to focus around those assets. Households that take up cash transfers also use these transfers to invest in similar types of asset. Most importantly, there has been a significant occupational shift out of wage employment and into self-employment for both in-kind and cash transfer treatment arms. In the short run, this is an encouraging result as occupational shifts is an early indicator of successful implementation of the program and its ability to shift households out of poverty. Following this, household earnings from self-employment, particularly from livestock related businesses have grown, as they divert away from low-paid labor.
At the aggregate household level, we also observe an increase in earnings for the two treatment arms. Given the rapid rise of earnings as observed in subsequent survey waves, we expect earnings from self-employment to improve further, and also potentially show some statistical difference in increase between the treatment arms.
We hope that with the help of data from subsequent follow up surveys, this evidence can play an important role in wider policy discussions on how anti-poverty can be most efficiently designed for both local and international stakeholders.