Let’s begin with a rich synthesis! Beazley et al summarize the performance of select social protection Covid-19 responses in Bangladesh, Ethiopia, Kenya, Pakistan, Sierra Leone and Uganda. Trade-offs were in full display: Pakistan and Bangladesh showed the widest coverage (reaching 50 and 13% of the population, respectively), but they also provided the least generous transfers (in the case of Pakistan 1-5% of the annual consumption of bottom quarter of the income distribution). Those two countries were also the fastest in response, including reaching half of target beneficiaries in less than a month (in its first week, Pakistan enrolled 1M people/day for cash transfers). Sierra Leone, instead, reached only 5% of households, but transfers’ adequacy was 23% of bottom quarter’s consumption; yet it took about 8 months to the main program to make its first payment. Somewhat surprisingly, Ethiopia’s preexisting programs (rural and urban PSNP) displayed a similar performance, i.e., 6-8 months to release payments. Other features: in Uganda’s public works, 20% of beneficiaries were refugees and required fewer paid work hours from women than from men; in Kenya, registration took place in person, through community structures and paper-based data collection tools; disability assessments in Sierra Leone relied on a visual assessment, which may have excluded people with less obvious disabilities; in Kenya and Pakistan, government responses were complemented by time-bound cash interventions of non-state actors. And on the more sobering note, the review calls for “… limiting the expectations about the use of social registries in responses to shocks” (h/t Chris Porter). Bonus: an interesting podcast with Lowe and Hebbar unpacks social protection delivery lessons (based on the ODI paper on Nigeria, Peru, Sri Lanka and Togo shared a few weeks ago).
But can a Covid-19 response fill a preexisting, structural social protection gap? Köhler and Bhorat assess the labor market impact of South Africa’s Covid-19 cash grant for the unemployed. In place from May 2020 to April 2021, the program not only “… brought millions of previously unreached adults into the system”, but its receipt increased the probability of job search by more than 25 percentage points. The paper recommended that a grant’s extension “… could either be temporarily in place conditional on the recovery of the labour market, or permanently made part of the country’s social assistance system as a means of filling the hole in the safety net and providing regular support to the transient and chronically unemployed”.
Speaking of employment, a paper by Hussam et al tested the psychosocial value of cash vs jobs in Rohingya refugee camps of Bangladesh. This included an experiment providing either weekly cash transfers or work paid at the same weekly rate. Results show that employment confers significant psychosocial benefits beyond the impacts of cash alone – in fact, the cash arm does not improve mental wellness despite transfers being more than twice the amount held by beneficiaries at baseline. In fact, 66% of those in the work treatment arm were willing to work for free to keep the job (h/t Truman Packard).
And what about regional assessments? Silva et al produced a rich, fascinating volume on the path to better jobs and social protection in post-pandemic Latin America. The discussion on social assistance is quite candid (p.96 onwards), indicating challenges related, for instance, to “tight poverty targeting and static identification systems” and “meager benefits”. Among others, really enjoyed the discussion on family allowances as de facto unemployment insurance in Argentina (p.99), the responsiveness of cash transfer programs to the crisis in 10 countries (p.100), the nice layering of responses presented in box 4.1 (p.104), and the way in which the report ties assistance, insurance and labor considerations all together (e.g., see conclusions p.123-125).
Let’s move to another shock in the region – rainfalls: in Brazil, Fitz found that rains increase wages and these attract children away from school. Bolsa partially mitigates these child labor effects (less so among boys and older children), although children’s intensity of school attendance is reduced and hindered academic progress (h/t Amber Peterman).
Bonus on LAC with a new old classic! Angelucci et al revisit the effects of the Mexican PROSPERA program and found that food consumption and food security do not change around the transfer payday, including for recipients with impatient or time-inconsistent preferences and households with higher than median transfer dependence (they argue that “the program transfers’ relative illiquidity may act as a commitment device, helping time-inconsistent and less experienced debit card holders smooth consumption”).
News from Asia? Dreze and Somanchi discuss India’s food security situation and social protection response in 2020. Examining results from 12 surveys, they argue that between 71-92% of respondents with a ration card received some food grains from the PDS, although data on adequacy and leakages is limited. In terms of cash transfers, the same sources seem to indicate that about 1/3 of the 200+M women with a JDY account didn’t receive any benefit. Overall, they argue that “… relief measures helped, but they were patchy, and their effective reach is uncertain (…) A second, stronger wave of relief measures is essential”. BTW, see also a rebuttal from Vyas on the CPHS survey in particular. Bonus on food: international food prices have increased by 47.2% since April 2020, reaching their peak in May 2021 (the highest level since 2014). Want to know more? Here are four facts from Bogmans et al.
Since I mentioned inkind transfers, an intriguing piece on welfare and paternalism by Williams and Medlock picks up on Liscow-Pershing findings that “… preference for in-kind benefits fades rapidly as you move down the income ladder [because] general population preferences are strongly correlated with beliefs about how the poor spend their money”.
Time for labor and skills issues (h/t Michael Weber)! A set of WBG case studies on Cote d’Ivoire, Kenya, Mozambique, Nigeria, and Rwanda identified successful emerging training models for the provision of digital skills; Arellano-Bover and Saltiel study the large disparities across firms with on-the-job learning for the youth in Brazil and Italy; Chen and Pastore show that in China, vocational upper secondary graduates face a wage penalty compared to academic upper secondary graduates; and some words of wisdom from the evaluation of a job training program for low-income workers in New Orleans, where Baird et al argue that “… findings encourage patience on the part of trainees and the government, as workers may not find their stable, target employment immediately”.
A couple of intersting resources at the intersection of climate and financing: Strand compares tax rebating versus carbon crediting for incentivizing carbon taxation in low income countries; while Busch et al have a comprehensive review of ecological fiscal transfers, a rapidly growing conservation finance mechanism (h/t Anit Mukherjee).
Final fireworks! WorldVision has a note on linking humanitarian cash transfers to social protection through social accountability (see for example the case study on Bangladesh on p.3, where WV leveraged humanitarian cash to raise awareness about social protection entitlements among women and ethnic minority groups). Caron and Tiongson discuss the implication of Covid-19 for measures of household well-being (h/t Mohamed Almenfi). And an interesting IPA event is coming up on forcibly displaced people with vouchers and cash in DRC (June 30).
Ugo Gentilini is from the World Bank’s Social Protection & Jobs global practice. The Social Protection Links newsletter, issued every Friday, distills and discusses a selection of curated resources on the topic, from academic articles to podcasts. The blog is republished on socialprotection.org each week, offering knowledge on social protection to helps you stay on top of it — succinctly, regularly and frequently. Previous editions can be found here.