Environmental Economics

AERE Happy Hour at the SEA meetings

I just sent this message:

AERE / SEA session participants,

Let's have our informal AERE Happy Hour after the Distinguished Guest Lecture ("Life as a Lab: Using Field Experiments in Economics" John List [see note below]) on Saturday. This is a great chance to meet old friends and new and organize your dinner party!

 

What: AERE Happy Hour

 

Why: Meet old friends and new and organize your dinner party

 

Where: Hotel lobby bar (i.e.., Pulse -- "At 5:00 pm, an audible heartbeat will welcome the beginning of an eight-hour social hour as Pulse transforms into a vibrant lounge. As the evening progresses, the bar staff intensifies the showcasing of mixology behind the bar by giving demonstrations on how to make classic and new trendy drinks. The color of the iconic sail will change with the mood of the evening ending with a heightened pulsating array of radiant colors as a salute to the evening.")

 

When: 6:15 pm

 

See you in Atlanta!

 

John W.

 

Note: Speaking of field experiments, I withdrew my field experiment paper just now. Buy me a beer at happy hour and I'll explain why!

 

I'll try to post on that note soon (if I can get ready for my other sessions). Here is the excuse that I gave the SEA "We've been waiting on some essential data and it has not arrived." Nor will it ever.

What is the optimal number of time slots on a Doodle Poll?

I don't know the answer, but I do know that 64 is too many.

Dubious and Dubiouser*

From the Miscellany Section of Economic Inquiry:

We collect contingent valuation data from 524 student survey respondents over a 3-day, 72-hour period. Data analysis of a hypothetical campus referendum focuses on time-of-day effects on willingness to pay for a renewable energy project. We find that subjects responding to the survey during the night-time hours (i.e., between 12 a.m. and 6 a.m.) do not display the law of demand, offering theoretically invalid responses to questions with important policy implications. Results from this research may have serious implications for the contingent valuation method (CVM). In short, just like your father said, nothing good happens after midnight when using the CVM. (JEL Q51)

via onlinelibrary.wiley.com

The last line of the introduction says it all: 

A guiding principle, consistent with the work of the CVM critics, is that a hypothetical valuation question, especially one used for major policy issues and natural resource damage assessments, should pass the most important theoretical validity test no matter what time of day the survey is administered.

I'll point out the funny parts if, like the folks we passed it around to for a quick read and the journal referees, you can't spot the humor (here's one: footnote 1 drops "cromulent" for the first time, I think, in the economics literature). And thanks to the editor for accepting the paper even when a referee said it should be rewritten as a real paper (and submitted where?). That was the original intention but we could never get motivated to do so. It was only during the hopeless to curious phase of my life that inspiration stuck. 

*Dickinson, D. L. and Whitehead, J. C. (2014), DUBIOUS AND DUBIOUSER: CONTINGENT VALUATION AND THE TIME OF DAY. Economic Inquiry. doi: 10.1111/ecin.12161

New paper on hypothetical bias in contingent behavior data

This has just been published in the journal Economics and Business Letters: Special Issue on Cultural Industries:

Criterion and predictive validity of revealed and stated preference data: the case of “Mountain Home Music” concert demand

John C. Whitehead, Douglas Simpson Noonan, Elizabeth Marquardt

Abstract

Despite a robust literature on nonmarket valuation of cultural assets, serious validity concerns remain.  We address this by estimating a demand model for a regional concert series.  We survey concertgoers during and then again after the concert season to gather ex ante and ex post stated and revealed preference data. Comparing ex ante stated preference data to ex post revealed preference data we find respondents overstate their concert attendance behavior. An ex ante revealed-stated preference demand model with a stated preference adjustment helps calibrate the results and avoid bias from using solely hypothetical, stated preference data. The results demonstrate how to improve predictive accuracy in contingent behavior models and improve our understanding of demand for live music performances.  

via www.unioviedo.es

The full text is available here: http://www.unioviedo.es/reunido/index.php/EBL/article/view/10289/10016. 

From the conclusions:

Despite its popularity in cultural economics, Hausman (2012) condemns SP data for hypothetical bias. Relative to contingent valuation there are very few contingent behavior studies where tests for hypothetical bias are even possible because of context and data limitations. We conduct a unique in-season and postseason survey to test the predictive validity of SP survey responses. This is a rare opportunity to apply it, and a novel opportunity in the cultural field. We find that SP concert attendance data lack criterion validity. Respondents tend to overstate their concert attendance behavior. Respondents are generally accurate, however, when predicting their own behavior after a statistical adjustment for hypothetical bias.

The sample size is only 38 so don't get too excited. 

 

"A Challenge to Three Widely Held Ideas in Environmental Valuation"

Matt Interis:

Environmental valuation is the branch of environmental economics in which researchers estimate the economic value of environmental goods and services. Environmental valuation has been practiced for decades. However, there are some ideas in the field of environmental valuation held by many environmental economists and nonenvironmental economists that appear to be outdated. This article discusses three such ideas: 1) that it is better to estimate willingness-to-pay values than willingness-to-accept values; 2) that stated preference valuation methods are questionable because they are based on hypothetical choices rather than real choices; and 3) that it is better to use a repeated-choice question format than a single choice format in choice experiments. We discuss the origins of each idea and why the idea became prevalent in the first place.We then review recent literature, which casts doubt on the idea. We conclude with a reminder for researchers—in environmental economics and in other economic fields—to periodically reassess ideas they currently hold in light of recent research developments and in light of the context in which they are used.

via ageconsearch.umn.edu

I agree with each of the three points.

Addendum:

Adam in the comments says: "You agree that the three points are valid critisms, or you agree that they are outdated?" So, here is some explanation:

1. If there are reasons why WTA is theoretically preferred, and WTP and WTA are expected to differ, then one should attempt to estimate WTA for the loss (in addition to WTP to avoid the loss)
2. Hypothetical choices contain too much information, some of it unobtainable with revealed preference data, to dismiss it out of hand.
3. I'm being a bit disingenuous with my agreement on this one. I don't think the implications of repeated questions in choice experiments has been explored enough to blindly accept the practice. In contrast, repeated questions in CVM studies are frowned upon (sometimes for good reason). I just had a revise and resubmit rejected at JEEM (grrr) and repeated questioning was one of the reasons given, even though the variation across questions was similar to choice experiments (i.e., the bid amount didn't change). My hunch is that many choice experiment studies won't pass a scope test if only the first question is used, which is the standard upon which CVM studies are judged. That said, I think it would be a shame if we go to all the trouble to develop a hypothetical choice scenario, ask one question and move on to the next section of the survey. 

New working paper

http://econpapers.repec.org/paper/aplwpaper/14-09.htm: 

Predictive Validity of Stated Preference Data: Evidence from Mountain Bike Park Visits Before and After Trail System Expansion

Kevin Atkinson and John C. Whitehead ([email protected])

No 14-09, Working Papers from Department of Economics, Appalachian State University

Abstract: This paper investigates the validity of stated preference data for use in recreation demand estimation. We use stated preference and revealed preference data from users of a mountain bike park collected before and after an expansion of the trail system. The ex-ante stated preference data elicited before the change exhibits hypothetical bias, but, it would provide useful information for demand prediction.

Key Words: hypothetical bias, recreation demand, revealed preference, stated preference

Kevin is an undergraduate economics major who'll be applying to graduate schools this year. This paper is revised from the paper he presented at the National Conference on Undergraduate Research. 

This guy (yes, I've been watching MNF) thinks I'm willing to pay about $16 per year to restore the environment of the southern Appalachians to a healthy state

Spencer Banzhaf:

In a new RFF discussion paper, “Valuation of Ecosystem Services in the Southern Appalachian Mountains,” RFF colleagues Dallas Burtraw, Alan Krupnick, and Juha Siikamäki and I, along with  Susie Chung Criscimagna of Eden Housing, Bernard Cosby of the University of Virginia, and David Evans of EPA,  estimate the monetary value of reducing acidity in the southern Appalachian region. Our estimates were based on the stated preference method, which allowed us to survey households about their willingness to pay (WTP) for the restoration of ecosystem services in the region. Based on their answers, we determined that households across the region would be willing to pay $15.67 per year each—or about $208 million in total—to restore the environment of the southern Appalachians to a healthy state. Aggregated and discounted into the future, and allowing for a 50-year delay for the ecosystems to recover, these annual values come to a present value of $3.7 billion. Yet it’s worth noting that these ecosystem service benefits still pale in comparison to the total economic benefits of reducing air pollution, which stem overwhelmingly from public health.

via common-resources.org

This isn't the sort of press the UMKC B-school wanted

We're number 1 in the world in generating bogus rankings!

The governor of Missouri has called for an investigation into the validity of rankings that prominently featured the University of Missouri at Kansas City’s business school. Gov. Jay Nixon sent a letter to the university’s Board of Curators after an investigation by The Kansas City Star revealed previously undisclosed relationships between the university and the two authors of a study that ranked the Henry W. Bloch School of Management as No. 1 in the world in innovation management research.

The university released a statement on Thursday that disputed the newspaper’s findings: “We have not violated the public trust; we say this with confidence, based on multiple reviews of these issues conducted in academic circles long before The Kansas City Star became involved—reviews that found these criticisms to be without merit.”

via chronicle.com

Here is a previous post. 

Are all citations good citations?

My dissertation, written under the direction Glenn Blomquist in 1989, considered whether information about alternative environmental goods was missing from contingent valuation (CV) scenarios. Subsequently we received funding to consider whether the CV method could distinguish amongst quality differentiated environmental goods. The paper got beat around a bit but eventually Blomquist and Whitehead (1998) was published. It turns out to be my second most cited paper. Here is the abstract:

Elicitation of valid statements of contingent value requires survey participants who are familiar with the environmental resource change. A primary purpose of the contingent market must be to assure familiarity by providing information. Information about resource quality is important when incompletely informed respondents, say nonusers, perceive resource quality which diverges from true quality. Differences in perceived quality and true quality can be influenced as respondents learn from information in the contingent market. By presenting survey participants with information about four wetlands of varying qualities we test for information effects in a dichotomous choice contingent market for wetlands allocation. We find that information about quality is a determinant of willingness to pay for wetland preservation. Information about resource quality presented in contingent markets will result in more valid valuations of changes in allocations of environmental resources.

During a navel gazing session I ran across this:

I don't think Glenn nor I walked away from Blomquist and Whitehead (1998) "suspicious" about the validity of the CV method. I've always wanted to be cited in the same parenthetical as Hausman (2012), for example, like this (Hausman 2012, Haab et al. 2013), but this one makes me feel yucky. For once, I wish one of my papers wasn't cited.

I received a response to my letter to Senator Vitter

It wasn't very responsive to my concerns.

Here is the whole thing (Download {Disarmed} Here to Help).