Links March 15 – Assessing the responsiveness of social protection to economic crises; the performance of Brazil’s cash transfers during the pandemic; connecting social protection and humanitarian assistance in Palestine; a recent survey on humanitarian access in Gaza; poverty dynamics among female headed households in MENA; the evolution of BISP in Pakistan; conflict-sensitive social assistance in Sudan; M&E of child grants in Namibia; drivers of poverty reduction in Kazakhstan; and three events coming up next week!

Let’s begin with a big question (and big answer)r: are social protection responses to economic crises countercyclical? A brilliant analysis by Pignatti et al offers plenty of fascinating insights. Covering 106 countries between 1980 and 2021, the report found that social protection is the most shock-responsive government measure: a 1 percentage point (pp) decrease in GDP growth rate increases social protection spending (as a share of GDP) by 0.11 pp – or over 10 times the responsiveness of subsidies (see figure 1A, p.10). Among social protection components, social assistance is four times more responsive than social insurance (0.08 vs 0.02pp respectively). Yet overall performance is driven by high-income countries, where the level of adaptiveness is six times greater than in low- and middle-income settings due to better delivery systems and access to finance (figure 2A, p.12). Were COVID-19 responses any different? Yes, they were so: compared to the financial crisis of 2009-2013, not only was pandemic spending more responsive to changes in GDP (0.08 vs 0.06 pp), but it was largely driven by low- and middle-income countries (see figure above). The latter doubled the use of social assistance compared to a decade earlier (social insurance was similar); by contrast, in high-income countries relied relatively more on social insurance than in the global financial crisis, while the responsiveness of social assistance spending to changes in GDP was close to zero. That’s not all: the paper also presents a juicy discussion of crisis forecasting (p.17-20), novel ways of identifying beneficiaries (p.21-23), and call for the establishment of a special fund for crisis responses (p.41).

Speaking of the pandemic, a great paper by Brollo et al assessed the performance of Brazil’s Auxilio Emergencial (AE) flagship cash transfers COVID-19 response (reaching about 34% of the population in 2020). In terms of incidence, the program was generally progressive (see figure 8, p.16), with eligibility of the 2021 version being slightly tighter than in 2020 (especially for the upper 60% of the distribution). In 2020, AE increased per-capita disposable income rose by 2.1% (consider that net market income dropped by 5.3% without any social protection support) and reduced the poverty rate from 28.4% to 21.6% (and extreme poverty was nearly eliminated, with the rate at just 2.3%). As the paper noted, “if only the pre-pandemic tax-benefits system had been in place, poverty would have risen significantly”. What about alternative designs? The paper’s simulations found that positive (although slightly lower) income and poverty effects could have been reached by maintaining the same eligibility criteria and reducing adequacy to one-third of the initial benefits. This would have halved AE’s cost, i.e., 2.3% of GDP as opposed to 4% GDP.

Since I mentioned crises, what’s new on the humanitarian front? Although released just before the war, a report by Smith et al may still present relevant findings and implications for better connecting social protection and humanitarian assistance in Palestine: their review scores the strength of 20 papers (with scores between 1 and 3) across seven questions. Recommendations revolve around thematic issues like targeting and coverage (see p.14-17, e.g., “enhance the accountability and transparency of the NCTP targeting process”), adequacy (p.22-23, e.g., “take the debt burden into consideration in transfer design”), shock-responsiveness (p.29-31, e.g., “define which ‘emergency’ needs are to be met”), and referrals (p.35-37, e.g., “overcome constraints to data-sharing”). Bonus: Stoddard et al published a report on humanitarian access in Gaza over the last six months based on phone surveys with residents conducted in mid-February (see figure 2, p.8 on aid need vs receipt).

More on MENA: does the definition of female-headed households (FHHs) matter for measuring poverty? Based on 20 rounds of cross-sectional surveys in six countries, a paper by AlAzzawi et al found that most types of FHHs are 1%-5% less likely to be poor (including whether self-reported, potential, and most-educated female-adult FHHs) (see discussion on p.20-21); and in terms of upward mobility, the chance of escaping poverty are higher for FHHs without children (42%) compared to those with children (37%) and non-FHHs (36%). Bonus on gender in MENA: “[i]t is beyond time that investing in women and girls is recognized to be at the core of development”, argues a blog by Saadah and Al Tuwaijri.

What about Asia? Guven et al document the rich evolution of the BSIP program in Pakistan. Specifically, the report highlights the spectacular delivery journey of the program, with a detailed overview of key features and innovations across the four classic delivery building blocks – assess, enroll, provide and manage (see rich analysis over p.35-72). See also the discussion on BISP budget and coverage trends (p.17), the challenge of keeping BISP’s purchasing power constant (p.21-22), and the process entailed by a dynamic registry (p.28). Among the key messages, “BISP can strengthen itself by building close links with early warning systems, developing a comprehensive communications strategy, incorporating human-centered design to make systems more easily accessible, integrating grievance redress mechanisms, incorporating tools to ensure performance of the increasing number of field staff, and deepening collaboration with provinces”.

News from Africa? A paper by Birch et al examines the “conflict sensitivity” of social assistance in Sudan: interestingly, they find lots of action on such theme, including around accountability (e.g., p.44 recounts how “UNICEF’s MCCT+ program adapted mid-rollout to deepen its engagement with tribal chiefs to overcome communities’ lack of trust in government and partners”), adaptiveness (e.g., see box 4.1, p.45 where Mercy Corps used the microfinance institution Ebdaa Bank for delivery to IDPs in Gadarif) and localization (e.g., p.46 mentioning how NRC provided cash-based transfers to locally managed shelters). Further South in the region, an article by Nshimyimana and Rabie identified several challenges with monitoring and evaluation arrangements of Namibia’s Child Support Grant, among others “stemming from a lack of shared understanding of the system, no dedicated theory of change that specifies results, [and] indicators that focus only on reaching the beneficiary rather than the results for the beneficiary”.

Moving to Europe and Central Asia, what drove Kazakhstan’s growth and poverty reduction in recent years? In one word, “productivity”: a report by Rigolini et al shows that other factors, like labor market factors, were also important but much less so compared to productivity (see figure 1.1, p.11).

And to conclude… McElhinney and Spencer warn that “the clock is ticking to build guardrails into humanitarian AI”; the European Humanitarian Forum is coming up next week (March 18-19); an event on the future of social protection facing a cascade of crises is planned for March 19-20; and on March 21, the OECD is launching a new version of its database on informality and social protection (KIIbIH).