Browsing by Author "Jeuland, Marc Allan"

This paper discusses results from a study of the use of cleaner cooking solutions and general health status of people in rural areas from the Battambang province of Cambodia. Data collection included 372 demographic, health and socio-economic surveys with households living in 6 villages in the Samlout district, general health examinations, and measurements of stove use and household concentrations of PM 2.5. The data reveal that health in this population is a major concern, with a very high prevalence of reported abdominal pain, nausea, chronic cough, chest pains, and fever during examinations.  At the household level, we find that clean stove ownership is significantly correlated with the educational status of household head and socio-economic status of a household. Respondents from households with clean stoves appear less likely (though not statistically significantly so) to report household individuals having health problems such as occasional cough, high blood pressure and tuberculosis. Concentrations of PM2.5 are positively correlated with prevalence of occasional cough, high blood pressure and tuberculosis. Based on these results, we advise field testing and evaluation of targeted health interventions in these villages to address the numerous concerns of the local population, including exploring the potential role of clean stoves.

Cooking fuel “stacking,” or the use of multiple types of fuels, can be problematic in interventions when households are using both clean and dirty fuels at the same time. Dirty fuels such as firewood contribute to indoor air pollution, cause detrimental health effects, and are inefficient forms of energy. In this study, cooking fuel preference data was collected from 360 households in peripheral-urban Kathmandu, Nepal during August 2019. Respondents provided fuel information and gave economic preferences for a contingent valuation experiment on their reported primary fuel type. We explored two aims through multiple regression analyses: the relationship between fuel stacking behavior and willingness to pay (WTP), and the household characteristics associated with fuel stacking behavior. The analyses showed that stacking does not affect WTP, and household expenses are a significant factor associated with WTP only among households using LPG as their primary fuel. The secondary aim found that the main household characteristics associated with fuel stacking are household size, firewood gathering behavior, and if the household was affected by the 2015 LPG blockade. The relationships of these characteristics are complex and depend on whether the household is using more LPG or more firewood when stacking. More research is needed to better understand fuel stacking, and why most people in peri-urban Kathmandu prefer LPG as their primary fuel.

This paper utilizes a mixed methods approach to evaluate household demand for reduced diarrhea risk through an assessment of willingness to pay (WTP) for water and sanitation technologies in Udaipur, India. Using data from a survey of 900 households, responses to a contingent valuation scenario were analyzed using multivariate logit regression to determine WTP for a water purification device. Furthermore, transcripts from 10 focus groups were analyzed for specific determinants of WTP for a ceramic filter, an open well chlorination tank, and a composting latrine by identifying preferences based on the frequency and type of responses given by group participants. The mean WTP estimated from the logit model was $1.03; this is the monthly WTP amount per person in this sample. The specific determinants for WTP include the randomized price levels, and factors like education level and household socio-economic characteristics. In addition, the qualitative work did not reveal a clear preference by all groups for any one of the three offered technologies, but demonstrated the influence of factors such as perceived health improvement and increased convenience on demand. Understanding what drives demand for these technologies could help implementers in designing more effective behavior change interventions, educational campaigns or social marketing schemes.

What farmers do matters. Even as the world’s population is increasingly urbanized, almost a billion people make their living in agriculture (World Bank, 2018) and 5 billion acres (38%) of the world’s land area is farmland (FAO, 2020). Moreover, agriculture accounts for a significant portion of global deforestation, forest fragmentation, and habitat loss (Defries, Rudel, Uriarte, & Hansen, 2010; Rudel, Defries, Asner, & Laurance, 2009), while consuming 70% of global freshwater and contributing substantially to water pollution (FAO, 2011). The sheer size of the land mass managed by smallholders means that interventions intended to improve environmental quality, sequester carbon, or foster economic development must reckon with the preferences and responses of rural landholders if they hope to meet their goals. Development economics is home to a rich literature modeling the actions and reactions of farmers, motivated by a desire to boost agricultural productivity and reduce rural poverty rates. Environmental economists have also contributed to an understanding of rural preferences and incentives as conservationists have increasingly seen the value of enlisting private land to increase wildlife habitat, sequester carbon, and provide ecosystem services. This dissertation examines farmer’s incentives, preferences, and interactions in relation to three policies in three different settings, with an eye to improving program design for agro-environmental policy in rural developing countries. Chapter 1 examines a change in water quality due to the expansion of a water treatment facility upstream of the Jordan Valley—the country of Jordan’s major agricultural region. I examine the substitutability of recycled water for freshwater, examining farmers’ water use and production using a model of agricultural production and exploiting the quasi-experiment to see how water use changes in the affected area. While I find no evidence that farm production was significantly impacted by the change—good news for urban water consumers whose water supply was increased by this substitution of recycled water in the agricultural sector—I do find that elements of water governance could improve the acceptability of the policy change from the farmers’ perspective. Chapter 2 examines farmer interactions when faced with a Payment for Ecosystem Services (PES) program in Central India. Specifically, it asks if program targeting should change if farmers’ acceptance of the program depends’ on their neighbors’ actions. Using the results of a choice experiment, I simulate private interactions of farmers’ across the landscape surrounding Pench national park, and place zones in priority order by total welfare both with and without private interactions. Consideration of private interactions changes both the priority order of the zones and the number of zones which pass a cost/benefit analysis. Chapter 3 asks the question, does land tenure registration cause, or prevent deforestation? Utilizing the implementation of a land registry called a DUAT in norther Mozambique, I observe land parcels before and after registration and estimate the effect on tree cover loss as measured by the Hansen et al. forest cover loss dataset (Hansen et al., 2013). I also show the results of a tree crown detection algorithm intended to estimate the presence of individual trees in agricultural mosaic landscapes. Findings indicate a modest increase in tree cover loss on parcels that are registered communally, whereas there is a modest decrease for parcels registered to individuals. The impact of DUAT registration also varies by pre-existing tree cover and population. These findings recommend caution regarding the possible environmental benefits of land registration and tenure reform, highlighting heterogeneity by both tenure type and pre-existing land cover.

Global energy-use patterns are characterized by deep inequality. Electricity is indispensable for households, clinics, schools and firms, yet over a billion people live without it. At the same time, nearly three billion rely on traditional stoves and polluting biomass fuels (such as firewood) for their basic energy needs. The resulting household air pollution causes four million deaths annually, a health burden borne disproportionately by women. The international community has hastened to respond to this global energy challenge. This dissertation highlights how—and under what conditions—policies that seek to ensure universal access to modern energy deliver expected environmental and development benefits.

In the first chapter, I ask what drives heterogeneity in the impacts of large-scale rural electrification. Prior evidence on the labor-market impacts of grid electrification is mixed. I hypothesize that variation in local economic conditions—which can complement investments in infrastructure—may help explain why, and combine two natural experiments in India within a regression discontinuity design to test this hypothesis. Most of the world's guar, a crop that yields a potent thickening agent used during hydraulic fracturing ("fracking"), is grown in northwestern India. The rapid rise of fracking in the United States induced a parallel commodity boom in Indian guar production, resulting in a large positive shock to rural economic activity. Leveraging population-based discontinuities in the contemporaneous roll-out of India's massive rural electrification scheme, I show that access to electricity significantly increased non-agricultural employment in villages located in India's booming guar belt. Where these complementary economic conditions were lacking, electrification had almost no discernible impact. Using a firm-level panel dataset, I then provide suggestive evidence that this growth in non-farm work is partly driven by the rise of electricity-intensive firms that complement agricultural production. In line with the prior literature, I show that electrification alone may not be sufficient to deliver economic benefits, but I also demonstrate that, when combined with complementary economic conditions on the ground, access to electricity can enable individuals, households and firms to take advantage of new opportunities in potentially welfare-enhancing ways.

In the second chapter, I turn to household-level energy use and empirically evaluate the role played by non-governmental organizations (NGOs) in delivering environmental, energy and development interventions in remote, rural settings. I develop a model of household decision-making to evaluate how NGOs address implementation-related challenges and influence intervention effectiveness. To test the model's predictions, I apply quasi-experimental methods to household-survey data from a randomized controlled trial designed to promote clean-cooking solutions in rural India. I uncover a large, positive and statistically significant "NGO effect": prior engagement with the implementing NGO increases the effectiveness of the intervention by at least thirty percent. These findings provide some of the first causal evidence on how NGOs directly influence outcomes, which has implications for the generalizability of experimental research conducted jointly with such local partners. In particular, attempts to scale up findings from such work may prove less successful than anticipated if the role of NGOs is insufficiently understood. Alternatively, policymakers looking to scale up could achieve greater success by fostering partnerships with trusted local institutions.

In the final chapter, I consider how heterogeneity in households' preferences influences demand for energy technologies. I conduct technology-promotion campaigns followed by second-price, sealed-bid ("Vickrey") auctions for two cleaner cooking technologies with over 1,000 households across seventy communities in rural Senegal. I induce exogenous variation in the extent to which these promotion activities cater to heterogeneous preferences by randomly assigning a subset of communities to an auction arm in which both devices are promoted jointly. Consistent with a model in which preferences are constructed—and not simply revealed—as agents make repeated choices, joint promotion lowers willingness to pay for the relatively less familiar alternative compared to settings in which the two devices are promoted exclusively. Rather than simply providing additional choices, implementers looking to enhance uptake of improved technologies must instead devise approaches to help potential end-users think carefully through trade-offs, crystallize and understand their own preferences, and identify solutions that fit their needs.