- The Welfare Cost of Lawlessness: Evidence from Somali Piracy, joint with Tim Besley and Hannes-Felix Mueller. Journal of the European Economics Association, Volume 13, Issue 2, pages 203 – 239, April 2015.In spite of general agreement that establishing the rule of law is central to properly functioning economies, little is known about the cost of law and order breakdowns. This paper examines this in a specific context by estimating the effect of Somali piracy attacks on shipping costs using data on shipping contracts in the dry bulk market. We study shipping routes whose shortest path exposes them to the risk of piracy and find that the increase in attacks in 2008 lead to an 8 to 12 percent increase in shipping costs. From this, we estimate the welfare loss due to piracy. Based on a fairly conservative estimate, generating around 120 USD million of revenue for Somali pirates led to a welfare loss in excess of 630 USD million, making piracy an expensive way of making transfers.
- Washington Post, Brad Plumer, The economics of Somali piracy, 13.03.2013.
- Financial Times, Tim Hartford, Lessons for pirates – from tax collectors, 18.01.2013.
- Foreign Policy, Paul Collier, It’s Lonely Being No. 1 – Is there hope for Somalia?, 18.06.2012.
- voxeu The Welfare Cost of Lawlessness: Evidence from Somali Piracy, 18.06.2012.
- IGC,The Economic Costs of Piracy, 18.06.2012.
- Group Lending Without Joint Liability, joint with Jon de Quidt and Maitreesh Ghatak. In Journal of Development Economics, pages 217-236, May 2016.This paper contrasts individual liability lending with and without groups to joint liability lending. We are motivated by an apparent shift away from the use of joint liability by microfinance institutions, combined with recent evidence that a) converting joint liability groups to individual liability groups did not affect repayment rates, and b) an intervention that increased social capital in individual liability borrowing groups led to improved repayment performance. First, we show that individual lending with or without groups may constitute a welfare improvement over joint liability, so long as borrowers have sufficient social capital to sustain mutual insurance. Second, we explore how the lender’s lower transaction costs in group lending can encourage insurance by reducing the amount borrowers have to pay to bail one another out. Third, we discuss how group meetings might encourage insurance, either by increasing the incentive to invest in social capital, or because the time spent in meetings can facilitate setting up insurance arrangements. Finally, we perform a simple simulation exercise, evaluating quantitatively the welfare impacts of alternative forms of lending and how they relate to social capital.
- Take what you can: property rights, contestability and conflict, joint with Samuel Marden, forthcoming, Economic Journal.Weak property rights are strongly associated with underdevelopment, low state capacity and civil conflict. In economic models of conflict, outbreaks of violence require two things: the prize must be both valuable and contestable. This paper exploits spatial and temporal variation in contestability of land title to explore the relation between (in)secure property rights and conflict in the Brazilian Amazon. Our estimates suggest that, at the local level, assignment of secure property rights eliminates substantively all land related conflict, even without changes in enforcement. Changes in land use are also consistent with reductions in land related conflict.
- Commercialization and the Decline of Joint Liability Microcredit, 2015.Numerous authors point to an apparent decline in joint liability microcredit, and rise in individual liability lending. But empirical evidence is lacking, and there have been no rigorous analyses of possible causes. In this paper, we first show using the well-known MIX Market dataset that there is indeed evidence for a decline. Second, we show theoretically that a plausible cause is commercialization: an increase in competition and a shift from non-profit to for-profit lending, both of which are present in the data, drive lenders to reduce their use of joint liability loan contracts. Third, we test the model’s key predictions, and find support for them in the data. Commercialization does indeed seem to be a contributor to the decline of joint liability.
- Social Insurance and Conflict: Evidence from India, EOPP Working Paper Number 53, 2014.Can public interventions persistently reduce conflict? This paper studies whether social insurance is effective in reducing conflict. Adverse income shocks have been empirically and theoretically identified as robust drivers of conflict. An effective social insurance system moderates the impact of adverse shocks on household incomes, and hence, could attenuate the link between these shocks and conflict. This paper shows that a public employment program in India provides social insurance. The program guarantees 100 days of employment at minimum wages providing an alternative source of income following bad harvests. This has an indirect pacifying effect. By moderating the link between productivity shocks and incomes, the program uncouples productivity shocks and conflict. An earlier version entitled “Can Workfare Programs Moderate Violence? Evidence from India” was circulated as Working Paper in October 2013 and as EOPP Working Paper Number 53, 2014.
- Fracking Growth, CEP Working Paper 1278, 2014.This paper estimates the effect of the shale oil and gas boom in the United States on local economic outcomes. The main source of exogenous variation to be explored is the location of previously unexplored shale deposits. These have become technologically recoverable through the use of hydraulic fracturing and horizontal drilling. I use this to estimate the localised effects from resource extraction. A key observation is that, despite rising labour costs, there is no Dutch disease contraction in the tradable goods sector, while the non-tradable goods sector contracts. I reconcile this finding by providing evidence that the resource boom may give rise to local comparative advantage, through locally lower energy cost. This allows a clean separation of the energy price effect distinct from the local resource extraction effects.Media coverage
- Royal Economic Society Media Briefing, Fracking Boosts Jobs, Wages and Energy-Intensive Manufacturing: Evidence from United States, April 2014.
- Britain in 2015, ESRC, Interview, November 2014.
- On the Comparative Advantage of U.S. Manufacturing: Evidence from the Shale Gas Revolution, joint with Rabah Arezki. CEP Working Paper 1399, 2016, OxCarre Working Paper, CAGE Working Paper 259 This paper provides empirical evidence of the newly found comparative advantage of U.S. manufacturing following the so-called shale gas revolution. The revolution has led to (very) large and persistent differences in the price of natural gas between the United States and the rest of the world owing to the physical properties of natural gas. Estimation results of gravity models show that the U.S. manufacturing exports have grown by about 6 percent on account of their energy intensity since the onset of the shale revolution. We also document that the U.S. shale revolution is operating both at the intensive and extensive margins with new manufacturing sector capacity being added in the energy intensive industries. These results are robust to an array of checks.
- Market Structure and Borrower Welfare in Microfinance, joint with Jon de Quidt and Maitreesh Ghatak.Motivated by recent controversies surrounding the role of commercial lenders in microfinance, we analyze borrower welfare under different market structures, considering a benevolent non-profit lender, a for-profit monopolist, and a competitive credit market. To understand the magnitude of the effects analyzed, we simulate the model with parameters estimated from the MIX Market database. Our results suggest that market power can have severe implications for borrower welfare, while despite possible information frictions competition typically delivers similar borrower welfare to non-profit lending. In addition, for-profit lenders are less likely to use joint liability than non-profits.
- More than an Urban Legend: The long-term socio-economic Effects of Unplanned Fertility Shocks, joint with Amar Shanghavi and Oliver Pardo. An earlier version of the paper had the title “An Urban Legend?! Power Rationing, Fertility and its Effects on Mothers”, and circulates as CEP Working Paper 1247, 2013.
This paper studies one dimension of the social cost of bad public infrastructure in developing countries. We use an extensive period of power rationing in Colombia throughout 1992 as a natural experiment and exploit exogenous spatial variation in the intensity of power rationing as an instrumental variable. We show that power rationing induced a “mini baby boom” nine months later. In particular, it increased the probability that a mother had a baby by five percent. We estimate that every tenth power outage baby was not adjusted for 12 years later, resulting in an overall increase in total fertility. This increase has indirect social costs, as women who were exposed to the shock and had an additional child find themselves in worse socio-economic conditions more than a decade later.
Work in Progress
Energy- and Resource Economics
- Energy Density and Local Comparative Advantage: Shale Gas Boom and the Allocation of
Industrial Activity,This paper studies the impact of energy density and relative transport cost on the allocation of industrial activity. Using the recent shale gas boom in the United States as a source of exogenous variation the paper demonstrates that the existing natural gas pipeline network in conjunction with the geographically shifting center of natural gas production creates distinct spatial variation in comparative advantage for energy intensive sectors. A reduced form measure of comparative advantage is obtained by simulating minimum cost network flows using mixed-integer linear programming routines. The measure correlates well with local natural gas prices and predicts growth in energy intensive sectors across the US. The incidence is driven by locations that were former net natural gas exporters.
- The Redistributional Effects of Environmental Policy: Evidence from Germany joint with Guo Xu and Christian Westermeier.Electricity consumption is relatively price inelastic and price increases can have redistributional effects. This paper analyzes the redistributional effects of renewable energy feed in tariffs in Germany on two margins. We study the direct redistributional effects due to spatial price heterogeneity exploiting variation due to the relative intensity of the expansion of solar versus wind power. The second margin studies the direct effect as ownership of assets are concentrated at the top-end of the wealth distribution.
- The Persistent effects of Private vs. Colonial Rule: Evidence from 19th century Indonesia, joint with Priya Mukherjee, Project funded with a grant from the History Project In this research project, we document the interplay between private enterprise and the state during the Dutch rule of Indonesia in the 1800’s, and study its long-term effects on the economic well being of the population on the island of Java. Combining detailed historical data and modern surveys, we create a unique panel dataset that allows us to compare regions that were controlled by private enterprises with those that were under direct Dutch rule. We exploit the fact that the timing of the establishment of the private estates as well as the duration of private rule was arguably exogenous to other factors, and use a regression discontinuity design to compare geographical units on either side of the private-public plantation borders over time.
- Weather, Income Shocks, and Crime: Evidence from India, joint with Jon Colmer and Benjamin Guin. Slides available upon request. This paper disentangles the direct and indirect effects of weather variation on crime in India. Evidence from behavioral psychology suggests that weather shocks, in particular temperature variation, may have a direct effect on criminal behavior, especially for violent crimes. On the other hand, the economics literature has widely emphasized the role of economic shocks in affecting crime. We attempt to separate these effects by examining variation across crimes, seasonal variation related to agricultural production, rural vs. urban differences and whether temperature effects are mitigated in states with strict prohibition laws. We find that an increase in the number of days in which temperature rises above 30 degrees C is associated with an increase in the number of murders and rapes, with no effect on other economic-based crimes. Furthermore, an increase in the number of days in which individuals are exposed to very hot temperatures (above 34 degrees C) is associated with a reduction in these crimes. Finally, we observe that increased rainfall is associated with reductions in property based and economic crimes in support of the needs based channel.
- Estimating Travel Cost through Mobile Phone Use Data, joint with Amadou Sy (Brookings). Blog Post about Big Data for Development at Brookings
Transport costs are of several orders of magnitude larger in developing countries (see Atkin and Donaldson (2014)). Rapid urbanization in Africa has created spatial agglomerations, while infrastructure has failed to keep pace. In this paper we study how mobile phone usage datasets can be used to estimate travel times in Dakar, Senegal. We show how the construction of a toll road affected the spatial distribution of travel times and distance traveled.
- Developing a Micro-Level Conflict Data Set for South AsiaThis paper describes the use of sophisticated natural language processing tools to develop a district-level conflict dataset for South Asia, with a particular focus on India. The methodology used requires only a limited amount of work of hand-coding data. As the procedure is objective, erroneous coding of data by researchers can be avoided or at least limited, using such automation techniques. The ideal combination would leverage human ability to discover nuances in textual notions, with the scalability and objectivity of automated procedures. I then show how the dataset correlates with the Global Terrorism Database (GTD), another major conflict database measuring terrorist or insurgency violence. I highlight that there appear to be systematic differences between these two datasets. These differences likely arise from the primary data sources that feed into the GTD.