Overcoming Bias

Future Self Paternalism

Via Tyler Cowen, we hear Robert Fogel did not trust his future selves:

When I graduated from college, I had two job offers.  One was from my father, to join him in the meat-packing business.  That would have been quite lucrative. The other was as an activist for a left-wing youth organization.  I chose the latter and worked as an activist from 1948 to 1956.  At the time I was making that decision, my father told me: "If you really believe in that cause, come work with me.  You will make a much higher wage and you could give your extra income to hire several people instead of just yourself."  I thought, well, that makes some sense.  But I was convinced that this was a way to get me to change my views or at least lessen my commitment to an ideological cause that I found very important.  Yes, the first year, I might give all of my extra money to the movement, but every year I would probably give less, and finally reach the point when I was giving nothing at all.  I feared I would be co-opted. I thought this was my father's way of indoctrinating me.

When I spent a few weeks at Oxford last summer, Toby Ord similarly said he wanted to commit his future selves to donating at least ten percent of income to third world charity; he did not trust his future selves to make that choice for themselves. 

These paternalism examples are striking, because paternalism is usually justified as a response to a combination of ignorance and irrationality, but Robert and Toby should expect their futures selves to be just as smart and rational, and even better informed than they.  How can they reasonably expect their future selves to be so much more biased that force is appropriate to constrain them?

Added: Tody Ord elaborates in the comments.

Bias on Self-Control Bias

Since Eliezer is concerned about our ability to resist modern temptations, let me summarize recent economic analysis:  Paternalism does not help people who are aware of their self-control problems, and are able to make future commitments.   To argue for paternalism regarding self-control, one has to assume we are biased to underestimate our self-control problems.  From a 2004 paper in Quarterly Journal of Economics:

We analyze the profit-maximizing contract design of firms if consumers have time-inconsistent preferences and are partially naive about it. We consider ... goods with immediate costs and delayed benefits (investment goods) such as health club attendance, and goods with immediate benefits and delayed costs (leisure goods) such as credit card-financed consumption. ... The predictions of the theory match the empirical contract design in the credit card, gambling, health club, life insurance, mail order, mobile phone, and vacation time-sharing industries. ... time inconsistency has adverse effects on consumer welfare only if consumers are naive.

From a 2002 paper by O'Donoghue and Rabin, who have many related papers:

We investigate the role that self-control problems - modeled as time-inconsistent, present-biased preferences - and a person's awareness of those problems might play in leading people to develop and maintain harmful addictions. Present-biased preferences create a tendency to over-consume addictive products, and awareness of future selfcontrol problems can mitigate or exacerbate this over-consumption, depending on the environment. ... For realistic environments self-control problems are a plausible source of severely harmful addictions only in conjunction with some unawareness of future self-control problems.

So, are we in fact biased to underestimate our self-control problems?   If so, why isn't it easier to just tell us we have self-control problems?   Why wouldn't we believe such advice?

Surely some of us do underestimate our self-control problems, but curiously I can't seem to find any papers in this (large) economics literature that consider people who overestimate their self-control problems.  This ethics paper by Tyler Cowen, however, does consider it and convince me that such situations are common.   Are we biased to assume others are biased toward too little self-control?

Blue or Green on Regulation?

In recent posts, I have predicted that, if not otherwise prevented from doing so, some people will behave stupidly and suffer the consequences:  "If people have a right to be stupid, the market will respond by supplying all the stupidity that can be sold."  People misinterpret this as indicating that I take a policy stance in favor of regulation.  It indicates no such thing.  It is meant purely as guess about empirical consequences - a testable prediction on a question of simple fact.

Perhaps I would be less misinterpreted if I also told "the other side of the story" - inveighed at length about the reasons why bureaucrats are not perfect rationalists guarding our net best interests.  But ideally, I shouldn't have to go to such lengths.  Ideally, I could make a prediction about a strictly factual question without this being interpreted as a policy stance, or as a stance on logically distinct factual questions.

Disagreement Case Study: Robin Hanson and David Balan

The basic challenge posed by Robin is this.  To support the imposition of a paternalistic government policy on Peter, one must believe that: (i) the government is sincerely motivated by Peter's welfare; and (ii) the government knows what's best Peter better than Peter himself does, even taking into account the fact that in the absence of paternalistic policy Peter need not rely only on his own knowledge, but is free to seek the uncoercive advice of anyone willing to give it to him, including the government itself.

To my mind, point (ii) is the easy part.  There really are people who left to their own devices will ingest poisinous miracle cures and the like, and who really would be better off if they didn't do so, and for whom actually existing paternalistic policies are the only hope of being saved.  Of course there are some cases where the government really just doesn't know better and gets it wrong.  Robin and I would agree that these are the cases where policy should be restrained, and if it's not restrained it's more likely to be an abuse of power issue than an ignorance issue.  So in my view the real action is in point (i), the extent to which the government will shun restraint and abuse or misuse the coervice power that paternalistic policies give it.

Disagreement Case Study - Balan and I

David Balan and I exchanged a few posts here over the last few weeks on paternalism, and we had a one hour debate a week ago (audio here).   David and I are similarly expert by conventional measures, so this is more disagreeing with an equal than disagreeing with a superior.  And being active contributors of Overcoming Bias, we are both well aware of many signs of bias.   David is employed by a U.S. agency much of whose policies are justified via paternalism, while I am employed in part because of an academic publication that leaned against paternalism.  So we have similar potential for self-interest biases. 

David seems to disapprove of most policies in most societies today and through history that have been justified on paternalistic grounds.  These include parents choosing kids' careers and spouses, bans on alternative religions, political groups and sexual orientations, rules about who can practice what professions, and limits on the freedoms of women, ethnic minorities, and lower classes.  So if he could only choose in general between paternalism or not, I think David would choose not.   

But David considers the rulers of our society, our democratic majority and the opinion elites they follow, to be much better paternalists than the rulers of all those other societies.   By "our society" David means the United States and nations with similar paternalism policies.  The main evidence David cites for the superiority of our ruling class is that we are the most prosperous society in human history.

Extreme Paternalism

(Though as Bryan Caplan points out, it might be better called Extreme Maternalism.)  In a comment on Friday, Michael Vassar suggested:

Why not simply extend the age of parental authority, or more optimally, designate a discrete set of "parental" authority types that individuals have periodic (annual) opportunities to review and to transfer to some other parental figure. Some sort of licensing might or might not be required for "in-loco parentis" candidates (and for biological parents, or perhaps not). The right to designate parental authority, as well as various reductions in such authority, might be gradually phased in with age, or as the result of various tests or maturity rituals, or possibly a combination of both.

This is very similar to a 1993 suggestion of mine, and both are attempts to find a minimal paternalism, realizing gains from decision review and veto, but retaining maximal flexibility and adaptation to individual variations.   

An opposite extreme would be extend paternalism to nations.  In civil rights, many argue national governments needed to overrule bad state decisions, for the good of those states.  So why not empower the United Nations to overrule what it sees as a bad decision by any nation, in the name of benefiting that nation?   

The reason for considering extremes is that simple arguments usually favor extremes. So you should either support an extreme, or find an argument complex enough to favor an intermediate position.  Why would national paternalism be a good idea, but international paternalism be a bad idea?

Burch's Law

Greg Burch said:

"I think people should have a right to be stupid and, if they have that right, the market's going to respond by supplying as much stupidity as can be sold."

Let's Get Ready to Rumble!!

Here's my opening bit for my debate with Robin tonight.  Needless to say since we're both all about Overcoming Bias, there will be no cheap point-scoring or sophistry and much gentlemanly conceding of valid points.  And maybe a little hitting each other with metal folding chairs.  At least in my day college kids liked that sort of thing.

Aside from basic functions like enforcing contracts and maintaining order, there are three justifications for governmental intervention in people's lives:

1. Correct market failures.
2. Redistribution (make Peter do or not do something, or take from Peter, for the benefit of Paul).
3. Paternalism (make Peter do or not do something for Peter's own benefit).

This discussion is about #3.  The traditional arguments against paternalism are:

Rational Agent Paternalism

While paternalism is about bias, it need not be about irrational bias.  In a 2003 Journal of Public Economics paper, I modeled product bans and warnings directed at completely-rational consumers, and chosen by a better-informed completely-rational regulator who only cares about economic welfare.   I found:

• Even ideal regulators want to lie about product quality, to correct for other market failures.
• Even a small temptation to lie makes consumers disbelieve most of what regulators say.
• Regulators who can ban sometimes do, as consumers won't believe severe warnings.
• Consumers believe more severe warnings from regulators who can only warn. 
• When regulators prefer to talk up quality (e.g., in health, school, investing), both regulators and consumers are better off if regulators cannot ban.

Yes, this model may be less relevant if irrational biases are central to paternalism.  But it at least gives us a concrete reference point.  Weather permitting, David Balan and I debate paternalism today. 

Paternalism Is About Bias

Contributor David Balan and I meet next week (6-7pm Wed. Mar. 7, GMU, Mason Hall, Rm. D3) to debate "Paternalistic Policy:  Altruism or Arrogance?"  (I say "Arrogance.")  Here let me start to set the stage.

Paternalism is policy intended to benefit some people by limiting their choices, like a parent who stops a kid from playing in the street.  Examples include laws requiring professional licensing and product safety features, or banning risky buildings, food, drugs, and financial investments.

A warning is usually a feasible alternative to a requirement or ban.  Parents could just say "Playing in the street is a very bad idea," and if the kid believed them, the result would be the same as a ban.  Similarly, governments could just tell us that certain doctors or drugs are unsafe, instead of outlawing them. 

Now one can imagine inefficient warning systems, such as having to go look up each drug at some badly organized government website.  But we can also imagine no-fuss government warnings: let anything the government would have banned be sold only at special "would have banned" stores, whose customers pass a test showing they understand that regulators disapprove.