Robin Hanson on Simple, Evidence Backed Models

post by Raemon · 2019-04-16T22:22:19.784Z · LW · GW · 11 comments

This is a link post for http://www.overcomingbias.com/2019/04/publish-tax-returns.html

This is link to a Hanson post that is primarily about tax returns, and whether they should be public. It was an interesting foray into the considerations that bear on that topic.

But I was particularly interested in the opening paragraph, which had a useful lens:

Our simplest model of an economy is: supply and demand. This model has many simple implications for policy. Now we know of many much more complicated economic models, which often have quite different policy implications. But often we are not sure which more complex models actually apply well to any given situation. So we have to worry that people favor more complex models mainly to justify their preferred policies. Knowing this pushes me toward recommending the policies implied by supply and demand, unless I see unusually clear evidence to support a different economic model. (FYI, the evidence that fixed costs exists seems plenty clear, so I really mean supply & demand with fixed costs.)
In our simplest modes of information, people are better off when they have more information, and also when information is distributed more symmetrically. There is a vast world of much more complicated models, but it is often hard to tell which more complex models apply well to given situations, and many probably favor particular models to justify preferred policies. So as with supply and demand, this uncertainty pushes me to favor the simplest info models, and their policy recommendations favoring more info and more symmetric info.

11 comments

Comments sorted by top scores.

comment by cousin_it · 2019-04-17T14:48:57.496Z · LW(p) · GW(p)

In our simplest modes of information, people are better off when they have more information

In what simple model of information are people never worse off when others have more information about them?

Replies from: Dagon, habryka4
↑ comment by Dagon · 2019-04-17T16:30:10.601Z · LW(p) · GW(p)

In the simple model that information is used only to make better predictions, more (correct) information is better. Models that add the complexity of adversarial motives generally do show that secrecy has value.

Replies from: RobinHanson
↑ comment by RobinHanson · 2019-04-17T18:34:07.982Z · LW(p) · GW(p)

Secrecy CAN have private value. But it isn't at all clear that we are typically together better off with secrets. There are some cases, to be sure, where that is true. But there are also so many cases where it is not.

Replies from: cousin_it, Dagon
↑ comment by cousin_it · 2019-04-17T22:51:52.313Z · LW(p) · GW(p)

It seems to me that removing privacy would mostly help religions, political movements and other movements that feed on conformity of their members. That doesn't seem like a small thing - I'm not sure what benefit could counterbalance that.

↑ comment by Dagon · 2019-04-17T20:28:39.554Z · LW(p) · GW(p)

Quite agree - depending on how you aggregate individual values and weigh the adversarial motives, it's quite possible that "we" are often worse off with secrets. It's not clear whether or when that's the case from the "simple model" argument, though.

And certainly there are cases where unilateral revelations while others retain privacy are harmful. Anytime you'd like to play poker where your cards are face-up and mine are known only to me, let me know.

I would love to explore whether private information is similar to other capital, where overall welfare can be improved by redistribution, but only under certain assumptions of growth, aggregation and individual benefits canceling out others' harms.

↑ comment by habryka (habryka4) · 2019-04-17T18:12:41.126Z · LW(p) · GW(p)

There is the classical example from Strategy of Conflict (ironically copied from your own book summary):

Imagine you and I have been separately parachuted into an unknown mountainous area. We both have maps and radios, and we know our own positions, but don't know each other's positions. The task is to rendezvous. Normally we'd coordinate by radio and pick a suitable meeting point, but this time you got lucky. So lucky in fact that I want to strangle you: upon landing you discovered that your radio is broken. It can transmit but not receive.
Two days of rock-climbing and stream-crossing later, tired and dirty, I arrive at the hill where you've been sitting all this time smugly enjoying your lack of information.
And after we split the prize and cash our checks I learn that you broke the radio on purpose.

(Edit: Nevermind, misread the above to say" In what simple models are people worse off when they have more information about others")

comment by Raemon · 2019-04-16T23:26:25.416Z · LW(p) · GW(p)

This feels related to a few different things that Benquo has written about, which I've been mulling over.

Replies from: ryan_b
↑ comment by ryan_b · 2019-04-17T14:31:52.360Z · LW(p) · GW(p)

I notice a distinct lack of examples of successful large-scale discourse. It feels like this is one of many areas where we pulled a sort of legerdemain where we envision the thing we wish were true as the natural state of the world or the state of the past, and then bemoan its sabotage or decline.

I currently cannot think of a single instance of good public discourse ever. This causes me to think it isn't sad that clever arguers are destroying good discourse; it is interesting that they are an obstacle to good discourse ever arising in the first place.

I'm happy to be wrong though. If we were to attack this problem by maximizing or duplicating the bright spots, what would they be?

Replies from: Raemon
↑ comment by Raemon · 2019-04-17T18:27:52.285Z · LW(p) · GW(p)

Yeah, this is closer to the way I look at the situation.

But one of the lenses into this area is the "Is Science Slowing Down? [LW · GW]" perspective, where it seems like in the early 20th century was a brief period where something about science was more productive, at least per unit of effort.

This doesn't feel like something I'd consider a natural-state, more like a bright spot to duplicate, and I don't know how much I expect it to be possible to scale.

comment by ryan_b · 2019-04-17T14:37:23.919Z · LW(p) · GW(p)

I agree strongly with the information symmetry argument, although I am less confident of the information volume one - this is because transaction costs apply to information too.

I particularly like it because information symmetry feels like an area which has gotten a negligible amount of attention compared to questions of supply and demand, and therefore is a good target for making a lot of progress relatively quickly for relatively little investment.

comment by Sherrinford · 2019-04-17T11:30:09.998Z · LW(p) · GW(p)

Note that "supply and demand" for economists means that the demand curve is derived from consumer optimization (price = marginal utility), that the supply curve is derived from firms' profit maximization (price = marginal cost), both assuming price-taking behavior, that it implicitly assumes that trade actually takes place where the curves intersect (my impression is that a large literature on adjustment processes has basically disappeared because the assumption that we only care about equilibria in this sense became the norm).

People who hear that this is simple may confuse it with the claims that consumers demand less when the price is lower and that firms offer less when the price is higher (and that trade actually takes place where the curves intersect), which are claims that can be backed by different underlying models. You can derive a demand curve either by assuming that everybody buys more when the price increases (the standard reasoning), or that everybody buys exactly one unit if the price is below her individual willingness to pay.

It seems that a model which assumes that firms are price-takers in this sense and supply at marginal cost is not simpler than a model which assumes that firms always have a mark-up of 20% on their marginal cost. Sure, the mark-up demands an explanation, but this way you don't need the profit-maximization arguments (which makes the model simpler).

Finally, often we are not sure whether a simpler or a certain more complex model actually applies well to any given situation. So we have to worry that people argue in favor of simpler models mainly to justify their preferred policies - because the policy-implications of certain simpler models are well-known.