Participants completed an online survey about their preferences over ways of reducing their risks of dying over time such that they obtained gains in life expectancy. The dataset includes the options they faced and their choices. It also includes some demographic information and other related preference questions (e.g. time preferences, risk preferences, sequence preferences).A key role of the UK government is to address causes of premature fatality. In the UK, air pollution leads to the loss of 340,000 years of life each year and workplace cancers led to the loss of over 140,000 years of life in 2010. Government policies can address the many causes of premature fatality, but these policies need to be evaluated to ensure they make the best use of public money. The question then becomes: what is the value of increasing a person's life expectancy? To address this question, researchers have introduced the concept of the Value Of a Life Year (VOLY). This VOLY is used in government policy evaluations as a measure of the benefits of policies including air pollution mitigation and workplace safety regulation, and thus it is crucial it is measured accurately. The VOLY is estimated using surveys of members of the public, in which people state how much they would pay for a given reduction in their risk of dying, or for a given increase in their life expectancy. The benefits being valued occur in the future. Crucially then, a key component of the VOLY is the effect of timing. Put simply, the further in the future something is, the less we tend to care about it. So a reduction in our risk of dying this year might be more valuable than a reduction in our risk of dying in the future, even if the effect on our overall life expectancy is the same. Unless we understand the influence of this 'discounting' for changes in life expectancy, we cannot accurately disentangle it from the true VOLY. This is the problem we aim to solve with our research. To solve it, our team of experimental economists will use an innovative mixture of experiments and surveys. Participants will play experimental games designed to include simplified models of the air pollution policies, so our team can learn the best ways to describe and measure discounting as it relates to delayed changes in risk. The survey will use the insights from the experiment and elicit individuals' preferences for reductions in their risks at different points in the future. Taken together, the experiments and survey will provide the first major investigation into how people discount their future life expectancy in the context of the VOLY. Our results will be important for policymakers in two ways. First, unless we can account for the effects of discounting on the VOLY, then policy estimates of the VOLY taken from current surveys might be wrong. If these incorrect estimates are used in the evaluation of policies aimed at improving life expectancy, then the value of the policies will be over- or under-estimated, which means public money is likely to be spent on the wrong policies. Second, when the government is evaluating policies where improvements in life expectancy happen in the future, as is the case for air pollution policies, they have to apply discounting to the value of the benefits. Our research will provide evidence about how governments should discount future gains in life expectancy, to make sure that public preferences are reflected in policymaking. Our research is also academically cutting-edge. It combines models from economics with insights from psychology to generate new methodological and empirical evidence about how discounting influences preferences for changes in risk, both for money outcomes (in the experiments) and for fatality risks (in the surveys). It also forges a new methodological agenda, which is the incorporation of incentivised experiments into policy-driven research projects. Overall, our research aims to provide the basis for changing the VOLY used in government policy, challenge existing guidance for discounting fatality risk reductions, and ultimately change how government money is spent, so that the policies implemented are those that improve the wellbeing of society.

Survey programmed by the research team in o-tree and conducted online using a sample of respondents recruited on prolific.ac. The sample sex and age band distribution was selected to match those of the UK population (although the respondents were not restricted to be UK residents).