Long Term Bets

May 8th, 2019
bets
When I have a disagreement with someone over something where we'll probably eventually know the answer, I'll often offer to bet. Coming up with terms we both agree on can clarify the disagreement, and putting even small amounts of money on the line can make people (including me) think a lot harder about what will happen. Examples of bets I have open right now:

  • We'll have driverless cars by 2027, where I can take one from my dad's house in Medford to the Concord Scout House.

  • No state will have banned human drivers by 2037.

  • If states start being permitted to fully ban abortion, some state will still allow it in 2040.

  • Slaughterhouses will continue to be legal in the UK in 2050.

How should the mechanics of long-term bets work though? Will we both still be around? I've used three different systems:

If I trust the other person a lot relative to the amount of money involved and expect to still be in touch, we'll just agree to the bet and confirm over email. When the date comes around I'm confident we'll be able to resolve the bet. This is by far the easiest, and what I've done the most of.

If there's more money or less trust you can have an investment account that you both put money into, and the winner gets that money at the end. This doesn't solve the problem that they might cheat you or disappear, but it helps some. One downside is that it locks up the money for the period of the bet, though that's also part of how it helps reduce risk.

The last system I've used is Long Bets, which makes the most sense if you have very low trust. It's kind of like a donor advised fund: you both make donations to the people who run the site (the Long Now Foundation) and you specify a charity to get your money if you win, and then when the bet resolves that charity gets the initial money plus half of the investment returns. The other half of the returns go to the foundation, and part of what you're getting is someone to review your terms and resolve disagreements. You are giving up a lot, though: if you make a 50-50 bet for 30y and we assume 2% inflation and 4% real return then you can think of this as the foundation taking 41% as their fee ($165 on a minimum bet of $200 each). [1] In the limit, as the duration of the bet gets very large, the fee approaches 50% ($200).

This makes me think a lower-fee not-quite-as-robust intermediate might be helpful. A trusted and interested friend could take the money from both parties, invest it, and the full amount could go to the winner's charity. I'd probably be up for being this person if anyone has long-term bets they'd want me to facilitate.


[1] The $400 will be $2,297 nominal ($1,297 in today's dollars). Returns are $1,897 nominal so the foundation gets $948 nominal ($523 in today's dollars) while the winning charity gets $1,348 nominal ($744 in today's dollars). So 41% goes to the foundation and 59% to the winning charity.

Referenced in: Long Bets by Confidence Level

Comment via: facebook

Recent posts on blogs I like:

Starting With Chords

A lot of people play fiddle. Basically nobody starts by learning chords before learning melodies. But that's actually how I learned. I started with chords. One of the nice things about learning to play violin this way is that you can go busking even…

via Anna Wise's Blog Posts November 15, 2024

Stuffies

I have some stuffies and I just have a bunny. Bunny is a rabbit. Woof is a seal. My favorite stuffie is bun bun. I play with my stuffies. Sometimes I jump up with them and I roll them. I can just throw them in the air when I want to play bthululubp wi…

via Nora Wise's Blog Posts November 15, 2024

You Can Buy A Malaria Net

2024 election takes

via Thing of Things November 6, 2024

more     (via openring)