Members of the NBER's Development Economics Program joint BREAD Conference met November 22-23 in Cambridge. Oriana Bandiera and Robin Burgess of London School of Economics, Research Associates Melissa Dell of Harvard University, Edward Miguel of University of California, Berkeley and Dean Yang of University of Michigan, and Program Directors Seema Jayachandran of Northwestern University and Benjamin A. Olken of MIT organized the meeting. These researchers' papers were presented and discussed:
Katherine Casey, Stanford University and NBER, and Niccoló Meriggi and Abou Bakarr Kamara, International Growth Centre
An Experiment in Candidate Selection (NBER Working Paper 26160)
Are ordinary citizens or political party leaders better positioned to select candidates? While the direct vote primary system in the United States lets citizens choose, it is exceptional, as the vast majority of democracies rely instead on party officials to appoint or nominate candidates. Theoretically, the consequences of these distinct design choices on the selectivity of the overall electoral system are unclear: while party leaders may be better informed about candidate qualifications, they may value traits -- like party loyalty or willingness to pay for the nomination -- at odds with identifying the best performer. To make progress on this question, Casey, Meriggi, and Kamara partnered with both major political parties in Sierra Leone to experimentally vary how much say voters, as opposed to party officials, have in selecting Parliamentary candidates. They find evidence that more democratic selection procedures increase the likelihood that parties select the candidate most preferred by voters, favor candidates with stronger records of local public goods provision, and alter the allocation of payments from potential candidates to parties.
Siddharth E. George, Harvard University
Like Father, Like Son? The Effect of Political Dynasties on Economic Development
Politics remains an exceptionally dynastic occupation in democratic societies, even though many countries democratised to end hereditary rule. Dynastic politics has theoretically ambiguous economic effects: bequest motives may lengthen politicians’ time horizons (founder effects), but heritable political capital may make elections less effective at holding dynastic descendants accountable (descendant effects). George studies how dynastic politics affects economic development in India, compiling detailed biographical data on all Indian legislators since 1862. Using three different identification strategies, George identifies (i) descendant effects, (ii) founder effects and (iii) the overall effect of a dynastic political environment. First, using a close elections regression discontinuity design, they find that dynastic descendants reduce earnings, asset ownership and public good provision of villages they represent. Descendants underperform
partly due to moral hazard: they inherit voters loyal to their predecessor, which dampens their performance incentives. Second, George shows that the incentive to establish a dynasty motivates politicians to perform better. Redistricting allows for the identification of villages that were exogenously exposed to founders but not descendants. Using this variation, it is found that dynastic founders reduce poverty and improve public good provision of villages they represent. Moreover, consistent with bequest motives, politicians with a son are twice as likely to establish a dynasty and exert more effort while in office. Third, George identifies the overall effects of a dynastic political environment using an instrumental variables strategy based on the gender composition of past incumbents’ children. Dynastic politics has negative economic effects and results in a “reversal of fortune” development pattern, consistent with positive founder and negative descendant effects. A simple overlapping generations model with heritable human and political capital explains these three empirical facts.
Vittorio Bassi, University of Southern California; Tommaso Porzio, Columbia University; Ritwika Sen, Northwestern University; and Raffaela Muoio and Esau Tugume, BRAC Uganda
Achieving Scale Collectively
It has long been argued that firms in developing countries enter small and struggle to grow large. This paper studies whether the small firm scale might itself hinder their labor productivity, thus spurring a vicious cycle of underdevelopment. The authors design and implement a novel survey to measure how firms produce output in three manufacturing sectors in urban Uganda. They find strong direct evidence that most firms operate at a scale that is too small to justify investment in modern high-capacity machines. However, an active inter-firm rental market has emerged in response, allowing firms to achieve economies of scale collectively. Interpreted through the lenses of an equilibrium model of firm behaviour, our data reveals that the rental market for machines limits substantially the productivity losses due to the small scale of production. Taken together, their results show that, while the small size of firms remains a concern, the design of effective development policies should take into account market level interactions between firms as a powerful mean to mitigate the cost of small scale.
Clare Leaver, University of Oxford; Owen Ozier, The World Bank; Pieter M. Serneels, University of East Anglia; and Andrew F. Zeitlin, Georgetown University
Recruitment, Effort, and Retention Effects of Performance Contracts for Civil Servants: Experimental Evidence from Rwandan Primary Schools
Accumulating evidence suggests that pay-for-performance (P4P) contracts can elicit greater effort from incumbent civil servants, but less is known about how these contracts affect the composition of the public sector workforce. Leaver, Ozier, Serneels, and Zeitlin provide the first experimental evidence of the impact of P4P on both the compositional and effort margins. In partnership with the Government of Rwanda, they implemented a 'pay-for-percentile' scheme (Barlevy and Neal 2012) in a novel two-tier experimental design. In the first tier, the researchers randomly assigned teacher labor markets to either P4P or equivalent fixed-wage contracts. In the second tier, they implemented a 'surprise', school-level re-randomization, allowing for the separate identification of the compositional effects of advertised P4P contracts and the effort effects of experienced P4P contracts. The pre-analysis plan sets out a theoretical framework that helps to define a set of hypotheses, and conducts simulations on blinded data to develop high-powered tests. The researchers find that P4P contracts did change the composition of the teaching workforce, drawing in individuals who were more money-oriented, as measured by a framed Dictator Game. But these recruits were not less effective teachers -- if anything the reverse. On the effort margin, the researchers observe substantial and statistically significant gains in teacher value added, mirrored in positive effects on teacher presence and observed pedagogy in the classroom. In Year 2, the researchers estimate the total effect of P4P, across compositional and effort margins, to be 0.21 standard deviations of pupil learning. One quarter of this impact can be attributed to selection at the recruitment stage, with the remaining three-quarters arising from increased effort.
Richard Hornbeck, University of Chicago and NBER, and Martin Rotemberg, New York University
Railroads, Reallocation, and the Rise of American Manufacturing
Hornbeck and Rotemberg examine the impacts of market integration on the development of American manufacturing, as railroads expanded through the latter half of the 19th century. Using new county-by-industry data from the Census of Manufacturers, they estimate substantial reduced-form impacts on manufacturing productivity from relative increases in county market access as the railroad network expanded. In general equilibrium, the railroads increased economic activity in marginally productive counties. The researchers therefore estimate substantial aggregate economic gains from the railroads, much larger than previous estimates, by allowing for the presence of factor misallocation. The estimates highlight how broadly-used infrastructure or technologies can have much larger economic impacts when there are inefficiencies in the economy.
Adam Aberrra and Matthieu Chemin, McGill University
Does Legal Representation Increase Investment? Evidence from a Field Experiment in Kenya
The legal system enforces contracts and secures property rights, thereby increasing the incentives to exert effort, invest, access credit, and grow. Yet, the high costs of access to the legal system may prevent these gains from taking place. Aberra and Chemin present the results of a randomized intervention offering the services of a free lawyer for 2 years in a rural setting with prohibitive lawyer fees and numerous land disputes. Not all cases were fully resolved after 2 years, but legal representation increased the security of property rights, which translated into greater access to credit and agricultural production.
Abhijit Banerjee and Esther Duflo, MIT and NBER; Arun G. Chandrasekhar, Stanford University and NBER; and Matthew Jackson, Stanford University
Changes in Social Network Structure in Response to Exposure to Formal Credit Markets
Banerjee, Chandrasekhar, Duflo, and Jackson study how the introduction of microfinance changes the networks of interactions among 16476 households in 75 Indian villages, and develop a new dynamic model of network formation to explain the empirical findings. None of the villages were exposed to microfinance by 2006, and 43 villages were through 2010. Using a two-wave panel of network data collected in 2006 and 2012, the researchers compare changes in networks in villages exposed to microfinance relative to those not exposed. Networks exposed to microfinance experience a significantly greater loss of links -- for credit relationships as well as advice and other types of relationships -- compared to those not exposed. Microfinance not only results in decreases in relationships among those likely to get loans, but also decreases in relationships between those unlikely to get loans. These patterns are inconsistent with models of network formation in which people have opportunities to connect with whomever they wish, but are consistent with a model that emphasizes chance meetings that depend on relative efforts to socialize coupled with conditional choices of whom to connect with, as well as externalities in payoffs across relationships between pairs and triples of people.
Emily Breza, Harvard University and NBER; Supreet Kaur, University of California, Berkeley and NBER; and Yogita Shamdasani, University of Pittsburgh
Labor Rationing: A Revealed Preference Approach From Hiring Shocks
Assessing unemployment levels is empirically challenging. Economists currently rely on survey self-reports, whose reliability is unknown and which make it difficult to disentangle those who have exited the labor force and the self-employed from rationed workers. Breza, Kaur, and Shamdasani develop an approach to obtain the first revealed preference estimates of labor rationing. They generate large transitory hiring shocks in Indian local labor markets -- hiring up to 35% of the male labor force for month-long work in external factories. The researchers examine the impact of these transitory shocks on the local labor market to diagnose the extent of rationing. They find evidence for severe labor rationing during lean seasons, which account for 6 months of the year. Specifically, "removing" a large portion of workers leads to (i) no change in the local wage, and (ii) no change in local aggregate employment levels (excluding external jobs). This is due to one-for-one positive employment spillovers on the remaining workers who benefit from decreased competition for job slots. The researchers further decompose rationing into involuntary unemployment and disguised unemployment (self-employed workers who would prefer wage labor). In contrast, they detect limited evidence for labor rationing during peak employment seasons: in these months, transitory external hiring shocks increase local wages and reduce local aggregate employment. In addition, the researchers show that traditional government surveys substantively underestimate unemployment in this setting. This approach can be extended to obtain revealed preference bounds on involuntary unemployment in diverse settings.
Matti Mitrunen, University of Chicago
Structural Change and Intergenerational Mobility: Evidence from the Finnish War Reparations
Mitrunen presents evidence that government industrial policy can promote new industries, move labor out of agriculture into manufacturing, and have long-term effects via increased human capital accumulation and upward mobility. Mitrunen uses plausibly exogenous variation generated by the Finnish war reparations (1944-1952) that forced the largely agrarian Finland to give 5% of its yearly GDP to the Soviet Union in the form of industrial products. To meet these terms, the Finnish government provided short-term industrial support that persistently raised the employment and production of treated, skill-intensive industries. Mitrunen traces the impact of the policy using individual-level registry data and shows that the likelihood of leaving agriculture for manufacturing and services increased substantially in municipalities more strongly affected by the war reparations shock. These effects were persistent: 20 years after the intervention, the reallocated workers remained in their new sectors and had higher wages. Younger cohorts affected by the new skill-intensive opportunities obtained higher education and were more likely to work in white-collar occupations by 1970. This result is consistent with higher returns to education. Finally, Mitrunen links parents to children to study how the policy affected upward mobility. Mitrunen shows that mobility in both income and education increased in the exposed locations, as people in lower socioeconomic groups benefited from the structural change.
Dennis Egger and Michael W. Walker, University of California, Berkeley; Johannes Haushofer, Princeton University and NBER; Paul Niehaus, University of California, San Diego and NBER; Edward Miguel, University of California, Berkeley and NBER
General Equilibrium Welfare Effects of Cash Transfers: Experimental Evidence from Kenya
How large economic stimuli generate individual and aggregate responses is a central question in economics, but has not been studied experimentally. Egger, Haushofer, Niehaus, Miguel, and Walker provided one-time cash transfers of about USD 1000 to over 10,500 poor households across 653 randomized villages in rural Kenya. The implied fiscal shock was over 15 percent of local GDP. The researchers find large impacts on consumption and assets for recipients. Importantly, Egger, Haushofer, Niehaus, Miguel, and Walker document large positive spillovers on non-recipient households and firms, and minimal price inflation. They estimate a local fiscal multiplier of 2.6. Egger, Haushofer, Niehaus, Miguel, and Walker interpret welfare implications through the lens of a simple household optimization framework.
Monica Martinez-Bravo and Andreas Stegmann, Centro de Estudios Monetarios y Financieros (CEMFI)
In Vaccines We Trust? The Effects of the CIA's Vaccine Ruse on Immunization in Pakistan
In July 2011, the Pakistani public learnt that the CIA had used a vaccination campaign as cover to capture Osama Bin Laden. The Taliban leveraged on this information and initiated an anti-vaccine propaganda to discredit vaccines and vaccination workers. Martinez-Bravo and Stegmann evaluate the effects of these events on immunization by implementing a Difference-in-Differences strategy across cohorts and districts. They find that vaccination rates declined 12 to 20% per standard deviation in support for Islamist parties. These results suggest that the disclosure of information that lends credibility to conspiracy theories about vaccines can have large effects on demand for immunization.