Presentations begin at 12pm SHARP and conclude at 1pm in Mulford Hall Room 240.
Spring 2019 schedule
16 January 2019
Presenter: Megan Lang
Title: “Understanding rural demand for solar energy in Rwanda and Kenya: Price versus flexibility.”
Abstract: While evidence suggests that demand for electricity among low income, rural consumers in developing countries is highly sensitive to price on the adoption margin, little evidence exists on the importance of price versus non-price factors on the use margin. I partner with a pay as you go (PAYG) solar company to investigate the importance of price and liquidity in shaping demand for electricity on the use margin in Rwanda and Kenya. I randomly offer solar customers incentives for using their solar home systems, varying the liquidity required to qualify for incentives as well as the price discounts customers receive when they qualify for incentives. This random variation enables me to measure consumer responsiveness to price and liquidity. I find that neither price nor liquidity substantially alters consumer demand for solar on the use margin. Finally, I present descriptive evidence from daily customer use data that suggests intertemporal substitution of energy consumption likely plays only a modest role in explaining my experimental results.
23 January 2019
Presenter: Andy Hultgren
Title: “Lobbying and R&D in the face of policy uncertainty”
Abstract: The field of economics invests extraordinary effort in measuring externalities and recommending appropriate policy corrections. Yet, the role of firms as strategic actors in the design and implementation of externality-correcting policy is rarely considered. Firms which are negatively affected by a potential policy may seek to profit from it by investing in R&D, or they may seek to kill the policy by investing in lobbying of government officials. In this paper, I present a simple model of firm behavior under information shocks on production externalities, in which firms can invest in R&D and/or lobbying in response to the shock. The model describes two key testable predictions regarding how firms will respond to externality-related information shocks. Empirically, information shocks are represented by scientific discoveries of adverse health effects associated with commonly used products. Early-stage empirical results on the R&D response suggest that R&D spending does indeed increase in response to an information shock of this type.
30 January 2019
Presenter: Peiley Lau
Title: “Estimating non-point source pollution in New Zealand waterways”
Abstract: Economic mechanisms that solve the Tragedy of the Commons are impeded in situations where polluters are anonymous. The effectiveness of Coasean bargaining, Pigouvian taxes, or cap-and-trade policy ultimately rests on the ability of a market or regulators to observe the pollution emissions of individual agents and assign responsibility for creating an externality. Unfortunately, in a large number of cases, policymakers are only able to observe aggregate emissions, such as the total flow of pollutants into a system, and thus struggle to design effective policies. Here we develop a scalable architecture that exploits existing public databases to identify local land-based sources of runoff pollution in river networks. We demonstrate the generalizability of our approach by allocating responsibility for the anoxic “Dead Zone” in the Gulf of Mexico to counties across the entire Mississippi River catchment and identifying highly localized pollution sources across the entire country of New Zealand, where reducing agricultural runoff is a policy priority. The output of our analysis also represents a novel data base of riverine pollution sources that can be used to study a variety of economic questions about the causes and consequences of pollution at a national scale. Furthermore, real-time extension of these data can be used to evaluate the effectiveness of modern policy interventions and to monitor compliance with current regulatory systems.
Presenter: Derek Wolfson
Title: “The effect of transportation network companies on alcohol consumption and drunk driving”
Abstract: Transportation network companies (TNCs), like Uber and Lyft, are changing the way consumers move around cities and their peripheries. This transformation in the way people consume transportation services has the potential to transform markets for other goods and services that require transportation to enjoy. This paper will focus on a distinct market that TNCs are likely to affect: alcohol consumption. The current literature on this topic assume Uber entry is exogenous, which I show is likely not the case. I propose solving these threats to identification by using unexpected firm exit instead of entry to properly identify the effects of Uber on drinking-related offenses and traffic fatalities. My results on fatal drunk accidents are inconclusive. The coefficient estimates suggest that TNCs may decrease drunk driving and drinking related fatalities, but these estimates are very imprecise so I cannot rule out large positive or negative effects. However, my results suggest that alcohol tax receipts decrease by approximately 6% when Uber is banned, suggesting that consumers may be visiting bars and nightclubs less frequently when they do not have access to TNCs
6 February 2019
Presenter: Molly Van Dop
Title: “Pricing groundwater in the Pajaro Valley: A saltwater intrusion story”
Abstract: Located on the Central Coast of California, the Pajaro Valley is an internationally recognized region for the production of berries and fruit. Growers in the valley rely primarily on groundwater to irrigate their crops. However, seawater intrusion due to the reduction of the groundwater table has led to decreased water quality along the coast and threatens water quality across the entire watershed. Following the guidelines of Proposition 218, the Pajaro Valley Water Management Agency has implemented a second-best pricing system for groundwater, where farmers pay for recharge facilities and the distribution of recycled water along the coast as a measure to reduce overdraft. Using a regression discontinuity design, we study how this spatially and time-varying pricing system affects land prices and water quality for the region. We discuss implications for the effectiveness of future recharge facilities on minimizing groundwater overdraft.
Presenter: James Sears
Title: “Forget it, I’ll Just Telecommute: the Value of Time for San Francisco Bay Area Commuters”
Abstract: As residents of one of the world’s most congested regions, San Francisco Bay Area commuters face difficult and potentially costly options when choosing between available transit modes. How individuals value travel time plays a key role in mode choice, and stands to play a major role in the introduction and proliferation of automated vehicles. Utilizing data from a recent survey of Bay Area commuters, we specify a discrete choice model of commute mode to estimate the value of travel time across a range of regional transit options. We explore the role of heterogeneity in commute type and commuter preferences to better understand the determinants of commute mode choice. Finally, we simulate the introduction of individual-owned and shared automated vehicles to comment on potential changes in welfare.
13 February 2019
Presenter: Katherine Wagner
Title: “Crowding-In with impure altruism: Theory with evidence from volunteerism in national parks”
Abstract: How does government provision of a public good affect private provision? The extent to which public supply crowds out private provision is the standard test between the models of pure and impure altruism, which provide different explanations for why people contribute to public goods. We show that this commonly used specification test relies implicitly on functional form assumptions that eliminate the possibility for government provision to crowd in private provision. Relaxing these assumptions yields two main theoretical findings. First, crowding-in is admissible in the standard models of privately provided public goods. Second, specification tests based on the extent of crowding-out are problematic, but tests based on crowding-in are more well-founded. We test the implications of the theory with a unique and comprehensive data set on volunteer hours in the National Park Service (NPS). We find statistically significant and economically meaningful crowding-in. Specifically, we estimate that an additional dollar of federal funding within a park crowds in 25 cents-worth of additional volunteerism. Consistent with theory, estimates of crowding-in are greater for parks and activities that are more likely characterized by joint production and thus impure altruism. The results contribute to our understanding of how policies may promote the efficient provision of public goods and provide a more robust test between models of the underlying motives of charitable behavior.
27 February 2019
Presenter: Ellen Lan Lin
Title: “Pollution and Fertility Intentions: Evidence from Online Purchases”
Abstract: China now has one of the world’s lowest birth rates, even after the government lifted the one-child policy. Survey results show that concerns over environmental conditions is an important factor affecting child-bearing decisions. In this paper, I use online purchases of products related to child birth to study whether and how pollution level affects fertility intentions. Using a distributed lag model and exploiting random variation in air pollution stemming from thermal inversions, I find that people have lower desire to have more children when air pollution worsens. Long-term change in air quality has stronger impact than contemporaneous fluctuations. I also find limited heterogeneity between high and low income cities, and that the negative relationship between fertility intentions and pollution is unlikely due to financial worries such as higher health care expenditure.
3 April 2019
Presenter: Hannah Druckenmiller
Title: “The impact of climate change on ecosystem services: Evidence from U.S. forests”
Abstract: The scientific literature documents large reductions in the health of U.S. forests over the last century, with tree mortality rates doubling every 20 years. In the coming decades, warming temperatures are expected to put further stress on forest health by increasing water deficits and enhancing the growth of insects and pathogens that attack trees. While declines in forest health are well documented, it is not well understood how these changes affect the social and economic benefits humans derive from forest ecosystems. Using an instrumental variables (IV) approach, this study investigates how climate-driven changes in forest health affect the provision of ecosystem services in the United States. First, I isolate plausibly random changes in forest health by exploiting a biological feature of bark beetles – a leading cause of tree mortality in the U.S. Second, I estimate the impact of changes in forest health on a wide array of ecosystem services, including the direct products obtained from trees such as timber, pulp, and fiber (provisioning services), the indirect benefits that trees provide through modification of the environment such as water purification and carbon sequestration (regulating services), and non-material benefits that people derive from forests such as recreation and aesthetic values (cultural services).
10 April 2019
Presenter: Susanna Berkouwer
Title: “Behavioral barriers to energy efficiency in development contexts: Evidence from cookstoves”
Abstract: We seek to quantify how behavioral biases and market frictions affect poor households’ adoption of energy efficient durables. While a substantial literature studies how behavioral anomalies and failures of rationality—self-control problems, incorrect beliefs, and limited attention—contribute to the energy efficiency gap in the US (see Gillingham and Palmer for a review), little work exists in a development setting. This is important because biases might operate differently in poor environments. It could be that biases are exacerbated by stress or cognitive load associated with poverty (see Haushofer and Fehr for a review), or it could be that because poverty makes large, durable purchases high-stakes, poorer individuals are more likely to make decisions more carefully. Market frictions common in developing settings, like liquidity constraints and asymmetric information, might further exacerbate any failures of rationality. We study this question in the context of energy efficient charcoal cookstoves. In a pilot RCT with 150 low-income households in Nairobi in 2018 we estimated that ownership of the energy efficient stove reduced weekly energy spending by 60% relative to control group spending ($2.50 per week, or $130 per year), but average WTP for the stove was only $15. We employ a 3x3 RCT design to understand how inattention and credit constraints affect adoption, and we use lab-in-the-field tests to measure present bias and risk preference parameters. This will be the first paper to rigorously quantify the energy efficiency gap, causally identify the mechanisms driving this gap, and estimate its welfare effects, and it will do so in a high-stakes development setting.