What do expert economists know?

Warning: thinking about this topic may blow your mind.

First some background before getting to my main point:

One thing environmental economists do is estimate the value of things we like about the environment (more birdies, cleaner air, etc.). The purpose of this is that the estimates can be compared to costs of obtaining these things we like about the environment. One tool environmental economists use for this purpose is stated preference surveys – surveys in which you put the respondent in a hypothetical situation in which he or she must make a tradeoff between money and some (usually) improvement in the environment.

These surveys have been used by decades, but some economists question the validity of the estimates derived therefrom. The debate is multifaceted (for more info, read the contingent valuation symposium articles here), but one suggestion made by super-famous economist/econometrician and noted stated preference survey detractor Jerry Hausman is that, rather than administering such surveys to the general public, it would be better to get a team of “experts” to assess these values. Remember this. (Key point: “value” refers to the value to individuals. It has always been counterintuitive to me that experts would be better judges of the value of something to people than those people themselves would be.)

Now to the story:

A team of stated preference survey practitioners (among them Richard Carson, who would probably be considered the foremost expert on the practice, and he’s an MSU alum to boot!) recently conducted a Delphi survey on the value of protecting the Amazon rainforest to people in countries outside of South America. A Delphi survey is one in which expert opinion is elicited about some topic, a summary of their responses are then provided in a second round of the survey, and the experts are then allowed to revise their original responses. In this case (read the study here), 216 expert environmental economists in stated preference methods from 36 countries (North America, Europe, Asia, Oceana) were asked to predict the mean and median value to the average person in the expert’s home country for protecting the Amazon rainforest from further destruction through the year 2050. (Your own Drs. Interis and Petrolia were among these experts.)

The data show that experts predict that the value, per year until 2050, range from $4-$36 in the Asian countries to near $100 in Canada, Germany and Norway, with other countries lying in between (note: the measure of value used was willingness to pay, which generally increases as income increases, so the higher values for wealthier nations makes sense). There are many more interesting findings of the paper which you can read about within the paper, but one of the most interesting things is that the Delphi method hasn’t really been used before for estimating environmental values.

Now back to the debate I mentioned above:

The authors state that they intend to conduct actual general-population stated preference surveys in several of the nations represented by the experts in this study. They say in this paper that “A large discrepancy between the Delphi exercises and the population surveys…may signal a need for further analysis of our overall valuation approach.” It’s not exactly clear what they mean by “our overall valuation approach” but when I read it I interpreted it as referring to stated preference valuation in general.

I think this may be a coy reference to Hausman’s complaint about stated preference methods that I mentioned above – if experts do not do a good job of predicting what the results of a general-population survey would be, then one would wonder if relying on a panel of experts (instead of going to the general population) would be a good approach. Of course, I’m assuming here that people are better judges of their own values than experts are (call this Stance 1), but the reason Hausman made the suggestion in the first place was because he believed the opposite: that experts are better judges of people’s values (at least when it comes to values of environmental public goods that are elicited with stated preference methods) than the people themselves (call this stance 2).

So this is a really interesting problem (if you don’t think so, you’ve probably stopped reading by now). All economists agree that “value” is a subjective measure, but the question is whether experts or people themselves are better judges of people’s values. This problem might be partially resolved if we had some other measure of value that people of both Stance 1 and Stance 2 could agree were correct – then we could compare estimates from experts and estimates from the general public to see which were more accurate (i.e. closer to this third measure that they agree on). BUT, there are two problems with this. First, there are other ways to measure values of public goods (generally called revealed preference approaches), but these have their own problems and weaknesses and can’t generally be judged to be more or less reliable (i.e. more or less “accurate”) than stated preference approaches. And second, when it comes to certain types of environmental values (e.g. value the rainforest to someone just because it exists – and hey, you can’t argue with this because value is subjective!), stated preference methods are the only way to obtain estimates in the first place, so there’s no revealed preference estimate to compare it to.

I don’t know about you, but my mind boggles just thinking about this. But, if you’ve understood everything, especially the previous paragraph, you, like me, might be inclined to think that there can never be a resolution to this debate (about whether experts or people themselves are better judges of people’s values for environmental things that can be valued directly only with stated preference methods). So why bother debating? I’d say for several reasons: it’s our job as economists to try to answer pertinent questions, it’s important to understand multiple sides of an issue, and – not of any lesser importance – it’s fun. If economists didn’t have fun, they wouldn’t be economists after all.

So we’ll have to stay tuned to see if there is a follow up study that actually compares the expert predictions to actual estimates from a general population survey. Can’t wait!

Hey New York! We evaluated a similar idea here in MS…

John Whitehead of env-econ.net posted this on his blog today:

Beware when federal officials request innovative ideas

Here is another innovative idea, don’t subsidize risky living:

A string of artificial islands off the coast of New Jersey and New York could blunt the impact of storm surges that proved so deadly during Superstorm Sandy, according to a new proposal.

It’s a big proposal – one that would cost up to $12bn – but it’s also the kind of innovative idea that federal officials requested as they consider how best to protect the heavily populated east coast from future storms. …

The “Blue Dunes” proposal is part of a competition sponsored by the US Department of Housing and Urban Development to come up with novel ways to protect Americans against the next big storm. It is one of 10 projects that will be evaluated and voted on next week, but there’s no guarantee any of them will receive funding. Other ideas include building sea walls around cities, re-establishing oyster colonies in tidal flats to blunt waves and creating water-absorbent nature and recreational preserves.

The artificial islands plan was created by Stevens Institute, along with the WXY architectural firm and West 8 Urban Design and Landscape Architecture. It is designed to blunt the worst effect of Sandy: the storm surge that pounded the coast. From Maryland to New Hampshire, the storm was blamed for 159 deaths, and New Jersey and New York alone claimed a total of nearly $79bn in damage. …

The islands, 10 to 12 miles off the coast, would be uninhabited, although day trips for surfing or fishing might be allowed, Blumberg said. They would be built by pumping sand atop some hard base made of rock, concrete or other material.

Steve Sandberg, a spokesman for Senator Robert Menendez, a New Jersey Democrat, said funding for at least some of the proposals is already available as part of the $60bn in Sandy aid that Congress passed last year. Other money could come from disaster recovery grants as well as public and private-sector funding.

A gap would be left between the New York and New Jersey island groups to allow water from the Hudson River to flow out into the ocean.

Blumberg also said computer modeling has shown such islands would have produced vastly lesser damage during Sandy, Hurricane Donna in 1962 and the destructive December 1992 nor’easter.

Aside from the formidable cost, many other obstacles remain. …

via www.theguardian.com

The most awesome part of the plan is that the $12 billion is already there! And every taxpayer in the U.S. has contributed! Why am I exclaiming like Mark Trail!

This reminded me of some research on a similar topic – restoring barrier islands off the coast of Mississippi in order to protect the coast from storms – by some researchers in this department….

Kim, G., D. R. Petrolia, and M.G. Interis (Journal of Agricultural and Resource Economics, 2011): A Method for Improving Welfare Estimates from Multiple-Referendum Surveys.

In this paper, we estimate the economic value of a project to restore barrier islands off the coast of Mississippi for the purpose of protecting the coast from storms. Our estimate, based on a population of MS of about 3 million people, is a little under $700 million.

It would be cool if someone (or team) estimated the value of the “Blue Dunes” project in New York to see if it exceeds the $12 billion cost of implementation. (Well, they say it will be “evaluated” but who knows what that means!)

By the way, in our paper, we propose a way of estimating the value of the project which increases the confidence we have in our estimate of that value (through a lower variance…for the statistics nerds out there).