Family, Friends and Framing : A Cross-Country Study of Subjective Survival Expectations

We implement a two-country online choice experiment on around 2000 respondents per country, first, to investigate the formation of subjective survival expectations; and second, to evaluate the relation between subjective survival expectations and attitudes to pension regulations. Each respondent provided subjective survival probabilities to a range of target ages in either a "live to" or "die by" framing, then subsets of respondents were sorted to conditions where information about current cohort survival, personalized information on same sex parent or grandparent survival, or both, were provided and the respondent updated their subjective survival expectations. Finally, respondents were asked their views on changes to pension eligibility ages. We find that giving people information about the longevity of peers does not induce them to revise their subjective survival expectations much, and neither does information about the relative longevity of their same-sex parent or grandparent. "Live to" and "die by" framing has a much larger effect on reported subjective expectations than personalized information about cohort or family longevity. However subjective expectations are relevant to explaining opinions on retirement policy and planning. Regulators trying to educate the public about longevity and the consequent need for delayed retirement or pension ages need alternative strategies