Brief Summary of Minutes of Annual Meeting

Brief summary of minutes of annual meeting:

The business meeting portion of the workshop included Amy Ando, Kathleen Bell, Steven Dundas, Ben Gramig, Matt Interis, Sonja Kolstoe, Craig Landry, Lynne Lewis, Frank Lupi, Dale Manning, Don McLeod, Klaus Moeltner, Jerrod Penn, Kim Rollins, Brent Sohngen, Brian Vander Naald, Roger von Haefen, Ken White, Seong Yun.

First order of business was to determine a new Secretary, whose term begins in 2020, and who follows a three-year cycle continuing on to Vice-President and President in the two subsequent years. Don McLeod nominated Steven Dundas and Amy Ando seconded. Steven agreed to stand. There were no additional nominations and Steven was installed by unanimous vote.

Don noted his appreciation for the new presentation format this year in which there was 15 minutes of general discussion and questions for any presenter at the end of each session. Others generally agreed.

Ken White, the new USDA liaison for the project, will notify the group of any upcoming deadlines.

Current Vice-President Craig Landry proposed Athens, GA as the site of the 2020 meeting and others seemed agreeable to this suggestion.

Amy Ando expressed she’d like the project members & friends to do a better job of communicating with each other throughout the year between meetings. She volunteered to set up a project Slack channel with Bryan Parthum (her student). Slack is a business sharing and collaboration application; the members were generally supportive of this idea on a trial.

At the conclusion of this meeting, Dale Manning concludes his three-year officer rotation (thank you, Dale!), Craig Landry rotates from Vice-President to President, Matt Interis rotates from Secretary to Vice-President, and Steven Dundas begins his term as Secretary.

Accomplishments:

Colorado State: We have begun an evaluation of the Conservation Reserve Enhancement Program in Kansas.  This includes working papers on spatial spillovers and the valuation of conserved groundwater. 

We have used USDA field level crop insurance data, merged with groundwater use at wells in KS, NE, and CO to examine the impact of crop insurance price on groundwater use.  We are currently finishing a manuscript for this work and examining how GW availability affects insurance demand and how insurance and groundwater restrictions interact. With colleagues at the Univ of Idaho, we have also published work on the complementarity between electricity and water at the household level.  This has important implications for the management of water across sectors in arid regions. We also examined the impact of learning on incentives to temporarily ban fracking in local communities. Finally, with collaborators at UN-R, we have begun to develop and link models of surface water flows under climate change with economic models of water value.  We are using this framework to examine the costs and benefits of alternative water allocation institutions in adapting to these changing flows.

Colorado State: We have completed a draft of a manuscript that proposes a stated preference method for linking a valuation function to a model of a natural system in order to examine the impacts of USDA policy on natural capital values. We apply this to the valuation of the CREP program in KS and its impact on market and non-market groundwater values.

Colorado State: In collaboration with others at the Univ of NE, we have completed a linked hydro-economic model of Finney County in Kansas.  This model demonstrates novel methods to link agronomic, hydrologic and economic models for policy evaluation.

North Carolina State: Received grant: von Haefen, Roger H. (PI) “Estimating the Benefits of Stream Water Quality Improvements in Urbanizing Watersheds: An Ecological Production Function Approach,” US EPA, $800,000, 2015-2020.

Oregon State: Dundas and collaborators (incl. w4133 member Roger von Haefen) published a paper that uses econometric methods to estimate non-market recreation costs of a land-use policy to protect coastal biodiversity. Results suggest the economic costs of the policy are relatively modest, ranging from $403,000 to $2.07 million annually and provide general support for the National Park Service's recently implemented off-road vehicle management plan in Cape Hatteras National Seashore. Lewis and a collaborator published a paper that develops an auction mechanism to implement the optimal provision of ecosystem services under climate change. The auction mechanism can be used as a subsidy for conservation (such as a payment-for-ecosystem services), or as a tax on using land. Lewis, Dundas and Kling developed a new choice experiment survey instrument to assess the non-market values of Pacific Northwest residents on restoration of threatened Oregon Coast Coho salmon runs. Kling and Dundas developed a new choice experiment survey instrument to assess the non-market values of Pacific Northwest residents on restoration activities focused on Oregon beaches. Kling published a paper that evaluate a selection of "myths", or popular conceptions about the role of uncertainty in decision models used in environmental and resource economics. Where conventional wisdom falls short, we argue practitioners should incorporate new advances in the economic uncertainty into their analysis.

University of Florida: Silver Springs Alliance and Florida Defenders of the Environment,  environmental groups dedicated to protecting the springs and restore the Ocklawaha River, have cited our study on the importance of recreation values provided by the Ocklawaha River. Presented findings on Valuing the Recreation Benefits of Springs in North Central Florida. Santa Fe River Springs Protection Forum in High Springs, Florida (2018 Feb). Produced 4 extension documents summarizing water resource management issues in Florida

University of Georgia: In support of a National Science Foundation-Couple Human & Natural Systems project, we conducted a pilot survey of coastal residents in Glynn County, Georgia. We tested an experimental design that permits assessment of risk preferences with salient (i.e. monetary) incentives, within the context of a mail or internet survey. Utilizing historical weather data, we present mail survey respondents with prior information about rainfall and temperature probabilities. We then allow them to “gamble” their $5 incentive payment on one of four possible weather outcomes, each choice is designed to bracket a range or risk preference parameters (via Constant Relative Risk Aversion formulation of Expected Utility). Responses permit inference on risk tolerance in the context of weather, which is relevant for the survey subject matter – flood risk and climate change. Provision of prior information and assessing posterior beliefs (that influenced weather lottery choice), allows for analysis Bayesian updating of weather beliefs in the context of location choice, insurance purchase, and perhaps other aspects of coastal resiliency. The survey also explores additional elements of individual risk perception; we utilize novel survey instruments to assess subjective perceptions of risk (likelihood of hurricane strike) and outcome (conditional damage to structure in hurricane). These data will play a prominent role in current PhD students dissertation on structural models of choice under risk.

University of Georgia: Received grant, Georgia Department of Natural Resources & National Oceanic and Atmospheric Administration “Economic Analysis of Green Infrastructure to Enhance Resiliency”, application to Liberty County, Georgia, $65,570, 2018-2020

University of Georgia: Received grant, National Oceanic and Atmospheric Administration “Hidden Infrastructure: Onsite Wastewater Disposal and Sea-level Rise”, application to Bryan County, Georgia, $16,928, 2-10-2019

University of Georgia: Received grant, Bureau of Land Management & United States Geological Survey “Benefit-cost Analysis for the Threatened and Endangered Species Program of the Bureau of Land Management”, $49,998, 2018-2019

University of Illinois-Champaign: Ando completed a manuscript to inform management of flood hazards and submitted it for review. She uses a contingent valuation survey to evaluate homeowner willingness to pay for a novel pre-flood agreement such that the homeowner pre-commits to relocating if a flood damages their home by more than 50% of its value, in exchange for which they gain an expedited and streamlined buyout process with payment equal to the full market value of their home. The study finds that nearly all homeowners would gain value from being able to enter into such a contract.

University of Illinois-Champaign: Ando worked with coauthors to complete and publish a paper that sets forth principles for optimal spatial targeting of conservation investment to reduce outcome risk with the smallest possible reduction in the expected level of ecological benefits when the decision maker is interested in multiple distinct environmental targets.

University of Illinois-Champaign: Ando worked with coauthors to complete and publish a paper that shows the kinds of conservation problems for which spatial portfolio theory analysis can be most useful.

University of Illinois-Champaign: Gramig worked throughout 2017 and 2017 with multi-state (some are members of NC-1190) collaborators in California, Iowa, Indiana, and Maine to review the recent literature on motivations and barriers to adoption of conservation practices on agricultural land in the US since 1980. A high-level initial analysis has been submitted for expedited review and five additional manuscripts are under development examining specific practices, actual adoption versus willingness-to-adopt and undertaking a statistical meta-analysis of the literature.

University of Illinois-Champaign: Gramig, along with W-4133 member Seong Do Yun at Mississippi State University began a cooperative agreement with NRCS and ERS economists to use econometric methods to measure the on-farm yield and cost of production effects of adopting NRCS conservation practices using public and private data sources. This work also partners with the Illinois Corn Growers Association’s Precision Conservation Management program led Dr. Laura Gentry that has partial support from the USDA Regional Conservation Partnership Program.

University of Illinois-Champaign: Ando has collaborated with researchers in Missouri to identify key attributes of water quality improvements in the Corn Belt that should be valued. She has worked with a graduate student to refine a choice-experiment survey to estimate how an individual’s WTP for water quality improvements in a watershed in Illinois is affected by their proximity to the area where landscape changes are implemented to accomplish the change and by whether they are in rural or urban parts of the watershed.

University of Illinois-Champaign: Gramig is conducting a synergistic valuation study to measure the public’s preferences for different mechanisms (point source nutrient abatement versus various approaches to non-point source pollution reduction) to achieve state Nutrient Loss Reduction Strategies in force throughout the Corn Belt region. Ando and Gramig have coordinated their WTP work to maximize insights from the resulting valuation estimates.

University of Illinois-Champaign: Gramig completed a study through Illinois-Indiana Sea Grant with a collaborator at Purdue University of the value of recreation fishing in Southern Lake Michigan along the shores in IL and IN. This work partnered with the Illinois Natural History Survey and the Indiana Department of Natural Resources to collect data using an angler intercept survey.

University of Illinois-Champaign: Ando collaborated with W4133 affiliate Noelwah Netusil at Reed College to write a review of the literature estimating the values of green infrastructure.

University of Illinois-Champaign: A team of researchers from W4133 partner universities UIUC and Oregon State, Amy Ando and Christian Langpap, wrote an analytical review of research on the economics of species conservation. They analyzed research on the values of species conservation and optimal policy design for species conservation.

University of Illinois-Champaign: grant received, Gramig (PI), B.M., S.D. Yun, G.S. Schnitkey and L.F. Gentry. “On-farm Benefits and Costs of Conservation Practices and Systems.” USDA-NRCS Cooperative Ecosystems Study Unit, 2018-2020, $462,403.

University of Illinois-Champaign: grant received, X Cai (PI), R Cusick, BM Gramig, G McIsaac, V Singh. “INFEWS/T1: Advancing FEW System Resilience in the Corn Belt by Integrated Technology-Environment-Economics Modeling of Nutrient Cycling.” NSF-Innovations at the Nexus of Food-Energy-Water Systems (INFEWS), 2017-2021, $2.43 million

University of Maine: Economists and social scientists from University of Maine, University of Massachusetts, Cornell University, and University Vermont conducted economic analyses to better understand the priorities and decision-making of private forest landowners in the Northeast as they negotiate changing forest product markets, land management policies, regional economies, demographics, and environmental conditions. Research focused on legacy planning indicated that the majority of landowners in our empirical sample believe controlling the future use of their land is important and are open to keeping their parcel forested and intact (i.e., not subdivided); varying legacy goals were associated with distinct behaviors and future intentions to designate or restrict the future use of their land (Markowski et al. 2018). Use of formal estate planning and conservation tools and participation in current-use taxation programs by family forest owners varies throughout the Northeast region; significant state and regional fixed effects suggest landscape context factors beyond owner and parcel characteristics influence spatial patterns in tool adoption and program participation (Bell et al. 2018).

University of Maine: To improve understanding of land management decisions and conservation opportunities on private lands in these landscapes, economists from University of Maine and Clark University developed a landowner typology that accounts for heterogeneity among landowners and investigated its usefulness in explaining differences in individual private landowners’ stated intentions to implement diverse land management actions. Cluster analysis based on measures of place attachment values, stewardship beliefs, and forest production objectives revealed four distinct landowner classes that, in turn, increased the explanatory power of our models of management intentions. Our findings support consideration of heterogeneity beyond production objectives to advance understanding of landowner decision-making and to improve the design and performance of conservation programs (Balukas, Bell, and Bauer, 2019).

University of Rhode Island: We conducted a non-market valuation survey to gauge the public preference and perception of wind power using as a case study the first wind farm established within the United States, near the cost of Block Island, Rhode Island. Our survey consisted of three parts. In the first part, we asked respondents questions about the characteristics of their trip(s) to Block Island. The second part of our survey included questions about knowledge and perceptions of the wind farm off of Block Island shores. This section also included Willingness to Pay questions for fishing, boating, sightseeing, beach, and bird and whale watching sites with and without a view of the turbines or without them. In the third and final section of the survey, we ask respondents to provide some basic demographic details such as age, household income and educational qualification. We also capture respondents' environmental attitudes by incorporating questions from the New Environmental Paradigm (NEP) (Dunlap and Liere 1978; Dunlap et al. 2000). We disseminated our survey in August 2018 through the Qualtrics platform to residents 18 years of age and above who reside in the states of California, Connecticut, Florida, New Jersey, New York, Maine, Maryland, Massachusetts, Michigan, Pennsylvania or Rhode Island. A total of 263 respondents completed the survey. Data from this survey is currently in the final stages of analysis and a report is expected to be finalized and submitted for publication within the next few months.

University of Rhode Island: A survey was designed to understand the impact of positive message framing on Willingness To Pay a premium for organic products among college students. The choice of college students for the sample is motivated by the fact that this is the demographic that will soon have households of their own and can potentially be the chief consumers of organic products. The team that developed the survey consisted of one faculty, one PhD candidate and three undergraduate students at URI. Two focus groups with 15 participants have been conducted and the team is now in the process of analyzing the feedback collected in these two sessions and refining or redesigning the survey. The survey will be administered- using Qualtrics, professional survey platform- to the URI undergraduate body of students.

University of Rhode Island: Used housing market data to examine aggregate responses to additional local government provision of open space. I also used housing market data to examine distributional impacts of brownfield remediation. During this reporting period, I wrote survey instruments for a 2018 statewide exit poll and three municipal exit polls. These data will help inform voter decision models and voter valuation of the environment.

University of Rhode Island: We conducted two studies related to China’s forest policies. In one study, we examine the role of risk and time preferences in how forest owners respond to forest certification. Using household survey data from Fujian Province, we test hypotheses from a two-period harvest model derived from prospect theory. This study was published in Land Economics. In another study, we evaluate the impact of a large-scale payments for environmental services (PES) program in rural China on forest structure. We couple satellite images detecting forest cover change for four different types of forests with county-level data on program participation. This study was published in Journal of Environmental Economics and Policy.

University of Rhode Island: “Aversion to environmental risk in stated preference: Implications for benefit transfer and scope response” (with Ben Blachly). The goal of this study is to learn about the relationship between environmental risk preferences and scale as it applies to stated preference research. This study builds on the limited literature on environmental risk preferences and makes three key contributions. First, we perform an explicit test of whether environmental risk aversion increases with scale. This is a well-known result in the financial domain, and some research suggests that a similar relationship exists in an environmental setting. Second, we compare the ability of different approaches to modelling the risk preference-scale relationship to reduce between-scale transfer errors. Preferences for environmental goods tend to be dependent on the scale at which they are elicited, so errors are likely when transferring values between scales. Finally, we explore how understanding the relationship between environmental risk preferences and scale can refine our interpretation of scope response. Scope response, the idea that willingness to pay should increase with the amount of good on offer, has evolved into a major test of theoretical consistency. Choice experiments, which are constructed to exhibit built-in (internal, or within-sample) scope response, have been subjected to few external (between-sample) tests of scope. We subject a choice experiment to an external scope test and explore risk aversion as a determinant of the degree of scope response.

University of Rhode Island: “Decision support modeling for multi-attribute water quality in the Narragansett Bay watershed” (with Robert Griffin). In this study, we develop an integrated assessment model for spatially simulating water quality changes and social welfare effects for recreational use and non-use as a function of common point and non-point source pollution management interventions. This extends prior modeling by incorporating a broader suite of water quality characteristics beyond phosphorus and nitrogen that are known to affect water quality. Beyond demonstrating the feasibility of such a model, we show the implications of omitting water quality factors on model estimates and find that in some cases a smaller set of water quality attributes may not significantly impact estimates of willingness to pay. We derive these lessons in a case study of Narragansett Bay, a coastal estuary that has recently experienced significant changes in nutrient loading due to a combination of point and non-point source changes and find that recent wastewater treatment upgrades are providing millions in annual value to adjacent residents.

University of Tennessee: Researchers identified optimal county-level targets for payments for forest carbon sequestrations with a total conservation budget optimally distributed among the counties and their resulting changes in forest carbon and economic benefits.

University of Wyoming: Received grant “Investigating Potential Impacts of Non-Attainment Risk on Conservation Exchange Outcomes.” Hansen, K. (PI), Co-PIs: C. Jones-Ritten, C. Bastian, A. Nagler.  Amount Requested: $74,317.  Submitted 9/30/2015.  Duration of Project: January 1, 2016 through December 31, 2018. Research completed regarding habitat conservation markets using experimental economics methods. Economic Experiments conducted and analysis completed.  Presentation made at Center for Behavioral, and Experimental Agroenvironmental Research Conference.  Paper submitted to Environmental and Resource Economics.  This paper has now been accepted and is in press for publication in 2019

University of Wyoming: “Incentivizing Open Spaces in Wyoming to Promote Pollinator Habitat: Applying Agglomeration Bonuses to Unite Fragmented Habitat.” Milestones: Research completed regarding pollinator habitat policy design using experimental economics methods.

Utah State University: A major study on the economic effects of landscape-scale National Monuments, specifically the Grand Staircase-Escalante NM (GSENM) of southern Utah, was completed, although research for other national monuments remains ongoing. The National Monuments work is a discrete, non-marginal contribution to our understanding of a major public lands management policy that has heretofore received little attention by economists (despite heated political debate). A review of economic benefit and cost studies of the BLM’s wildhorse and burro (WHB) program was also completed and published.

Virginia Tech: Developed a new econometric approach to process meta-data on water quality and wetland conservation, as used by the EPA for decision-making. This mode satisfies important theoretical and practical considerations, such as sensitivity to scope (higher willingness to pay (WTP) for better quality), and adding-up (total WTP for the change is not sensitive to the number of increments for implementation.) I expect this model to see widespread use by the EPA and other agencies in the context of benefit-cost and environmental impact analysis. I applied this model to a meta-dataset of willingness-to-pay for water quality improvements and shared results with agencies, peers, and practitioners at two conferences in 2018. I also applied the model to value wetland benefits.

Virginia Tech: Developed a new econometric approach (Bayesian matching estimator) to estimate the loss in home values from being located in a special flood hazard area. This model is less sensitive to misspecification errors, and widely applicable to other real-estate valuation contexts. Using a rich data set on home sales for five counties in Connecticut, controlling for a plethora of potentially confounding effects, and applying state-of-the art doubly-robust matching methods, we show that SFHA-related risk losses can vary dramatically by location relative to the coast line, with near-coastal losses exceeding interior effects by seven-fold. We take this as evidence that homebuyers hold beliefs of elevated flood risks in coastal zones, even though the official Flood Insurance Rate Map designation for those homes is identical to that of interior counterparts. To the extent that these beliefs align with objective risks, our results provide ammunition for calls for a more spatially refined rate setting policy for federal flood insurance.

Virginia Tech: Developed the econometric framework to analyze the data to be expected from the large-scale watershed improvement survey in New England. The highlight of the statistical model is that it allows for variable uncertainty and model-averaging. This is important given the large number of candidate explanatory variables that could explain a household's response to the binary (yes/no) willingness-to-pay question collected in the survey. The statistical model was tested with simulated data and found to perform well. I also contributed to the design of the emerging survey instrument via computational simulation of different bid designs to be used in the survey.

Publications:

Ahmadiani, Mona, Susana Ferreira, and Craig E. Landry. 2019. “Flood Insurance and Risk Reduction: Market Penetration, Coverage, and Mitigation in Coastal North Carolina” Southern Journal of Economics. DOI: 10.1002/soej.12332

Ando, A.W., A. Howlader, and M. Mallory. 2018. “Diversifying to reduce conservation outcome uncertainty in multiple environmental objectives.” Agricultural and Resource Economics Review 47(2): 220–238. doi.org/10.1017/age.2018.7.

Ando, A.W., J. Fraterrigo, G. Guntenspergen, A. Howlader, M. Mallory, J. Olker, and S. Stickley. 2018. “When portfolio theory can help environmental investment planning to reduce climate risk to future environmental outcomes – and when it cannot.” Conservation Letters. http://dx.doi.org/10.1111/conl.12596.

Ando, A.W. and N. Netusil. 2018. “Valuing the benefits of green stormwater infrastructure.” Oxford Encyclopedia of Environmental Science. doi: 10.1093/acrefore/9780199389414.013.605.

Ando, A.W. and C. Langpap. 2018. “The economics of species conservation.” Annual Review of Resource Economics 10: 445-467. doi.org/10.1146/annurev-resource-100517-022921.

Balukas, J.E., Bell, K.P., and D.M. Bauer. 2019. Classifying private landowners to improve understanding of management decisions and conservation opportunities in urbanizing forested landscapes, Journal of Environmental Management, 232: 751-758, DOI: 10.1016/j.jenvman.2018.11.128.

Bell, K.P., Markowski-Lindsay, M., Catanzaro, P., and J.E. Leahy. 2018. Family-forest owner decisions, landscape context, and landscape change, Landscape and Urban Planning, DOI: 10.1016/j.landurbplan.2018.08.023 (posted online 2018 October).

Bi, Xiang, Tatiana Borisova, and Alan Hodges.  2019. Economic Value of Visitation to Free-Flowing and Impounded Portions of a River: Implications for Management River Flow.  Forthcoming. Review of Regional Studies.

Byrd, ES, NJO Widmar, BM Gramig. "Presentation Matters: Number of Attributes Presented Impacts Estimated Preferences." Agribusiness: an International Journal, 34(2):377-389, 2018.

Cho, S., S. Moon, B. English, T.E. Yu, C. Boyer. 2019. “Targeting Payments for Forest Carbon Sequestration Given Ecological and Economic Objectives.” Forest Policy and Economics 100:214-226.

Doering, OC, BM Gramig and D Jeong. “Economic and Policy Implications of Nitrogen Management.” Soil Nitrogen Uses and Environmental Impacts, Advances in Soil Science: Soil Nitrogen volume, eds. R. Lal and B.A. Stewart. CRC Press, Taylor & Francis Group: 2018.

Dundas, S. J., R. H. von Haefen, and C. Mansfield. 2018. Recreation Costs of Endangered Species Protection: Evidence from Cape Hatteras National Seashore, Marine Resource Economics 33(1): 1-25.

English, E., Roger von Haefen, J. Herriges, C. Leggett, Frank Lupi, K. McConnell, M. Welsh, A. Domanski, N. Meade. 2018. Estimating the value of lost recreation days from the Deepwater Horizon oil spill. J. Environmental Economics and Management. 91:26-45.

Fu, Guanlong, Muna Shah, Emi Uchida, Xiangzheng Deng. (2018) “Impact of the Grain for Green program on forest cover in China.” Journal of Environmental Economics and Policy. https://doi.org/10.1080/21606544.2018.1552626

Gramig, BM and NJO Widmar. "Farmer Preferences for Agricultural Soil Carbon Sequestration Schemes." Applied Economic Perspectives and Policy, 40(3):502-521, 2018.

Jakus, Paul M., and Sherzod B. Akhundjanov. 2018. “Neither Boon nor Bane: The Economic Effects of a Landscape-Scale National Monument.” Land Economics, 94(3):323-339.

Jakus, Paul M. 2018. “A Review of Economic Studies Related to the Bureau of Land Management’s Wild Horse and Burro Program.” Human-Wildlife Interactions, 12(1):58-74.

Johnston, R., K. Moeltner (published online Jan. 2019). Special Flood Hazard Effects on Coastal and Interior Home Values: One Size Does Not Fit All, Environmental and Resource Economics.

Lang, C., & Cavanagh, P. (2018). Incomplete Information and Adverse Impacts of Environmental Cleanup. Land Economics, 94(3), 386-404. Lang, C. (2018). Assessing the efficiency of local open space provision. Journal of Public Economics, 158, 12-24.

LaRiviere, Jacob, David M. Kling, James N. Sanchirico, Charles Sims, and Michael Springborn. 2018. "The treatment of uncertainty and learning in the economics of natural resource and environmental management." Review of Environmental Economics and Policy 12(1): 92-112.

Lewis, D.J., and S. Polasky. 2018. “An Auction Mechanism for the Optimal Provision of Ecosystem Services under Climate Change.” Journal of Environmental Economics and Management, 92: 20-34.

Li, Jingyuan, Craig E. Landry. 2018. “Flood Risk, Local Hazard Mitigation, and the Community Rating System of NFIP” Land Economics 94(2): 175-198 doi: 10.3368/le.94.2.175

Liu, Tingting, James Opaluch, Emi Uchida. (2017) “The impact of water quality improvement in Narragansett Bay on housing prices.” Water Resources Research 53, doi:10.1002/2016WR019606.

Liu, Y, MR Langemeier, IM Small, L Joseph, WE Fry, JB Ristaino, A Saville, BM Gramig, PV Preckel. "A Risk Analysis of Precision Agriculture Technology to Manage Tomato Late Blight." Sustainability 10(9), 3108, 2018.

Markowski-Lindsay, M., Catanzaro, P., Bell, K.P., Kittredge, D., Markowitz, E., Leahy, J.E., Butler, B., Milman, A., and S. Allred. 2018. In Forest and Intact: Designating Future Use of Family-Forest-Owned Land, Journal of Forestry, 116(4):357–366, DOI: 10.1093/jofore/fvy015.

Moeltner, K., (2019). Bayesian Nonlinear Meta Regression for Benefit Transfer, Journal of Environmental Economics and Management, 93, 44-62.

Mueller, Julie M., Soder, Adrienne B., and Springer, Abraham E. (2019). “Valuing attributes of restoration in a semi-arid watershed.” Landscape and Urban Planning 184: 78-87.

Mueller, Julie M., Lima, Ryan E., Springer, Abraham E. and Schiefer, Erik. (2018). “Using matching methods to estimate impacts of wildfire and post-wildfire flooding on house prices.” Water Resources Research 54(9):6189-6201

Nohner, J., Frank Lupi and W. Taylor, 2018. Lakefront property owners' willingness to accept easements for conservation of water quality and habitat. Water Resources Research. 54(3):1533-48.

Netusil, N.R., M. Jarrad, K. Moeltner (2019). Research Note: The Effect of Stream Restoration Project Attributes on Property Sale Prices, Landscape and Urban Planning 185, 158-162.

Penn J, and W Hu. 2018. Understanding Hypothetical Bias: An Enhanced Meta-Analysis. American Journal of Agricultural Economics. 100(4): 1186-1206.

Penn JM, Hu W, and LJ Cox. 2019. Forced Choice with Constant Choice Experiment Complexity. Journal of Agricultural and Resource Economics. 44(2): 439-455.

Penn J, Penn H, and W Hu. 2018. Public knowledge of monarch conservation in Kentucky. Sustainability. 10(3). doi:10.3390/su10030807

Sullivan, Karen, Thomas Sproul, Emi Uchida, and Jintao Xu. (2018) “Prospect theory and tenure reform: Impacts on forest management”. Land Economics 94(3): 405-424.

Swallow, S, C Anderson, Emi Uchida. (2018) The Bobolink Project: Selling Public Goods From Ecosystem Services Using Provision Point Mechanisms. Ecological Economics 143: 236-252.

Trull N, Penn J, and W Hu. 2018. Public Support for Growth and Funding in Built Environments: Case of an Arboretum. Journal of Housing and the Built Environment. 33(4): 829-841.

Uchida, Emi, Stephen Swallow, Arthur Gold, James Opaluch, Achyut Kafle, Nathaniel Merrill, Clayton Michaud, and Carrie Gill. (2018) “Integrating watershed hydrology and economics to establish a local market for water quality improvement: A field experiment.” Ecological Economics 146: 17-25.

Wu, Qianyan, Xiang Bi, Kelly Grogan, and Tatiana Borisova.  2018. Valuing Recreation Benefits of Natural Springs in Florida.  Water.  10(10): 1379. https://doi.org/10.3390/w10101379.

Yang, H., Frank Lupi, J. Zhang, X. Chen and J. Liu, 2018, Feedback of telecoupling: The case of a payment for ecosystem services program, Ecology and Society. 23(2):45.

Zhong H, Hu W and J Penn. 2018. Farmers’ Willingness and Expected Economic Benefit to Adopt BMPs: An Application of Multivariate Imputation by Chained Equation Method. Journal of Agricultural and Resource Economics. 43(1): 78-102.

Forthcoming

Hess, Joshua, Dale Manning, Terry Iverson, and Harvey Cutler (forthcoming). Uncertainty, Learning, and Local Opposition to Hydraulic Fracturing. Resource and Energy Economics.

Lamb, K.*, K. Hansen, C. T. Bastian, C. Jones Ritten, and A. Nagler, “Investigating Potential Impacts of Credit Failure Risk Mitigation on Habitat Exchange Outcomes.”  Environmental and Resource Economics. Conference on Behavioral and Experimental Agri-Environmental Research Special Issue.  (Currently in Press).

Maas, Alexander, Chris Goemans, Jesse Burkhardt, and Mazdak Arabe (forthcoming). Complements of the House: Estimating Demand-side Linkages between Residential Water and Electricity, Water Resources and Economics.

Manning, Dale, and Jordan Suter (forthcoming). The Role of Well Capacity Constraints in Determining Gains from Groundwater Management. Journal of Agricultural and Resource Economics.

Mutyasira, Vine, Dana Hoag, Dustin Pendell, and Dale Manning (forthcoming).  Sustainable Intensification in Ethiopian Smallholder Farming Systems. Sustainability.

Mutyasira , Vine, Dana Hoag, Dale Manning, and Dustin Pendell (forthcoming). Assessing the Relative Sustainability of   Smallholder Farming Systems in Ethiopian Highlands, Agricultural Systems.

Shepler, Ryan, Jordan Suter, Dale Manning, and Chris Goemans (forthcoming).  Private Actions and Preferences for Coordinated Groundwater Conservation in Colorado’s Republican River Basin, Journal of the American Water Resource Association.

Soh, M., S. Cho. 2019. “Spatial Targeting of Payments for Ecosystem Services to Achieve Conservation Goals and Promote Social Equity and Economic Impact.” Natural Resource Modeling, In press.

Other Publications

Barfield, Ashley and Craig E. Landry. 2018. “Valuation of Beach Quality” in Oxford Encyclopedia of Environmental Economics, Oxford University Press, Earth & Environmental Science.

Miao, Haoran. 2017. Three Essays on the Impact of Information on Nonpoint Source polluters’ Behavior. Ph.D. Dissertation, Department of Environmental and Natural Resource Economics, University of Rhode Island.

Jakus, Paul M., Commentary, “Large National Monuments Don’t Damage Rural Economies--But They Don’t Help Either”, Deseret News (40,000 weekday circulation), August 23, 2018.

Jakus, Paul M., Radio Interview, “Large National Monuments Don’t Damage Rural Economies--But They Don’t Help Either”, KNRS 105.9 (Salt Lake City—125,000 drive time listeners), August 23, 2018.

Von Haefen. NC State University Press Release, “What’s the Value of Lost Recreation Days from the Deepwater Horizon Oil Spill?” August 2018.

Von Haefen. Extension article (with Steven Dundas), “Conflicts on Public Lands: The Case of Off-Road Vehicle Restrictions on Cape Hatteras National Seashore,” NC State Economist, Summer 2018.

Von Haefen. Interview, “Costs and Benefits of Cape Hatteras ORV Restrictions,” WUNC Radio, January 2018.

Von Haefen. NC State University Press Release, “Benefits of Off-Road Vehicle Restrictions May Outweigh the Costs,” January 2018.

Von Haefen. Extension article (with Steven Dundas), “Conflicts on Public Lands: The Case of Off-Road Vehicle Restrictions on Cape Hatteras National Seashore,” NC State Economist, Summer 2018.

Von Haefen. Interview, “Costs and Benefits of Cape Hatteras ORV Restrictions,” WUNC Radio, January 2018.

Von Haefen. NC State University Press Release, “Benefits of Off-Road Vehicle Restrictions May Outweigh the Costs,” January 2018.

Papers, Presentations, and Reports

Ando, A. “Environmental Economic Approaches to Soil Health as a Capital Stock.” Soil Health on Rented Lands in Indiana: A Workshop for Agricultural, Conservation and Landowner Service Professionals to Explore Potential Solutions convened by The Nature Conservancy at the NCAA Hall of Champions, Indianapolis, IN. March 23, 2018.

Jakus, Paul M. and Sherzod B. Akhundjanov. 2018. “The Antiquities Act, Large National Monuments, and Regional Income.” Presentations at Colorado State University (September), University of Tennessee (October), and Weber State University (November).

Penn J, and W Hu. 2018. Understanding Hypothetical Bias in Willingness to Accept: A Meta-Analysis and Experiment. Accepted presentation, SEA, Nov 18-20, Washington DC.

Penn J, and W Hu. 2018. Cheap Talk, Consequentiality, and Certainty Follow-up as Hypothetical Bias Mitigation Techniques: A Cross Country Comparison. Accepted presentation, AAEA, Aug 5-7, Washington DC.

Penn J, Hu W, and H Penn. 2019. Support for Native, Solitary Pollinator Conservation Among the Public Versus Hobby Beekeepers. Invited track presentation, ASSA, Atlanta, GA.

Sardana, Kavita, John C. Bergstrom, and Oleksiy Tokovenko. “A Latent Class Approach for Modeling Arbitrariness in the Definition of ‘Visitors’ for Individual Trip Behavior”, Selected Paper, 6th World Congress of Environmental and Resource Economists, Gothenburg, Sweden, June, 2018.

Steele, Amanda Harker and John C. Bergstrom. “Investigating the Effects of Energy Efficient Technology on Household Energy Security: A Stochastic Production Frontier Approach”.  Selected Paper to be presented at the 6th World Congress of Environmental and Resource Economists, Gothenberg, Sweden, June, 2018.

Steele, Amanda Harker and John C. Bergstrom. “Does Energy Efficiency Affect Energy Security? An Analysis of Energy Efficient Upgrades and Household Energy Security”.  Selected Paper to be presented at the Annual Meetings of the Agricultural and Applied Economics Association, Washington, D.C., August, 2018.

Steele, Amanda Harker and John C. Bergstrom. “Tackling Wicked Problems when Teaching Applied Economics: An Application to the Bears Ears National Monument”.  Invited Case Study and Selected Poster to be presented at the Annual Meetings of the Agricultural and Applied Economics Association, Washington, D.C., August, 2018.

Steele, Amanda Harker and John C. Bergstrom. ”Tackling Wicked Problems when Teaching Applied Economics: An Application to the Bears Ears National Monument”.  Selected Poster Presentation, University System of Georgia Teaching and Learning Conference, Athens, GA, April, 2018.

von Haefen, Roger H. “Weather Effects on the Demand for Coastal Recreational Fishing: Implications for a Changing Climate,” World Congress of Environmental and Resource Economists, Gothenburg, Sweden, July 2018.

von Haefen, Roger H. “Weather Effects on the Demand for Coastal Recreational Fishing: Implications for a Changing Climate,” Appalachian State University Experimental and Environmental Policy Workshop, Blowing Rock, NC, April 2018.

von Haefen, Roger H. “The Natural Resource Damages from the Deepwater Horizon Oil Spill,” Duke Kunshan University, July 2018.

von Haefen, Roger H. “Weather Effects on the Demand for Coastal Recreational Fishing: Implications for a Changing Climate,” Arizona State University, Tempe, AZ, April 2018.

von Haefen, Roger H. “The Natural Resource Damages from the Deepwater Horizon Oil Spill,” University of Alberta, Edmonton Alberta, February 2018.

2019 MEETING ABSTRACTS

Title: The Impact of Bison Reintroduction on the Local Economy

Authors: Liqing Li and Amy W. Ando (University of Illinois Urbana-Champaign)

W4133 task: 1-1

Presenter: Amy Ando

Presenter email: amyando@illinois.edu

Abstract:

This paper contributes to W4133 Task 1-1 by estimating the economic impact of changing rural environments by reintroducing bison.

Government agencies and non-profit conservation groups are restoring prairies and reintroducing bison to facilitate ecological restoration and reclaim this iconic mammal from the brink of extinction in the wild (Zumbach, 2015). Currently, there are forty-eight bison herds exist in the U.S., including both wild and non-wild bison herds. Bison reintroduction provides chances to help local tourism through wildlife marketing (Vasile, 2018). However, there is an opportunity cost to land dedicated to bison and ranchers have expressed concern about disease spreading from bison to cattle. Thus, the net economic impact of bison reintroduction on rural communities is unclear. This paper conducts a nation-wide assessment of the economic impacts of bison reintroduction so rural communities can take economic well-being into account when considering their future decisions regarding bison restoration.

Previous research has studied the effects of protecting nature on local economic activity Early studies raised concern that protected areas can be a threat to local economic well-being (Rasker,1993). Later research shows that it is possible that protected areas can change the structure of local economies in a positive way. For example, increased tourism can provide more employment opportunities (Andam et al. 2008; Andam et al. 2010). However, most of these studies focus on conservation and reintroduction of wildlife in developing countries. A study focuses on the northern forest region in the U.S. shows that conservation land has no significant impact on employment growth. (Lewis et al. 2002) and a small amount of work has studied the impact of bison or wolf populations on the desirability to visit particular recreational sites in the U.S. (Loomis & Caughlan, 2010; Duffield et al. 2008). There has been no nation-wide evaluation of how the reintroduction of bison into grassland areas affect the economic well-being of the residents. This paper fills this gap by providing a descriptive analysis of the presence of U.S. bison herds and estimating the causal relationship between bison reintroduction and local economic health.

First, we use a probit model to show correlations between the presence of bison herds in a county and the county’s physical and socioeconomic characteristics. Results indicate that the presence of bison herds is positively correlated with a county’s per capita income and (for non-wild bison herds) negatively correlated with unemployment rates. In this naïve analysis, bison herds are linked to economic prosperity.

Because correlation is a poor measure of causality, we carry out other analyses to estimate the impact of bison reintroduction on local employment, population, and per capita income. We use cross-sectional propensity score matching (PSM), matching with difference-in-difference (DD), and the synthetic control method (SCM).

We use cross-sectional PSM to evaluate the effects of bison reintroduction by comparing economic prosperity in the counties with bison herds to those of similar counties without bison herds. We match counties based on county characteristics that are fixed over time and unlikely to have been affected by bison reintroduction. Such variables include geographic and climate characteristics like elevation, precipitation, mean temperature, county size, the percentage of grassland in a county, and protected areas under different conservation mechanisms in a county. Then we compare the average outcomes from the treatment and control counties measured after bison reintroduction. We choose the year 2016 for our analysis since it is the year in which we can include the largest number of bison herds in the regression with the latest available county characteristics data. In the cross-sectional matching analyses, fifty-five counties with bison herds are considered as treatment counties. Different matching specifications, including nearest neighbor (NN) matching with caliper and kernel matching, are applied to identify the impact of bison reintroduction on local economic health.

One limitation of the PSM comes from matching on observables. If unobserved factors confound both bison reintroduction decisions and outcomes, the estimated impact of bison reintroduction gained by cross-sectional PSM is biased.  Applying PSM with DD can control for unobserved factors that are time-invariant. When we use 1980 as the baseline year and 2016 as the endline year, 24 counties that have bison herds established between 1980 and 2016 are included in the treated group. PSM is first applied to find controls for the treated counties, and DD is applied on the matched sample after matching to generate the causal estimates of bison reintroduction on local employment, population, and per capita income.

The SCM introduced by Abadie et al. (2010) extends the traditional DD framework by allowing for the presence of time-varying unobserved variables and constructs a better counterfactual for each of the treated counties. The basic idea behind the SCM is to construct a synthetic control unit as the counterfactual of the treated county by weighting average of several untreated counties (Abadie and Gardeazabal, 2003; Abadie et al., 2010). The weight of each control county is determined based on how closely the characteristics of the control match those of the treated counties in the pre-treatment period. As a comparative case study approach, this method provides more informative results not only on how bison reintroduction affects the local economy in each treated county, but also the trend of effects on outcome variables during the post-treatment period.

We find little evidence of a significant impact of bison reintroduction on local economic activity. Results from the cross-sectional PSM show that the effects of bison reintroduction on income, population, and the total number of jobs are positive in sign but insignificant, while bison reintroduction has a negative and significant impact on the unemployment rate. Results from our DD analysis show that bison reintroduction has a positive but insignificant impact on the total number of jobs, while the impact of reintroduction on the per capita income and the population are negative and insignificant. Results from the SCM are consistent with the findings of the DD analysis; for each county with bison herds, bison reintroduction has an insignificant impact on local economic health. Naïve cross-sectional analysis might imply that bison reintroduction is good for the local economy, but causal inference methods find no significant effect.

Title:  The Inefficiencies in Wildfire Suppression Resource Allocation

Authors: Jude Bayham

W4133 task: Task 1-2: Economic Analysis of Natural Hazards

Presenter: Jude Bayham

Presenter email: jbayham@colostate.edu

Abstract:

Federal, state, and local expenditures on wildfire suppression are reaching unprecedented levels.  While the underlying causes include climate change, a growing wildland urban interface, and historical management practices, the way in which wildfire suppression resources are allocated may also contribute to rising cost.  Economists are well aware of the problems that arise with common pool resources.  An analogous problem exists in the allocation of wildfire suppression resources.  Individual fire managers do not directly bear the cost of the resources they use, nor do they realize the opportunity costs imposed on other fire managers when resources are unavailable.

We develop a dynamic economic model to demonstrate the divergence between the incentives of individual fire managers and a social planner.  The model generates an important hypothesis: fire managers have an incentive to behave strategically by pre-emptively requesting resources to maintain excess suppression capacity in anticipation of increased fire activity.  We test this hypothesis using data compiled from several sources.  We collect daily data on the quantity and type of wildfire suppression resources ordered and assigned to fires from the federal Resource Ordering Status System (ROSS).  We also collect data on the so called preparedness level, which is a rating on a five point scale to indicate the level of suppression capacity in a region given expected fire behavior.  We integrate this data with fire conditions, weather, and other environmental conditions.  We test the hypothesis that fire managers strategically order resources and declare less suppression capacity when they believe that other managers competing for the same resources are likely to do the same.  We then use the empirical results to simulate the economically efficient allocation of suppression resources to demonstrate the consequences of ignoring opportunity costs.

This work addresses Task 1-2: Economic Analysis of Natural Hazards by studying inefficiencies in how federal agencies respond to wildfire.

Title: Environmental change and recreation demand: Assessing interactions between beach width and beach recreation demand

Authors: Kathleen P. Bell (University of Maine)

W4133 task: Task 3-2: Economic Analysis of Recreation Services

Presenter: Kathleen P. Bell

Presenter email: kpbell@maine.edu

Abstract:

In coastal areas of the US, where beaches serve as prominent recreation and tourism resources, mounting vulnerabilities to environmental change make the sustainability of recreation services uncertain. As natural resource managers face intensifying disturbances, multifaceted policy challenges, and, in some instances, shrinking budgets, numerous questions arise about the impacts of environmental change on coastal recreation services and the implications of these impacts for management strategies, communities, industries, and individuals. In this paper, we summarize the results of an economic analysis of the impacts of shoreline change on coastal beach recreation services (W4133 Task 3-2) at Popham Beach State Park, Maine. Our empirical analysis includes visual and statistical summaries of survey responses collected from beach visitors and recreation demand modeling of their beach day-trips. Working collaboratively with regional coastal manager, we designed our research to align with identified knowledge gaps about the impacts of shoreline change on visitors’ recreation experiences, awareness of changing shorelines, and attitudes’ about alternative management responses.

Our analyses of respondent survey responses provide insights about how visitors conceptualize shoreline change and assess distinct alternative management strategies. The majority of our survey respondents noted recognizing changes to the beach and shoreline at Popham Beach State Park. Notably, respondents acknowledged different types of change and offered diverse explanations for these changes. Further, analysis of open-ended responses revealed varied interpretations of the impacts of beach narrowing on recreation experiences as well as varied levels of support for alternative future management strategies. When asked about potential future management responses at Popham Beach State Park, respondents were split, with forty-nine percent of participants reporting a desire for the state to take management action, forty percent of respondents indicating the state should let nature take its course, and eleven percent of respondents uncertain about the state’s preferred strategy. Our findings have implications for how coastal managers in Maine and beyond engage with visitors about and respond to environmental change, and offer insights for recreation demand modelers interested in simulating visitor responses to different types of change.

Our recreation demand modeling of beach day-trips complements the descriptive and exploratory components of our work by simulating and quantifying the impacts of environmental change on coastal recreation services. We estimate a series of single-site, quasi-panel recreation demand models of day-trips to Popham Beach State Park to assess changes in beach recreation that might occur from beach narrowing and shoreline change. We integrate stated and revealed preference data collected onsite from visitors in August 2016, and intentionally follow conventions assessing trip responses to changes in beach width to bolster comparison with prior work. For example, our base model run produced a mean per-trip consumer surplus estimate or access value per trip of approximately $53.30 and predicted a mean loss per trip of $11.25 if the beach width at Popham Beach State Park were to be reduced by one half. Comparing the consistency of modeling and consumer surplus estimates from our Maine site with prior results, we found that our per-trip consumer surplus estimates fell in between estimates from Delaware Bay and North Carolina and that our predicted mean loss per trip exceeded prior estimates. We are investigating the sensitivity of model performance and valuation estimates to different modeling choices and assumptions, including the robustness of findings to different corrections for onsite sampling bias, measures of travel costs, and alternative treatments of heterogeneous recreation preferences.

In summary, by sharing insights about how coastal recreationists respond to changing beach width and other forms of environmental change, this work helps support management decisions and guide advances in economic analysis of recreation services.

Title: The Effect of Pollution on Aggressive Behavior: Evidence from Wildfire Smoke and Crime

Authors: Jesse Burkhardt, Jude Bayham, Ander Wilson, Ellison Carter, Katelyn O’Dell, Bonne Ford, Emily Fischer, Jesse Berman, Jeffery Pierce

W4133 task: Task 1-2: Economic Analysis of Natural Hazards (fire, invasive species, climate change)

Presenter: Jesse Burkhardt    

Presenter email: jesse.burkhardt@colostate.edu

Abstract: We estimate the effect of short-term air pollution (PM2.5) exposure on several categories of crime with a particular emphasis on aggressive behavior. To identify this relationship, we combine detailed daily data on crime, PM2.5, and satellite-derived smoke plumes for an eight-year period across the United States.  We develop two methods to correct measurement error in satellite-based wildfire smoke plume data and use the remaining exogenous variation as an instrument for overall changes in PM2.5. Our findings indicate a strong and robust positive effect of increased PM2.5 on violent crimes, and specifically assaults. However, we find no statistical relationship between increases in PM2.5 and property crimes or other non-violent crimes. The results suggest that a 1μg/m3 reduction in daily PM2.5 could save $1 billion in crime costs per year, a previously overlooked cost associated with pollution.

Title: TRAVEL COST MODELS USING CONTINUOUS NON-NEGATIVE DISTRIBUTIONS AND MEASURING OUT OF SAMPLE PERFORMANCE

Authors: Jeffrey Englin, Thomas Holmes and Octavio Valdez Larfarga

Presenter: Jeffrey Englin

Presenter email: jenglin@asu.edu

Abstract:

Big data is becoming more prevalent in the valuation of non-market goods. This is partly due to large environmental events with widespread impacts such as Deepwater Horizon, partly due to the ability to merge large detailed ecological data sets onto behavioral datasets but also due to the realization that administrative data can be used to infer environmental values. This analysis uses both large ecological and administrative data sets. The econometric analysis introduces the excess zero exponential and excess zero log normal distributions to travel cost modeling. Traditional excess zero count models are also employed for comparative purposes. The new distributions allow large national data sets to be applied to specific areas, in this case wilderness areas, and derive nationally valid welfare estimates. In particular, they allow for extremely high usage counts (popular sites) while dealing effectively with an excess number of zeros as well.

The data is driven by wilderness back country permits and the Environmental Protection Agency Level III ecosystem map. The wildernesses are located in the Cascades, Sierra Nevada, Rocky Mountains and the High Plains of the Midwest. The ecosystems studied include the Blue Mountain, Cascades Central Cascades and High Plains. The analysis proceeds by using five stratified sampling without replacement from a 70 million observation data set to repeated samples of 700,000 observations. The samples are stratified based on wilderness and year so as to retain a representative subsample.

Models using the new continuous distributions and the classic count models are estimated in the conventional way on one subsample. The estimated parameters are then applied to the remaining samples to analyze the potential of over fitting the models to first sample. It also allows the generalizability of the original set of estimates to be assessed. Finally, model averaging is applied since one cannot be sure of the correct distribution. The welfare results for the averaged model are also presented.

The performance of each model and method is analyzed and compared. This analysis works to achieve Objective 2 Task 2-1 and Objective 3 Tasks 3-1 and 3-2.

Title: Estimating Preferences for Water Quality when Water Quality Data is Scarce

Authors: Caroline Kelsoe, Matt Interis, & Seong Yun

W4133 task: Task 3-1, Task 3-2, Task 3-3

Presenter: Caroline Kelsoe

Presenter email: caroline.kelsoe13@gmail.com

Abstract:

In its 2012 National Lakes Assessment, the EPA determined that nutrients are the most widespread stressor of US Lakes, with one-third of US lakes containing excess nitrogen and/or phosphorus. In response, states are developing Numeric Nutrient Criteria to regulate the allowable amount of nutrients in surface waterbodies. Our study fits into the larger benefit-cost assessment of this policy in Mississippi as we attempt to estimate the benefits of improved water quality in Mississippi lakes. A major challenge, however, is the lack of water quality data for many Mississippi lakes. 

We present estimates of the benefits from improved water quality in Mississippi lakes based on travel cost data from Mississippi residents and introduce possible solutions to potential choice set specification issues caused by a lack of water quality data for Mississippi lakes. Specifically, respondents recorded trip information for over 100 lakes in Mississippi, but only 17 of those lakes have water quality data for the year the respondents reported trip information. In order to understand the extent to which a constrained choice set from inadequate data affects welfare estimates, we introduce a water quality prediction model from water science literature. Using a panel dataset of water quality data from Mississippi lakes both included and excluded from the choice set, we estimate a regression model to predict water quality in the 80+ lakes that lack water quality data. We then compare welfare estimates from a full and constrained choice set to determine evaluate the extent to which inadequate data could affect policy-making decisions.

Title: Hold the Line: Modeling Private Coastal Adaptation through Shoreline Armoring Decisions 

Authors: W. Jason Beasley, Steven J. Dundas

W4133 tasks: Task 1-2: Economic Analysis of Natural Hazards (fire, invasive species, climate change) Task 2-3: Advances in Spatial/Environmental Nexus Task 3-1: Economic Analysis of Ecosystem Services Flows

Presenter: Steven J. Dundas

Presenter email: steven.dundas@oregonstate.edu

Abstract

Managing and preparing for risks presented by sea-level rise (SLR) is an emerging issue of primary importance for coastal climate adaptation globally. Recent research suggest the majority of the expenditure on coastal adaptation in the U.S. by 2100 will be through private investments in shoreline armoring. Yet, decisions about public adaptation to coastal risk through beach nourishment has garnered most of the attention in the economic literature to date. This gap is significant because there are key differences between the public adaptation decision of nourishment and private shoreline armoring decisions.

This paper explores the economics of private coastal adaptation to climate change by modeling shoreline armoring decisions. We develop an analytical framework that suggests that negative spatial externalities, peer effects and neighbor interactions may lead to the oversupply of coastal protection when these decisions are made by individuals. To test our framework empirically, we assemble a parcel-level panel dataset of oceanfront land in Oregon from 1990 to 2015 where we observe the location and 1 timing of each armoring decision. For each parcel, we construct a detailed set of highresolution variables characterizing the geomorphological risk profile of land, peer effects of adjacent and proximate armoring of neighboring parcels, and indicators for seasonal shocks (El Ni˜no–Southern Oscillation) likely to impact armoring decisions. We use a correlated random-effects approach to estimate our econometric model to overcome endogeneity concerns related to modeling repeated decisions over time and unobserved time-invariant parcel level characteristics, while also allowing credible identification of the impact of neighboring land-use (i.e., peer effects) and land characteristics (i.e., risk conveyance) on the armoring decision over time. Our results show that land characteristics, peer effects and cost-sharing coalition effects increase the probability of observing an installation decision. Importantly, we show peer effects tend to dominate the impacts of risk characteristics of land.

We then construct Monte Carlo landscape simulations that capture the dynamic nature of the armoring decision. We first compare total armoring counts and spatial densities between a pure risk model (baseline model) and model that includes peer effects. Results suggest that the inclusion of peer effects leads a 93 percent increases in armoring counts and a five-fold increase in clustering of shoreline armoring. Then, using the peer effects framework as our preferred specification, we simulate a counterfactual policy that relaxes armoring restrictions under Oregon’s Goal 18. We find this policy change could result in an additional 350 parcels choosing to armor over the next 40 years. This suggests the current policy may be an effective tool to preserve the public good (beach access and natural landscapes) but highlights a growing tension between coastal preservation and risk to private property. Lastly, we run a SLR simulation by varying parcel elevation, distance from the parcel structure to the shoreline, and erosion rates based on NOAA projections of SLR. Our results indicate that under a moderate (worst-case) SLR scenario, we would expect a statistically significant 16 (26) percent increase in the count of armored parcels within three decades. Importantly, we find that the relaxation of current land-use policies combined with modest (worst-case) SLR predictions has potential to increase armoring in excess of 200 (230) percent over the same time period – highlighting the critical role policy is likely to have in shaping our future coastlines.

Title: “Preliminary Results from a Meta-analysis of US Farmer Adoption of Agricultural Best Management Practices”

Authors: B.M. Gramig (UIUC), K. Floress (USFS), L.S. Prokopy (Purdue), A. Singh (CSU-Sacramento), F. Eanes (Bates College), S. Church (Purdue), P Ranjan (Purdue), Y Gao (UC-Berkeley), and J. Arbuckle (Iowa State)

W4133 task: Task 1-1: Economic Analysis of Agriculture, Forest, and Rangeland Resources, Open Space, and WUI Zones

Presenter: Gramig

Presenter email: bgramig@illinois.edu

Abstract: We construct a database of 35 years (1982-2017) of agricultural conservation practice adoption studies to perform a quantitative meta-analysis of the determinants of adoption findings published in the academic literature. The database contains all reported model statistics and estimated parameters, including dependent and independent variable descriptions, from 171 separate quantitative studies conducted by economists and non-economists. Preliminary results of a statistical meta-analysis of these data are presented, and ideas for further and more refined analysis will be solicited from meeting participants. We apply conventional statistical methods to analyze study-level determinants of conservation practice adoption behavior under Hatch Multi-State project W-4133 Task 1-1.Co-authors Arbuckle, Floress, Gramig and Prokopy are also members of the NC-1190 multi-state Hatch project. Researchers from California, Illinois, Indiana, Iowa and Maine are engaged in this research, as well as one researcher from the US Forest Service’s Northern Research Station.

Title: Has the “Mighty 5” Ad Campaign Boosted National Park Tourism in Utah?

Tatiana Drugova, Man-Keun Kim, and Paul M. Jakus
Abstract:

Utah’s “Mighty 5” ad campaign was designed to attract out-of-state visitors to Southern Utah’s five National Parks (NP). Using the synthetic control method, we find the campaign to have been effective in raising visitation at three of Utah’s five NPs: Arches, Canyonlands, and Capitol Reef. No ad campaign effect was found for Bryce Canyon and Zion NP which implies that recent increased visitation in Bryce Canyon and Zion NPs has been driven by the national trends and not the promotional effects of the Mighty 5 campaign. Arches, Bryce Canyon, and Zion NPs currently suffer from excess demand (congestion), and the US National Park Service (USNPS) is actively engaged in visitor management efforts to mitigate the environmental pressures associated with over-visitation. This study provides evidence that an opportunity exists for state and federal officials to craft a coordinated “demarketing” campaign that relieves congestion at overcrowded sites, allows the USNPS to relieve pressure on sensitive ecological systems, and promote tourism to high quality alternative sites.

Title: Low snowpack and Wildfires: A Welfare Double-Edge Sword for Hikers

Author: Sonja Kolstoe

W4133 task: Objectives 1 (Task 1-1 & 1-2) & 3 (Task 3-2)

Presenter: Sonja Kolstoe

Presenter email: shkolstoe@salisbury.edu

Abstract:

The first decade of the 21st century provides us with early examples of the weather anomalies to expect in the future under climate change and thus a unique opportunity to look at the early-onset effects of climate change (Obj. 1, Task 1-2) on outdoor recreation such as hiking (Obj. 3, Task 3-2). Historically large wildfires were infrequent but have in recent times become the new normal in the western part of the United States (Barbero et al., 2015; Flannigan et al., 2013). Earlier studies have looked at how outdoor recreationalists’ value of forests and their attributes change in the presence of wildfires; however, they are looking at time periods with more infrequent wildfires and use a limited amount of spatial data (Englin et al., 1996; Loomis et al., 2001; Hesseln et al., 2003; Englin et al., 2006; Simões et al., 2013). I seek to address this void in the literature using data from the first decade and a half of the 21^st century that captures this shift in the frequency and intensity of wildfires. I also incorporate temporally varying spatial data to more accurately capture the state of the site attributes. In addition, I compare how the timing and magnitude of the fire changes hiker welfare, by estimating both monthly and yearly repeated count models.

I use U.S. Forest Service wilderness permit data from three wilderness areas, Indian Heaven, Mount Adams and Trapper Creek, located in the Gifford Pinchot National Forest and managed by the Mount Adams Ranger station, to explore how visits and hikers’ willingness to pay (WTP) change for forests and hiking trails as a result of wildfires in the study area (Obj. 1, Tasks 1-1 & 1-2, Obj. 3 Task 3-2). The permit data starts in 2001 when hikers were first required to fill out a wilderness permit to hike within these three wilderness areas. I include spatial data on land cover from the USGS, trail attribute data from the U.S. Forest Service through their FSGeodata Clearinghouse as well as monthly data on burned areas from MCD45A1, a dataset on Google Earth Engine. I look at how WTP changes over time as my analysis includes hiking trips from 2001-2014, a time period which includes three smaller fires (less than 500 acres) and two large-scale wildfires (approximately 8,000 acres and 20,000 acres) which occurred in 2008 and 2012. The 2008 wildfire was one of the largest of 29 major fires in the area since 1970 per the National Interagency Fire Report, Cold Springs Fire Long Term Suppression Strategy and Implementation Plan (2008).

The absence of a large wildfire (500+ acres) in any of these areas until July 2008, when the smaller of the two larger wildfires occurred, and then a subsequent larger wildfire affords me the unique opportunity to look at how the scale and intensity difference of wildfires change hikers’ welfare. Their welfare is affected through two different channels: site-closures during the wildfire, as well as long-term alternation of the landscape due to the significant resulting burn scars after each of the large wildfires. Results suggest hikers’ WTP decreased for trails along areas of timber affected by the wildfires while hikers’ WTP does not change for trails through alpine meadows within the area affected by the wildfires.

Title: Measuring Risk Preferences and Perceptions in the Field

Authors: Craig E. Landry, Department of Agricultural and Applied Economics, University of Georgia; Dylan Turner, Department of Agricultural and Applied Economics, University of Georgia

We devise an experimental design that should permit assessment of risk preferences within the context of a mail or internet survey, while allowing further exploration of risk perception. Utilizing historical weather data, we present mail survey respondents with information about rainfall and temperature probabilities. We then allow them to “gamble” their $5 incentive payment on one of four possible weather outcomes, each designed to bracket a range or risk preference parameters (via Constant Relative Risk Aversion formulation of Expected Utility). Responses permit inference on risk tolerance in the context of weather, which is relevant for the survey subject matter – flood risk and climate change. We also assess subjective beliefs about historical weather outcomes to produce information that is useful for Bayesian updating of weather beliefs. Lastly, we utilize novel instruments to assess subjective perceptions of risk (likelihood of hurricane strike) and outcome (conditional damage to structure in hurricane). We present empirical results to explore the utility and limitations of these approaches to assessing decision making under natural hazard risk.

Title: Biases of using aggregate data to infer individual voting preferences 

Authors: Corey Lang and Shanna Pearson-Merkowitz
W4133 task: Task 2-1: Advances in Stated/Revealed Preference Methods
Presenter: Corey Lang
Presenter email: clang@uri.edu
Abstract:
Voting is one of the main determinants of public good provision (Fischel 2001) and an important source of data for examining revealed preferences for environmental goods. Economists have often used aggregated voting data matched with aggregate census data to estimate demand for various environmental goods (e.g., Deacon and Shapiro 1975, Kahn and Matsusaka 1997, Kotchen and Powers 2006, Banzhaf et al. 2010, Holian and Kahn 2015, Burkhardt and Chan 2017).

The purpose of this paper is to assess the extent and causes of bias when using aggregate data to understand individual determinants of voting decisions. This research improves methods of using voting data as a measure of revealed preferences (4133 Task 2.1). Two distinct factors complicate the analysis of aggregate voting data. First, the voting population is fundamentally different than the non-voting population (Leighley and Nagler 2013), and using aggregated population characteristics essentially introduces systematic measurement error due to this mismatch. Second, analysis of aggregate voting data necessarily uses only spatial variation (i.e., between-precinct variation), whereas analysis of individual voting can use both within- and between-precinct variation. In the presence of omitted spatial variables, exclusively using spatial variation can lead to biased coefficient estimates. Monte Carlo analysis identifies conditions when mismatched population data and omitted variables lead to biased inference and the direction of bias.

To empirically assess biases of using aggregate data to infer individual voting preferences, we collect two datasets related to voting on a statewide environmental referendum in Rhode Island called the Green Economy Bonds (GEB). The first dataset mirrors prior studies; we match precinct-level, aggregate voting outcomes to demographic information from the American Community Survey. Second, to obtain individual voting data, we completed a statewide exit poll on Election Day. We enlisted 80 undergraduate and graduate student volunteers who surveyed at 37 sample precinct locations and collected over 2,000 surveys. We treat the individual data and the results that stem from them as correct, and compare the aggregate data and results against this baseline to assess bias.

From basic summary statistics, we find support for conditions leading to the two sources of bias. First, there are statistically significant and large differences among census and voting population. For example, while 22.4% of Rhode Island’s population has a college degree, 34.6% of our exit poll sample was at that education level. Second, there is a large degree of within-precinct variation among the voting and socioeconomic variables. On average, about 95% of variation is within precinct, meaning that aggregate models are using only 5% of total variation to estimate coefficients.

We estimate identical models of voting preferences for GEB using the aggregate data and the individual exit poll data, regressing GEB vote on presidential vote and a large set of socioeconomic characteristics (education, income, race, etc.). We find strong differences between the two models. Coefficients on presidential vote are substantially different: the aggregate model’s coefficient on Voted for Clinton is 28% larger than the individual model, and voting for third party candidates is off by orders of magnitude. The coefficient on college degree is 0.039 for the individual model and 0.073 for the aggregate. The upward bias of the aggregate model is consistent with the simulation finding that groups more likely to vote will have inflated coefficients, though the bias could also be a result of an omitted variable. In additional specifications, we are able to demonstrate the presence of spatial omitted variables and the biasing effect of a mismeasured voting population. However, no adjustment can mitigate the bias of the aggregate model. The conclusion of this paper is that aggregate data cannot infer unbiased individual voting preferences and should be used only with this caveat.

Title: Early Exposure to Nature and Willingness-To-Pay for Conservation

Authors: Liqing Li (University of Illinois at Urbana-Champaign); Amy Ando (University of Illinois at Urbana-Champaign);

W4133 task:

Task 3-2: Economic Analysis of Recreation Services

Task 1-1: Economic Analysis of Agriculture, Forest, and Rangeland Resources, Open Space, and WUI Zones

Presenter: Liqing Li

Presenter email: lli40@illinois.edu

Abstract:

[Include reference to w4133 task]

Current urban spatial structures reduce the interaction between nature and human beings. However, it is unclear how changes in people’s exposure to nature affect people’s preference for nature. Based on the standard assumption in neoclassical economics that individual preferences are fixed over time, the marginal value of nature increases with its scarcity. However, this pattern may not hold true if the changes in an individual’s interaction with nature affect his/her preference for nature. A negative change in an individual’s preference for nature shifts the demand curve to the left. People’s marginal value for nature may stay the same or even decrease with scarcity of nature. This paper uses a choice experiment (CE) and data on individuals’ childhood experiences with nature to evaluate which pattern holds true.  

An individual’s demand for local amenities affects people’s location choice decision. Existing research has examined factors that affect an individual’s demand for different types of local amenities (Bishop et al., 2018; Isen et al., 2017). Nature has been found to be an important amenity, but the drivers of people’s demand for nature amenity has been little examined. Childhood plays an important role in building up an individual’s experience with nature. Current urban spatial structures can affect the next generation’s interaction with nature and thus affect the individual’s willingness to pay (WTP) for nature via experience. This paper quantifies how the current generation’s early-life experience with nature affects their current WTP for nature.

A concept of amenity capital (human capital) (Krupk, 2009) may be used to explain the relationship between nature amenity and experience. The first generation’s choice of residential location affects the following generation’s ability to build up amenity capital by experience. Children build up nature related amenity capital through experiencing nature, and this amenity capital helps them enjoy nature more as it generates higher utility when visiting nature.

In our hypotheses, we expect that childhood experience with nature affects respondents’ WTP for open space conservation. Respondents’ early-life experiences are measured in different dimensions, including: how often the individual visited nature and engaged in nature-related recreational activities, how often the individual received environmental education, and how easy it was for an individual to interact with nature (measured by the percentage of land that was developed in respondents’ hometowns). To test our hypotheses, we carry out a CE in three states - Illinois, Iowa, and Minnesota - to estimate individuals’ WTP for different features and recreational activities of a hypothetical restored grassland. We also gather data on respondents’ early-life experience with nature and test whether those features of individuals affect their current WTP for preserving nature. We conduct this study in the context of hypothetical grassland restoration. Thus, this paper also expands our understanding of the values people place on preservation of grassland ecosystems; only Dissanayake and Ando (2014) have estimated the social value of grassland ecosystems and that work did not estimate values of recreation attributes.

To examine the relationship between early-life experience and individual’s WTP, we use a mixed multinomial logit model with interaction term to analyze responses from the choice experiment. We also use a random parameter latent class model as an alternative perspective to study the effects of childhood experiences on value for grassland.

The results reveal whether the demand of environmental goods is affected by an individual’s childhood experience and whether the impact of current urbanization on the next generation’s valuation for nature needs to be taken seriously (Task 1-1). Our findings also estimate respondents’ average WTP for the availability of different recreational activities of a hypothetical restored grassland (Task 3-2). That information can guide policymakers by indicating which recreational activities of a restored grassland should be prioritized.

Title: Estimating the Impact of Fires on Recreation in the Angeles National Forest Using Combined Revealed and Stated Preference Methods

Authors: Sophia Tanner, Frank Lupi*, Cloe Garnache

W4133 task: fits any of Obj 1-1, 1-2; 2-1 or 3-2.

Presenter: Frank Lupi

Presenter email: lupi@msu.edu

Abstract:

Wildfire frequency and severity are increasingly important issues in the western United States, as fires threaten lives, properties and outdoor amenities. This paper uses a novel combination of revealed preference data from site intercepts and stated preference data from online surveys to estimate the welfare impacts of different fire scenarios at recreation sites within the Angeles National Forest, which is a popular recreation destination on the outskirts of the Los Angeles metropolitan area. In 2016, we collected onsite visitation data at 32 sites in the Angeles National Forest. Site visitation was measured by randomly sampling recreation sites within three strata defined by expected use level (high or low), weekend or weekday, and morning or afternoon. The on-site visitation data includes on-site counts and short intercept interviews of 2,266 visitors. Next, we followed up with an online survey to collect contingent behavior data. Under eight different fire history scenarios which corresponded to the vegetation at the site they visited, respondents were asked contingent behavior questions to ascertain if they would still make the same trip as before. Our empirical strategy exploits both the onsite and contingent behavior data to estimate welfare effects of fires. The sampling design weights and site counts allow us to estimate total visits to each site as well as the number of visits a site receives from each origin zip code. With the zonal visitation data, we estimate a multi-site zonal travel cost model following the approach developed by von Haefen et al. for the Deepwater Horizon oil spill. The zonal model uses on-site sampling and intercept probabilities to estimate rates of visitation from each origin zip code, allowing us to estimate a multi-site recreation demand system with a full set of alternative specific constants (ASCs). This model provides estimates of visitation to each site under unchanged conditions which are then calibrated via contraction maps to estimates of the percentage of visitors who would have still visited the site at which they were intercepted under alternate fire history scenarios. The contingent behavior responses are thus embedded within the demand system and the implied fire preference parameters are estimated using contraction maps. This novel RP/SP estimation procedure allows us to value both site closures and the impacts of fire history on sites after they reopen. Results contribute to forest management when facing increasing threats of site closures by providing insight into potential impacts during and after closures in a popular urban national forest. Of the fire scenarios presented, recent forest fires are the costliest, causing estimated welfare losses of over $2.2 million per summer season for an average affected site. We find that recent forest fires cause larger trip and welfare losses than less recent forest fires or shrub fires, with forest fires decreasing welfare by roughly $29 per lost trip. Applying this method to a large fire that affected many sites in our study area, we illustrate how losses decrease over time, but can continue well after sites are re-opened due to lasting effects on the landscape.

Author: Taro Mieno 

Abstract:
In this study, we will seek to understand the how crop insurance and groundwater allocation limits with temporal flexibility affect agricultural producers’ groundwater use. Groundwater is a vital resource for high-productivity agriculture in U.S. Due to ever growing demand for water in the competing sectors, agricultural producers are under increasing pressure to conserve water. One of the popular policy tools to limit groundwater use is groundwater use quota: allocation limits. Several Natural Resources Districts (NRD) in Nebraska and LEMA region in Kansas has adopted allocation limits to conserve groundwater. Given the high annual variability in precipitation and evapotranspiration, major determinants of crop water demand, allocation limit policies typically have temporal flexibility. That is, instead of setting a hard groundwater extraction cap each year, farmers have a flexibility to satisfy allocation limits over multiple years. For example, Upper Republican NRD has a total of 65 inches per acre over 5 years. Producers are allowed to use however much water they would like in a year as long as the total groundwater extraction does not exceed 65 inches per acre. The problem producers face under such a policy is a dynamic optimization of groundwater use. Presumably, producers should save some water in early years so that they would have enough water to avoid a catastrophic crop failure in later years when a severe drought hits. While it is straightforward to imagine how producers behave under allocation limits with temporal flexibility, it is hard to imagine how allocation limits interacts with crop insurance to affect groundwater use behavior by agricultural producers. Since producers are safeguarded by catastrophic profit loss by crop insurance even when producers run out of allocation in a dry year in the later years, it seems that producers can behave more similarly to the case when there is no allocation limits.

The objectives of this study is two-fold: we will seek to examine (1) how irrigation pattern changes within an allocation period and (2) if total water use within an allocation period is greater under crop insurance compared to the case without crop insurance. To achieve this goal, we will employ a numerical dynamic simulation approach based on crop water production function generated by AquaCrop, a crop simulation model. Our preliminary results indicate that producers use more water in early years, and gradually reduces water use on average over the allocation period under crop insurance. On other hand, producers use less water in early years, and gradually increases water use on average over the allocation period without crop insurance. That is, water use pattern within an allocation period reverses when producers are under crop insurance. The impact of crop insurance on water use is mixed: it could either increase or decrease total water use.

This study contributes to the literature of crop insurance and water resource economics. Crop insurance is typically evaluated by itself without its potential interactive impacts with other resource use policies. This study is innovative in that we are the first to explore the interaction of a water resource use policy and crop insurance.

Waters of the United States: Upgrading wetland valuation via benefit transfer

Klaus Moeltner, Jessica Balukas, Elena Besedin, Ben Holland

W4133 Objectives:

Objective 2-2: Advances in BT Methods

Objective 3-3: Economic Analysis of Water Quality

Abstract:

With this study we respond to the ongoing debate on the legal protection of wetlands, and the concurrent need to understand the societal benefits created by them. Using updated meta-data on wetland valuation we illustrate how a state-of-the-art meta-regression framework that is consistent with economic theory can be adapted to generate benefit transfer predictions for incremental changes in wetland acreage over space and time. We apply this framework to estimate losses in benefits for realistic changes in wetland acreage for some sub-watersheds, as can be expected under the proposed re-definition of the “Waters of the United States” to be protected under the Clean Water Act.

Title: Spatial Leakages from Conservation Policies: An Application to the Kansas Conservation Reserve Enhancement Program

Authors: Mani Rouhi Rad, Dale Manning, Jordan Suter, Chris Goemans, and Zaichen Xiang

W4133 tasks: Task 1-1: Economic Analysis of Agriculture, Forest, and Rangeland Resources, Open Space, and WUI Zones

Task 2-3: Advances in Spatial/Environmental Nexus

Presenter: Dale Manning

Presenter email: dale.manning@colostate.edu

Abstract:

Conservation incentive programs provide a vital tool for federal policymakers to encourage the country’s agricultural producers to adopt practices and strategies consistent with local and national objectives.  For example, the Conservation Reserve Enhancement Program (CREP) partners with local managers and regulators to achieve local conservation goals.  CREP in the Arkansas River Basin (ARB) of Kansas aims to reduce groundwater use by paying producers to retire irrigated land.  While the policy prevents GW pumping at the retired wells, the responses of active neighboring wells determine the final impact of the program. In this paper, we examine the spatial leakages around this program, and how they affect the program’s impact over time.  Theoretically, a retired well leads to fewer competitors and decreased pumping at neighboring wells, conditional on water availability.  Yet, less pumping means that more water becomes available over time, leading to an increase in pumping.  Finally, as neighbors adopt conservation behavior, this may lead to a behavioral leakage as others become more conservation minded. 

The relative magnitudes of these effects determine the net impact of the program over time.  Therefore, to resolve the theoretical ambiguity, we use well-level data on pumping in the ARB to econometrically identify the spatial leakages associated with CREP.  We begin with a difference -in-difference specification to estimate the impact of CREP enrollment on pumping at wells that remain active.  Next, a model with well fixed effects and time controls allows us to explore the mechanisms driving the spatial leakages.  Finally, to examine the net impact of CREP on GW use over time, we link a MODFLOW model of the ARB with a water demand function to simulate the impact of well retirement on neighboring wells’ water availability and pumping.

We find that CREP enrollment causes neighboring wells to pump less water on average, and this holds conditional on water availability.  Further, we find no evidence of a move towards conservation-mindedness.  This suggests that the CREP impact is driven by changes in pumping resulting from fewer competitors using the common property resource.  Simulations using the linked model suggest that the competition effect dominates for X years, beyond which more water availability leads to more pumping than would have happened without the CREP program.

This result relates to the W4133 task 1-1 because it provides an economic analysis of agricultural use of a common property resource that creates substantial value for landowners and well operators.  Spatial spillovers from resource management policies has important implications for users of many public and private natural resources, including forests, rangeland, and residential development.  The spatial nature of aquifer resources means that modeling its uses also requires innovations at the spatial/environmental nexus (task 2-3).  Important policy implications include the potential for positive conservation spillovers from programs that reduce competition for shared resources.  In this case, policy goals may be reached at a cheaper cost than anticipated, as private responses have synergies with the direct impact of management programs.

Title:  Assistance Needed?  The Willingness to Pay for Flood Insurance

Authors:  Noelwah R. Netusil, Carolyn Kousky, and Howard Kunreuther

W4133 tasks: Task 1-2; Task 2-1

Presenter: Noelwah R. Netusil

Presenter email: netusil@reed.edu

Abstract:

Flooding is the natural disaster that causes the most property damage in the United States. Federal disaster aid for victims is currently limited, so to rebuild or replace damaged property without either incurring additional debt, drawing down savings, or diverting current consumption, those at risk need flood insurance.  Yet, nationwide, many people forgo flood coverage.  High premiums are likely a key driver of this decision, but the price elasticity of flood insurance has been difficult to estimate empirically due to a number of challenges. These include the following factors: homeowners in high risk areas are mandated to purchase a flood policy if they have a federally insured mortgage, premiums are correlated with risk, and price is not observed for the uninsured. This study draws on contingent valuation methods to elicit willingness-to-pay (WTP) for flood insurance from a sample of residents in two Portland, Oregon neighborhoods at risk of flooding.  We examine how WTP varies with income, risk perceptions, objective risk, previous experience with flooding, and the length of time residing in one’s current home and neighborhood, which relates to W4133 Task 2-1.  We also investigate how participant uncertainty about their responses influences their overall WTP for flood insurance.  Results from this study can inform the ongoing policy debate about whether and how premium assistance should be provided to help encourage the purchase of flood insurance for residents in hazard-prone areas, which is related to W4133 Task 1-2.

Title: The Price of Powder: Evidence on the demand for snow from short term property rentals 

Authors: Bryan Parthum, Peter Christensen
W4133 tasks:

• 2-1: Advances in Revealed Preference Methods
  • 1-2: Economic Analysis of Natural Hazards (climate change)
  • 3-2: Economic Analysis of Recreation Services

Presenter: Bryan Parthum

Presenter email: parthum2@illinois.edu

Abstract: We estimate structural demand parameters for wintertime recreation in 31 destination markets across the contiguous United States. Our first stage introduces a highfrequency hedonic method to instrument for the endogenous price of a short term property rental. By exploiting short-run shocks in weather and recreation demand, we derive hedonic price schedules for 237 ski resort towns and estimate daily state-specific implicit prices for mountain snowpack. We then estimate utility functions for snow tourists in each state. These resulting demand parameters allow us to project damages under two future climate scenarios, RCP4.5 and RCP8.5. We incorporate snow tourist information to map estimates of willingness to accept to each U.S. state and show that significant heterogeneity exists across states. Lastly, we estimate a lower-bound on total willingness to accept across the U.S. to be between $5.9 to $8.09 billion per snow season due only to reductions in mountain snowpack.

Title: Meta-Analysis to compare Certainty Follow-up and Cheap Talk to Reduce Hypothetical Bias

Authors: Jerrod Penn-Louisiana State University and Wuyang Hu-Ohio State University

W4133 Task: 2-Economic Valuation Methods

Presenter: Jerrod Penn

Presenter email: jpenn@agcenter.lsu.edu

Abstract: Hypothetical Bias (HB) continues to be a major concern among stated preference practitioners (Task 2). Two of the earliest and most popular methods to reduce HB are certainty follow-up and cheap talk. Certainty follow-up reduces HB by asking respondents how certain they are of their answer to the valuation question. Cheap Talk works by explicitly warning respondents about HB and to answer as if it were a real purchase decision. A number of studies have implemented either technique, a hypothetical control elicitation to establish their efficacy relative to an uncorrected hypothetical welfare estimate, as well as a real elicitation to demonstrate their efficacy to reduce actual HB. We use meta-analysis to investigate and compare the relative ability of both approaches using both sets of measures, the ability of certainty follow-up and cheap talk to alter welfare estimates relative to hypothetical control and real welfare estimates. At present, 19 studies certainty follow-up and 30 cheap talk studies meet the requisites for inclusion. Preliminary analysis shows that certainty follow-up is more effective than cheap talk at reducing the welfare estimate relative to hypothetical controls and correspondingly, the extent of HB when comparing against the real elicitation welfare estimates.  

Title: WATER AND CONCRETE: COMPLEMENTARITY (NOT SUBSTITUTION) BETWEEN NATURAL AND PRODUCED CAPITAL

Authors: Mani Rouhi Rad, Wiktor (Vic) Adamowicz, Alicia Entem, Eli P. Fenichel, and Patrick Lloyd-Smith

W4133 Task: Task 2-1: Advances in Stated/Revealed Preference Methods

Presenting author: Mani Rouhi Rad

Presenting author email: mani.rouhi_rad@colostate.edu

Estimating the substitutability of natural capital is important for the valuation of natural capital and the ecosystem services provided. The existing theoretical and empirical literature in natural capital focuses on the substitutability of natural capital and other forms of capital (Hartwick 1977) or the flows of ecosystem services and materials (Cohen et al. 2018) rather than the stock of natural capital.

We contribute to natural capital theory and derive conditions under which produced and natural capital are complements. In short, produced capital investment often occurs following a short-run cost curve, because market forces do not lead to increased production of natural capital in order to reach a new point on the long-run cost curve. Natural capital remains quasi-fixed - a defining feature of nature capital and short-run cost curves. This holds despite produced capital investments being made with a relatively long-term planning horizon. We then provide empirical evidence on complementarities between natural capital (fresh water) and produced capital (a new lock system) in the case of the Panama Canal expansion.

The Panama Canal consists of two sets of locks (produced capital) that connect Gatun Lake to the Pacific and Atlantic oceans. The depth of Gatun Lake provides clearance for ships’ draft and serves as the natural capital for ships passing through the canal. Passage of ships through the locks requires the release of water from the lake (utilization of natural capital). Under the pressure of increasing cargo ship sizes and investments by competitors such as the Suez Canal, the Panama Canal Authority invested in a new path including six new locks that opened in June 2016 to allow the transit of larger ships and water saving basins which reduce the release of water from the Gatun Lake per transit (investment in conservation).

We contribute to the literature on the complementarity or substitution between natural and produced capital in two ways. First, we distinguish between the capital space and the utilization space. The literature mostly considers utilization (or flows) of inputs and does not distinguish capital levels (stocks) from flows. We show that stocks and flows may not have the same relationship. We also show that the substitution or complementarity relationship depends on the type of investment. The nature of the relationship is not clear ex-ante. The observed construction of water saving basins leads the water and locks to be complements. We show that without these water saving basins, the natural and produced capital would have been substitutes in capital space.

Second, we provide empirical evidence of the complementarity between natural capital and produced capital. The main focus of the natural capital literature has been on substitutability of natural capital and other forms of capital, and Van den Bergh (1999) suggests that finding complementarity will be “very difficult to impossible”. We show that while the investment in produced capital has resulted in substitution in the utilization space, there is complementarity between produced capital and natural capital.

Our study contributes to the valuation of natural capital and ecosystem services provided using revealed production decisions of the firm (Task 2-1). Substitution or complementarity of natural and produced capital are critical for such valuations. Our theoretical analysis derives conditions for complementarity and substitution of these two forms of capital and our empirical analysis provides evidence for existence of complementarity between them. Importantly, our study finds that distinction between capital space and flow space is critical for valuation of natural capital and ecosystem services. Investments in produced capital in the form of infrastructure are often used for resource management. Our study sheds light on their effects on the value of natural capital and ecosystem services.

Title: Feeling the Burn: The Optimal Timing of Prescribed Fires for Recreational Benefits

Authors: Katherine Zipp, Pennsylvania State University; Yau-Huo (Jimmy) Shr, Iowa State University

W4133 task: 1-1: Economic Analysis of Agriculture, Forest, and Rangeland Resources, Open Space, and WUI Zones; 3-2: Economic Analysis of Recreation Services

Presenter: Yau-Huo (Jimmy) Shr, Iowa State University

Presenter email: yhshr@iastate.edu

Abstract:

Prescribed fire is getting more widely recognized and used in many parts of the United States. Although prescribed fire is an effective approach for fuel reduction and ecosystem restoration, it also comes with concerns such as (short-term) limited forest access and accidental spread. When making decisions about how often to use prescribed fires, public forest managers balance desired ecological and recreational outcomes. However, little is known about the optimal use of prescribed fires to achieve recreational outcomes.

In this paper, we fill this gap and make two contributions to the literature in answering the research question: how often should forest managers use prescribed fires to maximize recreational benefits? Our first contribution is that we use a Faustmann model, a long-standing dynamic optimization model for maximizing timber harvest profits, to estimate the optimal prescribed burn interval for maximizing recreational benefits. Our model builds on previous literature, such as Yoder (2004), that considers prescribed fire being used to maximize timber harvest conditional on the potential damage from wildfire. However, we recognize that the majority of prescribed fires occur on public land that is not managed for timber production but rather managed for recreational benefits. Our second contribution is that we estimate a recreational benefit function over time after a fire by coupling recreational welfare estimates with ecological estimates of forest dynamics after a fire. By using an ecological understanding of how various attributes of a forest change over time after a fire, we are able to improve estimation of the benefits that recreationalists receive from forests after a fire (Boxall, Watson, & Englin, 1996; Englin, Boxall, & Hauer, 2000). With these two contributions we are able to better inform management of forest and wildland urban interface (WUI) areas in terms of improving recreation services.

In 2017-2018, we conducted intercept surveys of 302 forest users in Central Pennsylvania and the New Jersey Pine Barrens. These two areas provide similar recreation services but have drastically different wildfire hazard. The survey includes a choice experiment to estimate the willingness-to-pay for different forest attributes including wildfire hazard, visibility in forest, and wildlife habitat quality. Travel time required to forest area is used as the cost attribute. To translate travel time into monetary cost, we adopt the approach proposed in Lloyd-Smith, Abbott, Adamowicz, & Willard (2017) to estimate the individual value of leisure time. Using a panel of ecological experts, we model how the three forest attributes change over time after a fire. The net benefit function in the Faustmann model is constructed using the information of individual benefits and the dynamic of forest attributes. The personal discount rate is estimated following the standard approach (Coller & Williams, 1999).

Under different model and parameter specifications, the optimal burn intervals are between 5 – 16 and 19 – 33 years in New Jersey and Pennsylvania, respectively. These estimates are in line with the optimal burn interval when solely considering ecosystems (5 – 10 years for NJ and 20 – 30 years for PA), but are greater than the optimal interval considered by fire managers. The results also shed light on the willingness-to-pay for the selected forest conditions and associated heterogeneity between two regions. In addition, our survey results provide one of the first assessments of the citizen perceptions of prescribed fires in the Mid-Atlantic region.

Title: A Physical Input-Output Table (PIOT) model of Illinois for estimating Life Cycle Environmental Impacts of Nitrogen Emissions

Authors: Shweta Singh

W4133 task: 2-1: Advances in Stated/Revealed Preference Methods

Presenter: Shweta Singh

Presenter email: singh294@purdue.edu

Abstract: Input-Output (IO) models allow to calculate total life cycle economic and environmental impacts of production in an economic region. Current methods of economic evaluation fail to capture the full cost of ecosystem services contribution due to limited models that capture the interdependence of economic sectors in physical units. Developing physical input-output table (PIOT) based models allow to capture this physical interdependence between economic sectors in a region. In this work, a PIOT model for Illinois capturing the interactions between agrobased sectors has been developed. The PIOT model captures the physical interactions between sectors in terms of Nitrogen flows. An environmentally extended PIOT is also developed that connects the production in sectors with nitrogen emissions in the state, thus allowing to calculate the total direct and indirect impact of production in the state of Illinois on nitrogen emissions. This physical model can form a basis of economic valuation of nitrogen related ecosystem services using cost of nitrogen emissions in the region. The reliability of the direct and indirect impact estimation is improved as the physical model is based on mass balances that ensures the accuracy of intersectoral dependence. Hence, PIOTs can provide an advancement to improve the accuracy of economic valuations.

Title: Assessing the private and social benefits of forest concessions in the Maya Biosphere Reserve

Authors: Corinne Bocci, Brent Sohngen and Daniela Miteva

W4133 task: Task 2-3: Advances in Spatial/Environmental Nexus

Presenter: Brent Sohngen

Presenter email: Sohngen.1@osu.edu

Abstract:

While forests provide multiple ecosystem services, it is often difficult to measure their amenity and private values. If land markets are efficient and property rights are well-defined, the private benefits of maintaining forest cover, such as harvesting timber and non-timber forest products, will be captured through market transactions.  In many developing regions, however, property rights are not well-defined and there is limited information on private forest values. Additionally, the value of the many ecosystem services that forests provide to the benefit of society is often not quantified and monetized. To promote local conservation of forest stocks, some developing country governments have provided local community groups with property or land-use rights if they agree to manage forest resources sustainably. Community-based common-property resource (CPR) management systems follow the advice of Gordon (1954) and Scott (1955), who recommended privatizing the resource to reduce over-extraction, but rights are vested with groups rather than single individuals or entities.

Most efforts to date to assess community-based tropical forest management strategies have focused on whether the projects have physically reduced deforestation.  Few studies have evaluated whether households involved in managing the resource benefit from the arrangement, and whether there are tradeoffs with providing other ecosystem services, such as carbon sequestration. To address this issue, this study evaluates whether households benefit from community-based forest concessions in the Maya Biosphere Reserve in Guatemala, and whether the arrangement increases carbon in the foersts. The household-level benefits are measured as the increase in income gained by community members involved in managing the concessions when their efforts to exert property rights and manage the landscape sustainably are successful.  The broader global benefits are the gains in carbon stored due to avoided deforestation. To calculate the household-level benefits, we use a dataset collected from households in the Maya Biosphere Reserve in 2012 and 2017 and a two-stage least squares, instrumental variable approach to estimate the effect of concession membership on household income.  We calculate the gains in carbon sequestration using a matched panel of concession areas and similar non-concession areas within the reserve. The results show that the value of forest concession management in the Maya Biosphere Reserve from 2012 to 2017 is about $17,166,005.  Most of this benefit is obtained privately by households, which suggests that common-property resource management systems can result in significant welfare gains for households that manage forests.

This study advances the literature in Task 2-3, Advances in Spatial/Environmental Nexus, in the W4133 project by developing methods to value the private and public benefits of a common-property resource management system. One contribution is that we use rigorous impact evaluation methods to determine whether the concessions are effective at reducing deforestation and increasing household income and compare the value of the additional forest carbon sequestered as a result of deforestation reduction and household income benefits to an estimate of concession implementation costs. Another contribution is that we highlight where trade-offs and complementarities between the ecosystem services exist and whether this effect varies based on the characteristics of the concession communities.  We find that, for households with forest-based histories, concessions increase the amount of forest carbon sequestered and increase concession member income.  However, for concessions managed by households from larger towns and cities, our results show that they are receiving the largest income benefit, but are deforesting areas with higher carbon sequestration values.  Nevertheless, our overall results suggest that the community-based forest management in the Maya Biosphere Reserve is an effective conservation and development strategy.

Title: Pursuit of Independence: A study into the households’ willingness to pay for reducing dependence on the local grid for electricity. 

Authors: Dilek Uz, Katie Champaigne
W4133 task: Task 2-1: Advances in Stated/Revealed Preference Methods
Presenter: Dilek Uz
Presenter email: dilekuz@unr.edu
Abstract

Deployment of residential solar photovoltaic (PV) panels have skyrocketed in the recent years in the US. In 2016 alone, PV market grew by 97 %, with more than 370,000 individual installations. Currently more than 40 states have regulatory settings in which a utility customer can get compensated for their exports into the electricity grid (unused portion of their electricity generation sourced by their rooftop solar panels.) Also in many states with renewable portfolio standards (RPS) customers get rebates on solar panel purchases while the local utility gets to count the generated electricity towards RPS compliance. At the federal level, residential solar panels are incentivized by a 30% investment tax credit. The justification for these policies is the public good nature of the environmental services associated with reducing consumption of electricity from the local grid, the majority of which is sourced from fossil fuels.

Having rooftop solar panels also provide various private benefits to the owners. They include savings on power bill, warm-glow, signaling environmental preferences (due to the conspicuous nature of owning solar panels), and the feeling of reduced dependence on the grid.

In this study, we design an online choice experiment to measure the willingness to pay for each of these attributes. Respondents, who are chosen from certain demographic groups, are offered a set of binary and trinary choices. Options are (i) “Default option” (i.e. fully relying on the local electricity supplier with the conventional fuel mix); (ii)“Green pricing” (i.e. still purchasing all electricity from the grid but paying a premium for an increased share of renewables); and (iii) “Solar panels” (i.e. installing a rooftop solar panel and using the electricity generated by the panel and supplementing from the grid when necessary).

There are many studies in the environmental economics literature that study the willingness to pay for renewable energy (e.g. (e.g. Roe et al. (2001), Borchers et al. (2007), Scarpa and Willis (2010)). Recent studies that review the literature include Oerlemans et al. (2016) and Sundt and Rehdanz (2015). A small share of this literature focuses on rooftop solar panels (e.g. Soskin and Squires (2013), Su et al. (2018)) To our knowledge, this is the first study that looks at willingness to pay for reduced dependence on the local electric grid as an attribute. Rooftop solar companies’ sales pitches are known to include the rhetoric that utility companies are monopolies who overcharge for their services and owning solar panels will deliver redemption from them.4 . According to Triandis et al. (1990), being independent is one of the top American values and the solar rooftop sales people seem to be appealing to that. The goal of this study is to quantify this value. Additionally, we look at the heterogeneity of this value with respect to demographic variables as well as political affiliation.

The results of this study will inform policy makers on the appropriate incentive levels for rooftop solar panel deployment. High willingness to pay for private benefits would imply that consumers would still purchase with lower incentives which would save tax payer money.

Title: Estimating a Demand Curve for Land Conservation

Authors: Alex Blanchette, Corey Lang, Michelle Peach, Jarron VanCeylon
W4133 task: Task 1-1: Economic analysis of Agriculture, Forest, and Rangeland Resources, Open Space, and WUI Zones
Presenter: Jarron VanCeylon
Presenter email: jvanceylon@uri.edu

Abstract:

Long periods of urban expansion have resulted in the significant loss of open space (US Forest Service 2019). Open space provides benefits to individuals and neighborhoods, from recreational and visual amenities to other ecosystem services like air and water quality (Irwin 2002, Anderson and West 2006, US Forest Service).  As a result, permanently protecting open space is a priority for local, state, and federal governments and NGOs. However, conserving land is costly, and in many cases the benefits of conservation are unknown or are imprecise.

In this paper we examine the effects of open space on housing values across numerous markets in an effort to construct a demand curve for protected open space. Several studies (e.g., Boyle et al. 1999, Netusil et al. 2010, Mei et al. 2017) have used multiple markets to estimate a demand function for an environmental amenity. However, the geographic scale of our data is considerably larger than prior work. We collect housing and population data and land cover data from 214 metropolitan statistical areas (MSA) in the Eastern half of the United States. Housing and population data come from the American Community Survey and are collected at the census block group level, the finest scale publicly available. Land cover data come from the USGS Land Use Land Cover database, USGS elevation rasters, and USGS Protected Areas Database. From these data, we discern protected and unprotected open space, steeply sloped land, and natural inland water features.

Our method involves a two stage framework. The first stage estimates marginal willingness to pay (MWTP) of protected open space for each MSA separately using hedonic valuation. Our specification controls for unprotected open space, a large number of housing and demographic variables, and county fixed effects and quadratic functions of latitude and longitude to guard against spatial unobservables that could be correlated with housing prices and open space. In the second stage, we estimate an inverse demand curve for protected open space by comparing MWTP estimates and geographic and population characteristics across all MSAs. Specifically, we regress estimated MWTP on the quantity of protected open space, average income, and geographic determinants of undevelopable land.

Our first stage results are consistent with previous literature and indicate positive housing premiums associated with proximity to open space. For example, a 1% increase in protected open space is associated with a 0.18% increase in housing prices in the New York City MSA. However, there is considerable heterogeneity in valuation, the 25th and 75th percentiles for estimated MWTP are about $100 and $1000, respectively. In the second stage, we seek to explain that variation in valuation by estimating an inverse demand curve and including multiple shifters of demand. Consistent with expectations, we find that demand slopes downward: a one standard deviation increase in protected open space decreases MWTP $42. We also find that protected open space is a normal good, a one standard deviation increase in average income increases MWTP by $134. Lastly, we include undevelopable area as an additional demand shifter, with the idea that naturally occurring undevelopable area is a substitute to preserved open space. We split undevelopable area into steeply sloped land and water/wetlands, and find that these geographic features have opposite effects on MWTP – steeply sloped land does act as a substitute, but water and wetlands act as a complement.

Title: A Bayesian Model for Small Discrete Choice Samples 

Authors: Lendie Follett, Brian Vander Naald
W4133 task: 2-1
Presenter: Brian Vander Naald
Presenter email: brian.vandernaald@drake.edu
Abstract
Task 2-1: Bayesian estimation has been used in the environmental economics literature to analyze discrete choice data. Train (2003) popularized the Bayesian technique for choice data. Rigby and Burton (2006) employed a Bayesian bounded logit model with an inverse Wishart prior on the variance to estimate a WTP value for genetically modified food. Balcombe et al. (2009) also use an inverse Wishart prior to investigate attribute non-attendance using choice experiment data examining beef produced from stem cell technology. Recently, Akinc and Vandebroek (2018) developed a Bayesian mixed logit model in which they noted that choice of prior in Bayesian estimation is important. They note that use of the inverse Wishart prior on the variance is inappropriate because it exerts too much influence on the results. They use an uninformative prior (LKJ) developed by Lewandowski et al. (2009), which has desirable properties. While they test this model with a large simulated dataset, they do not test it with actual data. We extend their work here, developing a Bayesian mixed logit model using the LKJ prior on the variance to extract inference from a small dataset from choice experiment data gathered in 2015. We estimate posterior probabilities for willingness to pay for reducing the rate of glacier recession, and examine variation in WTP based on sociodemographic differences. To our knowledge, this is the first Bayesian study in the WTP literature to focus on extracting inference from small discrete choice samples. This model is potentially very useful, as it could reduce the cost of conducting discrete choice studies.

We develop a data driven Bayesian framework for the estimation of mixed logit models for small discrete choice samples. In this framework, we build upon past research by incorporating demographics into the mean of the distribution of individuallevel preferences. For the covariance matrix on individual-level preferences, we note issues with the traditionally used inverse Wishart prior. We instead propose a prior on the covariance matrix that allows one to specify independent prior distributions on the correlation matrix and the standard deviations. We propose the use of the LKJ prior for correlation matrices in conjunction with the half Cauchy distribution for standard deviations as a way to be able to accommodate a wide variety of data-generating patterns.

We test the model using choice experiment data gathered in 2015 examining preferences for reducing the rate at which glacier ice is disappearing. 166 surveys were administered, 149 were completed, and after quality control measures were taken, 126 usable surveys remained. Of the 126 reliable surveys, 37 saw four choice scenarios and 89 saw two choice scenarios. Respondent characteristics were similar to those of other visitors during that time. To examine WTP differences along sociodemographic margins, the sample was further limited to those who provided answers for political preferences, sex, and environmental organization membership. After this final sample reduction, we are down to 98 respondents.

Results indicate that the mean monthly WTP for 0.15km3 rate of glacier loss is substantially higher than the WTP for 0.20km3 rate of glacier loss ($149 versus $24). Further, there is a 90% posterior probability that WTP for 0.15km3 rate of glacier loss exceeds zero, compared to a 63% posterior probability that WTP for 0.20km3 rate of glacier loss exceeds zero. These WTP values differ by political preference, membership in an environmental organization, and sex. Females are WTP more for both rates of glacier loss than their male counterparts, regardless of political preference, environmental organizational membership status. On the other hand, females are substantially less likely than their male counterparts to choose an alternative if it uses an insulated ice blanket to achieve a given level of glacier loss.

Title: Spatial and Temporal Dimensions of the Value of Coastal Recreational Fishing in US Waters

Authors: Alexandra Naumenko and Roger H. von Haefen, North Carolina State University

W4133 Objectives: Obj 2: Economic Valuation Methods, Task 2-2: Advances in Benefit Transfer Methods, Task 2-3: Advances in Spatial/Environmental Nexus; Obj 3: Integrated Ecosystem Services Valuation and Management, Task 3-2: Economic Analysis of Recreation Services

Presenter: von Haefen

Presenter Email: rhhaefen@ncsu.edu

Abstract:

The Bureau of Economic Analysis recently estimated that the outdoor recreation economy contributed 2.2% to Gross Domestic Product and was growing slightly faster than the overall U.S. economy.  Similarly, Siikimaki (2007) found that outdoor recreation time use has roughly doubled over the past four decades but has remained relatively flat in the 21st century.  However, how the nonmarket value, or consumer surplus, of outdoor recreation has changed over time as well as across space remains poorly understood.  The current paper aims to fill this gap in the literature by examining trends in the per trip value of coastal recreational fishing over a thirteen-year period from Maine to Louisiana (Task 2-3).  Leveraging data from NOAA’s Marine Recreational Information Program (MRIP) for approximately 300,000 shoreline angling trips to sites along the Atlantic and Gulf Coasts from 2004 to 2016 (Task 3-2), we estimate how the value of a coastal recreational trip varies across time and space.  A repeated discrete choice model of recreational participation and site choice is estimated separately by region (New England, Mid-Atlantic, Southeast, Florida and Gulf Coast) and two-month period, and per trip values of lost trips are calculated.  Our results suggest considerable within-year heterogeneity, but surprising homogeneity across years.  The per trip value of recreational fishing was largely unaffected by the Great Recession, although using an internal meta-regression analysis, we find suggestive evidence that per trip values have increased modestly during the recent expansion.  Further, our results reveal substantial spatial heterogeneity across the United States, with angling trips in the Southeast having the highest economic value.  Collectively, these results deepen our understanding of the temporal and spatial dimensions of recreational angling and may assist government agencies working in sparse data environments when transferring benefit estimates from one region and point in time to another (Task 2-2).

Title: The Impact of Drought on Crime in South Africa

Authors: Kyle Wilson

W4133 task: Task 1-2: Economic Analysis of Natural Hazards (fire, invasive species, climate

change)

Presenter: Kyle Wilson

Presenter email: kylew@uoregon.edu

Abstract

All societies rely on water as an essential input for life and economic activity. Droughts and other negative shocks to the water supply may contribute to negative societal outcomes such as crime. This paper tests one channel through which drought affects social outcomes: crime. Specifically, I estimate the effect of South Africa’s recent extreme drought (2015–2018) on crime. Using a police-station by year panel, I exploit variation in the timing of droughts and water management policies to explain changes in crime. I find that violent crimes increase by 10%, police-detected crimes fall by 20%, and that there is no discernible impact on sex crimes or property crimes. These findings suggest that in the future, especially as severe droughts become more prevalent due to climate change, crime prevention may be an important component of climate policy.

Title: Groundwater Use in Agriculture Accounting for Aquifer Flow Dynamics: An Analytical Model of Mississippi Delta

Authors: Kalyn Coatney, Department of Agricultural Economics, Mississippi State University;

Seong Yun, Department of Agricultural Economics, Mississippi State University

W4133 task: Obj 1: Land and Water Resource Management in a Changing Environment, Task 1-1: Economic Analysis of Agriculture, Forest, and Rangeland Resources, Open Space, and WUI Zones

Presenter: Seong Yun, Department of Agricultural Economics, Mississippi State University

Presenter email: seong.yun@msstate.edu

Abstract:

Irrigation is becoming increasingly important to Southeastern crop production. The increased demand and high prices for water-intensive crops (i.e., corn and soybeans) have led more irrigation and higher competition for groundwater. Unlike drylands in the US, the Southeast regions are known for abundant rainfalls. While annual rainfall totals are relatively high in the Southeast, most of the rainfall occurs during the winter months while precipitation during the growing season is often inadequate. Besides, the Mississippi River Valley Alluvial Aquifer (MRVAA) covering the large portion of the Southeast is shallow and has long been observed that the water level of MRVAA is steadily declining. Some Southeast regions have been experienced depletion of groundwater and the tragedy of commons in groundwater is expected to be worsened under climate change and variability.

Due to high concern on efficient groundwater management, economic models of groundwater management including aquifer flow dynamics have been studied well in the regions such as Ogallala Aquifer or aquifers in California. These models are, however, not sufficient enough to explain groundwater issues in MRVAA due to the two significant differences in groundwater flow dynamics. First, the MRVAA is a shallow aquifer with a relatively fast recharge rate. Because of the soils in MRVAA are composed of heavy clay, relatively little recharge can be attributed to percolation. Instead, rivers and streams provide most of the recharge relatively quickly from the side of this shallow aquifer. Second, the MRVAA is not a static aquifer, but appears to have some flow dynamics. From geo-survey, it turns out that the MRVAA generally flows from north to south. Because of the spatial and flow dynamics, producer wells located sequentially along the general flow creates an assignment problem. Producers located at the northern end of the MRVAA’s natural recharge entry point experience higher recharge rates than those to the south and central regions.

Reflecting these flow dynamics specific to the MRVAA, this study contributes to the literature by suggesting a bathtub model with spatially varying recharge rate with shallow aquifer structure. The proposed model is a discrete space and time dynamic optimization model to maximize the net benefit of agricultural production with irrigation subject to the flow dynamics of groundwater. The state equation of water level includes the unidirectional spatial dependence (North to South) recharge rate and extraction. This study first derived the analytical solution and economic implication of the given structure. Using the six producers’ simulation, this study numerically demonstrated how the downstream producers are disadvantageous in groundwater use due to their location. Also, this study showed how cheap altruism or falsely targeted regulation reached to faster and more worsened over-extraction.