Policy Beats Morality

post by Davidmanheim · 2018-10-17T06:39:40.398Z · score: 15 (15 votes) · LW · GW · 13 comments

(Crossposted from Medium)

This is a simple point, but one that gets overlooked, so I think it deserves a clear statement. Morality is less effective than incentives at changing behavior, and most of the time, policy is the way incentives get changed.

Telling people the right thing to do doesn’t work. Even if they believe you, or understand what you are saying, most people will not change their behavior simply because it’s the right thing to do. What works better is changing the incentives. If this is done right, people who won’t do the right thing on their own often support the change, and their behavior will follow.

I remember reading a story that I think was about Martin Gardner’s column (Edit:it was Douglas Hofstadter's - thanks gjm [LW · GW]!) in Scientific American in which he asked eminent scientists to write in whether they would cooperate with someone described as being “as intelligent as themselves” in a one-shot prisoner’s dilemma. He was disappointed to find that even many of the smartest people in the world were rational, instead of superrational. Despite his assertion that intelligent enough people should agree that superrationality leads to better outcomes for everyone, those people followed their incentives, and everyone defected. Perhaps we can chalk this up to their lack of awareness of newer variants of decision theory, but the simpler explanation is that morality is a weak tool, and people know it. The beneficial nature of the “morality” of non-defection wasn’t enough to convince participants that anyone would go along.

Environmentalists spent decades attempting “moral suasion” as a way to get people to recycle. It didn’t work. What worked was curb-side pickup of recycling that made money for municipalities, paired with fines for putting recyclables in the regular garbage. Unsurprisingly, incentives matter. This is well understood, but often ignored. When people are told the way to curb pollution is to eat less meat or drive less, they don’t listen. The reason their behavior doesn’t change isn’t because it’s “really” the fault of companies, it’s because morality doesn’t change behavior much — but policy will.

The reason politics is even related to policy is because politicians like being able to actually change public behavior. The effectiveness of policy in changing behavior is the secondary reason why — after donations by Intuit and H&R Block — congress will never simplify the tax code. To paraphrase / disagree with Scott Alexander, “Society Is Fixed, Policy Is Mutable.” Public policy can change the incentives in a way that makes otherwise impossible improvements turn into defaults. Punishment mechanisms are (at least sometimes) sufficient to induce cooperation among free-riders.

Policy doesn’t change culture directly, but it certainly changes behaviors and outcomes. So I’ll say it again: policy beats morality.

*) Yes, technological change and innovation can ALSO drive changes in incentives, but predicting the direction of such changes is really hard. This is why I’m skeptical that innovation alone is a good target for changing systems. Even when technology lowers the cost of recycling, it’s rarely clear beforehand whether new technology will in fact manage to prompt such changes — electric trolleys were a better technology than early cars, but they lost. Electric cars are still rare. Nuclear power is the lowest carbon alternative, but it’s been regulated into inefficiency.

13 comments

Comments sorted by top scores.

comment by romeostevensit · 2018-10-18T01:42:06.234Z · score: 18 (4 votes) · LW · GW

Related: don't use the few moments of agency you have to do object level things. Use them to alter your environment such that the object level thing gets done even by low agency you.

KonMarie is a good example of this.

comment by Joachim Bartosik (joachim-bartosik) · 2018-10-18T18:15:33.892Z · score: 1 (1 votes) · LW · GW

A link/ googleable phrase for KonMarie, phrase?

comment by Davidmanheim · 2018-10-19T11:13:41.005Z · score: 1 (1 votes) · LW · GW

Google "Konmari"

comment by Said Achmiz (SaidAchmiz) · 2018-10-21T22:47:04.674Z · score: 7 (5 votes) · LW · GW

I agree with Dagon [LW · GW] that recycling is a great example—of the opposite of your point.

Consumer/household recycling is actually bad [LW · GW]. It’s a great example of imposing a well-meaning measure, via regulation, that turns out to not only do no good but actually to do harm.

What can we learn from this? Perhaps that the reason “moral suasion” wasn’t convincing, in this case, is because it was a lie.

You say:

Telling people the right thing to do doesn’t work. Even if they believe you, or understand what you are saying, most people will not change their behavior simply because it’s the right thing to do.

The recycling example does not show this. What it seems to show is that if you lie to people, and tell them (stridently and accusingly) that the right thing is this-and-such, but those people can see perfectly well that the right thing is not what you say, then they may tell you that they believe you—mostly to get you to go away—but they will keep doing what they think is right… which is, of course, not the thing you told them.

So it’s not that “most people will not change their behavior simply because it’s the right thing to do”. It’s just that most people will not change their behavior simply because you tell them that it’s the right thing to do—especially when what you’re saying is not true, and they know it.

But, of course, if you force them to go along with what you’re saying, even though they are not the least bit convinced—well, people do follow incentives.

comment by Davidmanheim · 2018-10-31T10:46:18.375Z · score: 3 (2 votes) · LW · GW

I'm confused by your claim that it doesn't help, and would like some evidence other than Tierney's motivated reasoning, who claims that recycling is bad because it's annoying, that the impact is small, and that we try to recycle more than is economically optimal. That last one is probably true, and the first is subjective, and most people don't seem to mind.

But he says that excluding most recycling, the impacts are minor. Yes, excluding glass, plastic, and cardboard. Really: " Once you exclude paper products and metals..." https://www.nytimes.com/2015/10/04/opinion/sunday/the-reign-of-recycling.html

The claim that resource costs are low, so it's a bad idea ignores the fact that they are low because they mostly come from countries with no pollution restrictions, and almost all don't have GHG priced in.

In contrast, it usually costs nothing more than solid waste disposal, and when it is more expensive it isn't by a large margin - https://www.jstor.org/stable/3110116 and it reduces GHG emissions a tremendous amount - https://www.tandfonline.com/doi/abs/10.1080/10473289.2002.10470843

comment by Stuart_Armstrong · 2018-10-17T20:42:03.865Z · score: 4 (3 votes) · LW · GW

politicians like being able to actually change public behavior

But the ways in which they want to/can change them is strongly influenced by moral preferences among voters, donor, and civil servants. Why did they shift recycling or bring in clean air/water acts, rather than bringing in any of a million other policy changes they could have?

comment by Davidmanheim · 2018-10-19T11:02:33.769Z · score: 3 (2 votes) · LW · GW

I agree that there is an interplay between morality and policy, but it's not simple. My point is that when the two conflict, I think it's clear that policy influences the actions of the public much more directly than moral arguments.

Moral values are heterogeneous in the US, and depending on where you look, you could say morals among the legislature influences policy, sometimes moral concerns among the populace coerces politicians, and sometimes morals evolve based on changes elsewhere - such as the ongoing dissolution of privacy.

The implication is twofold; first, it's worth paying attention to policy, and second, policy change is (usually) a better instrumental goal for implementing change than convincing people of moral correctness. Both points are important for much of the EA movement, as well as for existential risk researchers looking to change the course of AI development.

comment by Dagon · 2018-10-17T18:42:54.866Z · score: 3 (4 votes) · LW · GW

I can't disagree that policy (aka threat of violence for non-conformers) changes things faster than reasoned thought.

I _DO_ disagree that it improves the world more than reason does. Recycling is a great example - the status quo had massive subsidies for undifferentiated garbage collection, so consumers could not see how much savings, if any, there would be from separating their recyclables. Moral suasion that directly contradicts daily experience (in paying my garbage bill) is going to have very low effect. So we passed laws, causing people to go to significant effort and _STILL_ see no benefit. Now we have a common pattern of separating and rinsing our recyclables, and a bad feeling that it all goes to the same place, and no evidence of actual savings or reuse.

So we've spent a bunch of energy in changing laws and in wearying consumers on the topic, but have not achieved the effects we might have if we'd just been transparent about the costs we are concerned about (by charging sufficiently for landfill and charging less for clean recyclables).

comment by Davidmanheim · 2018-10-19T11:09:22.576Z · score: 1 (1 votes) · LW · GW

I didn't make the second claim, that [policy] improves the world more than reason does. In fact, I don't think I discussed reason - I discussed moral suasion.

But for recycling, you're making a theoretical libertairan argument about the impact of imposed costs versus regulation. I'm generally sympathetic to using costs rather than regulation, but I'm pragmatic. As a factual point, this has been studied in various contexts, and sometimes imposing costs is more effective than regulation, and other times it is less so. For recycling, imposing small costs on consumers, such as bottle deposits and fees for trash collection, seems to have changed almost no behaviors, while fines for not separating recycling have, in my view, been incredibly potent. If there is research showing the reverse, I'd love to see it, and if other people have a different perception, I'd be happy to hear that as well.

comment by ChristianKl · 2018-10-21T21:42:44.014Z · score: 2 (1 votes) · LW · GW

When it comes to recycling, I understand the argument of why throwing batteries into the regular trash is bad.

On the other hand nobody made a convincing case to me that there's a high utility in other standard recycling behavior.

A large portion of the "recycled" garbage simply went for many years to China till China didn't want our trash anymore. Another portion gets recycled by burning much more energy than it would take to produce the products from scratch (and energy means more burned coal).

comment by Davidmanheim · 2018-10-31T10:54:05.889Z · score: 1 (1 votes) · LW · GW

I think you're reading some heavily motivated sources. See my reply above. There are significant environmental gains from recycling aluminum and cardboard, and some gains from glass, paper and some plastics.

And China does do a lot of recycling for US plastics - they weren't throwing it all away. Some of it was low quality and discarded, but at the very least numbers 1 and 2 (PETE, i.e. water bottles, and HDPE, like milk jugs,) are valuable. (But yes, 3-7 are too expensive to recycle, and get discarded with the garbage - feel free not to recycle them.)

comment by gjm · 2018-10-18T14:03:01.764Z · score: 2 (1 votes) · LW · GW

Pretty sure it was Hoftstadter (in his Metamagical Themas column; note that the title is an anagram of Gardner's Mathematical Games) rather than Gardner. You can find the relevant columns, and some further discussion, in his book of that title.

comment by Davidmanheim · 2018-10-19T11:10:22.357Z · score: 1 (1 votes) · LW · GW

Thanks! I've read the book, a decade ago, so I'm fairly sure that what I was mis-remembering.