Lots of new materials on cash transfers. Osorio et al have a new blog about how altering the benefit structure in CCTs can lead to significant long-term effects (see also full NBER paper here). By combining administrative records and data from a multi-arm experiment in Bogota, the authors show that forcing households to save one-third of the transfer until just before the beginning of the next academic year increases long-term human capital accumulation (i.e., graduation and tertiary enrollment). This is particularly significant since the traditional CCT payment structure didn’t increase tertiary education enrolment. A working paper on social pensions by Avila-Parra and Escamilla-Guerrero shows that, while Mexico changed the eligibility age from 70 to 65, the program both reduced the probability of the elderly being extreme poor and did not have short-term effects on the labor force participation of the elderly. In a brief K4D paper, Idris sets out issues and experiences with cash transfers in fostering social cohesion (hint: evidence is limited). Finally, a new volume on social protection in Africa was recently launched and edited by Awortwi and Aiyede. The book sets out perspectives on social protection from African scholars teaching in African universities. They take a generally positive view on the explosion of cash transfers in continent, although they portray the trend as one featuring excited donors and somewhat unenthusiastic governments. (Another book on non-state actors in providing social protection will be released next year).
Christiaensen and Demery have just published an awesome, myth-busting edited volume on agriculture in Africa. Each of the chapters begins with a ‘conventional wisdom’ statement, which is then dispelled with a series of hard facts. A total of 16 conventional wisdoms are examined, relating to four themes: the extent of farmer’s engagement in input factor and product markets; the role of off-farm activities; the technology and farming systems used; and the risk environment farmers face. Of particular interest is chapter 14 where Nikoloski, Christiaensen, and Hill point to the often-underemphasized price risks (which are as prevalent as droughts) and the widespread informal coping mechanisms (when formal social assistance is present, it is just as likely to target households in the top 60% as in the bottom 40%).
Some mobility-related papers. Does migration overseas reduce poverty at home? In Nepal, a paper by Shrestha increases in migration to Gulf-Malaysia explain 40% of the decline in poverty between 2001 and 2011. In particular, a $1 increase in remittance income increases consumption by $0.5, with the largest share going to expenditures on food. Dani Rodrik has a new paper on rebalancing globalization where he argues that economic integration has gone too far in financial globalization and regulatory harmonization, and it has not gone far enough in international labor mobility.
October 31st was the World Cities Day, so let me indulge with urban materials. Hanna and McIntyre have a piece on congestion in Indonesia – quite key also for social protection given the role that commuting plays in shaping urban livelihoods and how to design programs that need to account for high opportunity costs. Another article by Siddique in the Dhaka Tribune discusses that overlooked nature of urban poverty and the performance of the Old Age Allowance in cities (h/t Aneeka Rahman). The FAO has just released its flagship the State of Food and Agriculture report: tons of good stuff, including unbundling the concept of ‘urban spectrum’ (p.17) linking cities, towns and rural areas as part of structural transformation strategies (see nice visuals on p.39) – with value chains connecting the dots, e.g., aquaculture in Bangladesh being a stunning example (p.28).
Ravallion illustrates that poverty is declining and inequality is rising over 1981-2011 with the use of poverty incidence curves and growth incidence curves. More on poverty, this time among children: An Urban Institute brief by Isaacs et al asks which federal spending and tax programs in the US provide the most support for children. In 2016, the federal government spent $486 billion (12% of the $3.9 trillion federal budget) on children through programs and refundable tax credits. With $89 billion, Medicaid is the largest program benefitting children, followed by EITC ($61 billion). SNAP is 5th in the ranking with $51 billion.
In a HE article, Evans and Tarneberg investigate whether patients in Nigeria can evaluate health service quality effectively. Specifically, the paper demonstrates that although more than 90% of patients agree with any positive statement about the quality of their local health services, satisfaction is significantly associated with the diagnostic ability of health workers at the facility.
From people’s health to firms’ health. McKenzie and Paffhausen have a great new paper exploring the magnitude and causes of death among 14,000 small firms in 12 developing countries (with the analysis allowing to estimate the rate of firm death between 3 months and 17 years). So how many firms die? About 8.3% over the first 5 years. Why? Among many reasons, small firms run by women shut down because of illness and family reasons, suggesting non-separability between households and firms. See also the paper’s DI blog here.
Let’s conclude with an out-of-the-box piece. Recently, I was struck by accounts of unspeakable, traumatic experiences shared by displaced populations fleeing violence. Hence my attention was caught by a new machine-learning technique by Just et al which could help identify people suffering from suicidal thoughts. Researchers looked at 34 young adults, evenly split between suicidal participants and a control group. Each subject went through a functional magnetic resonance imaging and were presented with three lists of 10 words. All the words were related to suicide (words like “death,” “distressed,” or “fatal”), positive effects (“carefree,” “kindness,” “innocence”), or negative effects (“boredom,” “evil,” “guilty”). The researchers also used previously mapped neural signatures that show the brain patterns of emotions like “shame” and “anger.” Five brain locations, along with six of the words, were found to be the best markers to distinguish the suicidal patients from the controls. Using just those locations and words, the researchers trained a machine-learning classifier that was able to correctly identify 15 of the 17 suicidal patients and 16 of 17 control subjects. Can a variants of such technique be applied in tandem with social assistance among refugees and IDPs that experienced traumas?