Exploring temporal and spatial preferences for climate change adaptation

In recent years, the UK has faced severe weather that caused widespread and persistent flooding. This resulted in significant financial and environmental losses in a relatively short time-span, involving damages to infrastructure, real estate and coastlines. With a changing climate, such extreme weather is likely to increase, in terms of both duration and frequency. Adapting to the negative impacts from these climate change induced events is, therefore, a focal point in designing optimal climate change adaptation policy. A fundamental aspect in designing such policy is the need for an appreciation of societal views and perceptions of the risks associated with flooding, as well as a clear understanding of their preferences for strategies to mitigate the impacts of such risks. It is important for policy-makers who design climate change adaptation strategies to understand how the public contemplate strategies where the costs are incurred in the short-term but where the benefits may not be realised until the long-term. Indeed, the rate at which future benefits (and costs) are discounted is the key question when talking about climate change policy. A steady stream of theoretical literature has been produced to assess how uncertainty over future damages and future rates of economic growth affect the discount rate and thus cost benefit analysis. The large effect of different assumptions about discount rates is not surprising given that many of the effects of climate change are in the distant future. Although the general need for discounting is largely undisputed, debates about appropriate discounts have a long history. Indeed, much of the criticism of the well-known Stern Review on the economics of climate change has focused on the discount rate. To tackle this issue, and to establish the appropriate social discount rate, we make it a central element of our stated choice experiment by proposing that some of the attributes have a defined temporal dimension. In particular, we describe the length of time until and for how long the benefits under each policy are realised and define the payment mechanism as the value that the respondent will have to pay per year until policy completion (i.e., the time when the climate change adaptation measures are operational). This leads to a net present value utility that incorporates the average respondent sensitivity to annualised cost and the influence of the adaptation measures after discounting. The spatial distribution of the consequences of climate change are uneven. This unevenness maybe reflected in the public’s preferences and support for policies aimed at combating the effects of climate change. Indeed, it is likely that respondents who experience extreme weather events may have a different set of preferences and willingness-to-pay values to those who reside in a relatively unaffected area. Their risk perceptions and impatience for adaptation measures are also likely to differ, which could induce them to prefer more conservative projects. Moreover, while the benefits of climate change policy are enjoyed nationally (and, in fact, globally), there may be some ancillary benefits which are enjoyed privately on a local or regional scale. Accordingly, individuals may be more inclined to contribute to public goods that are provided locally rather than nationally (or, indeed, globally).This further underlines the importance of the specific context in which individuals experience the environmental externality, i.e., the intensity and severity of flooding—individuals who are in close proximity of where the extreme event occurs may value certain climate adaptation options differently compared to individuals who are ‘located’ further away. That is, while a policy aimed at reducing flood risk is a public good, not every individual will benefit equally from implementing the climate adaptation strategy, such as creating flood barriers. Thus, an individual’s location matters in relation to the location and size of the negative externality. For this reason, there is strong interest in studying the relative weight the public assign to adaptation measures delivered in the respondent’s local area and those in the rest of the country. To address this, our stated choice experiment focuses on the case where a respondent is asked to make choices that affect both their own locality and the rest of the country. In other words, each policy option entails different outcomes, costs and completion times in their local area and the remainder of the country. This represents an innovative style of stated choice and, therefore, makes an important contribution to the environmental economics and stated preference literature. Its main advantage is that the discrete choice model can retrieve the weight that respondents allocate to policy outcomes in their own area relative to those in the rest of the country. Armed with this information, we are, at last, able to ascertain the degree to which climate change adaptation measures are viewed as being a pure public good or an impure public good. In this paper, we report the evidence from a stated preference dataset conducted in Scotland. Our findings confirm that the public are highly concerned about climate change induced flooding and are willing to pay towards policies that lead to reduced flooding. Using various functional forms, we reveal that there is significant heterogeneity in the discount rates people have when it comes to climate policy. In particular, we identify the presence of a present bias, whereby some respondents place more emphasis on more immediate adaptation measures. The weight that people allocate to policy outcomes in their own area relative to those in the rest of the country is also observed to be heterogeneous. We discuss both the policy and methodological implications of our results.