Sensitivity of life evaluations
As noted earlier in this chapter, life evaluation data are sensitive to major life events and show strong associations with a variety of other well-being outcomes. However, there are a priori grounds not to expect large movements in life evaluations as a result of relatively small-scale policy initiatives or non-market factors. If the life satisfaction valuation technique is used to assess drivers that operate on a much smaller scale, this can risk under-valuing non-market factors that nevertheless people do regard as important. Fujiwara and Campbell (2011) also note that life evaluations may not be sensitive to the non-use value of items such as cultural monuments - so again, scale is important, and it may be more realistic to look at the collective impact of these goods.
One subjective well-being indicator that might be more sensitive to immediate surroundings and activities - for example, small changes in environmental quality, or the availability of green space - is affect. Kahneman and Sugden (2005) propose an approach to valuation based on “experienced utility”. This is estimated from short-term affect data collected through the day reconstruction method, which includes information about both activities and locations, as well as the affective states accompanying those activities. The effects of different activities or locations on positive and negative affect can then be reconstructed. Kahneman and Sugden do not propose a specific method for linking experienced utility and money - and further work is needed to address this - but they note that the life satisfaction valuation approach could be adapted for affect-based valuations.
One further limitation is that the valuation technique based on subjective well-being is retrospective, i.e. it cannot be used to project the potential impact of something that does not yet exist - in contrast to the hypothetical scenarios on which stated preferences are based. As policy-makers using cost-benefit analysis are frequently interested in assessing the potential effects of a policy that has not yet been put in place, analyses will often need to draw on examples of policy initiatives in other communities - where the generalisability of results to the population of interest may come with caveats.45