How much time does it take to scale-up social protection in crises? About 3.5 months on average according to a new AusAid paper by Barca and Beazley. Examining the performance of 9 countries affected by natural disasters, the analysis shows that response time ranges from 2 weeks to 14 months (see table 4 p.41) and discusses factors affecting performance. I really liked figure 3 (p.22) quantifying the different models and overlaps in response, as well as the final handy summary framework (p.48) compressing key actions in assessments, decision-making, preparedness and adaptation. Bonus: a new case study by Aldaba delves into the linkages between humanitarian assistance, social protection and disaster risk in the Philippines, while Lovell et al investigate the cross-sectoral approaches to resilience in Kenya and Nepal.
From disaster to age risks: there new evidence on social pensions in Vietnam. Those programs reach about 80% of eligible population (people above 80 years of age) with almost no leakage. An EDCC article by Nguyen shows positive economic-psychological impacts, and recommends expanding eligibility to those above 65 years of age. BTW, Eckardt and Ngoan reflect on how the country can avoid the middle-trap.
More on Asia: Bangladesh has 140M mobile subscribers… but do mothers in the primary education stipend program prefer digital or direct cash transfers? Because of paperwork, poor timeliness and high opportunity costs associated with manual disbursements, payments for the program switched to mobile phone payments in 2017. A CGD note by Gelb et al presents results from a survey of mothers in 100 households among 34 villages. Overall, 79% of respondents preferred the new system (less travel, more choice on when to draw money), with 95% finding it easy to register. However, a striking 45% wasn’t able to read and write SMSs.
A rich IPC research brief on Senegal by Ndiaye et al takes stock of the country’s National Social Registry (RNU). Incorporating data on about 461,000 households, the note outlines the evolution of the RNU, the cross-sectoral data sharing protocols, and processes behind the platform.
Cash and politics! McKenzie picked up the work (featured in The Economist) by Brollo et al on the politics of CCT enforcement in Brazil: “… Bolsa Familia households whose payments are suspended for not meeting the conditions of the transfer (e.g. kids not attending school enough) punish politicians at the polls – and knowing this, politicians are lax in enforcing the rules in the run-up to elections”. And speaking of Brazil, Orair and Gobetti provide an update of tax proposals under discussion.
From politics to jobs: when designing youth training programs, the quality of trainers can make a huge difference. A working paper by Hardy et al assesses the effects of a government-sponsored apprenticeship training program in Ghana. Results? The intervention shifts youth out of wage employment and into self-employment. However, the loss of wage income is not offset by increases in self-employment profits in the short run – but such shifts becomes a net-gain among participants trained with the most experienced trainers.
A VoxDev piece by Artuc et al takes on the automation-jobs debate and point out that “while robots compete with workers in the early stages of adoption, they complement them in subsequent stages” – basically, labor demand goes down, and later bounces back. However, their model shows – but doesn’t comment on – the somewhat damning fact for lower-income countries (“South”) where labor demand doesn’t bounce back (figure 2b). Is the implication that, once automation is introduced, low/middle income countries will pay the double-cost of (a) labor reduction and (b) not enjoying later returns in labor productivity that we observe elsewhere?
What happens in a low-income country when there are minimum wage hikes? Shrestha examines such case for Cambodia’s 4 such hikes between 2008 and 2015. The analysis finds that, except for immediate adjustments around the time of the hikes, the minimum wage hikes did not affect labor market participation rates in the affected sector — garments and footwear — or the unaffected sectors. However, the bumps increased wages modestly (3%) for workers in the affected sectors and modestly decreased them (1.5%) for workers in the unaffected sectors.
A new Urban Institute paper by Karpman et helps shed light on the economic hardships that people endure even at times of low unemployment. In the US, 9.3% of working-age people reported problems paying the rent or mortgage in 2018, 23.1% were food insecure, and 17.8% skipped medical care because of its cost—estimates that were not statistically different from 2017 estimates. In discussing these stats, they place jobs as a necessary but insufficient condition for poverty reduction: low pay/quality jobs plus economic hurdles that are not correlated with jobs (e.g., no disability/health insurance, housing costs) matter greatly.
Since I mentioned health… a key thematic areas where, in my humble view, social protection needs more work is the linkages to healthcare. So an insightful analysis like the new one by Chalkidou et al is a must read – in this case, connecting the ongoing debate on universal health coverage in India to global trajectories.
Assorted final mix: one of the most longstanding, tough and forgotten crises? It has 3 letters – IDPs. Some 40M Internally Displaced People are like refugees in their own countries, often persecuted by their own states, and – as a new brief by Huang and Graham shows – increasingly concentrated in dysfunctional urban areas; an interview with Zelleke covers gender and UBI, while Kasy contrasts the program to EITC (h/t Molly Dektar).