Let’s start with social protection in Africa. A new paper by Kagin et al. estimates that in Malawi, each Malawi Kwacha (MK) transferred through the Social Cash Transfer Program generates 1.88 MK, while multipliers of public works are between 2.9-3.24 MK. In the same country, the Malawi Economic Monitor by Kandoole et al. has a very crisp, insightful edition discussing safety nets, e.g., spending is only 0.6% of GDP compared to 2% of input subsidies, and almost 6% on humanitarian aid.
In another study, Prifti et al. examine the effects of cash transfers on employment: in Zambia and Kenya, households react to cash by supplying less labor off-farm and working more on their own farm. In Ghana and Lesotho, households instead engaged more in wage employment. Differences may stem from the varying frequency and predictability of payments. In Niger, Annan and Sanoh show that seasonal social assistance reduces the impact of shocks on households' consumption and vulnerability. Arruda and Dubois have an insightful brief on the history of social cash transfers in Zambia – see the three phases identified in table 3.
And now for our monthly dose of basic income papers. Political Quarterly has just published a series of papers devoted to Tony Atkinson’s Participation Income -- see Stirton’s introductory article. A short Nature article by Arnold reviews a sub-set of UBI and negative income tax pilots. Among the best lines, “… an affordable UBI is inadequate, and an adequate UBI is unaffordable”. California’s latest pilot in Stockton adds to the string of experiments dubbed as UBI, but which are not really so. Bonus: section 4 of a new AI paper by Furman and Seamans discusses UBI vs wage subsidies vs employment guarantee (with the second being their favorite).
Next an exciting collection of 7 papers on social protection in fragile contexts: Bruck et al find that assets-based transfers enhance nutrition in Niger; Aurino et al. show that school feeding led to increases in school enrollment by 11 percentage points in Mali, but gains were concentrated among children living in non-conflict areas; Schwab demonstrates that in Yemen both cash and food transfers led to productive investments, with cash mostly devoted to livestock while food recipients invested in higher-return crops; Valli et al. argue that cash, food and voucher transfers for Colombian refugees in Ecuador led to overall enhanced social cohesion as measured though six aggregate dimensions derived from 33 individual indicators; Schnitzer shows that proxy means tests perform better in targeting chronically poor households, while the household economy approach works better in targeting the transiently poor during crises; de Hoop et al. find that CCT for an after-school program for Syrian refugees doesn’t affect enrollment, but enhances attendance by 20%; lastly, Krishnan et al. quantify the welfare impact of a potential reform of the Iraqi PDS in urban areas, i.e., a complete PDS removal will require compensating poor households by 74 percent of their expenditures, while a targeted removal of the top 4 deciles will leave poverty unaffected and generate cost savings – but may create public discontent.
By the way, the Global Humanitarian Assistance report shows that cash transfers in humanitarian settings are estimated at $2.8 billion, a 40% increase from 2015. According to an evaluation by Metcalfe-Hough and Poole, cash transfers is indeed one of the areas where the humanitarian community has achieved most progress.
How to combine safety nets and health insurance for the poor? Hirvonen reflects on the experience with Ethiopia’s PSNP and Community Based Health Insurance benefits (less than 25% of beneficiary of the former were enrolled in the latter). Behrendt and Nguyen have an interesting review of approaches to adapt social protection to changing labor markets, including through a combination of contributory and non-contributory mechanisms.
Lasting effects on nutrition: Cahyadi et al find that 6 years after inception, the CCT program in Indonesia halved the share of children age 7-15 who are not enrolled in school, while wage labor for 13-15 year olds was reduced by at least one-third. Bonus on child-sensitive social protection: the IPC has two handy one-pagers on non-contributory transfers for children in Iraq and Iran.
Let’s conclude with two fascinating NBER papers on SNAP in the US: in the first, Ravallion et al find that economic growth in the US has come with a ‘sinking social floor’. The floor has fallen during the 1990s, and stabilized in the 2000s. The expansion of the food stamps program in the wake of the financial crisis was able to prevent a fall in the floor despite the inequality-increasing growth process. In another paper, Alsan and Yang show that, though not at personal risk of deportation, many Hispanic citizens in the US don’t participate in SNAP fearing that receiving benefits could expose non-citizens in their network to immigration authorities.
This monthly column draws from a weekly newsletter available here.