What happens 10 years after providing a package of cash transfers and assets? In India, Banerjee et al shows it improved consumption (1 standard deviations, SD), food security (0.1 SD), income (0.3 SD) and health (0.2 SD). Interestingly, they also trace how these are built up over time. In the first 18 months, beneficiaries are better off because of the transfers; at 3 years they have diversified into new non-farm businesses. By year 7, earnings from such microenterprises is nearly twice that in the control group. And here comes the underdog: wage employment wasn’t affected until year 3, but by year 7 is has the same impact as the self-employment enterprise. This is because members from beneficiary households migrated and became wage employed. And then the takeover: by year 10, the effects on micro-enterprises attenuates relative to year 7, but the effect on wage earnings is still large (120 days on average) and the effect on remittances are even larger than in year 7 (rising 3.7 times between month 18 and year 10). In other words, “… the household is increasingly specialized in labor earnings” (p.3).
Number “10” comes up also in another new paper: we know that cash transfers scaled up on average by 217% in response to Covid-19… but how are countries financing such coverage extensions? Our note jointly produced with UNICEF and OECD — “Where Does the Money Come From?” — lays out 10 stylized facts on financing sources, including domestic, external, and combinations thereof drawing from 31 countries. A range of strategies are pursued, including setting up special funds (e.g., Mauritania), public-private partnerships (e.g., Uzbekistan), reprioritization (e.g., South Africa), reductions in other services (e.g., Ukraine) and a generalized increase in deficit financing.
Moving to Africa, how did preexisting safety nets protect from Covid-19 in Ethiopia? Abay et al show that with no social assistance, food insecurity grew by 11.7 percentage points and the food gap by 0.47 months; these indicators soared by only 2.4 pp and 0.13 months among people covered by the Productive Safety Net Program.
More on the Horn: what are the effects of universal basic income during Covid-19 in Kenya? Based on a paper shared in past weeks, a blog by Banerjee et al argues that it helps “… but it’s not a silver bullet”. Bonus: based on high-frequency monitoring phone surveys in nearly 100 countries, a nice blog by Sanchez-Paramo and Narayan estimates that an average of 20% of households reported receiving social assistance.
Switching to Asia, new research by Dervisevic et al explores how community targeting for public works performs against a PMT method in Laos. In 2016, in each village there was a multi-day participatory process of PMT verification and adjustment by which households were assigned poverty ranks. During the registration process for program in 2018, village heads updated these rankings (24% of households changed their poverty rank as a result of this updating). Community leaders rely on “… a holistic impression of household welfare, rather than specific indicators like actual land ownership, nutrition, or economic shocks”. Such method is found to be quicker and more cost-efficient in updating lists as opposed to a full PMT plus community-ranking and validation process.
Let’s stay in the region, with an interesting twist to the climate-social protection interactions: Iqbal and Nawaz show that in Pakistan, BISP cash transfers help households switch away from traditional fuels like wood and bolster the use of existing electrical appliances like washing machines and refrigerators.
New research on gender! Stampini et al show that in four LAC countries (Chile, Colombia, Costa Rica and Mexico), women provide up to 84% of long-term family caregiving and account for up to 88% of total hours of long-term case. Women who provide long-term care are less likely to work (by 2.4 percentage points), and those who do so work less hours per week (-7%) and have a double burden of work plus care; Sánchez et al find that Chilean men would earn approximately 19% – 28% more than women as a result of the difference in labor supply elasticities by gender, ceteris paribus; and finally, Lassassi and Tansel present new estimates for female labor force participation in MENA (Algeria, Egypt, Jordan, Palestine and Tunisia).
Time for the skills and labor corner (with Michael Weber and Indhira Santos). Let’s start with evidence on work-to-unemployment insurance transitions: in Brazil, Gerard and Gonzaga find that unemployment insurance reduces incentives to return to a formal job, although such cost is lower than in more formalized countries like the US. In Germany, Fuchs et al find that the transition from gainful employment into unemployment benefit receipt leads to an average 3.2% decrease in reservation wages (such rate increases to 4.9% in and transition to social assistance, which has a stronger activation component). Bonus: Martín-Artiles et al have a great comparative analysis of unemployment benefits in Argentina, Uruguay, Spain and Italy.
A trio of skills papers: Das estimates that an on-the-job training program for disadvantaged youth in Bangladesh increased labor market participation by 16 percentage points and earnings by 23%: effects decline after 22 months, but (somewhat similarly to the India paper at the opening of the newsletter) the effect on earnings is sustained as the program induces a shift from casual work to wage employment. Examining adults’ skills and their labor market outcomes in Colombia, Acosta et al find that cognitive skills correlate with better jobs and completing tertiary education; instead socioemotional skills are associated with labor market participation but seemingly independently of schooling. Finally, in Indonesia Johnson et al evaluate the effects of a socioemotional learning program teaching “growth mindset” and “self-management” to adolescent students (the former is the belief that intelligence is malleable, while the latter is the ability to regulate behaviors and work toward goals): these proved effective some dimensions like study behaviors, but not on test scores or aspirations (h/t Noel Muller).
Databases! GiveDirectly launched a cool “Explorer” database of cash transfers evaluations, including a range of handy tags like type of programs, transfer frequency, payment amount, outcomes studied… super useful! Next week the ILO will launch its World Social Protection Data Dashboards (Nov 24), and the World Inequality Database now includes income distribution statistics for 170+ countries.
Speaking of inequality, Ravallion updated his lecture notes on “income” and “non-income” dimensions of poverty and inequality (Part 3.1) and Hvidberg et al show that in Denmark people underestimate the degree of inequality by believing that others are closer to themselves than they really are.
Final assortment: a compendium of UNICEF’s impressive social policy activities in Europe and Central Asia over 2014-2020 (h/t Pamela Dale); the SPARKS conference will feature social protection (Dec 2-3) (h/t Amber Peterman); and if you have 5 minutes, try predicting the outcomes of an unconditional cash transfer experiment in the US (h/t/ Berk Ozler).