Description and Objective of Research:
There is an increasing body evidence suggesting that a broad range of pollutants have the potential to alter human endocrine systems. One potential consequence of exposures to these endocrine disruptors is that they may significantly increase the incidence of infertility in exposed populations. The purpose of this research has been to develop and evaluate a methodology for applying stated preference techniques to assess the magnitude of economic values associated with reducing infertility risks from environmental sources. We have done this through the application of two small-scale pilot contingent valuation surveys.
This research is also intended to refine a conceptual framework for linking collective (household) decisions to the preferences of the individual members. Most previous research in non-market valuation of health and environmental risks has focused on individual decisions; however, infertility risks clearly present a context where the household (i.e. the couple) is the relevant decisionmaking unit. Our research explores a conceptual basis for deriving how measures of economic welfare based on households' observed or stated decisions relate to the preferences of the individual adult members living in that household.
Summary of Project Activities and Findings:
The following is a list and short description of the primary project activities and findings:
1. Literature Review
To establish a conceptual and empirical foundation for our work in this project, we began with literature reviews in three primary areas.
Household Decision Models. We have used this literature to inform the development of a conceptual model that examines how the separate preferences of individuals within a household can be recovered from household decisions.
Household Labor Supply. To better understand how time allocation decisions are linked to household composition, we conducted a brief review of the recent literature. The evidence from this literature indicates that women's labor supply decisions in particular and the time spent by both partners in nonmarket activities are significantly influenced by household characteristics related to children.
Environmental Causes of Infertility. The conclusion of this review is that although a number of pollutants are potentially hazardous to fertility, current evidence is inadequate to draw a strong connection between environmental exposures and increased infertility risk. Partly for this reason, we concluded that our initial framing of a CV scenario to assess WTP for a policy intervention designed to reduce infertility risk should not rely on developing the context of environmental exposures.
2. Development of Conceptual Model
Based on our literature review of household decision models and our other project activities, we have developed a conceptual framework that links the collective model of household behavior to the methods that are commonly used to estimate the economic value of non-marketed environmental resources.
3. Sessions with First-time Infertility Patients
To better understand the decisions and complications experienced by couples who are having difficulty conceiving a child, we attended sessions with first time patients as they discussed their reasons for seeking assistance and were confronted with alternatives for how to address their problems.
4. Focus Groups
We conducted a focus group with couples who did not yet have children to discuss (1) how they make typical household decisions, (2) their expectations with regard to starting a family, (3) their understanding of the sources and magnitude of infertility risks, and (4) different versions of a risk communication device and a WTP question for a hypothetical medication to reduce future infertility risks.
5. First Pilot Survey
The survey was initially implemented at a malls in Charlotte, NC; Seattle, WA; and Jacksonville, FL. We collected information from 209 respondents - 114 in Charlotte, 48 in Seattle, and 49 in Jacksonville. The sample was almost 40 percent female, with an average age of about 24 years, and an annual household income of roughly $40,000. When the male and female samples were analyzed separately, it was found that the internal validity of the female sample responses (based on the sign and significance of the price, income, and risk reduction variables) was much stronger. The results of this analysis were used to revise the survey instrument and collect a second round of data through a second pilot survey, which is described below.
6. Second Pilot Survey
Based on the results of the first pilot, we determined that the survey instrument needed to be revised in three ways in particular. First, we decided to focus exclusively on female respondents. Second, we needed to better account for the timing of the medication decision. Third, we needed to include more detailed information regarding the earnings and labor force participation of the two partners. The revised survey was implemented during the summer months of 1998 at four mall locations across the US - Tampa, FL; Tulsa, OK; Las Vegas, NV; and Freehold, NJ. The total sample of useable responses came to 188 women (72, 50, 38, and 28 at the respective locations) between the ages of 20 and 35 (with an average of about 25 years old). Based on the results of the probit analysis, the mean WTP for the medication was about $320 per year. The medication was described to delay the upward trend in infertility risk that a couple typically experiences (by up to five years). Depending on how long one takes the medication, this translates to reductions in the probability of infertility in any given year by about 1 to 10 percent. Under alternative assumptions regarding the discount rate and the length of time one takes the medication, this implies a total WTP for a "statistical" child of between $15,000 and $60,000. These estimates are very preliminary, but they do provide a point of comparison with other comparable empirical estimates
This research has extended the collective model of household decisionmaking (Chiappori, 1988; Browning et al., 1994) to demonstrate its relevance for measuring consumer surplus. To our knowledge it is the first time this collective approach has been used to generalize Hausman's (1981) approach for recovering Hicksian consumer surplus. In addition, the conceptual research demonstrated how a household production framework, public goods (internal and external to the household) and risk could be incorporated into the model. For each of these cases, the analysis highlights how the concept of an "income sharing rule," which is central to the collective model, can be used to distinguish household members' preferences.
The research also demonstrated that it is possible to elicit consumer preferences for reducing infertility risks. The two pilot surveys demonstrated that women are the most likely population to be sampled. An initial pilot with both men and women indicated that men seemed to have greater difficulty with the risk reduction scenario discussed in the instrument. However, in both pilots, female respondents provided answers to the WTP question that were, for the most part, consistent with economic theory and, thus, supported the internal validity of the instrument. Using the all-female sample and a number of simplifying assumptions, we are able to derive some preliminary WTP estimates for reductions in infertility risk. The ability to generalize these results to a broader population is greatly restricted by the size and non-random nature of the sample. However, we do estimate values for a "statistical child" that are considerably lower than those estimated by the most directly comparable study.
Findings from the focus groups and pilot surveys were also generally consistent with a "unitary" model of household infertility-related decisions. Under such a framework, household expenditure decisions are treated as if they were the result maximizing a single utility function subject to a single "pooled" budget constraints. Alternative models take into account the divergence of preferences within a household and the possibility that income is not entirely pooled. Our findings indicate a general convergence of preferences between partners regarding childbearing issues, which is consistent with an "assortative mating" view of partnering (e.g., marriage) and a single common preference structure. Furthermore, our analysis of the survey data found little evidence to contradict the income pooling hypothesis for infertility-related decisions. Nevertheless, we cannot firmly reject the collective view of the household, particularly for other types of household decisions. Under ideal circumstances, information would be collected from a larger sample, on a broader range of issues and from both partners in a household in order to more thoroughly evaluate which factors influence whether a collective view of the household matters.
Van Houtven, George. "Willingness to Pay for Reductions in Infertility Risks: A Contingent Valuation Study." Presented at the 1999 EPA Workshop entitled Valuing Health for Environmental Policy With Special Emphasis on Children's Health Issues, Silver Spring, MD, March 25, 1999.
Van Houtven, George. "Valuing Reductions in Infertility Risks: Individual Preferences and Household Decisions." Presented at the 1998 Decision-Making and Valuation for Environmental Policy Workshop, Washington, DC, April 3, 1998.
Smith, V. Kerry. "Non-Market Valuation and the Household: Individual Preferences and Collective Risks of Infertility." Presented at the Harvard University Center for Risk Analysis, March 1998.
Smith, V. Kerry and George Van Houtven. 1998. "Non-Market Valuation and the Household" Revise-and-Resubmit to the Journal of Environmental Economics and Management.
Smith, V. Kerry and George Van Houtven. "Non-Market Valuation and the Household: Recovering Individual Preferences with Collective Risk." Presented at the 1998 Meetings of the American Economic Association, Chicago, IL, January 3, 1998.