GDP and endangered species

Balmford et al recently suggested in Science that we should construct an index to measure the state of biodiversity conservation. Balmford et al would like the index to achieve a public significance comparable to GDP. Science has just published two letters commenting on this idea and a response by Balmford's team. Jurgen Brauer argues that GDP “is not rigorous and it is easily misunderstood.” He explains:

First, people become ill on account of pollution and have to seek medical treatment; more medical services are produced and counted in GDP at their market value. GDP rises. Economies grow. But we are not better off for having been polluted in the first place. Second, the more wars we fight, the more funds governments expend in the arms market, but we cannot argue that states are better off for fighting wars. Conversely, if we become healthier and fight fewer wars, GDP falls and economies shrink.

My understanding of GDP is that it grows only if people work more productively, or people work more hours, or a larger number of people get to work (or people leave illegal jobs for legal ones, or replace unpaid tasks with a paid job, etc). Do “people who become ill on account of pollution” thereby become more productive? Or does someone's illness compel family and friends to work more and better? If not, sick people simply change their spending pattern -- they buy more medical services but less of other things. There is no net effect on GDP (unless they do become less productive as a result of their illness, in which case GDP decreases).

In the second letter Brian Czech et al show that U.S. GDP has increased in almost perfect correlation with the number of species officially listed as threatened and endangered by the U.S. administration from 1973 to 2001. The authors conclude that inverse GDP is a good indicator of biodiversity. Although this letter looks like a half-joke, Andrew P. Dobson, Andrew Balmford et al very seriously respond that it would still be useful to have measures of biodiversity independent of GDP precisely to compare ecological and economic performance.

Now I quote Possingham et al [(2002) Limits to the use of threatened species lists, Trends in Ecology & Evolution 17:503-507]:

Most changes in the composition of threatened species represent changes in knowledge. When the causes of status change are not apparent, changes in knowledge and taxonomy mask true changes in conservation status. For example Master et al. reported increases in the number of listed species in the USA from ~200 in 1974 to ~1200 in 1998. The primary cause of the increase was the addition of numerous plants, reflecting backlogs in processing information through formal channels. […] We do not generally recommend using threatened species lists as they are currently constructed for indicating changes in the state of the environment, except where comprehensive data are maintained on well-studies groups (e.g. birds and mammals) allowing robust comparisons over time and space.