Environmental Economics

Last week I began a research project for @FoothillsOfNC

On the Monday of spring break I drove to Morganton, NC and visited Oak Hill Community Park and Forest: 

Located 10 minutes from downtown Morganton, the park will be a hub for outdoor recreation, education, agriculture and archaeological exploration for Morganton residents and visitors.

Foothills Conservancy's Executive Director Andrew Kota (one of my former MBA students) led us on a hike where you can get a view of the South Mountains. In the foreground of the picture you can see the plantings of native trees (I forget the species). 

We'll be putting up signs with a URL and QR code to an online survey when the park opens later this year. Combined with a trail counter we can develop estimates of economic impacts and do a benefit-cost analysis (an "estimate of probable costs" can be found in the Oak Hill Master Plan (see the park link above). The data is from a convenience sample but some data is better than no data. As these studies can be used to decide whether a more representative (and much more expensive) sample is necessary. The out-of-pocket cost estimate for this study is less than $1000. The signs will be about $250 and a trail counter is between $620 and $800. 

Back in 2021 I worked on a Beech Mountain trails study and this will be similar to that, although I'm hoping to get better at this sort of thing as more lessons are learned. For example, I need separate surveys for different data collection methods. At Beech we had signs with URLs and QR codes and a team of students passing out rack cards with the same information. It would have been very useful to see where the data came from but the links went to only one survey. Also, Beech Mountain is full of vacationers so it is difficult to estimate a recreation demand model. 

Today's example of how the critics of environmental regulations ignore the benefits: Clean Neighbor Rule

Coral Davenport in the NYTimes (E.P.A. Tells Dozens of States to Clean Up Their Smokestacks):

The Biden administration on Wednesday finalized a rule forcing factories and power plants in 23 Western and Midwestern states to sharply cut smog-causing pollution that is released from their smokestacks and fouls the air in Eastern states.

Known as the “good neighbor” rule, the new regulation strengthens and expands an earlier interstate air pollution standard that was enacted during the Obama administration. While that rule directed power plants to clean up their emissions, the revised rule enforces similar controls on mills, factories and other industrial facilities.

Here is EPA's website for the Good Neighbor Rule: https://www.epa.gov/... (that's where I got the map). Click on the Regulatory Impact Analysis link to see the numbers for the benefits and costs referenced below. 

The Environmental Protection Agency is required by the Clean Air Act to periodically review and revise the rule. After failing do so during the Trump administration, it is now strengthening restrictions under a court-ordered deadline.

The good neighbor rule holds that states should take measures to ensure that their pollution doesn’t affect downwind states. It directs coal-burning power plants and industrial facilities such as iron, steel, cement and concrete manufacturers in the Western and Midwestern states to reduce their emissions of nitrogen dioxide, a pollutant that causes smog and is linked to asthma, lung disease and premature death.

Soot and exhaust belched from those industrial facilities is carried by prevailing winds toward Eastern states, causing high levels of pollution in states with fewer industries. ...

The tighter rules on power plants will come into force later this year, while the new controls on factories and other industrial polluters will take effect in 2026. ...

The E.P.A. estimates that the updated good neighbor rule will cut emissions of nitrogen oxide in the affected states by 50 percent from 2021 levels by 2027, preventing 1,300 premature deaths, avoiding more than 2,300 hospital and emergency room visits, preventing 1.3 million cases of asthma and avoiding 430,000 lost school days and 25,000 missed work days. ...

If the EPA is using a value of mortality risk reduction estimate equal to $10 million per life saved (I used up my time budget and still couldn't find the number in the RIA) then those benefits are $11.3 billion. 

The critics are funny.

... Senator Joe Manchin, Democrat of West Virginia, has raised concerns about the costs to power plants and manufacturers to comply with the strengthened rule, saying it could force some plants to shut down or reduce operations, threatening the reliability of electric grids. Mr. Manchin wrote Mr. Regan last week to ask him to postpone action.

The costs are real but I'm fairly sure it won't threaten the reliability of electric grids. If I'm way out there on this one please let me know. 

Scott Segal, a lobbyist for Bracewell LLC, which advocates for coal plants and industrial manufacturers, said the cement industry would also feel the brunt at a time when its products are in demand because of a recent influx of federal investment in road and bridge construction.

“The bipartisan infrastructure law anticipates substantially increased use of building materials manufactured here in the United States, a policy objective seemingly at cross purposes with the rule,” Mr. Segal said.

The cement industry logic is that if the government increases the demand for cement then it shouldn't regulate the increased pollution the cement industry will cause. 

The E.P.A. estimates the cost of complying with the revised rule to be about $910 million annually from 2023 to 2042, while the net economic benefits of the associated public health gains would be up to $13 billion each year over the same time period.

That's not really a good summary of the present value of net benefits. First, delete "net" and "up to" since $13 billion is the annualized value equivalent to the $910 million number (see Table ES-15 of the RIA below the fold). The present value of the net benefits from 2023 to 2042 is $200 billion with a 3% discount rate and $140 billion with a 7% discount rate.

Other manufacturers, which have in the past joined industrial states in legal action to block the rules, have been wary of the updated versions.

“While the intent of this proposal is right and one that we share, it could have significant effects on American families if not thoughtfully implemented,” said Aric Newhouse, senior vice president of policy and government relations at the National Association of Manufacturers, when the E.P.A. first proposed the updated regulation last year.

“We must be careful with regulations that could further raise prices on all Americans, slow economic growth and threaten jobs,” Mr. Newhouse said.

When the benefits exceed the costs one play in the playbook is to raise the scare of macroeconomic impacts. I don't know of any credible evidence that environmental regulations have had much macroeconomic impact in the U.S. since 1970. They lead to lost jobs in polluting industries but these jobs move to other sectors of the economy.

Conclusion: The critics of environmental regulations (Manchin and the manufacturing industry) ignore the benefits of environmental regulations, especially when the benefits exceed the costs.

Our article in the Applied Economic Perspectives and Policy's "sizing up survey samples" symposium (?) is published online

This paper resulted from the BP/Deepwater Horizon damage assessment conducted for the state of Florida (Huffaker, Clouwer and Larkin 2012). We used the Knowledge Networks (KN) internet panel for the sample (besides me, the authors are Andrew Ropicki, John Loomis, Sherry Larkin, Tim Haab, and Sergio Alvarez). At one point I remember Sherry Larkin asking all of us if it was OK if KN filled out our sample needs (n=2000) with opt-in data (I don't recall what they called it back then). We all said fine, assuming KN knew what they were doing and it was the gold standard, right? The report sat without a paper carved out of it until I started using opt-in samples, noticed the different samples in the Florida data and compared models using the two different samples (*note). I presented the results at the 2019 Society for Benefit-Cost Analysis Annual Meeting and the paper sat a while longer. We finally submitted it to AEPP for a symposium on samples in stated preference research in the Fall of 2021 and it was accepted last month.

We find that the opt-in sample data leads to higher willingness to pay estimates and statistically insignificant scope effects (**note). The scope elasticity with the probability-based sample is 0.73 and statistically significant (***note). The income elasticity is higher with the probability-based sample too. 

There are three other papers in the symposium (not sure if AEPP is calling it a symposium, but I'm sure that it is a group of articles that are being published together) that are all good reads and should inform future research with internet panel data (which, I think, is the future). While I think opt-in data is inferior to probability-based panels, I'm not sure if the added cost is justified. This is especially true with more modern survey designs, relative to the NOAA panel era split-sample single binary choice scope test (repeated questions as CVM merges with discrete choice experiments), and methods to deal with the lower quality data (e.g., attribute non-attendance). One article in the symposium makes the comparison with this type of data (sans attribute non-attendance models). A third article compares high quality and convenience sample data and the fourth considers bots and other threats to data quality with internet data (Goodrich et al. 2023 -- the other two haven't appeared on AEPP's website yet). 

Here is the link to our paper: https://doi.org/10.1002/aepp.13352

Notes: 

*A recreation demand paper was previously published out of this report (Whitehead et al. 2018).

**This result might be due to low sample size. The scope elasticity is 0.37 which isn't crazy low. In other words, a huge sample size would reduce the standard error so that we could claim statistical significance. For example, see the federal study that has a sample size of n>3000 and a modest scope effect (the result becomes statistically insignificant if you randomly choose samples of n=1000, more on this someday). 

***I have another data set where GfK (formerly KN) did this. We are estimating CVM recreation trip models with repeated questions and variations in the choice attributes. I hope to write up the paper and submit it somewhere this summer. 

References:

Bishop, Richard C., Kevin J. Boyle, Richard T. Carson, David Chapman, W. Michael Hanemann, Barbara Kanninen, Raymond J. Kopp et al. "Putting a value on injuries to natural assets: The BP oil spill." Science 356, no. 6335 (2017): 253-254.

Goodrich, Brittney, Marieke Fenton, Jerrod Penn, John Bovay,and Travis Mountain. 2023.“Battling Bots: Experiences and Strategies to MitigateFraudulent Responses in Online Surveys.”Applied Economic Perspectives and Policy1–23.

Huffaker, R. G., R. L. Clouser, and S. L. Larkin. 2012. Contract for Analytical Services Related to the Deepwater Horizon Disaster: Estimation of Lost Indirect and Passive Use Economic Values to Floridians. Final Report to the Florida Legislature. Gainesville, FL: Food and Resource Economics Department, University of Florida. 

Whitehead, John C., Tim Haab, Sherry L. Larkin, John B. Loomis, Sergio Alvarez, and Andrew Ropicki. "Estimating lost recreational use values of visitors to northwest Florida due to the Deepwater Horizon oil spill using cancelled trip data." Marine Resource Economics 33, no. 2 (2018): 119-132.

 

The NYT's Willow article allows issue advocates to alternately assume perfectly elastic demand and supply

The decision to allow mining in Alaska is controversial (Biden Administration Expected to Move Ahead on a Major Oil Project in Alaska): 

In one of its most consequential climate decisions, the Biden administration is planning to greenlight an enormous $8 billion oil drilling project in the North Slope of Alaska, according to two people familiar with the decision.

Alaska lawmakers and oil executives have put intense pressure on the White House to approve the project, citing President Biden’s own calls for the industry to increase production amid volatile gas prices.

But the proposal to drill for oil has also galvanized young voters and climate activists, many of whom helped elect Mr. Biden and who would view the decision as a betrayal of the president’s promise that he would pivot the nation away from fossil fuels.

On one side is the "environmental activist" worried about a "carbon bomb" (in fairness to issue advocates, this is not a quote from an activist): 

Still, Willow would be the largest new oil development in the United States, expected to pump out 600 million barrels of crude over 30 years. Burning all that oil could release nearly 280 million metric tons of carbon emissions into the atmosphere. On an annual basis, that would translate into 9.2 million metric tons of carbon pollution, equal to adding nearly two million cars to the roads each year. ...

Environmental activists, who have labeled the project a “carbon bomb” have argued that the project would deepen America’s dependence on oil and gas at a time when the International Energy Agency said nations must stop permitting such projects to avert the most catastrophic impacts of climate change.

Increasing the amount of oil produced would increase the supply of oil if we consider new mines as firms entering the industry. But, the only way that the amount of oil produced would increase the amount consumed by the same amount (the "carbon bomb" scenario) is if the demand for oil is perfectly elastic. In the picture below the additional mining shifts (rotates) the supply of oil from S to S' and buyers in the market snap it up at the existing price. The increased quantity of oil consumed is equal to the increased supply at the market price (both arrows are the same size). 

But, if the demand for oil is sensitive to price (note: it is) then the amount consumed (the bottom arrow) will be less than the supply increase (the upper arrow): 

On the other side is the energy industry making even less sense: 

Kevin Book, managing director of Clearview Energy Partners, a research firm, ... argued that the emissions linked to burning oil drilled from the Willow project would not have been eliminated if Mr. Biden had rejected the project, but simply generated elsewhere.

This quote assumes that there is some demand increase, the source of which I don't know, and a perfectly elastic supply from existing miners. 

If the supply curve is upward sloping (it is) then the increase in consumption would be less than the demand increase. I'm not going to dignify this scenario by drawing this out because I'm not sure where the demand increase is coming from. As far as I can tell, if there is no oil from the Willow project nothing changes in the oil market. 

Happy to be corrected, what have I missed?

My SBCA slides stunk but I think I delivered on emphasizing a stated preference "robustness check"

I presented a paper at the SBCA meetings yesterday. The study was internally funded so I've been asked to present on-campus a couple of times. I started out with Google Slides for some reason (we're a Google campus but that is not a constraint). That was a mistake. I didn't have time to redo the slides in Powerpoint for the SBCA so I just edited the Google slides. The SBCA asked for the slides in PDF so I obliged. The PDF shrunk the slides to about 65% of the screen somehow (see below, with an added border). What? I wanted to try my Google Slides but our session was on Zoom so transition was time consuming so I just said let's go with the PDF.

If you are wondering what I might have said ... the main point of the presentation is that if you are using stated preference data and aren't testing for attribute non-attendance (ANA) then you are missing an important robustness check (I recently realized that I need to use the "robustness check" language to get some people's attention ([e.g., age < 40]). Our WTP estimates are cut in half when we use equality-constrained latent class models with ANA on the cost amount and scope variables. I have a list of several papers I hope to write in the next year that pushes this point.

Bottom line: add this to the list of ways I've embarrassed myself at conferences.

Here is the presentation: Download SLR Whitehead

I'm at the Society for Benefit Cost Analysis conference in DC

This is the first in-person SBCA conference since 2019 following a cancelled 2020 conference and online conferences in 2021 and 2022. I organized two sessions on environmental nonmarket valuation and, in my opinion, they turned out great. 

The sessions were put together following a "call for papers" on Twitter. Five guys were interested and with my paper that made 2 sessions. I'm hoping to do this again next year but know that a tweet isn't the best way to try to put a session together (e.g., 3 sessions is better than 2!). If anyone is interested in participating next year or has ideas about how to get the word out, please let me know (I'll probably risk the wrath of AERE with an email on the RESECON listserv). 

Here are the session details:

Estimating Environmental Benefits from Housing Markets
Thursday | 10:45 am-12:15 pm | GWU Amphitheater

Organizer:John Whitehead, Appalachian State University
Chair:John Whitehead, Appalachian State University

• The Economic Consequences of Local Gas Leaks: Evidence from Massachusetts Housing Market. .....Pengfei Liu, University of Rhode Island
• Environmental Valuation Across Time and Benefit Cost Analysis. .....Patrick Walsh, US EPA NCEE
• Untangling Willingness to Pay for Public Land Use Change: An Analysis of Greenways in Raleigh, North Carolina. .....Lee Parton, Boise State University

Estimating Willingness to Pay with Stated Preference Methods
Thursday | 2:15 pm-3:45 pm | GWU Amphitheater

Organizer:John Whitehead, Appalachian State University
Chair:Patrick Walsh, US EPA NCEE

• Distribution-free Methods to Estimate Willingness to Pay Models Using Discrete Response Valuation Data. .....Samuel Zapata, Texas A&M University
• A Meta-Analysis of Recreational Fishing Benefits. .....Matt Massey, US EPA NCEE
• Sea level rise, drinking water quality and the willingness to pay for coastal tourism. .....John Whitehead, Appalachian State University

This location MAY be unsafe?

When is the right time to measure economic damages from a disaster?

By now most know of the train derailment in East Palestine (pronounced pal-uh-steen), Ohio and the ensuing decision to burn the chemicals spilling from the train rather than wait for a potentially more disastrous--and perhaps more lethal--explosion. What is not yet known, and likely won't be known for some time, is the scope and extent of the full set of damages. Within days of the derailment I received calls asking something along the lines:

"Would you and your team please move forward with determining the agricultural, environmental and health impacts of the train derailment in Columbiana County?" 

My standard response is along the lines:

"That’s a big (and broad) ask. We have experts who may be able to estimate the economic impacts once the scope of the damages are known. Do we know the potential for short-term vs long-term effects at this point? Estimating economic impacts/damages will require a coordinated effort between those measuring and cataloguing the physical damages and then those who are well-versed in understanding the market and non-market values associated with damage assessment (the economics folks).

I realize this is a frustrating response and likely to make people wonder what economists really do, but the need for patience in determining the full value lost is important to ensure we capture all of the damages. For example, here's a story that came out yesterday highlighting the potential scope for damages:

The East Palestine train derailment has had a far reaching impact on businesses even outside of the village.

21 News reporter Leslie Huff spoke with one farm in Columbiana who says they have lost nearly half of their business, not because there is anything wrong with their product, but because of fear of the unknown.

For Lamppost Farm in Columbiana, their sales have dramatically fell by 50%.  Executive director Steve Montgomery says concerns about product contamination is the reason they are seeing a decrease. 

These types of damages are likely to persist and the full scope of health, social, economic, and environmental damages is likely to take months if not years to know with any sense of accuracy. Does this mean we should withhold aid until all damages are known?

 

No, of course not. Immediate aid should be focused on medical and income support for those directly impacted and documented. Once the urgency of the event subsides, the the real works begins to understand and estimate the full scope of damages and the associated loss of economic value.

 

Until then, the focus should be on taking care of the people and limiting future damages.

 

Now we just wait for the comment from Desvousges, Mathews and Train

Yesterday I spent 2 hours revising a final version of this paper so that we can say it is "forthcoming" in Applied Economic Perspectives & Policy. Here is Figure 1:

For the first time all semester I had a whole day to do some research

Here is what I accomplished (about 2 weeks ago)... six dichotomous choice contingent valuation logit curves with the probably of a yes (willing to pay the bid) on the vertical axis and the bid amount on the horizontal axis. Three of the data sets have a "fat/flat tails" problem and three don't. 

Head's up, you can't do much in the way of hypothesis testing if your data suffers from fat/flat tails because your standard errors are too wide (and asymmetric). If you estimate WTP with the nonparametric Turnbull you'll get tighter and symmetric confidence intervals but you're cheating, especially if your data is smoothed to hide those non-monotonicities. More on this in about 2 years ...