Environmental Economics

Ummm, @PeterGleick, I addressed this 5,958 days ago: Free markets are, in fact, not free

"Dr. Peter Gleick is a leading scientist, innovator, and communicator on global water and climate issues." He is also a frequent Tweeter and sometimes instigator. Yesterday he tweeted this:

I'm fairly certain Dr. Gleick knows there are a large number of 'atypical' economists who understand the non-market values of water (I'm not sure what a non-economic value is because value is inherently economic). Anyway, here's what I wrote on this exact topic (Are Free Markets Free?) almost 15 years ago...

Pragmatic and existential concerns about the social cost of carbon

There is a great new policy forum article in Science defending the use of the social cost of carbon (SCC) in climate policy analysis, written by four great environmental economists (Joe Aldy, Matt Kotchen, Robert Stavins, and James Stock).  While I enjoyed reading the article, I am not sure that I endorse its conclusion.

The background: for the last thirty-plus years, the economics profession has been studying climate change through integrated assessment models (IAMs), pioneered by William Nordhaus, who was awarded the Nobel Prize in 2018.  These IAMs include specifications of the costs and benefits of reducing greenhouse gases, and so they can be used to calculate the "efficient" path of emissions reductions (i.e. the path that maximizes net benefits according to the model) and to calculate the marginal damages of each unit of carbon emitted along the efficient path, the so-called SCC.  Then, this SCC can be smacked on to carbon emissions in the real world to get the real world to behave like the efficient outcome in the model.

Then, there have been some critics of the use of IAMs and the SCC, including from among the economics profession.  Robert Pindyck has been very critical of the false level of certainty implied by IAMs. (He has a paper titled "Climate Change Policy: What do the Models Tell Us?", and the first sentence of the abstract answers "Very little.")  Pindyck writes:

These models have crucial flaws that make them close to useless as tools for policy analysis: certain inputs (e.g., the discount rate) are arbitrary, but have huge effects on the SCC estimates the models produce; the models' descriptions of the impact of climate change are completely ad hoc, with no theoretical or empirical foundation; and the models can tell us nothing about the most important driver of the SCC, the possibility of a catastrophic climate outcome. IAM-based analyses of climate policy create a perception of knowledge and precision, but that perception is illusory and misleading.

More recently, Nicholas Stern and another Nobel-prize winner, Joseph Stiglitz, have released a working paper that also criticizes IAMs and the SCCs they produce for neglecting, among other things,

...the immense risks and impacts on distribution across and within generations; the many failures, limitations or absences of key markets; and the limitations on government, both in offsetting these failures and distributional impacts.

Stern and Stiglitz propose an alternative to the SCC, which can be called a "target-consistent price."  The SCC uses the costs and benefits in the IAM to calculate the efficient level of emissions and associated price.  The target-consistent price treats the target as exogenous, whether that target is a specific emissions reduction level, or a specific temperature threshold, and then uses the IAM to calculate the cost-effective (i.e. least-cost) path to achieve that target.  The target-consistent price is the associated price of carbon for that cost-effective path.  So the difference between the SCC and the target-consistent price is that the SCC uses a cost-benefit analysis while the target-consistent price uses a cost-effectiveness analysis.  

These critiques of IAMs and the SCC by economists are similar to critiques by non-economists, who have argued that IAMs like Nordhaus's DICE model have yielded policy prescriptions that are not nearly ambitious as they should be.  

The recent article by Aldy, Kotchen, Stavins, and Stock defend the use of IAMs and calculation of the SCC.  The basic idea is that SCCs are fundamentally sound, and we (scientists and economists) should be working to improve the calculation of the SCC and its use in climate policy design rather than throw it out and settle for mere cost-effectiveness analysis.  It was the Trump administration, after all, that effectively rewrote the rules to set the SCC to near-zero, while the Biden administration is bringing back legitimate policy analysis and science in its calculation.

I like the article and agree with most of the points it raises but am not sure that I endorse its conclusion.  There are two ways of defending or attacking the SCC: on pragmatic grounds and on existential grounds.  In either case, I don't know that we need or want an SCC, or that an SCC is preferable to something like Stern and Stiglitz's target-consistent price.

The pragmatic question is: which type of analysis is going to make the world a better place, which in this context means which is going to lead to more aggressive climate action? (Assuming here that we are all on the same page and want aggressive climate policy, relative to what's actually been enacted.)  The implication in the recent Science piece is that relegating ourselves to setting exogenous policy goals and only using economics for cost-effectiveness analysis will put us at the mercy of political whims or administrations who can willy-nilly throw out any justification of aggressive policy.  But that is true when using an SCC too!  That's just what the Trump administration did.  Furthermore, the longstanding critique of DICE and other IAMs is that they are not aggressive enough because they omit many difficult-to-quantify costs of climate change, like tipping points, and they tend to ignore or downplay distributional and equity concerns.  This is the heart of Stern and Stiglitz's, as well as Pindyck's, critique of IAMs.  Setting clearly-defined policy goals, like a maximum temperature increase, seems likely to increase the support and possibility of getting aggressive climate action done.  Of course, that is an open question that will be hard to answer (which type of policy analysis will lead to more aggressive policy?), but it is by no means obvious that the SCC wins.

Beyond the pragmatic justification of the SCC, there is an existential issue with it.  Namely, is the SCC in fact a real thing, a meaningful concept for which a concrete value exists in the world (say, $45 per ton), and we just need to find out what it is?  Or, is the SCC itself a construct, which exists only within and because of the methodological framework of neoclassical welfare economics?  The Science article takes for granted that the SCC is a real thing.  But that claim assumes that a valuation of damages to the environment exists.  The existence of that valuation is essential to neoclassical welfare economics, but a large literature in ecological economics and philosophy questions its existence.  

I am especially interested in this issue since I'm currently in the middle of reading Elizabeth Anderson's 1995 book Value in Ethics and Economics. This is a must-read for economists interested in alternative ways of thinking about policy analysis.  Like many ecological economists and some political philosophers like Michael Sandel, Anderson is critical or skeptical of the authority of economic valuation techniques across numerous domains.  She and others defend value pluralism, or the incommensurability of values - it is not appropriate to simply put a dollar value on the damages caused by climate (even if those dollar values come from revealed preference studies of people's willingness to pay) and toss that into a benefit-cost analysis.  She writes

... Goods can be plural, ... they can differ in kind or quality; they differ not only in how much we should value them, but in how we should value them

It could be that the SCC is bad, not because of the problems and limitations of IAMs that are raised by Stern and Stiglitz and other critics, but because the entire philosophical infrastructure of the valuation of environmental damages is unsound.  It's hard for me to fully accept this view, since I've been trained for the past 20 years in neoclassical economic orthodoxy.  But there are many compelling arguments in favor of value pluralism that are worth struggling with.

Or, it could be that the existential questions about the SCC aren't important and what matters is the pragmatic effects of using or not using an SCC or cost-benefit analysis in climate policy.  

In either case, I think it's an open and important question to ask whether the SCC is worth pursuing or not.

A great job opportunity for an undergraduate economics student (deadline May 28)

Assessing the Benefits & Costs of Environmental Policies that Improve Water Quality: 

A research training opportunity is currently available at the U.S. Environmental Protection Agency’s (EPA) Office of Policy (OP)/National Center for Environmental Economics (NCEE), located at EPA Headquarters in Washington, DC.

The participant will be involved with NCEE economists on a research project that will improve methodologies to assess the benefits and costs of environmental policies that improve water quality. This project will have two components. The first component involves assessing the validity and reliability of studies that estimate social benefits of improving outdoor water quality. The participant will engage in research efforts to review literature of all relevant studies, evaluate the results of published validity and reliability tests, and conduct additional retrospective evaluation tests on results, making use of raw survey data where available. The second component is an evaluation of affordability metrics for costs of drinking water that are passed on to customers. The participant will engage in research efforts to review literature on proposed affordability metrics (including those for similar services, such as electricity), apply these metrics to existing market data and evaluate their performance. This research project will provide the participant the opportunity to be involved with large, administrative data sets. The participant will also have the opportunity to participate in additional research projects that are mutually beneficial.

The mentor for this opportunity is Chris Moore ([email protected]).

This program, administered by ORAU through its contract with the U.S. Department of Energy (DOE) to manage the Oak Ridge Institute for Science and Education (ORISE), was established through an interagency agreement between DOE and EPA. The initial appointment is for one year, but may be renewed upon recommendation of EPA contingent on the availability of funds. The participant will receive a monthly stipend commensurate with educational level and experience. No funding will be made available to cover travel costs for interviews, relocation costs, costs of tuition/school fees, or a participant's health insurance. Proof of health insurance is required for participation in this program. The appointment is full-time at EPA in the Washington, DC, area. Participants do not become employees of EPA, DOE or the program administrator, and there are no employment-related benefits.

Qualifications

The qualified candidate should have received a bachelor's or master's degree (preferred) in one of the relevant fields. Degree must have been received within five years of the appointment start date. 

Preferred skills:

• • Experience using analytic software (e.g., STATA, R or Matlab) 
  • Strong quantitative skills
  • Understanding of benefit-cost or economic impact analyses

I wish my benefit-cost analysis students from last fall would stop by so that I could say "I told you so." 

ERE Special Issue on Benefit Transfer

Well, when I drafted this, it was a recent publication I wanted to announce, but now that I have some free time, I can finish it...

 

The March issue of Environmental and Resource Economics is a special issue on "Benefit Transfer: Current Practice and Future Prospects." Most of the papers arose from an EPA-sponsored workshop on benefit transfer; either through direct funding, as comments, or to introduce the issue. Kerry Smith put a ton of intellectual capital into identifying topics and authors for commissioned papers, and he wrote the introduction to the issue:

This paper introduces a special issue devoted to the benefits transfer methods used as part of benefit costs analysis for policy analysis. Benefits transfer methods, as they are applied for environmental policy analyses, use economic concepts together with existing empirical estimates to predict the incremental benefits from a change in some feature of an environmental resource. After giving two examples of the decisions that analysts confront in performing these analyses, I discuss the interconnections between the papers in this issue and the research challenges that emerged from discussions of them.

I helped contribute to organize the workshop and contribute to an EPA overview paper on benefit transfer challenges. I also discussed a paper by Laura Blow and Richard Blundell using non-parametric revealed preference to value environmental quality changes: 

We develop an approach to valuing non-market goods using nonparametric revealed preference analysis. We show how nonparametric methods can also be used to bound the welfare effects of changes in the provision of a non-market good. Our main context is one in which the non-market good affects the marginal utility of consuming a related market good. This can also be framed as a shift in the taste for, or quality of, the market good. A systematic approach for incorporating quality/taste variation into a revealed preference framework for heterogeneous consumers is developed. This enables the recovery of the minimal variation in quality required to rationalise observed choices of related market goods. The variation in quality appears as a adjustment to the price for related market goods which then allows a revealed preference approach to bounding compensation measures of welfare effects to be applied.

This was actually one of harder things I have had to do at EPA. I volunteered to be a discussant because I could recall being exposed to the technique in micro theory (Varian was one of the earlier developers and it's in his textbook). This method has not been used much in environmental economics (I only found four examples) but allows bounding of indifference curves--and therefore welfare measures--without functional form assumptions. Let me try to sketch the logic. In the figure below, there are two goods (superscripted) with one observed choice (subscripted). The shaded area RW is revealed worse, because any choice in that area was attainable but not chosen. The shaded area RP is revealed preferred, because any choice in that area would have strictly more of both goods. The indifference curve must therefore line in the unshaded area. 

Now, if you have more choices, you can narrow the indifference curve. 

There are further details in the paper on how to tighten the estimates of the indifference curves and then bound welfare estimates. The idea for benefit transfer (and let's be clear, this was Kerry's idea) is to use these kinds of estimates to help us benchmark parametric estimates of welfare. But it's a really cool paper!

Also in the issue are a paper by Cathy Kling and Dan Phaneuf on scope and adding up; and a paper by four current or former EPA authors (Steve Newbold, Patrick Walsh, Matt Massey, and Julie Hewitt) that improves a meta-regression framework by imposing theoretically-consistent assumptions on the functional form. And Kevin Boyle and Jeff Wooldridge have a great paper on how to handle errors structures and panel data in meta-regressions. And there are more, but I don't have a lot to say about them.

Important note: Whitehead (2016) gets a lot of love in this issue.

 

This work is not a product of the United States Government or the United States Environmental Protection Agency, and the author is not doing this work in any governmental capacity.  The views expressed are those of the author only and do not necessarily represent those of the United States or the US EPA.

 

 

It's like, I shouldn't count the health benefit from my oatmeal. It's delicious, that should be enough, boy.

Coral Davenport and Lisa Friedman at the NYTimes (The E.P.A.’s Review of Mercury Rules Could Remake Its Methods for Valuing Human Life and Health):

Last week the Trump administration took a crucial step toward de-emphasizing the life and health benefits in this calculus when the Environmental Protection Agency said it would rethink a major regulation that restricts mercury emissions by coal-burning power plants.

The 2011 mercury rule — based on decades of research showing that mercury damages the brain, lungs and fetal health — is among the costliest but most effective clean-air policies put forth by the Environmental Protection Agency. Utilities estimate they have spent $18 billion installing clean-air technology, and mercury pollution has fallen by nearly 70 percent.

Modifying the rule could have an impact far beyond any immediate concerns about the release of toxic mercury into the air and water. In fact, the re-evaluation fits into a far-reaching administration strategy to loosen environmental rules affecting countless other industries for years to come by adjusting the factors used to judge the benefits to human health that the rule has brought.

“This goes way beyond just weakening the mercury rule,” said Alan Krupnick, an economist at Resources for the Future, a nonpartisan Washington research organization. “This is part of a change that would give the Trump administration a way to more easily justify loosening many other pollution regulations, such as rules on smog, and rules on climate-change pollution.” ...

The electric utilities that operate the nation’s coal-fired plants, and thus are heavily affected by the mercury rule, are urging the administration to leave the rule alone. They have already spent billions of dollars to become compliant, the utilities say, so changes are of little benefit to them. ...

The calculations that the E.P.A. conducts for every major new air or water regulation lie at the philosophical heart of the agency’s work. And its process for that analysis has long been in industry’s cross hairs.

 

That’s because the agency has long counted not just the direct health benefits of pulling a certain pollutant out of the atmosphere, but also what are called the “co-benefits” that occur when, as a result, other toxins are also reduced.

For example, in the case of the mercury rule, the Obama administration found between $4 million to $6 million in health benefits directly from curbing mercury. But it further justified the regulation by citing an additional $80 billion in health benefits a year by, among other things, preventing as many as 11,000 premature deaths. Those savings come from a reduction in particulate matter linked to heart and lung disease that also occurs when cutting mercury emissions.

So, you can see why industry and the Trump administration is so upset. The co-benefits of 11,000 premature deaths should be worth zero in the mercury rule because mercury ain't whats killing them.

Common Resources: Ancillary Benefits Redux

Alan Krupnick in toto (emphasis added):

At a press briefing in mid-August, Bill Wehrum, the appointed head of the US Environmental Protection Agency’s (EPA’s) Air Office, reiterated the Trump administration’s position that ancillary benefits are not to be counted in cost-benefit analysis of major rules. The context this time was the Affordable Clean Energy Rule (ACE) proposed by the current regime to replace the Obama administration’s Clean Power Plan (CPP). If only the forgone carbon dioxide (CO2) benefits of pulling back on the CPP are counted, the cost savings from ACE outweigh these forgone benefits. But adding the ancillary health benefits that are lost with the ACE rule—the value of 1,400 fine particulate matter (PM2.5)-associated deaths related to greater coal use under the ACE rule—turns ACE into an economic loser, with net social costs relative to the CPP.

At the briefing, Wehrum said: “We’re not dealing with [sulfur dioxide] SO2. We’re not dealing with [nitrogen oxides] NOX. We’re not dealing with particulate matter. … We have abundant legal authority to deal with those other pollutants directly, and we have very aggressive programs in place that directly target emissions of those pollutants. So our view is, if we want to regulate PM, we regulate PM straight up. If we want to regulate SO2, we regulate SO2 straight up.”

The overarching response to this statement, as we have written on previously (here and here), is according to (1) theory (basic economic welfare analysis); (2) practice (OMB’s and EPA’s own guidance documents on the conduct of the analysis of the impacts of rules and statutes); and (3) case law—allbenefits and costs of a rule are to be considered. Just because the stated purpose of a rule is to reduce CO2 does not give license to ignore other impacts that occur. Consider how much the administration cares about looking at employment impacts of a rule. Surely environmental rules are not issued to change employment, but are written to protect public health and the environment, making the employment impacts ancillary. Yet the administration certainly wants these ancillary impacts on jobs to be considered, including not only employment impacts but also the ancillary health damages that occur from employment loss.

Wehrum’s statement further suggests that he wants the public to trust EPA to “aggressively” target NOX and SO2 emissions in their own right and with the knowledge that they cause PM2.5 concentrations that harm health. Yet, the administration’s proposed rule on the use of science at EPA is targeted at relaxing rules on PM2.5 and its precursors by challenging the science linking PM2.5 to deaths and the process of including these and other scientific studies in RIAs. If Wehrum really believes that we should regulate PM2.5, SOX, and NOX “straight up” to make up for the increases in these pollutants caused by ACE, then the issuance of the proposed ACE rule should be accompanied by the issuance of a PM2.5, SOX, and NOX mitigation rule that would bring these pollutants down to where they would have been under the CPP. Otherwise, the planned actions under ACE are merely about reducing the costs of regulations with little concern about the health impacts on the American people.

Note: The new name of the Environmental Economics blog (my posts, at least) is "Common Resources Echo".

A: Because it's better than acting like an obnoxious teenager

When I mention what I do (come up with ways to estimate the value of environmental goods and services), I sometimes get a response like this:  

It really is irritating that some people need to use economic measures to value our environment.

The reality is our environment, air, water, seas, national parks & Great Barrier Reef are truly priceless & must be treated as such ie no cost benefit BS studies needed.

My response is really quite simple.  If we don't measure the value of environmental goods and services and place them on par with marketed goods, we risk some acting as if the environment is free. 

It's like the nice neighbors who are going to be out of town on Halloween, so they leave a bowl of 'free' candy out for the neighborhood kids.  Most of the kids (mainly the supervised ones) are good citizens and only take one or two pieces of 'free' candy.  But it only takes one obnoxious teenager to recognize that 'free' means take all you can, and everyone loses out.  

So think of those who treat the environment as 'free' as the obnoxious teenage neighbor who needs more adult supervision.  

Or, start charging for your Halloween candy.

My benefit-cost analysis reading list for Fall Semester

I'm teaching benefit-cost analysis this semester. The course is for upper level economics majors (ECO 4660) and non-economics graduate students (ECO 5660).  I have 6 books on my shelf but none of them are satisfactory for a variety of reasons (mostly, the theory is too deep and the estimating benefits chapters are dated). So, I'm using the EPA's Guidelines for Preparing Economic Analyses supplemented with the Haab and Whitehead Encyclopedia. Here is the reading list (suggestions welcome!):

	Topic	EPA Guidelines (Required)	Encyclopedia (Recommended)
1	Introduction & Background	Chapter 1, 2; (Boardman et al. Chapter 1)	Benefit–Cost Analysis, Executive Order 12291, Guidelines for Preparing Economic Analyses, Welfare
2	Market Failure	Chapter 3, Appendix A	Externality, Public Goods
3	Decision Rules & Sensitivity Analysis	(Boardman et al. Chapter 7)	Risk and Uncertainty
4	Baseline	Chapter 5
5	Discounting	Chapter 6	Discounting
6	Estimating Benefits	Chapter 7	Averting Behavior, Benefit Transfer, Choice Experiments, Contingent Behavior, Contingent Valuation, Experimental Economics and Valuation, Hedonic Price Method, Meta-Analysis, Nonmarket Valuation, Passive Use Value, Travel Cost Method, Value of Statistical Life
7	Estimating Costs	Chapter 8	Pollution Abatement Costs
8	Economic Impacts	Chapter 9, Appendix C	Green Jobs
9	Distribution	Chapter 10	Environmental Justice, Welfare and Equity
10	Case Studies

RFF: "Questions about the Trump Administration’s Cost-Benefit Analysis for its Proposal to Freeze the CAFE Standards"

Alan J. Krupnick, Joshua Linn, and Virginia McConnell:

In April, we wrote a blog post responding to the Trump administration’s announcement that it was preparing to weaken the fuel economy and greenhouse gas (GHG) standards hammered out in 2011 by the Obama administration. We argued then that although some things have changed since those vehicle rules were established, the agencies should not simply roll back standards, but pursue a strategy to ensure steady long-term reductions in passenger vehicle GHG emissions to address the threat of global climate change.

Now, the Trump administration is proposing to freeze the fuel economy and GHG standards at model year (MY) 2020 levels until MY 2026 (the standards are set based on model years, which typically begin in the Fall of the preceding calendar year). This is an abrupt change from the Obama standards, which would have continued reducing emissions. Given current policies, this change would increase greenhouse gas emissions during that period and beyond. Fuel economy regulations were first established to reduce fuel use for more energy security (this rule would increase fuel use in new cars) and then concerns about climate change became an additional consideration (this rule would increase GHG emissions). The Trump administration downplays the GHG benefits of the Obama standards and claims that those standards would increase traffic accidents.

We will issue a series of blog posts discussing various aspects of the new proposal and the cost-benefit analyses underlying it. ...

The National Highway Traffic Safety Administration (NHTSA) is conducting the benefit-cost analysis instead of the Environmental Protection Agency:

The Trump administration’s modeling, inputs, and underlying assumptions behind the analysis of this new rule are quite different from those behind the analysis of the earlier rule finalized in January of 2017 at the end of the Obama administration. EPA models, assumptions, and focus on greenhouse gas emissions reductions tended to dominate the earlier analysis of the proposed rules for 2021 to 2025 MYs. The current proposed rule under the Trump administration uses models developed by National Highway Traffic Safety Administration (NHTSA) to evaluate benefits and costs. The NHTSA analysis emphasizes the effect of the new rule on fuel use, vehicle miles traveled, and accidents by the entire fleet. The effort to set standards for vehicles has always been plagued by the problem of having two masters—improving fuel economy on the one hand and reducing GHG emissions on the other. The two are closely linked, but the agencies that regulate them have different jurisdictional authority and perspectives. What is needed is a unified effort to address the societal costs of consuming fuel with clear goals and consistent modeling and analysis.

To its credit, NHTSA attempts to expand the scope of the benefit-cost analysis, which could improve the accuracy of the evaluation. For example, NHTSA has introduced a dynamic stock model that includes effects of the rule not just on new vehicle technologies and costs, but on the number of new and used vehicles predicted to be on the road and how much they are driven. The model also allows for variation in the mix of cars and light trucks in the fleet in response to changes in the rule.

The NHTSA is making stuff up:

However, while expanding the scope of the analysis is good in theory, in reality there are important questions about both the way the model is used and the underlying assumptions. For example, the analysis finds that in going from the Obama rule (stricter standards each year through MY 2025) to the Trump rule (no change in standards after 2020), the overall size of the vehicle fleet falls even though new vehicle prices are lower. On its face, this is inconsistent with economics. Further, NHTSA predicts that total vehicle miles travelled also falls during this period, appearing to assume that households choose how many miles to drive their vehicles independently of the number of vehicles they own. This assumption is also inconsistent with economics. The agency is arguing that looser standards decrease the size of the on-road fleet and decrease total driving. We’re not aware of any real-world evidence supporting this argument. These new elements of the model are key because they result in significant reductions in external costs (such as congestion) and increases in safety from the change in vehicle ages and decreased driving under the proposed new rule.

Costs of CAFE regulation are inflated:

In addition, changes in how and when new technologies are introduced may have a significant effect on estimates of total technology costs. In the current NHTSA analysis, the technology costs are predicted to be more than two times higher for the Obama standards than the Obama analysis predicted for its rule in 2016. However the per vehicle costs in the new assessment are only about 50 percent higher ($1300 vs. $1950 in 2016 dollars). The NHTSA analysis provides some explanations for why these per vehicle costs might be higher, but it is not clear why total technology costs would be so much higher. ...

And, it ain't over:

In future blogs we will address other topics related to the proposed freezing of the standards, such as the effect of gasoline prices on fuel economy standards, assumptions about greenhouse gas emissions and other air pollutants, and the possible use of a safety valve if there is concern over high costs of meeting the standards.

Here is the link: http://www.rff.org/blog/2018/questions-about-trump-administration-s-cost-benefit-analysis-its-proposal-freeze-cafe

What’s the value of a clean beach? Here’s how economists do the numbers

I was asked to write an article for The Conversation that explains how economists place values on environmental goods (with examples).  Here's what I came up with.   Apologies to all of those involved in the research if this comes across as self-promoting--the perils of writing with an editor.  All of the examples mentioned in the article were team efforts--most were not my idea--and I was a minor part of each team.  It takes a village, and every village needs an idiot.

Millions of Americans head outdoors in the summer, whether for a day at a nearby lake or a monthlong road trip. For environmental economists like me, decisions by vacationers and outdoor recreators offer clues to a challenging puzzle: estimating what environmental resources are worth.

In 1981 President Ronald Reagan issued an executive order that required federal agencies to weigh the costs and benefits of proposed major new regulations, and in most cases to adopt them only if the benefits to society outweighed the costs. Reagan’s order was intended to promote environmental improvements without overburdening economic growth.

Cost-benefit analysis has been so successful as a tool for policy analysis that every administration since Reagan has endorsed using it. However, it requires measuring benefits that are not “priced” in typical markets. Fortunately, putting a price on non-market environmental outcomes, such as safer drinking water and fewer deaths from exposure to dirty air, has proved to be possible, and highly valuable. These estimates help to make the case for actions such as cleaning up beaches and protecting scenic areas as parks. ...