I'm buying a house! A very exciting thing for me. In the process of finding and buying this house, I've come to some interesting game-theoretical realizations.

Observe the three-step process:

  1. We find a house that we like.
  2. We contact the agency to organize things between us, the bank, and the homeowners.
  3. We do everything on our own, with the help of the homeowners.

Putanumonit Jacob has some great insights into how frequently you'd expect to see people not behave according to their incentives (trolls), and it's something like 4% of cases. I guess that's true in my experience. My added data point is that people follow their incentives but not reliably - or at least not reliably enough for me. The classification of people into incentive-followers (non-trolls) and incentive-ignorers (trolls) can be useful, but this experience acquainted me with mixed trolls. Sometimes they'd follow incentives, on other days, not so much. That's because they're playing other games as well, so they have to balance out the payout for my game with their other payouts.

That is to say, I cannot reliably count on the agent to do the things I need done (on time), even though him doing these things means that he gets paid 1% of the selling price. He also gets 1% on like twenty other transactions, which are maybe much more lucrative. Not all one percents are equal.

The unexpected side-effect of this story is making friends in unexpected places. The previous homeowners and us, we share a bond of interest in the house and a common dislike for the agent and the bank. It's circumstance that forces us to go through these proxies, instead of trading directly.

And the expected lesson is that when you have people do the work for you, you still have to do work. It's an instance of the principal-agent problem. No amount of money will allow you to outsource things perfectly. So it makes sense to try to do things on your own, learn the skills in the process, and pay for repairs if you mess things up too badly.

New to LessWrong?

New Comment
4 comments, sorted by Click to highlight new comments since: Today at 4:06 PM

Interesting. Care to elaborate? Why don't more people do this? 

Most agents and bank employees are competent, dedicated, professionals. And by most I mean 51%. And only half of sellers are competent, dedicated amatures. Likewise buyers. And that is correlated. So there are only ~1/25 transactions that involve a confluence of

  1. An on-the-ball personal buyer
  2. A slacking agent and bank
  3. An on-the-ball personal seller

Going outside of the world of house buying, I find that 50% of interactions I had with repair people were like that as well. So that definitely makes sense.

Other explanations are that people are afraid of trying things out, don't have time, or insist on a reality that simply isn't there (that when you pay someone to do something, they should reliably do it).

Would you stand by your advice for legal work too?

I think it might not be too terrible! In my (thankfully limited) experience, the actual advice of legal professionals can be extremely helpful, but the 'outsourcing' overall seems far from perfect. Think basic project management skills, or a lack thereof. It's not possible to know if the legal services one is receiving (or not) are good or useful, or what their value is relative to the same services from someone else, without also – at least partially – thinking thru the relevant legal issues oneself.

And it feels like this is worse than necessary but cloaked in 'dangerous mysteries beyond the mortal ken' like in medicine. And yet one still can't 'morally', or even sometimes practically, outsource perfectly even to specific individual doctors. Always, it seems to good to reflect, e.g. 'Do I trust this person? Should I find someone else to help me instead?'.

My favorite 'outsourcing vendors' are those that are eager and happy to explain their craft, even to paying customers. It's an under-appreciated area of charity!