Over the last few weeks, there have been a lot of rumblings from folks in the coal industry about the big West Virginia University study that concluded the adverse health effects of coal mining in Appalachian far outweigh the industry’s economic contribution to the region.
I give Roger Nicholson, vice president and general counsel of International Coal Group, credit for actually trying to respond in writing in a public forum, rather than grumbling in private.
But in his Sunday Gazette-Mail commentary, Nicholson hardly lays a glove on the study. And because he spends little time or effort in actually critiquing the methods of the study, his criticism doesn’t amount to much as far as really discussing the costs and benefits of coal — or what role coal could or should play in the region’s future.
Roger starts out by saying my initial coverage of the study ignored “some key facts” and held WVU researcher Michael Hendryx and his co-author, Melissa Ahern of Washington State University, to a “much lower standard of factual rigor than [the Gazette] would a pro-coal industry study.”
Well, it’s worth noting that this study appeared in a respected scientific journal, Public Health Reports — a journal that puts such articles through an accepted peer-review examination before they are published. Like the previous coal-related articles by Hendryx, this one went through the process that science recognizes for an examination of factual and methodological rigor. That doesn’t mean it is a perfect study. There is no such thing. But unlike the pro-coal studies that industry officials prefer the public to focus on — see examples here and here — the work Hendryx is doing is being reviewed and approved by other scientists as meeting accepted standards.
IÂ will say that I and many other journalists probably do far too many stories about “reports” and “studies” that were put together by various interest groups, rather than looking more to the peer-reviewed literature to help us educate the public. For example, there is much to be learned from reviewing the peer-reviewed scientific literature about mountaintop removal’s environmental impacts.
But this study, Mortality in Appalachian Coal Mining Regions: The Value of Statistical Life Lost, is not simply a report by one side or the other in the endless debate over coal. As the chart above shows, the study reports that mortality rates in coal producing counties are higher than those in other Appalachian counties and the nation as a whole.
Its findings are important, as I wrote in my initial story on the study:
Writing with co-author Melissa Ahern of Washington State University, Hendryx reports that the coal industry generates a little more than $8 billion a year in economic benefits for the Appalachian region.
But, Hendryx and Ahern put the value of premature deaths attributable to the mining industry across the Appalachian coalfields at — by one of their most conservative estimates — $42 billion.
“The human cost of the Appalachian coal mining economy outweighs its economic benefits,” they wrote.
Roger writes that the Hendryx-Ahern study “is deeply flawed and does not withstand the scrutiny that its authors would apply to their own students.” But he offers little evidence to support such a strong criticism.He writes that the authors are biased. To support that, he makes fun of the purpose of Ahern’s Northwest Climate Change Center in Spokane, Wash., but admits he knows little about that organization. And, to show that Hendryx is biased, he cites his feeling that another one of his studies was also “fatally flawed” (again offering little evidence of that).
Then, Roger belittles the Hendryx-Ahern study for including in its footnotes references to an Ohio Valley Environmental Coalition article by Vivian Stockman and a book about the coal industry by Jeff Goodell. He doesn’t point out that neither citation is a major point of the study and that neither really has anything to do with the methods or data used by Hendryx and Ahern. (Of course, Roger also doesn’t point out that Goodell is a very established and respected author and journalist, either).
In his entire commentary, Roger offers only one real criticism of the study itself, the data it uses, or the methods behind it. Here it is:
Finally, the report ignores taxes other than severance taxes. Unmined mineral taxes and other real and personal property taxes generate multimillions of dollars in public funds annually. These taxes, as well as severance taxes that benefit both coal-producing and noncoal-producing counties, are essential to the provision of governmental services and the development of infrastructure required to diversify the economy.Â
As best I’m able to tell, that criticism is on the mark. Hendryx and Ahern used as their starting point in figuring coal’s economic benefits a previous study done at the University of Kentucky. This U.K. study itself did not attempt to include such impacts, citing “the complexity of modeling these local tax rates and coverage in all of 118 major coal-producing counties.”
Hendryx and Ahern added to the economic impact numbers and made some important adjustments. But they did not add in figures for property taxes on unmined coal. Roger is right about that. UPDATED,3:45 P.M.TUESDAY — However, even if hundreds of millions of dollars a year in property taxes on unmined coal were added, it would still not bump the cost-benefit ratio in coal’s favor.
This is not the study’s only shortcoming. In fact, the authors themselves outlined some relevant problems with their work … right there in the study itself:
Despite the significant associations between coal-mining activity and both socio-economic disadvantage and premature mortality, it cannot be stated with certainty that coal-mining causes these problems. It is not possible to determine what the economic and public health outcomes would be in these areas in the absence of mining.
I included, by the way, a discussion of these and other problems in my initial story for our print edition and in my blog post on the study:
The study is far from a complete cost-benefit analysis of the coal industry, the authors report. But, the things it leaves out, they say, are mostly costs that they haven’t been able to completely account for yet.
“They do not consider reduced employment productivity resulting from medical illness, increased public expenditures for programs such as food stamps and Medicaid, reduced poverty values associated with mining activities, and the pros and the costs of natural resource destruction,” the study says.
“Natural resources such as forests and streams have substantial economic value when they are left intact, and mining is highly destructive of these resources,” the study says. “For example, Appalachian coal mining permanently buried 724 stream miles between 1985 and 2001 through mountaintop removal mining and subsequent valley fills, and will ultimately impact more than 1.4 million acres.
“Coal generates inexpensive electricity, but not as inexpensive as the price signals indicate because those prices do not include the costs to human health and productivity, and the costs of natural resource destruction.”
But Roger and others in the coal industry haven’t so far offered any constructive criticism aimed at improving the ability of researchers to do a true cost-benefit study of the coal-mining and coal-burning industries.
You think perhaps the industry folks would rather not have that kind of discussion? Maybe Hendryx and Ahern were on to something with this statement:
The reliance on coal mining in some areas of Appalachia constitutes a de facto economic policy: coal is mined because it is present and because there is a market for it. However, other economic policies could be developed if reliance on this resource was not in the best interest of the local population.
Â
Â
Â
Â
Â
Â
Subscribe to the Coal Tattoo
The Dr Hendryx study states the following: “Despite the significant associations between coal-mining activity and both socio-economic disadvantage and premature mortality, it cannot be stated with certainty that coal-mining causes these problems.â€
Given the above quote from Dr Hendryx’s study, what would allow anyone to conclude that coal mining in an area causes premature death?
The study also stated “The results suggest, but do not prove, that a coal mining dependent economy is the source of these continuing socioeconomic and health disparities.â€
These statements cause me to believe that perhaps some “mining†of the data “suggests†and maybe “indicates†a relationship but much more medical related research would be needed to be able to actually prove a relationship. And if this relationship was proven then much needed corrections could be addressed and engineered to correct the problem. As it stands now what exactly needs to be corrected? Perhaps saying that the costs exceed the benefits with coal is only wishful thinking for those eager for the demise to commence.
Not a perfect analogy by any means but it seems like there are a lot of premature deaths from auto accidents. At least most researchers and officials aren’t trying to ban cars but instead work to improve safety through engineering, improved laws and physical research.
Just where are we going to get reliable base-load electrical power?
Response to Mr Powell “W.Va. and climate: Toward a more honest discussion”:
I agree that nuclear is a option and it needs to be pursued but I never see that suggested on this blog. That’s really the only choices we have, coal and nuclear for base load power. Realistically if you are against coal then you have to be for nuclear.
My question remains “What are the causes of premature deaths in coal mining areas and what exactly is the link to mining?†This research would not be done by a social scientist. I can’t even find out which counties were used in the study in my review of this research.
It’s my understanding that peer review is more than getting published but involves independent researchers performing similar experiments to verify the hypothesis, so that it may become a scientific theory. Has this been done? This is required so that conclusions are not used to support a position that is not legitimately justified by the totality of the evidence.
Some of Dr Hendyrx’s comments from his on-line chat include:
“I think there are enormous technical and financial barriers to clean coal technology. It can work on a small scale but to make it widely implemented would be incredibly expensive, and maybe not workable. The money we are investing in this technology should be invested in renewal energy in my opinion.â€
“Also, air and water quality around mining should be more carefully monitored and controlled. And mountaintop mining should be eliminated.â€
“I think we need to consider a mix of solutions including ecosystem restoration, sustainable timber, small agriculture, development of renewable energy like hydro and wind, and investments in entrepreneurial ventures.â€
I guess if you agree then there’s no bias.
[...] för kolindustrin, Roger Nicholson, undersökningen och sättet som tidningen rapporterade om den. Enligt Ken Ward faller det mesta av kritiken platt, även om både han och forskarna är medvetna om studiens [...]
Casey, you comments reflect a lack of understanding of how science works.
(Casey) “It’s my understanding that peer review is more than getting published but involves independent researchers performing similar experiments to verify the hypothesis.”
First of all, there are two fundamental ways in which science is done. One is experimental, or hypothesis-driven science. The other is observational science, in which some aspect of the world is studied via observation (not experimentation). Observational science does not involve an hypothesis. The results simply report what was observed, and conclusions may be drawn from those results. The Hendryx/Ahern study is of the latter type; they had no hypothesis and did no experiments.
Second, science is not in the business of “proving” things; there is no standard of “proof” in science. I’m therefore a little surprised that the Hendryx study used the word, but so be it. What science does is to offer EVIDENCE. “Proof” is in the eye of the beholder. What “proves” something to you might not “prove” it to me, and vice versa.
(Casey) “Given the above quote from Dr Hendryx’s study, what would allow anyone to conclude that coal mining in an area causes premature death?”
Answer: the chart Ken includes above would allow it, because it shows, with statistical data, that mortality rates are higher in coal mining counties. That’s EVIDENCE that coal mining causes the problems, but it’s not CERTAINTY. Some people might even say it constitutes “proof.” Others might not.
Very well put Vickie. Thanks for that. I agree that Casey is confused by what the scientific process is and isn’t — but I appreciate him trying to read the Hendryx study and understand where it’s coming from.
I believe that the “Data” section of the study itself lists where they obtained data they used (including where they got the list of Appalachian counties). Then, the footnotes to that section include Web links to the data.
Ken.
Well, I guess I crossed the line and must modify my comments. It’s true I am pretty ignorant on the scientific method (although I have stayed at a Holiday Inn Express). I agree that observational science should NOT have a hypothesis to confirm. Again maybe this is due to my ignorance, but isn’t more research needed to verify these results to be fully peer reviewed and accepted? And if verified wouldn’t the most important thing be to actually discover the actual physical/emotional reason and address it? But anyway check out part of Dr Hendyrx’s project summary:
As stated in our original RRI grant application: “this research will test hypotheses stating that, after statistical adjustments for social and economic confounds, population hospitalizations and mortality rates for health conditions that are sensitive to exposure to coal mining byproducts will be elevated as a function of the amount of coal mined in Appalachian counties, while hospitalizations and mortality rates for other health conditions will not be elevated.â€
Hmmmm, hypotheses.
I think Dr Hendyrx believes that one of the biggest issues is the negative health effects from poor socio-economic conditions. The graph certainly shows increased mortality in Appalachia versus the rest of the nation. I am concerned that WV being ranked as the worse state to do business in is causing this increased mortality.
I agree with Vickie that proof is in the eye of the beholder and I am not sure that the study accounts for all factors to allow the conclusion that was suggested, but I’m no scientist. Furthermore it is unfair to not look at the whole cost/benefit picture of coal without including all the benefits including steel products and electricity which are substantially responsible for out standard of living and health.
Casey,
The peer review process doesn’t, as a matter of course, involve actual verification of a researcher’s results–i.e., RE-PERFORMING the research to see if the same results are obtained. Rather, typical questions reviewers ask would be things like “Were proper techniques followed?” “Are the conclusions the researcher(s) arrived at reasonable, given the data that were obtained?” Etc.
Now, as for having an hypothesis in observational research, there’s nothing PREVENTING a researcher from forming an hypothesis; but a) it’s not NECESSARY; b) at best, it’s an educated guess as to what your results will be; and c) in some cases, it would be pointless. Say I was going to do a census of frog species in a given area–an observational study. There’s nothing PREVENTING me from saying, “I hypothesize I’ll find 8% species A, 42% species B, etc.” But a) that’s not necessary, since I’m not actually TESTING an hypothesis and I’m going to report my results no matter what I found; and b) it would be pointless if I have no idea at the outset what the results might be–which is often the case in observational research.
On the other hand, if I wanted to know whether mercury causes cancer in frogs, the best way to answer that question would be through EXPERIMENTATION, and the hypothesis being TESTED via experiment is “Mercury causes cancer in frogs.”
Finally, as for “discovering the actual reason,” that would be a good topic for another, follow-up study.
Hope that helps.