Working Papers

Job Market Paper: The Value of Nature-Based Adaptation – Evidence that Tree Cover Protects Urban Revenue from Heat

[Link to Working Paper.] Climate change is increasing the frequency and intensity of extreme heat events, particularly in cities. This paper estimates the causal effect of temperature on daily revenue using over 15,000 consumer-facing storefronts in the 49 largest U.S. cities between 2019 and 2023. Above 35 °C (95 °F), revenue falls steeply, averaging 9 percent lower on days above 37.5 °C (99.5 °F) days than on 20 °C (68 °F) days. Consumption smoothing across days mitigates some damage from an extreme heat event, but a 1.3 percent revenue drop persists for two weeks following a hot day. Therefore, I estimate how effective urban green space is as a climate adaptation strategy that can prevent revenue losses caused by extreme heat ex ante. I find that a one percent increase in tree cover surrounding a storefront increases revenue by 0.94 percent on hot days. A 10 percent increase in surrounding tree cover eliminates revenue losses at 37.5 °C and reduces them by half at 40 °C, relative to areas without canopy. Storefronts located in the Southwest and South have the most to gain from urban tree cover. On average, businesses in this region can cover the cost of moving to a 10 percent tree cover scenario in four years with the avoided revenue losses. This length of time shortens under projected climate scenarios. These results suggest that green infrastructure can improve firm resilience to heat, providing evidence of a private incentive to finance urban green space that could simultaneously provide a positive externality.

The National Value of Local Outdoor Recreation over the Past 20 Years

[Link to Working Paper.] I measure the benefits of local outdoor recreation in the United States over 2003-2023 and show how conclusions depend on how travel time is priced. Using the American Time Use Survey to estimate a travel-cost model, I estimate (i) welfare-based willingness to pay, which values travel time at one-third of own wage, and (ii) accounting values that price the realized service (travel) at a replacement wage consistent with household production accounts. Average value per trip is stable at roughly $15-$18 (2023 USD) regardless of pricing technique, with temporary increases around the Great Recession and the COVID-19 pandemic. Scaling by trips and population, national annual value rises by about $100 billion over two decades. In 2022, it totals $216-$241 billion, far exceeding typical municipal park investments. In contrast to average and aggregated value, distributional results depend on the modeling choice of how to price time. While aggregate time trends are similar under both approaches, using a replacement wage compresses the richest-to-poorest quintile ratio in per-trip value from about 10:1 to 1.5:1. The method produces a scalable, repeatable measure of local recreation benefits and connects non-market valuation with accounting-consistent pricing.

Measuring the Value of Outdoor Recreation for National Environmental-Economic Statistics

With Jorge Forero Fajardo and Eli P. Fenichel

[Link to Working Paper.] Accurately measuring welfare changes from outdoor recreation requires accounting for both market expenditures and household-produced services, yet current U.S. satellite accounts capture only market transactions. Ignoring the value of recreation under the household production accounting boundary can create a measurement gap for benefit-cost analyses of environmental protection or infrastructure investments, where recreation benefits would be systematically understated. We find that outdoor recreation is primarily a household-produced service that combines purchased inputs (gasoline, equipment) with travel time: the major welfare-relevant cost. Using detailed U.S. coastal recreation data, we partition travel costs across the System of National Accounts boundary (market expenses) and the household production boundary (travel time) to quantify this measurement gap. We find that market expenditures represent only one-third of total recreation costs, while household production of recreation services through travel time comprises two-thirds. This means current assessments using only market expenditures underestimate outdoor recreation’s welfare-relevant value by approximately threefold. Including travel time would increase the Household Production Satellite Account by 4-5 percent nationally, corresponding to $12.4 billion in the U.S. Gulf Region or 0.7 percent of regional GDP. For benefit-cost analysis, jointly considering both the SNA and household production boundaries is essential for accurately measuring welfare-relevant expenditure changes over time and properly valuing natural assets that enable outdoor recreation.

Work in Progress

Measuring the Value of Outdoor Recreation for National Environmental-Economic Statistics (with Eli P. Fenichel and Jorge Forero Fajardo).

Publications

Creel A. M. (2025). Muller et al., eds., Measuring and Accounting for Environmental Public Goods, Chap. 7 Comment. National Bureau of Economic Research Book, University of Chicago Press.

Creel, S., Creel, N. M., Creel, A. M., & Creel, B. M. (2016). Hunting on a hot day: effects of temperature on interactions between African wild dogs and their prey. Ecology, 97(11), 2910-2916.

Research Statement

I am an environmental economist who estimates the non-market value of nature-based adaptation solutions to mitigate climate change, ecosystem services, and biodiversity loss. My research examines how natural resources influence people’s well-being and economic activity in ways that are often invisible in traditional economic statistics. My work estimates these unknown values using revealed preference techniques, which involve observing how people trade off money and time to access nature, and how firms and governments, in turn, capture value from those choices. My research agenda aims to integrate the value of nature that I estimate into the systems that measure welfare, business performance, and fiscal health. This integration is important because it highlights the necessity of sustainable management of natural resources for the well-being of people and economic outcomes.

My research requires building novel datasets that link human behavior to the surrounding environment. I combine nontraditional behavioral data, such as credit card transactions, mobility records, and time-use surveys, with high-resolution environmental measures (e.g., remotely sensed GIS data). By building these large datasets, I can estimate how environmental conditions causally affect welfare and economic activity, reveal who gains and who loses from those effects, and integrate those insights into systems of economic measurement and policy. Additionally, this data-intensive approach enables my work to carefully examine how environmental benefits can vary across demographics, geographic space, and time.

In my job market paper, “The Value of Nature-Based Adaptation: Evidence that Tree Cover Protects Urban Revenue from Heat,” I find that urban green space sustains local business revenue during extreme heat events. I combine daily credit card transaction data from over 15,000 storefronts in the 49 largest U.S. metropolitan areas with high-resolution temperature and tree canopy measures. Leveraging within-store variation in daily maximum temperatures, I find that revenue drops by 9 percent on extremely hot days (above 99.5 °F). Using spatial variation in green space across businesses that are a part of the same brand in the same city, I find that moving from no tree canopy to high tree canopy cover (zero to 10 percent coverage) fully offsets the revenue loss from a 99.5 °F day and halves the loss from a 104 °F day. This paper outlines my broader research approach, which involves observing people’s revealed preferences for nature, examining how these preferences influence private returns, and identifying situations where businesses have a private incentive to finance investments in green infrastructure.

This paper contributes to several strands of literature on climate adaptation, nature-based solutions, and urban economics. It addresses a central gap in the climate adaptation literature: the lack of evidence on private, ex-ante adaptation returns outside of agriculture, health, and labor. My job market paper provides the first large-scale, daily, storefront-level evidence on how heat shocks affect the service sector—a major component of the economy. I then show that urban tree cover protects firms’ revenue from extreme heat. By quantifying how the magnitude of tree cover’s return varies across regions and under warmer climate scenarios, my job market paper identifies where adaptation is currently most valuable and how its benefits scale with continued warming. In doing so, it links the literatures on adaptation and natural capital. Together, these contributions position urban green space not only as an environmental amenity but also as a productive asset that sustains private-sector performance under climate stress.

My interest in ecosystem services is motivated by the need to incorporate non-market services into national statistics. A significant portion of my research agenda is focused on outdoor recreation, particularly local recreation, which is one of the most widely consumed ecosystem services but is often overlooked in traditional economic statistics. For example, the U.S. Bureau of Economic Analysis (BEA) has a satellite account for outdoor recreation, but it does not include the value of time that people give up to engage in recreation. Similarly, the Household Production Satellite Account includes services that individuals produce for themselves using their own time, but excludes recreation (even though it meets the definition of household production).

In a working paper, “Measuring the Value of Outdoor Recreation for National Environmental-Economic Statistics” (with Eli P. Fenichel and Jorge Forero Fajardo), I show how current BEA accounting boundaries obscure the value of recreation. Using coastal recreation data, we partition expenditures between the System of National Accounts and household production accounting boundaries. We find that market spending (e.g., gas, equipment, lodging) represents only one-third of total travel costs, with the majority being individuals’ travel time. As a result, local recreation (which accounts for 77 percent of recreation trips) is barely represented in the Outdoor Recreation Satellite Account, as it only considers market expenditure and overlooks travel time. The Household Production Satellite Account does include travel time when estimating the value of different services people produce for themselves, but it does not include outdoor recreation. If the travel time for all recreation trips were included in the Household Production Satellite Account, the value of household production in the U.S. would rise by 4 to 5 percent (equivalent to increasing GDP by 0.6 to 0.7 percent). This work exemplifies my approach to environmental-economic measurement: using revealed-preference methods to make nature’s contribution to welfare visible, measurable, and ultimately incorporated into official statistics.

In another paper, “The National Value of Local Outdoor Recreation over the Past 20 Years,” I provide the first long-run, nationally representative estimates of willingness to pay for local recreation in the U.S. I find that the average value per trip remains stable at roughly $15–$18 (2023 USD), with temporary increases around the Great Recession and the COVID-19 pandemic. Scaling by trips and population, the national annual value rises by about $100 billion over two decades, totaling $216–$241 billion in 2022, far exceeding typical municipal park investments. In contrast to these aggregate values, the distributional results depend on how travel time is priced. Using a replacement wage, rather than an opportunity cost based on income, compresses the richest-to-poorest quintile ratio in per-trip value from about 10:1 to 1.5:1. This approach yields a scalable, repeatable measure of recreation benefits that aligns non-market valuation with national accounting principles, advancing efforts to integrate ecosystem services into official measures of economic welfare.

My interest in how environmental benefits and costs are distributed across demographics, space, and time developed from formative research experiences. As a research assistant for Kerwin Charles, I examined whether the closure of Black hospitals during integration narrowed the Black–White infant mortality gap and found no measurable effect. This was an early lesson in testing whether well-intentioned policies deliver equitable outcomes. In my master’s thesis, I utilized nationwide mobility data to examine park use during the early months of the COVID-19 pandemic and found that people of color and immigrants reduced their park visits more than white domestic recreators, suggesting a greater reliance on social forms of outdoor recreation. Later, as an intern at Resources for the Future, I analyzed Land and Water Conservation Fund grants and found that they were equitably distributed across income groups but disproportionately favored white counties. Together, these experiences underscored the need to pair efficiency-based measures with distributional analysis to understand how public goods affect welfare in heterogeneous ways.

Going Forward: My research will continue to quantify the welfare gains from ecosystem services and the costs of environmental degradation, illustrating how these values are distributed through households, firms, and fiscal systems. I have three primary tracks for future research.

First, I have multiple research questions about how to adapt to climate change. This work extends from my job market paper. First, I plan to use credit card and mobility data to estimate the public costs of food deserts under extreme heat, analyzing how environmental stress amplifies the burdens faced by low-income households and marginalized groups in securing everyday necessities. Second, I have found preliminary evidence that extreme heat leads to additional tax revenue raised from alcohol sales. I plan to ask how heat interacts with drinking, and how that interaction influences heat-related crime. Third, I will evaluate the cost-effectiveness of alternative adaptation strategies to extreme heat, comparing interventions such as urban tree planting and reflective surface treatments. Together, these projects foster an understanding of how climate adaptation strategies can mitigate inequities while being economically efficient.

Second, my future work will advance the measurement of recreation’s value for national statistics. I plan to translate recreation trips into land-based measures of service flows for integration into natural capital accounts. I will also examine how recreation functions as a welfare “backstop” during crises (e.g., recessions or pandemics) when budgets are constrained, but access to local green space remains accessible. Finally, I will investigate how social forms of recreation generate positive externalities in parks, thereby challenging congestion-based views in the recreation literature.

Third, I aim to expand my research agenda to the valuation of biodiversity by examining how conservation actions, such as wetland protection by birders, generate spillover benefits for surrounding communities, including flood protection and enhanced climate resilience.

Together, these projects will advance my research agenda: developing environmental-economic statistics that reveal the hidden value of nature, inform equitable policy design, and support investment in natural infrastructure.