pageok
pageok
pageok
Political Ignorance and Libertarian Paternalism:

"Libertarian Paternalism" is all the rage in law and economics circles these days. To slightly oversimplify, libertarian paternalists claim that people systematically make mistakes as a result of cognitive errors and biases. Afterwards, they end up with outcomes that they themselves consider inferior to at least some of the alternatives they could have gotten by making a different decision in the first place. As a result, third party intervention (usually, but not always, by the government) can help people make the "right" decision. The difference between the new paternalism and the old is that the former defines "right" by reference to the actor's own preferences, not by some "objective" theory of morality or utility imposed by others. Thus, claim libertarian paternalists, we aren't saying that government should override people's preferences; we just want to help people get what they themselves want. For an introduction to the concept by Richard Thaler (a leading proponet) and a critique by NYU economist Mario Rizzo, see this debate in the Wall Street Journal Econoblog.

At least tentatively, I agree with libertarian paternalists that cognitive errors often lead people to make mistakes that they afterwards regret. The question is, however, compared to what? To justify paternalistic policies ("libertarian" or otherwise), advocates must prove not only that autonomous individuals make mistakes, but that the government will make fewer mistakes if you let it constrain individual choices.

If government policy is subject to democratic control, the key question is whether people are more irrational and ill-informed when they act as consumers than when they act as voters. Regular VC readers won't be surprised to learn that my answer to this question is an emphatic "no." Ignorance and irrationality heavily influence voting decisions, and voters are routinely ignorant about very basic things, such as the mere existence of extremely important government policies. They also hold irrational and ill-founded beliefs about even simple political issues (e.g. - believing that the economy is a zero-sum game and that free trade reduces national wealth instead of increasing it). While direct comparisons with markets are hard to come by, few if any market errors are as widespread as numerous voter errors (of which the free trade example is a notable case). Similarly, few consumers are likely to be as ignorant of the basic characteristics of the products they buy as the 70 percent of eligible voters unaware of the very existence of President Bush's medicare prescription drug plan, the largest and most expensive new government program voters have "bought" over the last 40 years.

Overcoming bias and cognitive error requires time and effort. In political markets, voters have only an infinitesmal chance of influencing the outcome (less than 1 in 100 million in a presidential election, for example). That gives you very little incentive to do the hard work of increasing your knowledge and overcoming your biases. By contrast, when you choose to buy a product in the market, your individual choice is highly likely to be decisive in determining what you get. That gives you a much stronger incentive to try to choose wisely. Casual empiricism bears out this hypothesis. I have yet to meet a person of any ideological persuasion who spends more time and effort deciding which candidate to support for president than they do deciding which car to buy. And it's certainly not because the presidency is less complicated than your car, or less important!

Libertarian paternalists may be right to be pessimistic about how well people make decisions in market settings (though my colleague Josh Wright has an excellent article questioning this). They are, however, implicitly overoptimistic about the quality of the decisions we make as voters.

For some libertarian paternalists, of course, the alternative to market decisionmaking is not democracy but decisionmaking by unelected experts. I will take up this possibility in a follow-up post.

Mark Field (mail):
Sympathetic though I am to the idea of constraining some choices, I'd prefer to begin with proper pricing. If government would start by ending subsidies and by forcing prices to reflect externalities, I'd be prepared to test that outcome for awhile and see if more is needed.
5.29.2007 12:35am
Sean O'Hara (mail) (www):
God save us from those who want to save us from ourselves. .

I'm fine with my mistakes, because they're my mistakes. I'd rather be wrong than have the government make the right choice for me.
5.29.2007 12:49am
David Krinsky (mail):
Isn't it a bit misleading to compare consumer decisionmaking to voter decisionmaking, when the paternalist laws in question are virtually never enacted by referendum, but rather by elected representatives or even by appointed regulators who have been delegated authority by elected representatives?

Even leaving aside the unelected experts--regulation by whom is commonplace today, and not necessarily incompatible with democratic legitimacy--it strikes me as entirely plausible that a Congress of professional politicians might well make more informed decisions than either an average voter or an average consumer. That Congress is advised by professional policy wonks (and lobbied by lobbyists on various sides who, though obviously biased, at least can be expected to know something about the groups they represent). And representatives' longevity in office is dependent on voter agreement with representatives on "important" issues, but also with overall perceptions about the state of the nation, and thus, at least at the margins, on (perceived) good governance.

Whether this model actually leads to prudent decisionmaking is quite another question, but it's hardly as crazy as assuming that people are more informed when wearing their voter hat than wearing their consumer hat.
5.29.2007 1:12am
Dave Hardy (mail) (www):
It could also be phrased as "the people en masse make more rational choices than each individual, acting on his/her own." This would be consistent (on most non-moral issues) with modern liberal thought, and (on most moral ones) with classical republicanism, i.e., a "rules republican."

The problem wih the formulation is that as a practical matter, it really works out to "the people, acting thru their elected representatives, who in turn usually act by enacting general statements of policy to be implemented by not-at-all unselfish executive branch implementors, make more rational choices than each individual would make on his/her own." This is a ... less clear conclusion ... than the original premise might suggest.

DTH, former GS-14.
5.29.2007 1:19am
Hey (mail):
It's not as bad as it normally is, since they at least claim that you'll be able to make different choices, they're just changing the default options. Opt-in vs opt-out pricing on cable plans and student fees does show how people might actually prefer one option but don't take it because the benefit is much lower than the amount of work required. Of course student fees and cable plans rely on this to extract money unwillingly, while this approach is claiming that they'll make the better decision easier to make by reducing paperwork. Seems like a decent idea, auto-enrolling you in an IRA/401(k), maximizing your pay deductions...

Of course, any use of the word paternalism sets off my skepticism, as does the co-opton of the word "libertarian" by lbierals/leftists to camouflage their policy activities. I don't trust the people promoting this, but it is possible as the least worst way to implement questionable to evil things.
5.29.2007 1:32am
Brian K (mail):
A lot of people are incredibly ill informed about the consumer choices they make. link

I don't see the reason why we should assume people are more informed in consumer markets over political markets. Just look at the discrepancy between what organic actually means and what many consumers think it means. Or how people act in the stock market. Or the mess with mortgages.
5.29.2007 1:57am
Ilya Somin:
Isn't it a bit misleading to compare consumer decisionmaking to voter decisionmaking, when the paternalist laws in question are virtually never enacted by referendum, but rather by elected representatives or even by appointed regulators who have been delegated authority by elected representatives?

Yes, but that still leaves open the question of how good the voters are at picking the elected representatives. Doing so well requires some knowledge of their proposed policies and their likely effects.

If the advantage of elected officials over referenda is the supposed expertise of the former, it's worth noting that market consumers can rely on experts too. That's what you do when you hire a lawyer or doctor, for example. And, for the reasons noted in the post, people are likely to do a better job of hiring experts in consumer markets than in elections.
5.29.2007 2:01am
American Psikhushka (mail) (www):
It all depends on how they frame things - the choices, the information, the default choices, etc. There is some real possibility for abuse here. Like when a choice is banned rather than requiring that additional information be required - trans-fats, for instance.

Coming up with good default choices, providing information, outlining choices, etc. may be positive as long as no choices are banned and it is easy to opt-out of the default. But this kind of thing could be just as dangerous as regular government intervention, and needs to be monitored just as closely. (Which isn't saying much, because normal government intervention isn't monitored nearly as closely as it needs to be as things stand now.) "I'm from the government and I have selected a good default choice for you." shouldn't put anyone very much more at ease than "I'm for the government and I'm here to help you."
5.29.2007 2:23am
American Psikhushka (mail) (www):
Correction: Should be "from" in the last sentence above.
5.29.2007 2:25am
George Weiss (mail):
I agree with david krinsky..the discussion of voting was a far cry from the issue of whether gov regulation can be more or less intelligent than a consumer decision.

the idea that we vote less intelligently than we shop..even though you've made the point beautifully and a number of times..and even though the point is true...its also not exactly a news flash...and your discussion of market experts vs the "experts" in congress adds little to your argument..

the fact is...gov can and does make more intelligent decisions that the average person...all the time..based on its consultants..studies etc..true..maybe those decisions would be better if there were better voters...but they are still better decisions that the average person


i gotta tell you though...if this form of paternalism is "OK" becuase it gives us what we want and not what sociaty wants...we must ask not just whether the gov is better at getting what we want...but also..is the government better at determining what we want?

isnt it really only possible for the person to know what he wants?
5.29.2007 2:25am
Henri Le Compte (mail):
I have to confess ignorance of "paternalistic libertarianism," but doesn't it just look like the same old "Nanny-State knows best" dressed up in new verbiage and new rationalizations? And isn't this just the same problem that has existed forever-- with freedom comes responsibility? If you wish to shield people from the consequences of their choices, very soon-- one way or the other-- you will have to take away their ability to make those naughty choices.

And again, you face the notion that a group of "experts" (who in previous systems were known as bureaucrats) can make choices more "expertly" than a open and free market that consists of the accumulated wisdom of millions of independent players. Hasn't history shown, over and over, that free markets yield better results? More prosperity, less tyranny?

An last, but by no means least, "Who watches the experts?" You know, allowing power to amass in the hands of a few experts is only likely to teach us again that "power corrupts, and expert power corrupts expertly," or something like that.
5.29.2007 2:30am
juris_imprudent (mail):
If you wish to shield people from the consequences of their choices, very soon-- one way or the other-- you will have to take away their ability to make those naughty choices.

That's certainly the impression I get - these folks aren't starting from a libertarian point and their major justification for co-opting the libertarian label is that their proposals are non-coercive. This feels very much like a repackaging of New Deal/Great Society liberal nostrum.

For those who have not read it, I strongly recommend Thomas Sowell's Knowledge and Decisions, which undercuts the notion of paternalism be it libertarian or otherwise.
5.29.2007 2:56am
Chrisper (mail):
Where are all these 'libertarian paternalists' lining up to dictate my choices? About the only example I can think of is the rhetoric decrying the choices of 'the sheeple' but that comes with admitting the speaker is hopelessly unable to get enough power to dictate choices anyway.
5.29.2007 2:57am
ReaderY:
It's fun to construct castles in the air, isn't it?
5.29.2007 2:58am
A.C.:
Another problem is that the people making decisions on behalf of the government may not be representative of the population as a whole. They probably will not be completely unrepresentative, but it is easy enough to imagine a government that decides society will work best if the choices of Group X are severely contrained, while members of Group Y can do anything they want. It may even be true that society will work well that way, but we still do not want that outcome.
5.29.2007 3:51am
Blar (mail) (www):
1. His name is Richard Thaler, and he is an economist at the University of Chicago business school.

2. What makes "libertarian paternalism" libertarian is that the paternalistic policies that it advocates are minimally coercive, such as default rules that individuals can opt out of (e.g. 401k enrollment). It has nothing to do with matching the person's own preferences. As Sunstein and Thaler (2003) write, "The libertarian aspect of our strategies lies in the straightforward insistence that in general, people should be free to opt out of specified arrangements if they choose to do so. Hence we do not aim to defend any approach that blocks individual choices."

3. Laws are enacted by elected representatives, after input from their staff and other advisers, not by the populace at large. And since libertarian paternalists often advocate specific policies, not just government intervention in general, it seems that they should primarily be concerned about whether those specific policies are likely to produce a net benefit, not about the effectiveness of government in general.
5.29.2007 4:58am
David Welker (mail) (www):

The question is, however, compared to what? To justify paternalistic policies ("libertarian" or otherwise), advocates must prove not only that autonomous individuals make mistakes, but that the government will make fewer mistakes if you let it constrain individual choices.


My problem is that you are trying to answer this question in the abstract and as a general matter. What you really need is a more nuanced account that analyzes this on a case by case basis, not in some super abstract manner.

It is not about voter decisions versus consumer decisions either. Voters hold government accountable -- they don't dictate its choices. So, this is an entirely misleading way of framing the issue. Government expertise is constrained by democracy, it is not dictated by it. So, I would say your failure to address this basic point makes your analysis thus far close to useless.

Will government make better decisions? The answer is sometimes. When better decision-making can be informed by expertise that an individual just would not ever have the time or resources to make.


If the advantage of elected officials over referenda is the supposed expertise of the former, it's worth noting that market consumers can rely on experts too. That's what you do when you hire a lawyer or doctor, for example. And, for the reasons noted in the post, people are likely to do a better job of hiring experts in consumer markets than in elections.


This only works when the individual: (1) Has the resources to hire the expertise in question -- the harder the question, the more expensive it is to hire expertise -- few people can afford all the expertise they can benefit from. (2) They realize that they would benefit from expertise. It is often the case the people do not realize when they would benefit from expert advice, nor do they have any idea how to evaluate the advice of different individuals who hold themselves out as experts. (3) Has the discipline to follow through on the advice. (4) Manage to overcome the inertia associated with making decisions in a complex world.

There are myriad problems here.

Let us get specific. What is wrong with having an opt-in for say 401k participation? This overcome the inertia that causes people to unduly delay savings. If someone does not want to participate, they can do so with a little effort.

Instead of arguing in the abstract, I would be more interested in you addressing these questions in the concrete. What is the wrong with the 401k proposal?

In general, these questions are not really capable of being answered in the abstract. I know you don't like solutions to problems that involve government generally. In fact, it appears you have an irrational prejudice against government. For that reason, I think you can usefully contribute to the analysis of specific problems, but I am skeptical that you have anything useful to say about the more general case. Nonetheless, you probably would probably have something useful to say about specific proposals.
5.29.2007 5:24am
Russ Mitchell (mail) (www):

As a result, third party intervention (usually, but not always, by the government) can help people make the "right" decision. The difference between the new paternalism and the old is that the former defines "right" by reference to the actor's own preferences, not by some "objective" theory of morality or utility imposed by others.


There is absolutely nothing new about this: it's the recipe that gave rise to the Progressives: modify public behavior (the public not being able to know better) by using legal social controls to alter the human environment. The Progressives, muckrakers, etcetera, may have argued that they were performing an objective good (and who doesn't like cleaner water?), but in the end their standards were as generationally-arbitrary as anything the current crop has come up with.
5.29.2007 10:12am
Sean O'Hara (mail) (www):
2. What makes "libertarian paternalism" libertarian is that the paternalistic policies that it advocates are minimally coercive, such as default rules that individuals can opt out of (e.g. 401k enrollment).

In Soviet Russia, policies opt you out.
5.29.2007 10:19am
Ken Arromdee:
They also hold irrational and ill-founded beliefs about even simple political issues (e.g. - believing that the economy is a zero-sum game and that free trade reduces national wealth instead of increasing it).

I think that at this point, instead of criticizing voters for being ignorant, you're criticizing them for taking the opposite political position from yours. Are you seriously suggesting, not just that free trade increases wealth, but that it's *so obviously true that anyone who doesn't think so must be ignorant, and that educated disagreement on the issue does not exist*?
5.29.2007 10:40am
RealityCrutch (mail) (www):
"Libertarian Paternalism"

This sounds like a general misuse of a technically detailed definition. Since we live in a time when disfavored Ideologies like to go by more favored names, I'm thinking "Anarcho libertarianism" here, I become very suspect when another seemingly counterintuitive usage comes into some much play.

I think the general libertarian gut response is healthy. Death to experts or some such pithiness.
5.29.2007 11:34am
Montie (mail):

Are you seriously suggesting, not just that free trade increases wealth, but that it's *so obviously true that anyone who doesn't think so must be ignorant, and that educated disagreement on the issue does not exist*?


Ken, you have somewhat of a point. Under some very limited and specialized assumptions, you can show that free trade does not increase wealth of all parties to trade.

However, voters tend to believe that this is the case generally. It is not. Trade policy is an area in which the rare and perceived exceptions all too often make the rules.
5.29.2007 11:39am
David Krinsky (mail):
Yes, but that still leaves open the question of how good the voters are at picking the elected representatives. Doing so well requires some knowledge of their proposed policies and their likely effects.


Maybe, maybe not. It might be sufficient to choose on the basis of generalized bundles of opposing positions--a/k/a party affiliation. It might be sufficient to choose on the basis of perceived character and/or experience in leadership positions. Or it might even be sufficient for anyone who has a pulse and a vague interest in reelection to be overseeing a staff of policy experts.

If the advantage of elected officials over referenda is the supposed expertise of the former, it's worth noting that market consumers can rely on experts too. That's what you do when you hire a lawyer or doctor, for example. And, for the reasons noted in the post, people are likely to do a better job of hiring experts in consumer markets than in elections.


Experts are expensive; regulation provides efficiencies where information and the choices based on it are likely to be common to a large group of people.

It makes sense for me to hire a lawyer when a fair amount of money, or issues of great emotional import such as family, are at stake. It makes no sense for me to hire a lawyer to sue half the neighborhood for nuisance when the pooper-scooper ordinance is repealed and the amount of dog poo on my lawn increases. And it makes still less sense for me to hire a nutritionist to analyze the food I buy, when required nutrition labeling can give me the same information at next to no cost.
5.29.2007 11:48am
whackjobbbb:
Hmmmmm, "libertarian paternalism"... is it?

This sorta reminds me of that book that came out a few years ago, "What's Wrong with Kansas?" or something like that, which was basically some doctrinaire lefty's musings as to why the people in Kansas keep making the "wrong" decisions in their elections. Per this work, those stupid Kansan voters were obviously too stupid to understand that they were constantly voting against their own personal interests. But now thankfully this author had sat down and detailed all their foolishness, and given them a roadmap out of their stupidity. Thanks due to the author, for sure, for his missionary work performed for those stupid Kansans.

And now, finally, some eggheads have picked up the mantle and want to bring that author's work into public policy... FINALLY. Afterall, why speak only to the stupid Kansans? The rest of us are AT LEAST as stupid as they are, and we demand those same tender ministrations. Don't we deserve them?

Our political system was designed to drive people to the middle (anybody truly think there's much difference between George W and his political "opposition"?). We don't really get much of a choice, by design, IMO. Broad concensus is required, and that's what we get out of this system.

In contrast, our consumer/market system was designed (theoreticallY) to provide a broad range of choices. The eggheads are comparing apples and oranges when they compare political and comsumer choices.
5.29.2007 12:12pm
Justin (mail):
Don't forget the war on terror. We need to stop voters from voting out the GOP, because then the terrorists win.

Seriously, this is less a post about "liberterian paternalism" (which is different, and discussed better outside the paternalist framework by, amongst others, Michael Dorf), and more about declaring your political preferences correct, and then stating "damn the voters." It's not a facially invalid argument, just a simplistic one, and already well-dealt with in the political philosophy field over the last century and a half.
5.29.2007 1:02pm
Maureen001 (mail):
Nothing sets my teeth on edge nor raises a mental red flag more than someone telling me "I know what is best for you..."
5.29.2007 4:15pm
PGofHSM (mail) (www):
"And it's certainly not because the presidency is less complicated than your car, or less important!"

But as elections occur in America, it is less important. At the election, only two candidates have a plausible chance of winning, i.e. those nominated by the two major parties. Those candidates have gone through a vetting process via primaries. Because neither party wants to lose the White House, they will have done their best to nominate a candidate who is capable of bringing out the party faithful without alienating moderate voters. I can't think of a candidate who has been truly outrageous to the opposing side since about 1860, when the country was about to split into two anyway. Even FDR, in proposing to have the federal government regulate securities in the wake of a stock market crash that helped to precipitate the Depression, didn't dare go so far as to have the federal government dictate *what* publicly traded corporations could do, only that whatever they did, it had to be disclosed.

I was deeply troubled by the results of the 2000 and 2004 elections, both in the immediate aftermath and over the last several years, but probably not as troubled as I would be on a daily basis if I got a bad car. Certainly getting a new battery because the used car I purchased last year had a bad one (something that managed to escape my attention because the car salesman was careful always to be the one to start it, so I didn't notice that he had to pump the gas to do so) probably has cost me more even than the effect of the Iraq war on my present taxation and gas prices.

While there's some limitation to how bad a car can be on the market, the number of choices nonetheless is much larger and of much greater range, and of price-to-quality tradeoff, than any presidential election can be.
5.29.2007 4:56pm
Eli Rabett (www):
Ah, I understand now, there is really no issue about illegal voters, because, taa daa, individiual votes are unimportant. Make up your minds, at least what is left of them.

Besides which voting is not the sum of citizens interaction with their government. Any fool would know that.
5.29.2007 5:58pm
byomtov (mail):
I'm not at all convinced that people are so wonderfully rational in their consumption choices. (I'm not even sure what "rational means in this context). The claim that they are rational is based partly on circular reasoning - purchases reveal preferences, so are presumed "rational."

So if I buy an unreliable but attractive car that's rational because I am presumed to value appearance more than reliability.
5.29.2007 7:58pm
juris imprudent (mail):
I'm not at all convinced that people are so wonderfully rational in their consumption choices.

If people were utterly rational in their consumption choices, there would be no advertising industry.
5.29.2007 8:58pm
Sebastian Villarreal:
There are certain choices where it is possible to assume that most people, or the vast majority of people, would choose the same thing if they were well informed. Such as a cheap fix that vastly improves gas mileage, for instance. Sure, it is possible to educate everyone as to the virtues of said quick fix, and assume that the same majority will eventually choose to enact it. However, it is certainly possible that it would be cheaper for the government (and hence, to all of us), to simply mandate said cheap fix.
In other words... what of those cases where it is simply more efficient to mandate than (gasp!) to educate? I will preemptively respond to the cries of 'fascistĀ”' by reminding you that if the majority ends up being ill content with the mandate, they can always (and it is these cases where voters are educated) vote the decision maker out of office.
5.29.2007 10:08pm
whackjobbbb:
If there were a "cheap fix" for fuel economy, as per your example, then somebody would be offering it up in the marketplace. By definition, mandates drive expense, generally in an UNcheap way.

If it were easy, it would already be in place.
5.30.2007 2:28am
Justin (mail):
"I can't think of a candidate who has been truly outrageous to the opposing side since about 1860."

I actually disagree about this in 2004, though oddly not in 2000.
5.30.2007 1:47pm
Mojo Jojo:
Ilya's initial post and most of the comments seem to miss the point of libertarian paternalism. Libertarian paternalism does not (or at least should not) "constrain individual choices", as Ilya asserts that it would. One of Sunstein's favorite examples is setting default contribution rates for 401k accounts. Companies that set a default contribution rate of 5% tend to get much different overall results than those that set a 0% default, even though all employees retain complete freedom to adjust their individual contribution rates however they please. Libertarian paternalism is concerned with trying to find socially beneficial non-coercive defaults in scenarios like that. If you remove coercion from the picture, most of Ilya's criticisms (and those of commenters) don't make a lot of sense.
5.30.2007 2:17pm