LOMBORG-ERRORS                       Home
                        List of chapters
Chapter 25:

Predicament or progress ?




The use of cost/benefit analysis and the calculation of the costs of saving a human life-year are subject to many methodological flaws. As a counterbalance to Lomborg´s uncritical accept of such calculations, one should consult the critique advanced by Lisa Heinzerling & Frank Ackerman, e.g. (2002): The humbugs of the anti-regulatory movement. Cornell Law Review 87(2): 648-670; and (2002): Pricing the priceless: Cost-benefit analysis of environmental protection. University of Pennsylvania Law Review 150: 1553-1584.

P. 328-329: (COMMENT)

" . . a steady stream of progressively more serious ecological catastrophes that will be repeatedly proffered to us." Comment: The quote is interrupted in the middle of a sentence. The cited text continues as follows: " . . proffered to us and will, sooner or later, arouse us to action and convince us to fight back." Lomborg purports that the quoted sentence is an example of the mood that doomsday is nigh. But the missing half of the sentence shows that this is not said by a defeatist pessimist, but by a politician who wants us to take action.

P. 328 top left: FLAW

"The pursuit of happiness and comfort is paramount". Flaw: Lomborg´s quote is slightly misleading. The full text reads: "Our industrial civilization makes us a similar promise: the pursuit of happiness and comfort is paramount, and the consumption of an endless stream of shiny new products is encouraged as the best way to succeed in that pursuit.". So, Al Gore does not postulate that modern people in general think like this. He postulates that industry (e.g. via advertisements) wants people to think like this. And who could deny that ?

P. 328 left: (COMMENT)

"We have forgotten our `direct experience with real life.´" Comment: Lomborg´s quote is slightly misleading. Actually, the quoted text reads: "In our frenzied destruction of the natural world and our apparent obsession with inauthentic substitutes for direct experience with real life, we are palying out a script passed on to us by our forebears."

P. 329 left: ERROR

". . we cannot see the prison walls that surround us." Error: When checking Lomborg´s reference, that is pages 230 ff. in Al Gore´s book, it is not possible to find any mention there of prison walls or any similar expression.


"Ironically, Al Gore believes that the way to escape this dysfunctionality is by means of `the harsh light of truth´": Comment: The actual text in Al Gore´s book reads: "But there is a way out. A pattern of dysfunctionality need not persist indefinitely, and the key to change is the harsh light of truth." In this quote, and other sentences close to it in Al Gore´s book, there is focus on what can be done to change the present situation. Al Gore tells us what are the steps towards the situation when we come to terms with the new story of what it means to be a steward of the earth. On this basis, it is hardly correct to call Al Gore a `cultural pessimist´. And there is no reason to derogate Al Gore by using the word "ironically". Al Gore does not try to escape the harsh light of the truth.

P. 338 top left and Table 8: FLAW
" If  we drink water which contains pesticides  . . we face the same death risk as if we . . cycle 15 km . . "
Flaw: Table 8 lists "actions which increase the risk of dying by 0.000001". From this, one would understand that cycling 10 miles would give a net increase in death risk. But this is not so. A study published by the British Medical Association in 1992 demonstrates that  the health benefits of cycling outweigh the risks. The gain of  `life years´ through improved fitness among regular cyclists - and thus their increased longevity - exceeds the loss of `life years´ in cycle fatalities by a factor of around 20 to 1.

P. 338 bottom right: FLAW OF OMISSION
". . have computed the cost evaluations and the number of life-years saved and rendered the figures comparable".
Flaw: Lomborg ought to have mentioned that future costs and life-years saved were all discounted to their present value at a rate of 5 %. This means that if e.g. a safety belt saves a life the day after it is installed, then this counts 3.4 times as much as if the ban of a toxic chemical spares a case of mortality from cancer after 25 years. In other words: immediate effects count more than delayed effects. This produces a systematic bias in favour of preventing immediate accidents over preventing diseases with a long time of latency.

P. 339 top left: ERROR

"It is important to point out that only interventions that have as the primary stated political goal to save human lives are included." Error: This is not true. For instance, the purpose of the EPA regulations according to the Clean Air Act is stated thus: the act requires EPA to regulate air pollutants whcih may reasonably be anticipated to result in an increase in mortality or an increase in serious irreversible, or incapacitating reversible, illness". As an example, regulations of formaldehyde are primarily meant to avoid painful and irritating, but nonfatal, skin conditions. Incidentally, it also prevents a few cases of nasal cancer; if the cost of formaldehyde regulation is held up only against the avoidance of fatal cancer cases, and not against the avoidance of irritated skin, the regulation will appear much more cost-inefficient than it is. Another example is that the phasing out of lead in fuel is motivated to a large extent by the wish to avoid reduced intelligence in affected children. Altogether, the public wants restrictions on toxins not just to avoid deaths, but to avoid dread, to avoid long lasting painful diseases, and to avoid harms to future generations. Restrictions concerning radioactive wastes are motivated not just by the avoidance of deaths, but also to avoid that certain areas become pemanently uninhabitable.

P. 339 bottom left and figure 171: FLAW

"Median cost per life-year saved for different sectors of society . . " Flaw: It ought to have been clear from Lomborg´s text that the interventions include both those that are fully implemented, those that are partially implemented, and those that exist only as proposals. So the prices shown do not necessarily indicate actual expenditures. For instance, as stated in Tengs (1997), the EPA´s proposal to control benzene emissions during waste operations at $19 million per life-year saved has been set in force, but benzene emission control at chemical manufacturing process vents at $530 million per year of life saved has not been implemented. Out of the 587 possible interventions listed by Tengs et al. (1995), many are very similar and just describe more or less extensive versions of the same type of regulation. A general ban on all uses of asbestos (which was only in force for five years, and never had time to become effective - see Heinzerling (1998): Yale Law Journal 107 (7): 1981-2070) has been artificially subdivided into the use of asbestos in 30 different products, some of which do not actually provide any substantial health hazard, whereby absurd costs of up to $ 1,400,000,000 per life-year saved are made to appear in the list. Some items in the list are proposals that an existing exposure standard should be lowered further, and so the cost is not that of having a restriction, but the projected cost of strengthening an already existing restriciton.

P. 339 top right and figure 172: ERROR

"Median cost per life-year saved for different government sectors . . " Error: As is evident from Lomborg´s source, the items in this figure include only proposed interventions, not interventions that have actually been implemented.

P. 339 right: FLAW

"Thus, the extremely high typical cost of $7.6 million for the EPA area is fairly representative . . " Flaw: This cost cannot be said to be fairly representative, when it mostly refers to interventions that have not been implemented and, in some cases, to interventions that have not even been proposed, but only discussed informally.

P. 339 bottom right: (COMMENT)

"It was not that the most efficient programs were fully implemented . . " Comment: The figures for percent implementation of the 185 programs are not reliable, see comments to p. 341 left.

P. 340, table 9: (REMARK)

Remark: Notice that reducing the lead content of gasoline, and controlling SO2 by desulphuration of fuel oil, actually pay off in economic terms, i.e. the net economic expenditure for these measures is below zero.

P. 341 left: ERROR

"The Harvard study did make such calculations and discovered that instead of saving 592,000 life years, 1,230,000 life-years could have been saved for the same money." Error: As is evident from the paper of Heinzerling & Ackerman in Cornell Law Review (cited above), the "Harvard study" is grossly manipulated. Out of the 185 interventions included in the paper by Tengs (1997), 90 were toxin control measures. The status of these 90 interventions has been studied by Heinzerling & Ackerman in detail, and they conclude that out of them, fully 79 were never implemented - in many cases exactly because they were too cost-inefficient. On the other hand, Tengs postulates that less than 20 were never implemented. So there exists a disagreement concerning at least 59 interventions. This disagreement arose because Tengs did not consider formal implementation, but instead had a number of anonymous "independent experts" to evaluate to what degree each intervention had been implemented, i.e. how many percent of the target population were actually covered by the intervention. When Tengs (in a letter) was asked how regulations that had not been adopted were nevertheless considered by her to be implemented to some degree, she just answered that "Toxin control interventions that were never promulgated (or even considered) by the EPA might neverthelss have some percent implementation, at least according to the experts we interviewed" and "Some degree of implementation can exist even in the presence of a `no-go´ decision." That is, many of these toxin regulations that were never promulgated, do nevertheless count as public expenditures in Tengs´ accounts. For instance, two extremely cost-inefficient interventions where the price per life-year is nearly $ 100 billion - i.e. at least a factor of c. 1,000 larger than all others of the 185 interventions - were considered by Tengs to have been implemented to a degree of approximately 50 %, although these interventions have never been promulgated. Then, by postulating that the degree of implementation for these two interventions could be reduced to 0 %, she obtains a large but completely fictive amount of money which allegedly might be saved. The same principle applies to many other cost-inefficent measures that allegedly have been implemented to some extent. On the other side of the balance, Tengs has only been able to document one extra cost-efficient measure that could be implemented. Altogether, the postulate that 600,000 more life-years could be saved by wiser allocation of the resources is a lie. A wiser allocation could be made, but the effect would be very much smaller than indicated. Most of the alleged 600,000 life-years are simply fictive.

P 341 bottom left: WRONG STATEMENT

"Even so, the results are so unambiguous . . " Error: As will be understood from what is said above, the results are extremely ambiguous and so flawed that they may more properly be designated as fictive.

P. 341 - 342: ERROR

"We have also seen above that pesticides . . .other areas where the same resources could do much more good." Error: Actually, the Harvard study of 587 interventions includes only four interventions regarding pesticides. There is a ban of chlorobenzilate on noncitrus and on citrus fruits, and a ban of amitraz on apples and on pears. Out of these four proposals, two (one for each pesticide) save human lives for costs below zero. So, when Lomborg in his chapter 25 purports to discuss the rationale of restricting pesticide use on a life-saved-per-dollar basis, the limited evidence is that pesticide bans have a very high success rate in this respect.