No offense intended, but my guesss is that if the fraction of US GDP to be devoted to dental care was decided by experts on dental health (read, "dentists"), it would be higher than if ordinary citizens could meaningfully be polled as to the importance of improved dental health relative to other uses of their own money. Similarly, while environmental experts have useful information as to the value of environmental projects, whether it be cleaning up after an oil spill, preserving a wetlands, or foregoing natural resource extraction so as to preserve the habitat of an endangered species, I am typical in believing that there is a role to play in budgeting for environmental policies by information as to the value of environmental projects to consumers, taxpayers, outdoorsmen, and parents.
Information as to how people evaluate many projects arises from market prices. The value of a Toyota Prius or a fitness club contract is well estimated by the price paid. Market prices are almost no help in estimating the value to indivuals of environmental or other public projects. The standard approach is called the Contingent Valuation method, or CVM. It is a survey method, asking respondents how much they would pay to preserve a wetlands, or whether they would be willing to pay $X to clean up after an oil spill. The principal problem with the CVM and other survey methods is that the answers do not associate with any real financial commitments, and are thus overstated by some unknown degree of hypothetical bias. (A page on problems of CVM is here.) Recent developments in value-eliciting laboratory methodologies and in auction theory can be applied to design environmental valuations based on observations of actual, finaancially incentivized behavior