July 13, 2018 (links edition #65)

Let me start with one of the themes close to my heart, namely food subsidies. In a new paper, Dreze et al review the issues and performance of the Public Distribution System in 6 lower-income states in India (h/t Harold Alderman). In line with recent research on the matter, they found that over the past ten years the scheme “… has turned from a morass of leaky and poorly-targeted transfers to a critical form of social support for a vast majority of poor households”. The market value of monthly PDS transfers for eligible households is now roughly equivalent to one week’s earnings under the NREGA program, without having to work. The paper notes that the current list of ration cards will soon be outdated. This begs the question on whether the future of PDS eligibility will rest on survey-based targeting, self-declaration, or universalization: “… the answer is far from obvious”. The PDS system in Iraq is, instead, fully universal: in a Brookings blog, Livani and Graham find that living in a household that is the recipient of food rations has a significant and positive relationship with life satisfaction for the poor (h/t Ruslan Yemtsov).

Speaking of universality, there is a growing and intriguing global debate on ‘UBI vs jobs guarantees’ (actually the debate has been there for decades…). This very debate is unfolding within the FT microcosm: it ended in a tie in Harford’s elegant piece shared last week (here). No doubt, instead, that Sandbu’s new piece favors UBI: why? His view is that, whether it’s unemployment or precarious jobs, the problem lies in the demand-side of the labor market. This requires macroeconomic policy combined with redistribution, and jobs guarantees are a “distant second-best on either account”. More on UBI: Salmon’s review of Lowry’s new UBI book portrays her as “… too clear-eyed and empirical to be an ardent true believer, yet also too Utopian to be a policy wonk”. Guess it could be both a good and bad thing?

Two big themes that don’t talk to each other: migration and cash transfers. Would cash deter or encourage mobility? What does the evidence say? Does design matter? A new paper I co-authored with Samik Adhikari classifies social assistance programs into three clusters: (i) social assistance that implicitly deters migration centering on place-based programs, (ii) measures that implicitly facilitate migration by relaxing liquidity constraints and reducing transaction costs, and (iii) interventions that are explicitly conditioned on spatial mobility. We find that impacts on migration generally align with the implicit or explicit goals of interventions. Under cluster (i), the likelihood of moving declined between 0.22 and 11 percentage points; among schemes in clusters (ii) and (iii), the probability to move soared between 0.32–25 and 20–55 percentage points, respectively. The analysis also finds spillover effects within households and communities. While social assistance seems not to determine migration decisions per se, it nonetheless enters the broader calculous of mobility decision making. As such, social protection can be an important part of public policy packages to manage mobility.

More on mobility. Clemens and Postel have a succinct IZA summary of their recent paper on the relationship between development assistance and migration. From an cross-sectoral perspective, Jobbins et al have a nice ODI paper on how migration intersects with the water and sanitation agenda, while Scott et reflect on those links in the energy sector. Bonus on forced mobility: as part of their research on a Solidarity Compact for the Rohingya refugee crisis, Huang and Gough discuss how partners – especially regional ones – could contribute and their potential incentives to do so.

A quick look at insurance: Tafere et al estimate the causal effects of index-based livestock insurance coverage on subjective well-being among herders in Ethiopia. The paper is interesting in that it identifies ex ante welfare gains from insurance that reduces risk exposure and ex post buyer’s remorse effects that may arise after the resolution of uncertainty – in other words, was there any regret about insuring if there were no payouts. They find that current insurance coverage generates subjective well-being gains that are significantly higher than the buyer’s remorse effect of an insurance policy that lapsed without paying out. Hence, failure to control for potential buyer’s remorse effects can bias downward estimates of welfare gains from current insurance coverage. Bonus on poverty in Africa: a paper by Jones and Tvedten asks “… what does it mean to be poor?” Based on insights from Mozambique, they examine how quantitative and qualitative analyses relate to one another. The paper also goes deeper and identifies three underlying poverty philosophies, namely the ontological character of poverty, its generative mechanisms, and epistemological priorities. In unbundling these, they offer an interesting discussion on interpreting poverty as a process of social marginalization, as power relations, and other explanations.

From Africa to Asia. Beyer et al use satellite data on evening hour luminosity to measure monthly economic activity in South Asia at the district level – an analysis that they think could be extended to estimate, among others, the economic costs of natural disasters. In a new WIDER working paper, Torm shows that in Vietnam, unionized workers’ wages are 9-22% higher than those of non-union workers. The wage gain is substantially larger at the upper end of the wage distribution.

The OECD recently released the Global Revenue Statistics Database. It’s a great resource with data on direct and indirect taxation, although it lacks data on the low and middle income countries in the Middle East and Asia (except for Indonesia). So the ICTD revenue dataset might be more complete, but the OECD GRSD will soon add more countries. Since I mentioned the Middle East, here is a handy set of three IPC one-pagers on child-sensitive social assistance in the region, including Lebanon, Kuwait and Jordan. BTW, Barca’s recent rich paper on social and integrated beneficiary registries is also summarized in a single page.

From taxes to tax credits: a report by Scally et al examine the low-income housing tax credit in the US. In some context, the program is quite significant: LIHTC has created or preserved an estimated 2.3 million affordable units since 1987. In 3% of US counties, more 10% of rental units are financed by LIHTC. These 51 counties, mostly rural communities, have larger-than-average shares of black or Hispanic residents and higher poverty and unemployment rates.

Assorted mix! Cottam argues that the welfare state in the UK needs be reinvented, similarly to the transformation sparked about 70 years ago by the Beveridge report; Bilal has a blog on rethinking gender is social protection, with some references to the Philippines’ experience. In another VoxDev blog, Azulai explores party-based connections between local level politicians (mayors) and national politicians in charge of grant allocations in Brazil. This is done by examining cases where local and national politicians are of the same party, and what happens in allocations when either changes party line. He finds that the welfare cost from connections is relatively small, or amounting to 0.24% of the typical ministry’s budget for grants.

Finally, Ravallion has a JEL article reviewing two classic books on inequality, i.e., Bourguignon’s Globalization of Inequality and Milanovic’s Global Inequality: A New Approach for the Age of Globalization, both published in 2016. Ravallion’s view is that “… unlike poverty, less of which is surely always better, inequality can also be too low from the point of view of our valued social objectives. Bourguignon clearly recognizes this point, and makes some references to goals such as poverty reduction and human development. However, neither book provides a convincing case for the intrinsic importance of global inequality as a concern that warrants action in its own right”.

p.s. links will be back after a short Summer break