The much awaited social protection delivery systems sourcebook is out! Penned by Lindert et al, the 500-page flagship sheds light on the complexities of delivering social protection and lays out a framework for thinking about the delivery chain in stages – i.e., a set of 9 sequenced, logical activities across 4 core functions. Drawing extensively from real-world processes, case studies and practical illustrations (103 countries are featured to different degrees), it also provides quantitative insights on select issues – for example, by comparing the costs of alternative registration methods in Pakistan’s NSER (i.e., door-to-door approach ($2.3/person) versus temporary-desk solution ($2.7/person), p.388), offering tips for estimating administrators’ staff time (p.392), and quantifying transaction costs faced by beneficiaries for accessing benefits (p.53 on human centered design).

Let’s zoom into a specific (and central) delivery issue: a new ID4D toolkit provides guidance for qualitative research with end users of identification systems. Really juicy and practical stuff here… for example, what’s the ideal number of field teams in a study (p.37)? How to ask questions to explore marginalization (p.23)? How to store and maintain confidentiality of data collected (p.67)? (see also webinar slides).

And now into the evidence! Can the combined provision of cash transfers and assets work in war-torn Yemen? Brune et al find there were “… positive impacts on asset accumulation and savings behavior, albeit substantially less than the amount the household originally received”.

Moving to East Asia, what’s the effect of small-value, “use-it-this-week-or-lose-it” digital coupons in China? Xing et al find an immediate increase in weekly consumption by $3 out-of-pocket spending for every $1 in subsidy (spending mostly occurs in larger commercial outlets). Bonus on China: an ILO brief by Walker and Lichao examines the country’s performance on relative poverty.

… and onto South Asia: what’s the effect of social pensions and labor in India? Unnikrishnan and Sen show that having a person eligible for the Indira Gandhi National Old-Age Pension Scheme in the household increases the probability of working by 3.2 percentage points for women aged 20–50, with the effect stronger for urban women. Why? Because the income effect of the scheme leads to reduced labor supply by the elderly, allowing them to provide greater childcare support. The increase in women’s labor force participation is mostly in flexible employment, e.g. self-employment and agricultural and non-agricultural wage employment, and is particularly evident for poorer households.

More on the region, where Covid is amplifying pre-existing challenges such as urban poverty: for instance, a survey led by the BMC, Niti Aayog, and TIFR estimates that half of Mumbai’s slums dwellers are exposed to C19. In Mongolia, results from the first round of the World Bank’s Covid19 Household Response Phone Survey shows that nearly 3/4 of households, and 85% percent of people living in poverty (as of 2018 HSES) experienced some sort of shocks since end-January. But there is also lots of creative action: have you ever seen an ATM dispensing… food? Covid sparked the introduction of rice ATMs in Vietnam and mobile food storage units in India.

What’s new in Africa? An in-depth review of 53 studies (with 3 exceptions, they all cover African countries) by Bakrania et al takes stock of lessons on pandemics/epidemics on child protection outcomes (the ‘evidence map’ on p.25-26 is a gem!); Lawson et al’s book “What Works for Africa’s Poorest Children: From Measurement to Action” is open access! Save the Children produced an interesting social protection sector review in Somaliland; a media piece examined South Africa’s possible grant for vulnerable populations currently not covered by other programs: how much would it cost? About 3.5% of GDP. And when would it be operational? Within 2 years (h/t Ian Orton); and Bowles et al evaluate how WhatsApp messages have substantively large effects on Covid behaviors in Zimbabwe. BTW, Madden and Kanos review the latest data on digital skills and the future of work in Africa… some indicators are better than you might think!

What’s up in high-income countries? One of the best pieces I have seen on Covid response plans, with Furman et al setting out a proposal to increase federal Unemployment Insurance (UI) benefits in the US, link its phasedown over time based on economic circumstances, and adopt automatic triggers for UI extended benefits. A paper by Cassidy on homelessness in NYC finds that families placed in shelters in their neighborhoods of origin remain there longer than those assigned to distant shelters, access more public benefits, and are more apt to work. And in Europe, a paper on fiscal austerity in Italy and Spain by Toubeau and Vampa finds that, over 2003-2015, healthcare spending was better ringfenced than social assistance.

Whether it’s a high or low-income country, hard choices writ large…. Hausmann and Schetter ponder the “horrible trade-offs” in a pandemic, namely the interplay between lockdowns, compliance, transfers, and fiscal space. Through general equilibrium analysis they identify an “optimal lockdown” level, one that is stricter for more severe pandemics and in richer countries… but limited fiscal space lowers the optimal lockdown and welfare.

Three great blogs on gender! Ghils discusses gender and social protection in Covid times; Beegle presents estimates on how the lockdown in India impacted violence against women (spoiler: it increased domestic violence and cybercrime complaints, while it reduced sexual harassment and rape/sexual assault); and Erten and Keskin argue that refugee inflows from Syria to Turkey led to the displacement of women in the labor force, but also reduced the risk of intimate partner violence. Bonus on forced displacement: a controversial IZA paper by Green and Iversen suggests that in Norway, an increase in the share of refugee students in classrooms by 5 percentage points worsens math test scores of Norwegian students by an equivalent 3% of an expected year’s progress.

Since I covered displacement… there is an explosion of materials at intersection of humanitarian assistance and social protection. Last week I shared the new CaLP cash report and Development Initiative’s GHA flagship, so here is more: Longhurst’s et al’s high-level paper on humanitarian-social protection linkages; Metcalfe-Hough et al’s Grand Bargain annual independent report 2020; Cabot-Venton et al’s piece on embedding localization in Covid19 response; a case study on Venezuela refugee/migrant crisis by USAID and CaLP; a ‘crib sheet’ on the MENA region by Smith, and her case studies on IraqLebanon and YemenOxfam’s partnership toolkit; and Altangerel’s blog on Covid lessons and opportunities. Bonus: Maxwell et al lay out their views on determining famines via multi-dimensional analysis.

From humanitarian action to societal perception: an intersting study by Caviola et al estimated perceptions of charities’ effectiveness among different audiences: the “general public” considered the most effective charities for global poverty alleviation to be 1.5 times more effective than the average charity. “Effective altruists”, in contrast, estimated the difference to be a factor of 30, while “experts” assessed such factor to be 100. The authors then found that participants donated more to the most effective charity, and less to an average charity, when informed about the large difference in cost-effectiveness.

Let’s look at agriculture, with an intersting twist to the energy subsidy literature: in India, Ryan and Sudarshan document the common practice of states rationing power by switching off the electricity grid for farmers during most of the day. Effects? It limits farmers’ water use to a nearly efficient level on average, but also lowers agricultural productivity. Speaking of farmers, in Zambia Pelletier et al estimate that the modern seed inputs for smallholder farmers reduces deforestation (without those inputs, predicted forest cover loss would be double). And in Nigeria, Takeshima et al show that public expenditures in agriculture improve consumption, poverty reduction, nonfarm capital investments, and household dietary diversity.

Final selection: Baumann and Wohlrabe ask what happens to working papers: 67% are published in journals, 8% in books, and 26% are… seemingly lost! (h/t Shwetlena Sabarwal); a podcast with Oxford’s Hugo Slim covers human rights and duties, humanitarian diplomacy, moral dilemmas, localizing aid, and ethical leadership; and FaiVLive hosted a great webinar on of the future of pro-poor digital financial services.

Enjoy the Summer safely, links are back in a few weeks!

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 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.

To sign up to the newsletter or share materials, you can contact Ugo by email ([email protected]), Twitter (@ugentilini) or LinkedIn.

Social Protection Programmes: 
  • Social assistance
    • Social transfers
      • Cash transfers
  • Social insurance
    • Old-age pension
  • Labour market / employment programmes
    • Passive labour market policies
      • Unemployment benefits
        • Unemployment insurance (contributory)
Social Protection Building Blocks: 
  • Programme implementation
    • Benefits payment / delivery
  • Programme design
Cross-Cutting Areas: 
  • Agriculture and rural development
  • Gender
  • Labour market / employment
    • Unemployment
  • Poverty reduction
  • Migration
    • Remittances
  • Humanitarian assistance
  • Nigeria
  • Zambia
  • Zimbabwe
  • United States
  • Venezuela
  • China
  • India
  • Iraq
  • Lebanon
  • Mongolia
  • Pakistan
  • Syria
  • Turkey
  • Vietnam
  • Yemen
  • Norway
  • Spain
  • Global
The views presented here are the author's and not's