Most people who have studied behavioural science of various stripes will be familiar with the difference between i) asking someone to choose between 100 euro now versus 110 euro tomorrow and ii) asking them to choose 100 euro in a year versus 110 euro in a year and a day. Many very bright people have contributed to understanding why more people choose to wait for the 110 when the choice is in the future. Why do you think? One of the greatest attempts came from George Ainslie who was studying pigeons in behaviourist paradigms partly so he could help people deal with alcoholism. His idea of hyperbolic discounting was adapted by economist David Laibson who used it to develop a lot of modern ideas about how people consume and make pension decisions etc,.. In general, the reasons why people make decisions in these formats are up for discussion and a lot of recent work has examined the role of things like cash & liquidity constraints, trust in institutions, etc., alongside traditional preference explanations. Also a lot of work linking to psychological traits and several recent papers on whether time discounting more generally is a global phenomenon (e.g. here https://lnkd.in/e-geU9AM and here https://lnkd.in/eT6cga6b). Again, it will be a frustrating literature if you are looking for very unambiguous answers.
A recent meta-analysis (currently available as a working paper pre peer-review) by Cheung, Tymula & Wang on estimates of hyperbolic discounting. So-called beta-delta preferences imply that if beta is less than one and delta is less one then people both discount the future but also discount the future more in early periods e.g. money tomorrow is worth less than money today to a greater extent than money in 31 days is worth less than 30 days. https://lnkd.in/e2GeeJ3q Another recent meta-analysis also finds strong evidence of present-bias https://lnkd.in/eq3hAmSb Both reviews find little to modest evidence of publication bias and selective reporting driving the findings. There are differences between the papers and in the literature particularly on estimates of delta and clearly a lot to sort out about what generates differences in present bias across people. Again, I think it is fair to say that these are mature literatures theoretically and empirically. They will evolve over time but the main findings are emerging from many teams and studies and a vigorous and interesting empirically-grounded theoretical discussion with a lot of practical implications.Humans are awkward & complex but it is absolutely fascinating and very rich as an area of study and at least partly underpins many real applications such as pension autoenrolment/autoescalation. Great to discuss it every year with students here.
No comments:
Post a Comment