Volume CXXXIII, Number 11 November 30, 2001 f

Arsenic in drinking water: How much is too much?
A. MYRICK FREEMAN III
FACULTY CONTRIBUTOR

Many observers were surprised when President Bush's Environmental Protection Agency (EPA) recently announced that it would uphold the 80% reduction in the maximum allowed concentration of arsenic in public drinking water supplies first announced by the Clinton administration last January. The reason for the surprise was that EPA announced last March that it was blocking implementation of the Clinton standard pending reviews of the basic science and the analyses of the benefits and costs of the proposed standard.

Since EPA's own analysis showed that the quantifiable benefits of the proposed standard were outweighed by its costs, and because President Bush's appointees to environmental posts generally supported the use of benefit-cost analysis in environmental decision making, it was widely believed that Christine Todd Whitman, Bush's appointee as Administrator of EPA, would use this review as an opportunity to justify a less protective standard for arsenic in drinking water.

The Bush decision has not settled the controversy over the standard. Opponents of stricter controls announced that they would fight the decision to implement the Clinton standard. And environmental organizations such as the Natural Resources Defense Council announced that they would fight for an even stricter standard.

Economists would argue that if we are to make the most of our scarce resources, we should compare what we receive from environmental protection activities in the form of increased well-being with what we give up by taking resources from other uses. We could measure the values of what we gain (the benefits) and what we lose (the costs). We could undertake environmental protection only if the results are worth more, in terms of individuals' values, than what is given up by diverting resources from other uses.

The standard for arsenic has been at 50 parts per billion (ppb) since 1942. The Clinton proposal, now accepted by Ms. Whitman, is to lower the standard to 10 ppb. There is substantial evidence from exposed populations in some developing countries (rural Taiwan, Chile) that drinking water containing arsenic in the range of 500 ppb results in a significant increase in the risk of cancer. What is not known is whether there is a threshold concentration below which there is essentially no risk, or whether lower concentrations result in proportionately lower, but still positive risks of cancer.

Assuming a proportionately lower risk (no threshold), EPA estimated that the benefits of a 10 ppb standard would be between $140-200 million per year, while it would cost about $206 million per year to meet this standard. Opponents of reducing the standard from the present 50 ppb point to this excess of costs over benefits. They also say that the benefits could be zero, since it is likely that there is a threshold for arsenic at some level well above 10 ppb. Supporters of the proposed standard say that it is likely that there are other adverse health effects besides cancer and that it is worth the extra cost as a safety margin. Some also argue that it is immoral to make policy decisions on the basis of benefit-cost analysis when people's lives are at stake.

The most vocal opponents of the 10 ppb standard are the elected and appointed officials in those cities where drinking water presently contains more than 10 ppb of arsenic and whose citizens would have to bear the costs of meeting the new standard. Most of the arsenic in public water supplies is of natural origins. There is no set of "black hat" polluters who arguably should be made to bear the cost of cleaning up the arsenic. Rather those who are currently bearing the risks of arsenic will wind up bearing the costs of removing the arsenic in the form of higher taxes and/or higher water bills. This opposition to the standard is consistent with EPA's finding that the benefits (which are the willingnesses to pay for reduced arsenic by the affected people) are less than the costs.

In conclusion:

1. Economists would argue that if the benefits of the 10 ppb standard are clearly less than the costs, it is bad public policy to force the standard on an unwilling population.

2. However, the excess of cost over benefit is relatively small; and there are enough uncertainties in the estimates of both benefits and costs that the excess cost might be worth it in terms of purchasing a safety margin.

3. Because of these uncertainties, benefit-cost analysis can rarely provide an unambiguous answer about a difficult policy choice. Therefore choices have to be made by politically responsible officials. Given this, the 10 ppb standard is arguably a reasonable judgment in the face of uncertainty. And it is a judgment arrived at by both a Democratic and Republican administration.

4. But there is another way to look at this issue. The cost of meeting any standard for arsenic is high because all of the water going through the system must be treated to remove the arsenic. However, only a small percentage of this water is used in a way that poses any health risk to people. Most of it goes down the toilet or shower drain, is sprinkled on the lawn or is used to wash the car, etc.

The costs of protecting people from the risks of drinking water containing arsenic would be much lower if water suppliers simply gave their customers bottled water for drinking. The benefits of this option would likely exceed the costs by a substantial margin. But this alternative could not be considered by EPA because of specific language in the Safe Drinking Water Act. Rethinking our approaches to improving environmental health can show less costly and more beneficial ways of promoting desirable environmental goals.