Published on Let's Talk Development

What’s new in social protection – April edition

This page in:

Let’s start with the perennial question on whether cash transfers affect work incentives… the answer is yes but not by much. A review by Baird et al shows that programs tend to result in little or no change in adult labor decisions. The exceptions are adults living with seniors receiving pensions and on select refugee programs (although to a limited extent and in risky locations). Check out tables 1 and 2 (p.26-27) for handy summaries of the evidence. Similarly, Daidone et al. found significant impacts of the Zimbabwe Harmonized Social Cash Transfer Program on beneficiary agricultural activities, the share of households owning livestock, and non-farm enterprises. 

Speaking of Africa, Owusu-Addo et al have a systematic review on the impact of cash transfers in the region. Programs bolstered food security up to 25 percentage points (pp), but had limited impacts on malnutrition. Positive effects were detected on health care, e.g., in Ghana cash increased the use of curative health care by 24 pp. Other findings include improvements in birth registration (effects ranging from 1.5 to 37 pp), beneficiaries’ involvement in community decision-making (from 3.3 to 22 pp), and social cohesion (from 6 to 11.8pp)
The State of Social Safety Nets 2018 Report is out! Social assistance is now ubiquitous – e.g., globally, about 70% of countries have unconditional cash transfers in place. About 36% of beneficiaries escaped poverty because of safety nets. Country spending averages amount to 1.5% of GDP, while 82% of poor households in low-income countries are uncovered. Average transfers are less than one-fifth of income/consumption of the poor in most developing countries. 
Do the effects of cash transfers fade over time? The question, sparked by a Kenya paper, generated a huge stir across the blogosphere and twitterverse. The GiveDirectly-Ozler exchanges centered on both technical issues, and on whether evidence is communicated in a nuanced way. GD’s reply acknowledges part of the criticism. In a CGD blog, Sandefur tried to reassure the ‘cashonistas’ that it is fine “… to admit when the new evidence is not quite as clear as we’d hoped”. An ensuing rebuttal by Ozler is here. The storm is certainly not over…
In a similar vein, Mvukiyehe has a blog on the impacts of public works programs on employment and violence in three countries – Cote d”Ivoire, Egypt and Tunisia. He finds that programs had positive impacts on economic welfare in the short-term. In Cote d’Ivoire, total employment increased by 12 percentage points (pp) as a result of the program – with similar increases in Egypt’s social services scheme (15pp) and in Tunisia (8pp). However, welfare impacts in the medium/long-run (12-15 months after program completion) seem limited.
More on public works. Drawing on lessons from SNAP and TANF, a paper by Hahn argues that work requirements in the US often fail to create economic mobility. A working paper by Cho and Ruthbah find that in Bangladesh, public works have increased household consumption and health investments, as well as reduced outstanding loans. However, the paper is “… inconclusive whether the workfare would achieve the same objectives as efficiently as unconditional cash transfer”. A review by McCord cautions  against the role of social protection in addressing labor market challenges, but calls for wider-scale social protection as a basis for complementary interventions.
From employment to child labor: a working paper by Canelas and Niño-Zarazua found that the Bolivian Bono Juancito Pinto program increased school enrolment rates, but had no impact in reducing child labor. Two explanations are proposed: the size of cash is too modest to compensate for opportunity costs of schooling, while various institutional and legislative factors may allow children to combine schooling with work.
Speaking of children, an AER article by Akee et al evaluates the effects of a cash transfer program among Cherokee Native American children. They found large effects on reducing behavioral disorder symptoms and increased conscientiousness. These impacts were most pronounced for children with the lowest initial endowments.
What are the intergenerational effects of social protection reforms? In a thought-provoking discussion paper, Dahl and Gielen studied a reform in Dutch disability insurance in the mid-1990s. They found that children whose parent’s insurance was curtailed were less likely to be themselves on insurance as adults. They conclude that ignoring parent-to-child spillover effects understates the long-run cost savings of the Dutch reform by as much as 40%.
Can cash transfers decrease violence among partners? A review by Buller et al examined 23 studies on the matter. Out of these, about half showed that cash helps to reduce violence. Programs appear to reduce physical and/or sexual violence more consistently than emotional abuse. Only two studies showed mixed or adverse impacts.
More on a variant of cash transfers, universal basic income: how would firms respond to a UBI? In a new article, Calnitsky sets out two scenarios: a UBI would be an employer subsidy leading to low wages, or it would tighten labor markets and pull wages up. Examining the Canadian Minincome experiment in the 1970s (all residents in the town of Dauphin received cash), he found support for the latter.
And finally, Finland is unwinding its UBI experiment even before results are published: Why? While the pilot wasn’t a UBI to start with, it wasn’t extended because of political push-back on the notion of ‘unconditional’ transfers and the desire to explore alternative reforms (e.g., universal credits and negative income tax).

This monthly column draws from a weekly newsletter – to join send me an email: ( The previous edition of the column can be accessed here.”


Ugo Gentilini

Global Lead for Social Assistance, World Bank

Join the Conversation

The content of this field is kept private and will not be shown publicly
Remaining characters: 1000