My research is seated in environmental and resource economics, including nonmarket valuation, hedonic price analysis, and benefit-cost analysis of government programs. A driving force behind my research is a desire to increase our understanding of the linkages between humans and the environment, and to examine how policies that affect these linkages distribute benefits (and costs) to subgroups of people.
Climate Change and the U.S. Market for Snow
Current Version Here
Job Market Paperwith Peter Christensen (University of Illinois Urbana-Champaign)
Many mountain towns rely on climate amenities such as wintertime precipitation to generate local economic activity. However, climate models predict large reductions in annual snowfall that could greatly reduce tourism flows to these markets. Harnessing a unique panel of daily transactions from the short-term property rental market, we combine daily weather, daily resort snowpack, and daily resort snowfall to estimate the causal effect of changes in snowpack on visitation in 219 resort markets across the United States. We make three primary contributions to the study of climate change: 1) we develop a new method to estimate elasticities for climate amenities by matching the spatial and temporal variation in the level of the amenity with the frequency of related market transactions; 2) we derive state-specific snowpack elasticities for all major markets across the United States and find significant heterogeneity in the behavioral response across states; and 3) we estimate year-to-year variation in the recreation revenue from snowpack under current and future climate scenarios. We predict that resort markets could face reductions in local snow tourism of -40% to -80%, almost twice as large as previous estimates suggest. This translates to a lower-bound on the annual willingness to pay to avoid reductions in snowpack between $1.55 billion (RCP4.5) and $2.63 billion (RCP8.5) by the end of the century.
Local Benefits and Willingness to Pay to Reduce Hypoxia in the Gulf of Mexico
Current Version Here
R & R 2019 (Land Economics)with Amy W. Ando (University of Illinois Urbana-Champaign)
Improvements in local surface water quality can contribute to regional environmental goals such as reducing hypoxia in the Gulf of Mexico. However, an environmental policy or program that affects both rural and urban populations might be inequitable if the policy benefits one group more than the other. We estimate willingness to pay for changes that would improve local water quality and reduce hypoxia far downstream, and test for differences between rural and urban residents. We find both groups have similar preferences, and simulations suggest rural areas of the watershed stand to benefit two to three times more than urban clusters.
The Recreation Demand for Snowpack
Paper in Progresswith Peter Christensen (University of Illinois Urbana-Champaign)
We estimate utility functions for winter recreationists in 236 resort markets across the United States to estimate the marginal willingness to pay for climate amenities such as snowpack, temperature, and precipitation. In order to address a dimensionality problem in the computational estimation of the likelihood function, caused by the large choice set that the consumer faces when making a decision of where and when to make a trip, we randomly sample from the available set of outside options to relax restrictions surrounding conventional (conditional) logistic estimation (IIA). Using this approach, we recover a distribution of preference parameters for climate amenities that allow for more realistic substitution patterns when solving the consumer's problem. However, simply allowing for substitution only provides part of the story. We extend the analysis by incorporating the demand estimation framework of Berry, Levinsohn, and Pakes (1995) to estimate own and cross-snowpack elasticities. We recover a substitution matrix which provides us with the ability to refine estimates of damages under future climate scenarios by predicting substitution patterns of snow tourists. We find that failing to account for spatial and temporal substitution misestimates the spatial distribution of welfare, but aggregate welfare is not significantly different under the two assumptions. We also find that welfare damages are substantially larger than simple estimates of lost revenues that we derive in the first paper. We estimate that by end of the century (2100) total welfare losses across the U.S. are between $9.35 billion and $14.2 billion per ski season due to reductions in mountain snowpack. Site substitution is an important phenomenon, but will not reduce the overall threat that climate change poses to this ecosystem service.
Willingness-to-Volunteer and Stability of Preferences between Cities: Estimating the Benefits of Stormwater Management
Current Version Here
Forthcoming in Journal of Environmental Economics and Management JEEMwith Catalina Londoño Cadavid, Amy W. Ando (University of Illinois Urbana-Champaign), and Noelwah R. Netusil (Reed College)
Urbanization places pressure on existing stormwater systems, yielding high flood rates, degraded urban aquatic habitat, and low water quality in lakes and rivers. Because of their lower cost and multiple benefits, cities increasingly rely on decentralized stormwater facilities, such as green streets and bioswales, that use citizen volunteers to help with maintenance. This paper uses a choice experiment survey in two major U.S. cities to estimate the benefits associated with stormwater management improvement in terms of both willingness to pay money and willingness to volunteer time, and to test how stable preferences for these environmental goods are across cities. We find that preferences are somewhat stable across cities. We also find that people are willing to devote more time volunteering to provide amenities in their neighborhoods than one would expect from their wages and the monetary values they place on environmental goods because they gain positive utility from volunteering.
Benefits of a Fire Mitigation Ecosystem Service in the Great Dismal Swamp National Wildlife Refuge
Current Version Here
Published January 2017 - Journal of Environmental Managementwith Emily Pindilli, and Diana Hogan (U.S. Geological Survey)
The Great Dismal Swamp (GDS) National Wildlife Refuge delivers multiple ecosystem services, including air quality and human health via fire mitigation. Our analysis estimates benefits of this service through its potential to reduce catastrophic wildfire related impacts on the health of nearby human populations. We used a combination of high-frequency satellite data, ground sensors, and air quality indices to determine periods of public exposure to dense emissions from a wildfire within the GDS. We examined emergency department (ED) visitation in seven Virginia counties during these periods, applied measures of cumulative Relative Risk to derive the effects of wildfire smoke exposure on ED visitation rates, and estimated economic losses using regional Cost of Illness values established within the US Environmental Protection Agency BenMAP framework. Our results estimated the value of one avoided catastrophic wildfire in the refuge to be $3.69 million (2015 USD), or $306 per hectare of burn. Reducing the frequency or severity of extensive, deep burning peatland wildfire events has additional benefits not included in this estimate, including avoided costs related to fire suppression during a burn, carbon dioxide emissions, impacts to wildlife, and negative outcomes associated with recreation and regional tourism. We suggest the societal value of the public health benefits alone provides a significant incentive for refuge mangers to implement strategies that will reduce the severity of catastrophic wildfires.
Works in Progress
You are here: Bringing New Life, and Methods, to Stated Preference Research
Environmental goods can be valued with stated or revealed preference methods. Some researchers have raised concerns about stated preference methods, such as choice experiments, surrounding hypothetical scenarios, respondent panels, unobservables, measurement error, internal/external validity, and modeling selection. We address these concerns to generate a reproducible and generalizable approach for maximizing the validity of stated preference tools. We develop a workflow that bridges theory with practice by employing a wide range of tools and novel software integrations (Stata, ArcPy, MS Office Suite, python, AWS, Qualtrics.com, html, and R) in order to address the concerns listed above, while providing a transparent and freely available toolbox that can be used in a variety of settings. Lastly, we carry out an exercise in valuation to explore preferences and urban/rural preference heterogeneity for water quality in Midwest river systems. We show that even in the world of big data, stated preference research is still a useful and competitive tool for valuing nonmarket environmental amenities.