Grantee Research Project Results
Final Report: An Integrated Valuation Model Linking Nutrient Reductions to Changing Ecosystem Services in Freshwater Systems
EPA Grant Number: R836168Title: An Integrated Valuation Model Linking Nutrient Reductions to Changing Ecosystem Services in Freshwater Systems
Investigators: Lupi, Frank , Stevenson, R. Jan , Herriges, Joseph A.
Institution: Michigan State University
EPA Project Officer: Packard, Benjamin H
Project Period: September 1, 2016 through August 31, 2020 (Extended to August 31, 2023)
Project Amount: $799,074
RFA: Water Quality Benefits (2015) RFA Text | Recipients Lists
Research Category: Water
Objective:
The overall objective of this research was to estimate the use and non-use values stemming from changes in nutrient loadings to the freshwater systems (rivers, lakes, and the Great Lakes) of the lower peninsula of Michigan and, in doing so, address a number of the challenges in non-market valuation when applied to water resources. The proposal identified specific three project objectives. Objective 1 focused on quantifying changes in water quality stemming from changes in nutrient loadings and developing water quality metrics that could be used to estimate changes in ecosystem services that consumers value. Objective 2 sought to use these metrics to assess total willingness-to-pay (TWTP) for changes to streams, lakes, and Great Lakes in Michigan. Finally, Objective 3 was concerned with the various approaches used to apportion total value into use and non-use components.
Conclusions:
To address the project objectives, the key tasks for the project centered around designing and fielding two nonmarket valuation surveys. A baseline survey (survey 1) examined how WTP for changes in water quality differs in terms of the reported ecosystem service affected (water contact, fish biomass, biological condition and a modified WQI). In this first survey, we conducted a split sample test of two versus three water quality indices. The two-index treatment conveys the impact of water quality changes via: (1) a Recreation Score and (2) a Wildlife Score. The former is a variant of the traditional water quality index, with the accompanying swimmable, fishable, and boatable categories. For three index treatment, we segmented the Recreation Score into two sub-indices: (1) a Fish Biomass Score and (2) a Water Contact Score, with the former measuring the diversity and biomass of game fish in Michigan waterways, while the latter captures water quality as it pertains to recreational activities involving water contact (e.g., swimming, boating, etc.). The reason for this decomposition is that fish-related ecosystem services and other recreational opportunities are likely to change differently as nutrient loadings, the primary stressor being examined in this project, change.
As reported in Lupi et al. (2023), we found that willingness to pay (WTP) for improvements in each of the individual water quality indices are consistently positive and statistically significant in both the 2- and 3-index versions of the survey. The largest WTP in the 2-index version of the survey is for changes in the overall recreation score. We also find that changes in our index reflecting changes in fecal bacteria and water clarity are valued differently from changes in our recreational fishing index. Aggregating changes in these two distinct recreational services using a single WQI yields consistently lower benefit estimates across a range of underlying changes in our experiment. In valuation scenarios with small changes in overall water quality, the WQI-based benefit estimates can differ substantially from benefits measured by decomposing the index and valuing the disparate sub-indices, differences which might change the balance of benefits and costs in regulatory evaluations. Moreover, our study confirms our hypothesis that for some underlying trade-offs between key ecosystem services, it matters whether those services are valued by combining ecosystem services within a single index or separating the services in separate indices. We also found that when a marginal change in our water contact index is mapped through the WQI function in the model with WQI alone, the value is significantly smaller than the respective marginal value in the model with water contact modeled separately, which is relevant for valuation of recreational services related to nutrient loadings.
In a second set of surveys, we compared water quality valuation results from an address-based probability sample and two opt-in non-probability samples, MTurk and Qualtrics. The address-based survey (ABS) sample of 12,000 individuals included a random set of addresses from Michigan using the postal service’s delivery sequence file. We implemented the same survey instrument and experimental designs to a sample of Qualtrics panel respondents and to a sample of MTurk respondents to compare the general population sampling data to these convenience samples. As summarized in Sandstrom-Mistry et al. (2023), we found that the samples differed in some key demographics, but measured attitudes were strikingly similar. For valuation models, most parameters were significantly different across samples, yet many of the marginal willingness to pay values were similar across samples. Notably, for non-marginal changes, there were some differences by samples: MTurk values were always significantly greater than the probability sample, as were Qualtrics values for changes up to about a 20% improvement. Overall, the evidence is mixed, with some key differences but many similarities across samples. We generally conclude that the non-probability methods generate different valuation results than the probability-based sample, with most of the differences stemming from the MTurk sample. Previous literature supports using probability-based samples for population-based applications. Our findings do not contradict this evidence, and we recommend that probability samples be used for studies that aim to make inferences to a broader population. Considering that most policies are for non-marginal changes, and that we find differences between the samples in WTP for non-marginal changes, we also recommend that probability samples are used for non-marginal policy applications and any high-stakes decisions (Sandstrom-Mistry et al. 2023).
To address our third objective, as reported in (Kim and Lupi, 2023), we developed a utility-theoretic structural approach for decomposing and measuring use and non-use values by combining revealed preference date on recreation site choices and stated preferences from a single consequential referenda question. Monte Carlo methods were used to test the econometric properties of the estimators, and we showed that the structural method is generally robust but has bias under certain conditions. The upshot is that researchers pursuing this approach should be cautious with its power if they have preliminary data or priors that non-use values are small relative to use values, as our simulations show those cases are prone to some bias, whereas bias is small for a broad range of cases. In addition, using data from our second survey, the structural model was applied but had difficulties with convergence (Kim 2023) so was not successfully implemented empirically with our available empirical data.
Finally, the project also developed estimates of the value of water-based recreation using multiple-site recreation demand models specified as nested logits (Kim et al. 2024). That paper also derives valid estimation procedures in recreation demand models with incomplete data on trip locations, which is common in many recreation demand datasets or with data collection strategies that only elicit the location of a subset of trips (or in some cases only one trip). The Monte Carlo results and empirical results show that a convenient trip-weighting strategy that can be implemented in existing nested logit software that approximates true values and values from a more complex structural model that fully accounts for the censored trip data. The results are summarized in Kim et al. (2024), which shows that depending on the specific estimator, the valuation results ranged from $32 to $44 dollars per water-based recreation trip.
Journal Articles on this Report : 2 Displayed | Download in RIS Format
Other project views: | All 12 publications | 3 publications in selected types | All 3 journal articles |
---|
Type | Citation | ||
---|---|---|---|
|
Lupi F, Herriges JA, Kim H, Stevenson RJ. Getting off the ladder:Disentangling water quality indices to enhance the valuation of divergent ecosystem services. Proceedings of the National Academy of Sciences. 2023;120(18). doi:10.1073/pnas.2120261120. |
R836168 (Final) |
not available |
|
Kim H, Lupi F. Testing the robustness of a structural model for discerning use and non-use values of ecosystem services. Agricultural and Resource Economics Review. 2023;52(2):406-21. doi:10.1017/age.2023.26. |
R836168 (Final) |
not available |
Supplemental Keywords:
Nonmarket valuation, random utility modelsProgress and Final Reports:
Original AbstractThe perspectives, information and conclusions conveyed in research project abstracts, progress reports, final reports, journal abstracts and journal publications convey the viewpoints of the principal investigator and may not represent the views and policies of ORD and EPA. Conclusions drawn by the principal investigators have not been reviewed by the Agency.
Project Research Results
- 2022 Progress Report
- 2021 Progress Report
- 2020 Progress Report
- 2019 Progress Report
- 2018 Progress Report
- 2017 Progress Report
- Original Abstract
3 journal articles for this project