Why do richer countries spend more on social protection as % of GDP than lower income states (see left graph)? Is it a “luxury” that gets demanded when incomes rise? Or is it that perhaps social protection in low-income settings is only one among many priorities? Or maybe it’s about different societal preferences? And what about contextual factors inhibiting delivery? Answering these questions is easier said than done, but here is the good news: a thoughtful paper by Lokshin et al formally tests 4 groups of hypotheses about the income-spending relationship over 1995-2020 and during the pandemic. For the former period, the paper found that when accounting for various factors (like government effectiveness, internet access, tax revenues, etc.) it turns out that poorer countries actually spend more in relative terms than richer nations – put differently, social protection “is more like a necessity than a luxury” (see middle graph, solid line). And in the pandemic? Basically, when controlling for government effectiveness, spending is similar across the country income spectrum (see dotted line in the right graph) – that is, “… [r]elatively low public spending on SP among poorer countries during the pandemic appears to stem mainly from weak government effectiveness in public service delivery (…) rather than low income per se”. Improving such state capabilities alongside providing access to technology may go a long way in increasing spending. The “luxury good hypothesis” is rejected!
Speaking of the pandemic, what may have caused a better recovery in the second half of 2020 and the first half of 2021? A great paper by Demirgüç-Kunt et al estimates the relationship between economic outcomes and country spending on two types of social protection and labor interventions – i.e., income support or support for jobs. Two main findings stand out: first, job protection policies were more effective in promoting economic recovery (like in terms of robust GDP, higher employment, and lower inactivity and poverty rates); and second, income protection had a significant economic impact in countries with weaker pre-pandemic social insurance systems.
Bonus on social protection and health: a short piece by Laverty and Jindal calls for “… unconditional cash transfer directly to beneficiaries giving them a choice to select physician and hospital for surgical treatment”. And WHO’s new policy brief recommends policymakers to “… support communities experiencing hardships (…) and ensure social protection and mitigation measures such as cash transfers, provision of housing and food and support to access essential health services”.
From formal to informal social protection: back in July, I showed a paper that pointed to the social pressure that low-income people get when their income rises (see here). A brand-new paper by Carranza et al confirms such important pattern: in Cote d’Ivoire, 77% of workers in their study “… believed that if they increased labor supply to earn money they would be subject to more transfer requests”. But what happens if they are provided with a direct deposit of earnings and a savings instrument? Both labor supply and earnings increase, revealing a “social tax” rate of about 9-14% on their earnings. But hey, there is no evidence of reduced transfers to others while earnings rose!
More on labor supply: an evaluation of a short-term cash for work program in Tunisia by Leight and Mvukiyehe presents some interesting medium-long term effects. The pilot intervention, which provided $180/month (minimum wage was $114/month), was implemented between April-September 2015 and increased labor market participation (by 0.1-0.2 standard deviations) after one year from treatment. Similarly, the pilot enhanced “… assets, consumption, and financial inclusion, as well as increases of comparable magnitude in civic engagement, psychosocial well-being, and women’s empowerment”. Yet, 5 years after the program effects largely dissipated, except for a small effect on assets (h/t Amber Peterman).
Let’s move to energy reforms: in Burkina Faso, energy subsidies (in this case transfers from the government to national energy companies) represent a little over 1% of GDP: a paper by Vandeninden et al estimates that their phase out would have no direct impact on poor households (because their energy consumption is low, see figure 3). But they may be affected indirectly via price increases in products included in their typical consumption basket. What about redistributing the savings from the energy reform as cash transfers? The authors discuss 5 scenarios with programs using different targeting and adequacy parameters (table 3): the most poverty-reducing modality combines geographic and needs-based criteria, which however also leads to lower coverage and generosity. Trade-offs! (h/t Rebekka Grun).
Final round up! Kelly just produced the latest edition of the humanitarian evidence and discourse summary; Timor-Leste is expanding its child benefits program Bolsa da Mae first introduced in 2008; in the US, the Child Tax Credit led to a 46% reduction in child poverty since 2020 (h/t Ian Orton); GiveWell has a clever way of checking for mistakes in cost-effectiveness analyses (h/t Mattias Lundberg); and “states and markets must be treated as the complementary institutions they have always been” argues Rodrik.
And before you go… some great events here: check out the recordings from yesterday’s webinar by WFP and ODI on digital financial Inclusion and cash transfers in the Asia Pacific region; an interesting seminar on interoperability of France’s social protection system is coming up the 20th; and Norad is hosting an event on cash transfers on the 28th.
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 help you stay on top of it — succinctly, regularly and frequently. Previous editions can be found here.