Interesting academic articles for July 2019

Here’s what I’m looking forward to reading these days!

Emmanuelle Auriol, Julie Lassebie, Amma Panin, Eva Raiber, and Paul Seabright.  2018. “God insures those who pay? Formal insurance and religious offerings in Ghana.” Working paper.

This paper provides experimental support for the hypothesis that insurance can be a motive for religious donations by members of a Pentecostal church in Ghana. We randomize enrollment into a commercial funeral insurance policy, then church members allocate money between themselves and a set of religious goods in a series of dictator games with significant stakes. Members enrolled in insurance give significantly less money to their own church compared to members that only receive information about the insurance. Enrollment also reduces giving towards other spiritual goods. We set up a model exploring different channels of religiously based insurance. The implications of the model and the results from the dictator games suggest that adherents perceive the church as a source of insurance and that this insurance is derived from beliefs in an interventionist God. Survey results suggest that material insurance from the church community is also important and we hypothesize that these two insurance channels exist in parallel.

Sudhanshu Handa, Silvio Daidone, Amber Peterman, Benjamin Davis, Audrey Pereira, Tia Palermo, and Jennifer Yablonski.  2018. “Myth-Busting? Confronting Six Common Perceptions about Unconditional Cash Transfers as a Poverty Reduction Strategy in Africa.”  World Bank Research Observer.

This paper summarizes evidence on six perceptions associated with cash transfer program- ming, using eight rigorous evaluations conducted on large-scale government unconditional cash transfers in sub-Saharan Africa under the Transfer Project. Specifically, it investigates if transfers: 1) induce higher spending on alcohol or tobacco; 2) are fully consumed (rather than invested); 3) create dependency (reduce participation in productive activities); 4) in- crease fertility; 5) lead to negative community-level economic impacts (including price distortion and inflation); and 6) are fiscally unsustainable. The paper presents evidence refuting each claim, leading to the conclusion that these perceptions—insofar as they are utilized in policy debates—undercut potential improvements in well-being and livelihood strengthening among the poor, which these programs can bring about in sub-Saharan Africa, and globally. It concludes by underscoring outstanding research gaps and policy implications for the continued expansion of unconditional cash transfers in the region and beyond

Apollo Kaneko, Thomas Kennedy, Lantao Mei, Christina Sintek, Marshall Burke, Stefano Ermon, and David Lobell.  2019. “Deep Learning For Crop Yield Prediction in Africa.”  Presented at the International Conference on Machine Learning AI for Social Good Workshop.  

Lack of food security persists in many regions around the world, especially Africa. Tracking and predicting crop yields is important for supporting humanitarian and economic development efforts. We use deep learning on satellite imagery to predict maize yields in six African countries at the district level. Our project is the first to attempt this kind of prediction in Africa. Model performance varies greatly between countries, predicting yields in the most recent years with average R2 as high as 0.56. We also experiment with transfer learning and show that, in this data sparse setting, data from other countries can help improve prediction within countries.

David McKenzie and Dario Sansone.  2019.  “Predicting entrepreneurial success is hard: Evidence from a business plan competition in Nigeria.”  Journal of Development Economics.

We compare the absolute and relative performance of three approaches to predicting outcomes for entrants in a business plan competition in Nigeria: Business plan scores from judges, simple ad-hoc prediction models used by researchers, and machine learning approaches. We find that i) business plan scores from judges are uncorrelated with business survival, employment, sales, or profits three years later; ii) a few key characteristics of entrepreneurs such as gender, age, ability, and business sector do have some predictive power for future outcomes; iii) modern machine learning methods do not offer noticeable improvements; iv) the overall predictive power of all approaches is very low, highlighting the fundamental difficulty of picking competition winners.

Pia Raffler, Daniel N. Posner, and Doug Parkerson.  2019.  “The Weakness of Bottom-Up Accountability: Experimental Evidence from the Ugandan Health Sector.”  Working paper. 

We evaluate the impact of a large-scale information and mobilization intervention designed to improve health service delivery in rural Uganda by increasing citizens’ ability to monitor and apply bottom-up pressure on underperforming health workers. Modeled closely on the landmark “Power to the People” study (Bjorkman and Svensson, 2009), the intervention was undertaken in 376 health centers in 16 districts and involved a three wave panel of more than 14,000 households. We find that while the intervention had a modest positive impact on treatment quality and patient satisfaction, it had no effect on utilization rates or health outcomes (including child mortality). We also find no evidence that the channel through which the intervention affected treatment quality was citizen monitoring. The results hold in a wide set of pre-specified subgroups and also when, via a factorial design, we break down the complex intervention into its two most important components. Our findings cast doubt on the power of information to foster community monitoring or to generate improvements in health outcomes, at least in the short term.

Thomas Calvo, Mireille Razafindrakoto, and François Roubaud.  2019.  “Fear of the state in governance surveys? Empirical evidence from African countries.”  World Development.

The need to collect data on governance-related issues has been growing since the 1990s. Demand gained momentum in 2015 with the adoption of SDG16 worldwide and Agenda 2063 in Africa. African countries played a key role in the adoption of SDG16 and are now leading the process of collecting harmonised household data on Governance, Peace and Security (GPS). Yet the possibility has recently been raised that sensitive survey data collected by government institutions are potentially biased due to self-censorship by respondents. This paper studies the potential bias in responses to what are seen as sensitive questions, here governance issues, in surveys conducted by public organisations. We compare Afrobarometer (AB) survey data, collected in eight African countries by self-professed independent institutions, with first-hand harmonised GPS survey data collected by National Statistics Offices (NSOs). We identify over 20 similarly worded questions on democracy, trust in institutions and perceived corruption. We first com- pare responses from AB survey respondents based on who they believe the survey sponsor to be. No systematic response bias is found between respondents who believe the government to be behind the AB survey and those who consider it to be conducted by an independent institution. Our estimations suggest that the observed residual differences are due to a selection bias on the observables, which is mitigated by propensity score matching procedures. The absence of a systematic self-censorship or attenuation bias is further evidenced by means of an experimental design, whereby responses from GPS surveys conducted by NSOs (the treatment) are compared with AB surveys sponsored by reportedly independent bodies. Our results provide evidence, at much higher levels of precision than other existing data sources, of the capacity and legitimacy of government-related organisations to collect data on governance as a matter of national interest and sovereignty.

Maya Berinzon and Ryan Briggs.  2019.  “Measuring and explaining formal institutional persistence in French West Africa.”  Journal of Modern African Studies.  

Colonial institutions are thought to be highly persistent, but measuring that persistence is difficult. Using a text analysis method that allows us to measure similarity between bodies of text, we examine the extent to which one formal institution the penal code has retained colonial language in seven West African countries. We find that the contemporary penal codes of most countries retain little colonial language. Additionally, we find that it is not meaningful to speak of institutional divergence across the unit of French West Africa, as there is wide variation in the legislative post-coloniality of individual countries. We present preliminary analyses explaining this variation and show that the amount of time that a colony spent under colonisation correlates with more persistent colonial institutions.

Benjamin Rubbers.  2019.  “Mining Boom, Labour Market Segmentation and Social Inequality in the Congolese Copperbelt.”  Development and Change.

The study of the impacts of new mining projects in Africa is generally set in a normative debate about their possible contribution to development, which leads to a representation of African societies as divided between beneficiaries and victims of foreign investments. Based on research in the Congolese copperbelt, this article aims to examine in more detail the inequalities generated by the recent mining boom by taking the processes of labour market segmentation as a starting point. It shows that the labour market in the mining sector has progressively been organized along three intersecting lines that divide it: the first is between employment in industrial and artisanal mining companies, the second is between jobs for mining or subcontracting companies and the third is between jobs for expatriates, Congolese skilled workers and local unskilled workers. Far from simply reflecting existing social in- equalities, the labour market has been actively involved in their creation, and its control has caused growing tensions in the Congolese copperbelt region. Although largely neglected in the literature on extractive industries, processes of labour market segmentation are key to making sense of the impacts of mining investments on the shape of societies in the global South.

Benjamin Chemouni.  2019.  “The rise of the economic technocracy in Rwanda: A case of a bureaucratic pocket of effectiveness or state-building prioritisation?”  Effective States and Inclusive Development working paper #120.

The Rwandan Ministry of Finance and Economic Planning (MINECOFIN) is recognised as the most effective organisation in the Rwandan state. The objective of the paper is to understand the organisational and political factors influencing MINECOFIN’s performance since the genocide and link them to the wider conversation on the role of pockets of effectiveness (PoEs) in state-building in Africa. It argues that, because of the Rwandan political settlement and elite vulnerability, MINECOFIN is not a PoE but only a good performer in a generally well functioning state. The Ministry overperforms first because, unsurprisingly, the nature of its tasks is specific, requires little embeddedness and allows a great exposure to donors, making its mandate easier to deliver in comparison to other organisations. MINECOFIN also performs better than other state organisations because it is, more than others, at the frontline of the elite legitimation project since it is the organisation through which resources are channelled, priorities decided, and developmental efforts coordinated. Given the rulers’ need for an effective state as a whole, MINECOFIN appears only as the lead climber in a wider dynamics of systematic state building.