Calculations like those in Tables 2 and 3 are controversial. How can impacts like the loss of ecosystems, or an increased likelihood of premature death be expressed in monetary terms? The notion is not as exotic as it may seem. Monetization is frequently used in many branches of economics.
Since welfare cannot be measured directly, like temperature or weight, an indirect measure has to be used. The most common measures are based on the concepts of willingness to pay (WTP), and willingness to accept compensation (WTA). WTP measures the amount of income a person is willing to forego in exchange for an improved state of the world. For example, it measures what fraction of their income people would be willing to spend to preserve an endangered species or a unique ecosystem. Correspondingly, WTA estimates the compensation that would be required for people to accept its loss. In this way, WTP/WTA can serve as a monetary measure of people's preferences.
There are several techniques to obtain empirical estimates of people's WTP/WTA. In many cases the easiest way is to elicit estimates directly in surveys or interviews (the contingent valuation method). Alternatively, estimates can be deduced from observed differences in the price of market goods with different environmental characteristics, e.g. from real estate prices in noisy and quiet neighborhoods (the hedonic approach). In the same way, the wage differential between high-risk and low-risk jobs is often used to estimate people's WTP for safety. The money spent to visit a park gives an indication of people's valuation of the visited sight (travel cost method). Similarly, money spent on defensive measures (e.g. on noise reduction) can sometimes serve as an indication of people's WTP.
The empirical estimation of WTP and WTA is often rather tricky. The analysis of climate change, with its variety of different, and usually uncertain impacts, is particularly complex. WTP/WTA estimates are therefore not always available, and proxies, such as the return on input factors (e.g. on capital or land), and other indicators are frequently used to approximate the welfare impacts of climate change (see Table 1). Moreover, where estimates are available they indicate the willingness to pay of the current generation. However, climate change impacts will only occur gradually, with significant effects not expected for another thirty to fifty years. By that time the demographic and socio-economic structure will have changed, and the welfare impacts of global warming will have changed with them.
Available damage figures can therefore only indicate the likely order of magnitude of the true welfare costs. Even so, they provide useful insights in the relative importance of different climate change impacts on human welfare.