ARSENIC IN DRINKING WATER - WERE BUSH & WHITMAN RIGHT? -------------Warren D. Smith Sept 2001---------------- The EPA in the final days of Clinton's presidency directed the US legal limit on Arsenic in drinking water should be lowered from 50 ppb (by weight, i.e. 50 ug/L; this legal standard had been codified in 1975 based on a 1942 recommendation) to 10 ppb. Then the Bush administration's EPA, headed by Christine Todd Whitman, reversed that decision. The final move in his game remains to be played since Whitman says the reversal was "preliminary" and there will be a later "final" ruling. [Later note: after "further study" Bush & Whitman reversed themselves and went with the Clinton ruling.] The World Health Org. (WHO, part of UN) meanwhile had recommended a limit of 200 ppb in 1958, reduced to 50 ppb in 1963, and to 10 ppb (as a "provisional guideline") in 1993. What is the right level? There are several possible answers. First, the LEGAL ANSWER. The "Safe Water Drinking Act" (SFDA) as amended 1996 directs for EPA to establish enforceable limits of contaminant concentrations at which no adverse health effects in humans are expected to occur. It has been shown that adverse health effects can occur at 10 ppb. There is evan a new National Academy of Sciences report on this, that came out slightly after the Bush/Whitman ruling, saying so. Furthermore there is no question 10 ppb concentrations are detectable, hence this law is "enforceable." Hence, legally speaking, there is no question Clinton was right. Bush has sworn to obey the law and thus broke that oath here. But more important is the non-legal answer, which is: what actually makes sense? And that answer involves a cost-benefit tradeoff. I daresay a level even of a few Arsenic ATOMS per liter would still be enough to cause cancer (with some extremely low probability, all those atoms would end up in a single one of your cells, would cause a mutation, and lead to a cancer). But the cost of treating the water to force As levels that low, would be enormous. It would far, far exceed the cost of treating that few cancers. (This cost could well be measured in lives wasted as well as money, since "time is money." Also: since the money spent to save lives by reducing As levels to ridiculously low limits, could instead have been spent to save lives in some more cost-effective manner.) Similarly: a speed limit of 10 MPH instead of 55 or 65 would undoubtably save lives, but it is not worth the cost to society. Implementing that low a speed limit would destroy our economy immediately and consequently far more lives would be lost than would be saved. Moral: We have to strive for the RIGHT level. OK, let's try to do the calculation. The EPA has estimated that reducing the As limit from 50 ppb will cause... new limit US Pop affected Estim. Annualized #cancers prevented on (who currently cost of water per year (D=death As/H2O drink higher As) treatment ($M/yr) B=bladder L=lung) --------- --------------- ----------------- -------------- 3 ppb 35.7 Million 645-756 25B(=7BD) 50-125L 5 ppb 22.5 379-445 20B(=5BD) 40-100L 10 ppb 10.7 166-195 13B(=3BD) 26-75L 20 ppb 4.4 65-77 7B(=2BD) 14-35L How much would you be willing to pay to avoid getting bladder or lung cancer? Probably a lot. In fact, to avoid death from bladder cancer, you'd probably be willing to pay everything you can, which, shall we say, is 1/2 of your average annual salary, extracted for your whole life. Let us say, to make a very liberal estimate, on average, that means $20K/year per person. OK, so the total amount the <=150 people who are doomed to get cancer (if As legal limits are 50 ppb instead of 3 ppb) are willing to pay to avoid it, is <= $20K * 150 /year = $3 Million/year and, say, they are willing to pay that much for 75 years each to avoid cancer in only 1 year (!), so then <= $20K * 150 * 75/year = $225 Million/year Note this estimate is an upper bound; with lower (and more realistic) estimates on death rates and per capita salaries (the per capita US salary in 2001 really is only about $29K not $40K and apparently <1/3 of the cancers lead to death) this number might be only $50M/year. Meanwhile, the water treatment cost, even merely for reducing As levels to 20 ppb, in the US is at least $65M/yr. (I have checked the EPA cost estimates independently and I agree they are of roughly the right magnitude at today's cost levels. I have not checked the EPA cancer rate estimates independently.) So: based on this data, it seems that the right value for the As limit should indeed be set somewhere between 10-50 ppb, i.e. somewhere between the Clinton and pre-Clinton values, and I personally would vote for 20 ppb as a good tradeoff somewhat biased toward the side of safety.