Saturday, February 4, 2012

Discounting the Future Can Be a Rational Strategy

We discount the future when we place lower value on the same event or object at some future time than we place on it right now. In other words, a dollar today is worth more than a dollar tomorrow. How much less it's worth tomorrow is how much you discount the future.

Experiments about discounting the future typically take the form of "Would you rather get twenty dollars today, or twenty-five dollars a month from now?" If you take the second one, your discount rate is 25% or lower per month. All humans do this to some degree, but countries differ in how much their residents, on average, discount the future. The residents of developed countries discount the future less than the residents or poorer countries.

In its relationship to economic development, discounting the future is often thought of as a proxy indicator for ability to delay gratification or even general intelligence (and therefore accumulate capital) and this is probably useful. However, ability to delay gratification has a different value in different environments. If someone offers twenty dollars today or twenty-ve dollars a month from now, it makes more sense to go with the future money if you're in Japan than in Somalia. In Japan, things are predictable; when you go to collect your money, the train will come at the same time as the last time you went, the experimenters can be trusted to be there, there are institutions to enforce the contract if they don't honor it, and most importantly, no one will try to kill or rob you on the way and the Psychology Department will still be there, rather than having closed or been blown up. Also, you will not have been killed or lost all your money in the meantime. In Somalia (or places like it) these assumptions do not hold. In Somalia, you're stupid to make those assumptions; you can't know what's going to happen a month from now, or even tomorrow. Discounting the future heavily in unpredictable places is actually a rational strategy.

This is unfortunate. It stands to reason that everyone choosing a heavily discounting strategy in unpredictable places would reinforce the very unpredictability it's accounting for, creating a vicious circle.


Grego said...

I've always thought that my answer to that question would be to immediately take the $20. Because the idea of such a transaction is unorthodox to me, I will of course take the immediate benefit instead of a possible delayed, slightly greater benefit. It's that "possible" part. I don't view it as $20 vs $25, but as $20 vs ($25 or $0), so I'd rather take the sure bet. I don't live in Somalia, and I certainly hope that I'm a rational person.

Maybe my numbers are just different. If you asked me if I'd take $20 now or $1000 in a month, I'd almost certainly wait.

Overall, though, I suspect the way this problem is posed makes a big difference in people's answers.

Michael Caton said...

So you have discount rate that's between 25% and 4900% per month.

The *amounts* are even more important than the wording. With very small amounts you run into two problems: transaction costs (is your time and trouble worth more than the extra amount you'll get by coming back next month, or just remembering to check your bank account?) Also The utility we gain from money is partly relative to how much we already have - check out the St. Petersburg lottery paradox.

Alistar Reeves said...
This comment has been removed by a blog administrator.