October 20, 2017 (links edition #30)

Some new cash transfers materials. In a new WD article, Gao and Mills estimate that rainfall shocks have severe effects on households’ consumption, and that the PSNP significantly mitigates such adverse impacts. Panelists from Mexico’s Prospera and Coneval discuss the current role of evaluations in CCT’s different institutional and operational components – see the event’s ppt and video. (Btw, the website that hosted the discussion, socialprotection.org, just issued its second annual report). More videos? CaLP released a new one on the effectiveness of cash transfers during the Ebola crisis. The IPC produced four handy one-pagers, including on behavioral change communication in Mozambique, social protection and agriculture synergies, CCTs and reform coalitions in Mexico and Brazil, and public works in South Africa.

 

Two stylized facts about poverty – having a job may not be enough, and a meal may cost you more than what you earn. In the US, Shambaugh et al. conducted an interesting profiling of the 40.6 million people, or 12.7% of the population, currently living in poverty. Over half of the poor were working-age adults, and about 40% of them were working or actively seeking work in 2016. One in ten working-age adults living in poverty is working full-time full-year. A new joint report by WFP and Mastercard shows that the poorest people pay more than they earn in a day for a single meal. They determined how much an average person in New York would have to pay for a plate of rice and beans if they spent the same proportion of their daily income as people in other countries. That $1.20 in New York, which is just 0.6% of the New Yorker’s average daily income, costs $8.27 in Guatemala, $27.77 in Nepal, and $72.65 in Haiti. In South Sudan, the worst-ranking country in the study, the meal costs $321.7, or 155% of a person’s daily wages. More on food assistance: Inspired by a trip to Liberia, Sandefur and Wadhwa have a new CGD piece – kind of midway between a blog and a paper – making the case for school feeding.

 

In a new ALNAP report, Obrecht reviews common donor challenges in using evidence to allocate humanitarian resources (i.e., unclear priorities, gaps in data and analysis, high costs, and evolving operating contexts). The paper suggests two ways to think about resource allocation, borrowing models from healthcare and the ‘charity evaluator’ approach. The former uses several measures to calculate and compare the cost-effectiveness of interventions, while in the latter charity evaluators’ concepts of expected value and marginal utility provide private donors with an approach to considering and comparing the impacts of different interventions.

 

A couple of materials on displacement. Colombia has the world’s highest number of IDPs (7 million): in such context, a very interesting JDE article by Morales estimates that the inflow of displaced populations leads to a short-run negative impact on wages, but that subsequent out-migration from receiving municipalities helps to mitigate these effects. A new paper by Calabria et al. examine refugees’ challenges around jobs and self-reliance more broadly. For example, they argue that many Afghan refugees residing in Delhi have a high level of education and professional skill-sets, but are restricted to working in the informal economy. This happens because most employers don’t recognize their Refugee Cards and Long Term Visas as valid documentation, and they struggle to open bank accounts (a prerequisite for formal jobs). I was also intrigued by the suite of available displacement tools, such as the Hall’s Multi-Dimensional Integration Index, the Joint IDP Profiling Service Indicator, RefugePoint’s Self-Reliance Measurement Tool, and the Women’s Refugee Commission Well-Being and Adjustment Index.

 

Let’s turn to economic mobility. Testaverde et al launched an impressive report on migration in ASEAN countries. Migrating to Opportunity presents a range of juicy findings – for example, while legislative steps to facilitate mobility among countries exist, these only cover select skilled professions (doctors, etc.) accounting for just 5% of jobs in the region. Reducing  barriers to mobility would, they estimate, improve workers’ welfare between 14 and 29% depending on skills. In a new brief, Vos et al look at migration (and displacement) from a food security perspective and discuss resilience, productivity, and mobility measures to both attenuate and facilitate migration.

 

More on labor: Packard and Montenegro, examine the relationship between businesses’ use of digital technology and statutory minimum wages, employment protection, and contributory insurance. Following and enriching the Alesina-Battisti-Zeira empirical strategy, they find a significant relationship between the use of digital technology by firms and labor market policies. Talking of labor, an independent evaluation of ILO’s social protection portfolio (2012-17) rated the organization’s work as ‘satisfactory’, with recommendations, mostly to be implemented by 2018, spanning seven areas.

 

A Brookings blog offers my personal reflections on the limits and value of universal basic income pilots (hint: there are no true UBI pilots, yet they help elicit relevant debates).

 

Three great new pieces on Africa’s structural transformation pathway. In a new PS commentary, Dani Rodrik dusts off his ‘premature deindustrialization’ argument and shows that in countries such as Ethiopia, Malawi, Senegal, and Tanzania, labor is moving from agriculture to services rather than manufacturing (see full paper here). In other words, we see structural transformation without industrialization, and he warns about the limits to how far this process can carry the economy. Similarly, a new paper by Gelb et al. uses data from 5,500 African firms in 29 countries and found that factories in Africa were almost always more expensive to start and run. Looking at overall costs, small African firms had a 39% premium over comparative firms elsewhere, while medium and large firms were around 50% more expensive. For example, the labor cost per Kenyan worker is $2,118 compared with Bangladesh where it’s $835. The capital cost per Kenyan worker is nearly $10,000, but it’s less than $1,100 in Bangladesh. Ethiopia, instead, is one of the few African countries whose labor costs ($909) are close to Bangladesh’s (relatedly, you may recall the Blattman-Dercon sweatshops article on Ethiopia shared in a past links edition). Finally, Stuart reflects on the implications from the African demographic change, including the 18 million high-productivity jobs needed each year through 2035 to absorb new labor market entrants, i.e., around six times the number that are currently created in the continent each year (full paper by Samman and Watkins is available here).

 

Let me conclude with an assorted mix of materials. Kar et al have a quantitative ESID paper revisiting the institutions-growth debate. They develop a measure called ‘episode magnitude’, which is the difference between the level of output at the end of the episode and a counterfactual. They apply such measure to 210 data points and show that higher institutional quality, irrespective of the measure of institutions, is a significant determinant of growth. In an NBER paper, Foster and Rosenzweig show India’s farms are too many and too small to exploit local equipment-capacity scale economies. An IDS briefing by Apgar et al shows that significant investments are being made in Climate Information Services (CIS) tailored to Kenya’s pastoralists, and recommend a more nuanced understanding of constraints for minority groups in accessing and using CIS. Riccio and colleagues at MDRC published a piece laying the foundations for future evaluations (in 2018-19) of tenant-based housing vouchers. These are subsidies provided to qualifying families to cover for their rent and utilities, and the evaluations will assess the impact of different voucher structures on work incentives. Looking forward to the final reports. Lucchetti has a technical paper proposing a non-parametric method to produce point estimates of intra-generational economic mobility in the absence of panel data sets that follow individuals over time. The method is successfully validated by producing welfare mobility estimates out of cross-sectional data and by comparing them with actual mobility from panel datasets on Chile, Nicaragua, and Peru.