Bryan Caplan  

Extending the Enlightened Preference Approach: Proceed With Caution

Climate Change vs. Entitlement... Robin Hanson to Technophiles: ...

How would people's preferences change if they knew more? Political scientists usually attack this question using the so-called "Enlightened Preference" method. (See Scott Althaus' Collective Preferences in Democratic Politics for a fantastic survey of this large literature).

The gist of the method:

You start by administering a two-part survey.

Part 1 is a test of objective political knowledge. (How many senators does each state have? Name as many Supreme Court justices as possible. Etc.) Think of it as a "Political IQ" test.

Part 2 asks respondents about their policy preferences. (Should we rely more on the market or government? Should we invade Iran? Etc.)

The survey also collects standard information about respondents' income, gender, party i.d., race, etc.

Once you've got all this information, you're ready to discover the public's "Enlightened Preferences." In essence, you estimate policy preferences (from part 2) as a function of Political IQ (from part 1), plus a bunch of control variables that you think might influence political preferences holding knowledge constant. If you're finicky, you can even allow Political IQ to have a separate coefficient (and sign) for various subgroups of the population. (Perhaps knowledge makes the rich more pro-market, but makes the poor more pro-government).

The final step is to use these results to simulate what public opinion would look like if you raised Political IQ up to the stratosphere but kept all other characteristics the same. The resulting distribution of opinion is what we call the public's Enlightened Preferences. It's what people would want, if everyone knew a lot more.

Overall, I am a big fan of the Enlightened Preference literature, and I won't conceal the fact that Enlightened Preferences are generally more socially liberal and economically conservative (in short, more libertarian) than the actual distribution of opinion.

Still, this approach has some potentially awkward implications for me. You could just as easily use it for consumers or workers as you could for voters, right? Just estimate consumer demand as a function of people's score on a "Consumer IQ" test, plus a bunch of control variables you think might influence consumer preferences holding knowledge constant.

For example, I strongly suspect that this approach would reveal that if people knew more about financial markets, they would be far more likely to invest in index funds that they actually are. It wouldn't surprise me if we found parallel results for health care.

I have no principled objection to this, but if we're going to generalize the Enlightened Preference approach, we should proceed with caution for at least two reasons.

First, in politics, you're generally picking a policy that everyone's got to live with. As a result, you really only need to understand how greater knowledge would change the distribution of preferences. In contrast, in the market, purchases are individually customized. Knowing that the average person would change his consumer behavior if he knew more doesn't mean that many consumers wouldn't change no matter how much he knew (or change in the opposite direction). From a slightly different perspective, what statisticians call "the error term" might be better-described as "person-specific preferences."

Second, in the market, there is a much bigger role for reverse causation. People who know a lot about reptiles buy more reptiles, but it's probably not knowledge changing preferences. It's preferences changing knowledge. In contrast, people rarely learn a lot about international trade purely because they think that "trade is fun." The best counter-example I can think of is that people who know a lot about politics might be more supportive of things like PBS because PBS shows programs that political junkies enjoy.

Thoughts? Examples? Counter-examples?

Comments and Sharing

TRACKBACKS (3 to date)
TrackBack URL:
The author at Trade Diversion in a related article titled Bryan Caplan says I'm rare writes:
    I'm special: [P]eople rarely learn a lot about international trade purely because they think that "trade is fun."... [Tracked on May 31, 2008 3:37 PM]
COMMENTS (7 to date)
Ross Williams writes:

A problem I see with the enlightened preferences approach (I have not studied to literature myself) is the type of knowledge that you describe it testing. Memorization of facts is far different from analysis, it is the difference between a high school and college education. My 9th grade self would probably be able to ace one of these tests, but I would fear for the world if my beliefs then influenced policy decisions.

On the market side I am a fan of this approach, and it is exactly why I read Consumer Reports, Consumerist, and Lifehacker to seek advice for market decisions. The editors BTW are probably much better at finding these bright people to share their opinions than a quiz. After all, these companies have so far survived the market test.

jsalvati writes:

Two things:

First, I don't know if you are aware of "Deliberative Polling," ( but they take a different approach to enlightened preferences. They take a group of people, ask them their opinion on a variety of policy questions, and then make them sit through a carefully balanced debate on those issues. Then they ask the same people about their opinion. This way they get the change in opinions.

Second, I notice that "Enlightened preferences" does not have a wikipedia article. I think it would be a good thing if someone familiar with the literature created a good wikipedia article on the topic.

Les writes:

It seems to me that Enlightened Preference is to politics what Socialism is to economics.

No matter how much data is centrally collected, the centralized political planners can never approach the efficiency of the political analogue to the invisible hand.

Contributor X writes:

Les gets the gold star.

More thoughts:

1) This sounds exactly like the argument I hear all the time, from both left and right; essentially, "If people only knew about the true amount of [harm we're doing to the planet || gains from global free trade] they would change their position and support [emissions caps || NAFTA]. So frequently would I hear this argument that I took to labeling it the, "If Only People Weren't so Dumb/Ignorant Argument." This argument is not very persuasive to those who disagree with you.

2) Assuming all the conclusions of Enlightened Pref.'s are true, how could the American political system utilize this knew information? If a position is to be decided through voter participation, all the interested parties are free to educate and advocate the public. Surely it wouldn't be used to overturn the result. If the issue is decided indirectly, through our rep.'s, then this would be but one consideration out of many.

3) Distribution of costs and benefits is key, think farm bill and earmarks. Think about dems calling oil execs to DC to harass them on TV about being greedy.

4) Perhaps this is one of the key diff's between libertarians and conservatives. I think the "enlightened preference" is the one reached through democratic practices and institutions that are widely viewed as free, fair, and thus, legitimate. The process is more important than the outcome.

Finja writes:

I second the need for a good Wikipedia article... or could you suggest a general introductory text/book about Enlightened Preferences?

michael gordon writes:

Introductory Comments

That's a good post, Bryan --- and it sent me back to re-read the criticism you made in 2005 (Economic Journal Watch) of Donald Wittman and, to an extent, of the Univ. of Chicago political economy school: Stigler, Posner, Becker, Pelzman (and probably otghers) . . . all of whom found 1) that voter ignorance was not a major problem for a democracy like the US; 2) that such ignorance --- in line with all political science public opinion surveys and collective action theories in economics --- was rational, given the insignificance of one peson's vote in electoral outcomes, plus the costs in time and effort to be more informed; 3) and that large numbers of voters ignorant or uninformed beliefs owing to various problems and biases cancelled out one another, to the point that --- according to the Chicago school of political economy --- a democratic system like the US performed over time as well as a market economic-system. In the process, of course, such reasoning has infuriated libertarians and still does (apparently).

You disagreed, set out a very good summary of Wittman's (classical?) argument, and showed in line with your enlightened preference theory (developed by others) that the Chicago school was wrong.


Because if you look empirically at what people actually believe about the workings of a market economy, then on three scores --- later enlarged by you in your recent book to four --- voters were actually not "rationally ignorant", but rather "rationally ignorant" and hence showed both explicitly and more by implicit extension, that American democracy was, in its workings, erratic and wrong about the role of government in the economy and that we would probably be better off (as I infer from your arguments, in that article and your book) with a "benevolent despot".

Meaning? As used by French philosophes in the 18th century who knew France was unlikely to develop peacefully into a constitutional monarchy with freedoms of the sort Voltaire and others found in Britain, the term meant a tyrant who more or less in knowlege fitted your simulated error-free computer-genius.

It was a very good exchange, but in my view Wittman came across more successfully, and still does in the light of your recent book and extended argument. And leaves the Chicago School of political economy sound in its overall estimation that, in an imperfect world --- judged by pragmatic criteria (including the enormous wealth and extremely high per capita income of the US population) --- our democracy does, on balance, about as well as could be humanly expected.


Rational Irrationality = Democratic Voters Are Systematically Biased and Uninformed and Vote Against Their Interests, and So Democratic Politics Performs Blatantly Worse Than A Model of a Free Market (Laissez-Faire?) Economy. Democatic Politics Is Therefore Suspect and Should Be (Drastically?) Limited or Curbed

The long sentence after the = summarizes, I think fairly, your view of rational irrationality . . . and presumably not just in economic matters. But how do you arrive at such a judgment?
You don’t show, in your book --- which I looked at one day at the UCSB library (I wish I had a copy now, mind you) --- that voters are irrational except in a strange way: you compare their beliefs in four areas of economic life with the beliefs of experts: Ph.D. economists . . . the surrogate perfectly enlightened computer-genius or, in real life, presumably, the totally enlightened despot who would curb and limit democratic politics to the point that something like a laissez-faire economy of the English sort in the mid-19th century (certainly the US was never laissez-faire as England was). But is this not a strange criterion?

Start with macroeconomics.

To start here is to note that there is no consensus whatsoever among Ph.D. economists on such things as the Ricardian equivalence in fiscal policy, or whether menu costs create a system-wide rigidity of prices, or whether business firms are irrational to not lower wages in a recession to the existing reservation wage, or whether money is endogenous or exogenous . . . the latter a big lively issue reawakened by the Bernacke-led Fed Reserve. I wager that you and your colleagues as GMU recoiled from each and every one of his measures, not just lowering interest rates, but taking action to try reassuring investors and creditor-institutions that involve a considerable amount of quasi-governmental actions.

In microeconomics, there is no doubt more consensus --- but disagreement still exists, with probably free trade the one policy on which there is near-unanimity in the economics profession. That said, to then claim that the public is especially uninformed and even stupid to not like this policy is to overlook the fact that there are, in a period especially of rapid technological and globalizing flux, lots of individual losers: witness the US mfg. sector’s labor force size. Yes, the economy overall benefits, and you could argue that the laid-off workers benefit too from lower costs at Walmart and elsewhere, but it does not seem either irrational or stupid for the laid-off workers and those who fear their turn is next to not like the policy.

As for what you call “xenophobia” as at work here, I could just as well argue that libertarian economists lack patriotism and presumably think that Chinese workers or those in India are going to volunteer to fight against our enemies in the next war.


Experts In General

More generally, I think you badly go astray in crediting experts with far more accurate, valid, true, and pragmatically or operationally sound knowledge and policies at their disposal. For well over four decades now, Philip C Converse’s path-breaking studies in public opinion have been vindicated in dozens of studies, most recently in Gordon Tetlock’s Expert Political Judgment. What have these studies found?

 The so-called experts in political matters --- pundits, political campaign leaders, talking-heads, opinion-setters, and so on --- tend, like experts in virtually every field, to overestimate their understanding, their knowledge, their ability to analyze a situation, and their predictions of what will happen. In fact, they do no better than a simple linear equation, and practically any large group of people chosen at random if you find the median of their own views.

 It’s the most educated --- the graduates of the best universities, graduate school products, and the like --- who tend to have the most ideological system of beliefs. They tend, accordingly, to be more disposed than the average person to reject any information that doesn’t fit into their preconceived belief-systems and to seek reinforcement . . . a combination, in short, of reducing cognitive dissonance and, in groups, succumbing to group-think.

Are economic Ph.D’s any different?

Do they not, as Kahneman’s and Tversky’s work as well as others have found, 1) exaggerate their knowledge, 2) think they have more control over events (which they analyze or predict in terms of), and 3) exaggerate their ability to control things. These are widely accepted, experimentally tested findings: hence 80% of drivers think they’re better than average, and probably that’s true of 90% of investors in the stock market.


Democracy in Practice: Comparative Evidence

(i.) No country in the world has a per capita income above, say, $25,000 (PPP) --- leaving aside oil-rich dictatorships (and Singapore) --- that doesn’t have a democratic system of politics, an independent judiciary, limited corruption and elite predatory behavior, and a largely market economy.

(ii.) Among democratic countries, those with the most trust in the population across family, class, ethnic/racial, and regional division have the highest per capita income. They also have the least political corruption --- brought out in annual surveys for 20 years now, using a variety of cross-checking evidence. Only Singapore again --- 4-5 million Chinese, under a dictatorship --- is in the 20 countries least corrupt. The US and Britain are the least corrupt of countries larger than 30 million, Germany usually next to them (the three probably not distinguishable at the 5.0% or less level, statistically speaking)

(iii.) Among these top-performing 20 democratic countries, the US --- leaving aside Norway (5 million, whose oil riches drive up or down its per capita income depending on the price of oil) and Luxembourg (250,000) --- the US has by far the highest per capita income (roughly $47,000), a good 30-35% higher than Germany’s, France’s, Britain’s, or Japan’s; and with only Ireland (4 million) having a per capita income near it. The US also tends to have invented and innovated most of the high-performing cutting-edge industries of the last 35 years of the long-wave Schumpeterian creative-destruction cycle, as it did with the previous two (going back to the internal combustion engine, electrification, and mass –production mfg. of the assembly line sort in the first two to three decades of the 20th century).

(iv.) On top of that, the two great liberal democracies --- Britain and the US --- have defeated in war all major non-liberal authoritarian or totalitarian countries since the industrial revolution of the late 18th and early 19th centuries: Napoleonic France, militarist Germany and its allies (WWI), Nazi Germany, Fascist Italy, militarist Japan (WWII), and together in NATO with others the Soviet Communist empire . . . with the destruction of communism world-wide save in North Korea and Cuba.

In short, these same impressive achievements, and they seem a fairer way to gauge what is humanly and historically possible, rather than using some standard of a theoretical model and its alleged practioners (Ph.D. economists, who seem to be divided on almost all major macroeconomic issues and some notable microeconomic ones).


Preference Ordering and Aggregation
 No doubt whatsoever: markets generally produce a quicker feedback and have --- if the market industry is competitive --- in the price mechanism, qualified by consumer preferences for quality and brand names --- than happens to be the case for politics: elections are less frequent and the attitudes of opinion survey respondents are subject to less checks and feed-back mechanisms --- and are far more ambiguous than opinion surveys and the media portray, depending in the former case on how the questions are formulated, whether they are open-ended or forced (list of alternatives), whether the respondent is informed or not, whether the question is free of misinterpretation, whether the questions tap deep beliefs, and so on

 That said, it’s because preference formation is dealing with far different matters in both areas of life. Private goods are private because they are excludable and divisible and involve individual choice. Public goods are not divisible, not excludable, and lack a similar quick feed-back if things go astray.

 Does that mean market preferences are not subject to systematic (non-random) errors? Stock-market investors come to mind; money illusion among workers pops to mind too; the failure of home owners to read --- assuming you could make sense of the jargon anyway and have the time and patience --- their mortgage contracts and understand them are another example.

 In any case, to conclude this long reply, there is no one theory of democracy I know of --- other than those who initially created and whooped for public opinion surveys (including the initial academic “experts”) --- who said that democracy is a system that is measured by its competence and efficiency because its policies match some aggregate sum of them. And for that matter, Althaus --- whose book you cite in your initial sentence or two --- has admitted this as well in a fairly recent article of his. For my part, the best measure is the practical one that involves a number of criteria, comparatively analyzed as in the previous section here in this post, across countries and historical periods.

The rest seems largely to me to be a mainly abstract academic exercise, based on a largely idealized image of an ideal economic system . . . a libertarian failing that is matched equally on the Marxist and radical left.

Michael Gordon, AKA, the buggy professor.

michael gordon writes:

The next to last sentence in my post was a little garbled, banged out --- as was the entire commentary --- at bursting speed and fairly late at night.

The original reads:

 In any case, to conclude this long reply, there is no one theory of democracy I know of --- other than those who initially created and whooped for public opinion surveys (including the initial academic “experts”) --- who said that democracy is a system that is measured by its competence and efficiency because its policies match some aggregate sum of them. And for that matter, Althaus --- whose book you cite in your initial sentence or two --- has admitted this as well in a fairly recent article of his. For my part, the best measure is the practical one that involves a number of criteria, comparatively analyzed as in the previous section here in this post, across countries and historical periods.

It should be revised and read:

 In any case, to conclude this long reply, there is no one theory of democracy I know of --- other than a version created and whooped for by the inventors of public opinion surveys (including the initial academic “experts”)in the 1940s and 1950s --- that claims that the competence and efficiency of democracy should be measured by how well its political leaders respond to an aggregate sum of the citizenry’s individual preferences. That transient enthusiasm looks in retrospect silly and shallow, based on logical fallacies. The hard fact is that there is no way to satisfy the contradictory and often ambivalent preferences of over 100 million American voters by weighing and aggregating them into a sum that guides all policymaking, and fortunately so. Nor is there any evidence that most people want to spend their leisure time actively engaged in political learning and practical politics except for voting and, presumably, listening to the often lamentably bad discussions in election periods that appear on radio and TV.

 For that matter, Althaus ---whose book you cite in your initial sentence or two --- has admitted this as well in a fairly recent article of his. And for my part, the best measures of our democracy’s performance are the practical ones touched on earlier in my comments here: they involve a number of criteria, economic, civil liberties, and foreign policy, comparatively analyzed across countries and historical periods. Not to forget, among these criteria, that the political leaders of a huge country, 300 million in number and of diverse ethnicity, does very well in international studies of comparative corruption.

 Does that make our democratic system perfect? Hardly, but then neither is our economy, though it does extraordinarily well too if viewed comparatively and historically across countries.

And so? And so, the rest --- including your measures, Bryan, of dogmatic "rational irrationality" --- seem largely to be a mainly abstract academic exercise, based on a an image of an ideal economic system . . . a libertarian failing on the right that is matched equally on the Marxist and radical left.

Michael Gordon

Comments for this entry have been closed
Return to top