Why are the Swiss so happy?

Because I head to George Mason University tomorrow, this will be my last post for a while.  You might want to follow the discussion on Cato Unbound, where I will be posting replies to Hamilton, Selgin and Hummel.

Right after my last post extolling the virtues of Swiss-style democracy I read Bryan Caplan’s persuasive attack in The Myth of the Rational Voter.  He argues that voters aren’t just misinformed; rather they hold deeply ingrained biases that lead them to make poor public policy decisions.  Naturally I wondered if this refuted my argument.  I don’t think it does, but it weakens it a bit.

Bryan’s book is not an empirical study of various forms of democracy; it is an attack on the theoretical underpinnings of the argument for democracy, especially unconstrained democracy.  Switzerland is not even mentioned.  On the other hand the theoretical arguments are impressive, and supported by survey findings.  For instance, he shows how the “wisdom of the crowd” argument breaks down if voters have systematic biases, and then he presents a lot of evidence that they do have such biases in areas like free markets, free trade, etc.   He concludes the book with an outstanding chapter on “market fundamentalism” and “democratic fundamentalism,” which shows that there is no merit to the frequent accusations that free market economists are market fundamentalists, but that there are lots of democratic fundamentalists.  Caplan basically defines a ‘fundamentalist’ as someone who supports a societal institution regardless of whether it works better than alternative institutions.  If we adopted the terminology that I use, most free market economists are pragmatic libertarians, not dogmatic libertarians, as they favor a role for government in certain areas.  Hence they are not market fundamentalists.  (I use the term ‘pragmatist’ as a philosopher would use ‘consequentialist.’)  In contrast, he finds numerous examples of political scientists who say they support democracy even if it doesn’t work as well as other systems.

One commenter argued that my full-throated defense of the Swiss system of direct democracy made me a democratic fundamentalist.  But this can’t be true, as I support democracy entirely on consequentialist grounds.  I was first attracted to Switzerland by its economic success, not by its democratic system.  Indeed, I was initially skeptical about the “wisdom of crowds” argument.  Only much later did I take a fresh look at their political system, and start to entertain the possibility that it might play a role.

Although I accept Caplan’s theoretical arguments against direct democracy, these arguments don’t sway my belief in Swiss-style democracy, which combines decentralization and referenda.  Here are a few reasons:

1.  Caplan focuses on a few areas where voters have strong cognitive biases, such as trade.  But there are many examples of public policy decisions where voters would probably make better decisions.  Consider pork barrel spending.  Would American voters have voted for a “bridge to nowhere” that served a few dozen people at the cost of over $100 million?  Would Alaska voters have supported the plan if they had to pay for it?  How about the voters of Ketchikan?

2.  Caplan argues that voters are not just misinformed on trade, but hold cognitive biases that are not even shaken by college economics courses that extol the virtues of free trade.  I agree, indeed I’m just as frustrated as Bryan.  But I also think there are many false beliefs that can be easily shaken.  If I talk to an average person about deregulating taxis, I find they are initially skeptical.  “Wouldn’t a lot of unlicensed taxis be dangerous?  What if the drivers are not well trained?  Or if the cars are not well-maintained?”  When I explain that the debate is about market access, not safety, and that it is about whether a cartel should be able to artificially prop up prices by making taxis much harder to find, there is an almost universal change in mind.  Thus in the area of occupational licensing, the problem really is ignorance, not simply biases.  Peoples’ instincts favor free entry into an occupation, at least if minimum quality standard are met (I admit, that’s a tougher problem.)

3.  And I also think that ignorance plays bigger role in even trade than Bryan suggests.  I seem to recall that NAFTA was trailing in the polls until Gore debated Perot in 1993.  Gore bought out slides showing how Smoot-Hawley was enacted at the beginning of the Great Depression, and dramatically contracted world trade.  The poll numbers swung in favor of NAFTA (and stock prices rose after the debate— something investors pay attention to.)  When people focus on issues in a serious way they become more rational.  An important part of this argument is the way people defer to authority when the stakes are high.  Sometimes they do this too much as when they (and I) assumed Bush must have had good intelligence on Iraq.)  But many middle-brow voters (who are much more likely to vote than the totally uninformed) listen to “experts” for advice.  Don’t laugh, but Rush Limbaugh is a sort of expert—and I believe he subtly pushes conservative voters toward the more free market Republicans, and away from the high-tax, protectionist Republicans.  Or take the Sunday news shows.  Even if voters have protectionist instincts, when they hear both liberal and conservative pundits agreeing that a xenophobe like Pat Buchanan would be dangerous for the country, it has an impact.  In France someone like Le Pen can get a lot of protest votes, but in an election where he made it to the final round he was absolutely destroyed (by a Turkmenistan-like margin of 82.2% to 17.8%.)

4.  It is true that most voters oppose a unilateral move toward free trade, but the same could be said for non-economist elites.  And if we are to have rule by elites, it won’t be by economists, it will be by lawyers (economists are mostly too idealistic and naive for the rough and tumble world of politics.)  But there is also good news in Bryan’s book; a graph on page 69 shows that most American voters support free trade agreements with other countries.  If this is to be believed, then the reason we don’t yet have a free trade agreement with the EU is not the voters, rather it is opposition by powerful special interest groups such as farmers.

5.  I also think that poll results must be interpreted with extreme caution.  Bryan criticizes the public’s responses to 11 survey questions about whether this or that factor plays a “major” role in hurting the economy.  In 10 of 11 cases the public leans more toward the “major” end of the spectrum than economists.  And indeed their answers do seem silly if you think of “major” in a relative sense.  Economists are used to minimizing big things–i.e. defense spending is “only” 4% of GDP.  But average people don’t think that way.  Obviously if they named 10 of 11 factors as having major roles, they meant something more like “important in an absolute sense”, or “has a multi-billion dollar impact on GDP.”  Indeed, how could it logically have been otherwise?  Another example occurred on page 51 where Bryan argued that a poll (conducted in the late 1970s) found majority support for the proposition that the government had a responsibility to “keep prices under control.”  Given that inflation was very high at that time, and given that most voters don’t understand the difference between relative and absolute prices, it is not at all clear that this was a call for Soviet-style price fixing.  It might have merely reflected frustration that the government wasn’t controlling inflation.  Bryan also mentioned a poll showing most drivers consider themselves “above average.”  Isn’t it likely that people differ in their view of what constitutes “good driving?”  I think most drivers are above average–by their own criteria.  Most young men are above average at aggressive driving, and most 55 year old women are above average in terms of safety.

Bryan mentioned four major cognitive biases, but left out the one I view as most important, elasticity pessimism.  I find almost all students think demand curves, and the supply of labor, are very inelastic.  If so then there is no good argument against the welfare state on supply-side grounds, which are the only valid grounds of opposition for a pragmatic libertarian.  But this bias toward the welfare state is (very fortunately) partially offset by conservative values, especially values like hard work and “just deserts.”  Pragmatic libertarians like me may not put much weight on these values, but without them there would be little opposition to massive tax increases.  The supply-side argument is just too counterintuitive.  Of course this is the pragmatic libertarian/social conservative coalition that the Republican Party struggles to hold under one big tent.)

Even with all these caveats, Bryan’s book is fairly persuasive at a theoretical level.  So then why do I still cling to my support for direct democracy despite Bryan’s claim (p. 195) that the combination of cognitive biases are “as bad as it gets” in terms of making sound public policy decisions?  Perhaps because I am not a democratic fundamentalist.  I don’t support democracy because it is “fair” but rather because it seems to work well.

This led me to rethink the Swiss model.  Maybe Switzerland’s success has more to do with decentralization than with direct democracy.  Or maybe decentralization and direct democracy fit together like a hand and glove.  Maybe you need a decentralized political structure, where 95% of goods are imported, for voters to see the merits of free trade.  As we will see, both factors seem to play a role in Switzerland’s success

Part 2:  The Swiss Experience.

There are no perfect examples of direct democracy in the modern world, but one country comes far closer than any other, Switzerland.  During the 20th century, almost one half of all national referenda in the entire world occurred in Switzerland.  And there are two other factors that further contribute to Swiss democracy, it is a relatively small country and it is highly decentralized.  Aristotle said that no country could remain well-governed if its population exceeded 100,000.  As we will see, recent European history provides some support for Aristotle’s hypothesis.

Switzerland is still a fairly neoliberal economy, but no longer exceptionally so.  Swiss voters never moved as far toward socialism as most other countries, but have also been less aggressive with neoliberal reforms since 1980. In addition, Switzerland has long had much lower tax and spending levels than its immediate neighbors, which suggests that under direct democracy voters may prefer less egalitarian policies than those prevalent in Western Europe.  And surveys suggest that Switzerland is the second happiest country on earth.  Are these facts related?  A recent study by Inglehart (1990) found a strong correlation between happiness and democracy, and later studies suggested that the causation ran in both directions.

Until recently, happiness researchers assumed that aggregate happiness in most countries remained stable over time.  An important recent study by Inglehart, et al (2008, p. 266), however, reports strong evidence that happiness has been rising throughout much of the world between 1981 and 2007, and also suggests some reasons why:

Like democratization, social tolerance broadens the range of choices available to people, thus enhancing happiness.  Accordingly, Inglehart and Wetzel (2005) found that support for gender equality and tolerance of outgroups were strongly linked with happiness””not just because tolerant people are happier, but because living in a tolerant society enhances everyone’s freedom of choice.  Similarly, Schyns (1998) argued that gender equality is linked with happiness.

During the late 1980s and early 1990s, dozens of societies experienced transitions to democracy that enhanced freedom of expression, freedom to travel, and free choice in politics.  Moreover, from 1981 to 2007, support for both gender equality and tolerance of outgroups increased substantially in most of the countries monitored by the [World] Values Surveys (Inglehart & Welzel, 2005).  Furthermore, during the past two decades, low-income countries containing fully half the world’s population have experienced one of the highest rates of economic growth in history, allowing them to emerge from subsistence-level poverty.  By a favorable combination of circumstances, societal changes of the past two decades have increased both the prosperity of people in less-prosperous societies and the political and social freedom of people in middle-income and high-income societies, enhancing the extent to which people in both types of societies have free choice in how to live their lives.  We hypothesize that these changes have been conducive to rising levels of happiness within entire societies.

Thus it is no surprise that I found happiness to be much more highly correlated with liberal values than with economic variables, others have found the same.  One potential problem with their findings is that the causation between democracy and happiness could run in either direction.  Frey and Stutzer (2002, p. 424) noted that:

“For Latin America and Russia, one study (Graham and Pettinato 2001b) indeed identified a mutual dependence of pro-democracy and pro-market attitudes with well-being: both raise happiness, but happier people are also more likely to have pro-democratic and pro-market attitudes.  With due caution, it may be hypothesized that, for the respective respondents, there is a virtuous circle in which attitudes favorable to democracy, to the market, and to life satisfaction, reinforce each other.”

Let’s assume that there is at least some causation running from democracy to happiness, which seems plausible given that much of the huge wave of democratization in the late 1980s and early 1990s was exogenous, as events in one country quickly impacted its neighbors.  This still wouldn’t tell us whether or not moving from representative democracy to direct democracy further boosts happiness.  Fortunately, the extent of direct democracy varies considerably across the 26 Swiss cantons.  Frey and Stutzer (2002, p. 425) report that:

“the extent of direct democratic participation possibilities exerts a statistically significant, robust, and sizable effect on happiness over and above the demographic and economic determinants normally taken into account.  When the full variation in the institutional variable is considered, i.e. when individuals in the canton with the highest democracy index (Basel Land) are compared to citizens in the canton with the lowest direct-participation rights (Geneva), the former state with [has?] an 11-percentage-points higher probability that they are completely satisfied.  This effect is larger than living in the top rather than in the bottom income category.”

If the result in the last sentence is valid, it would suggest a very powerful relationship between direct democracy and happiness.  And they also report (p. 425) results from a number of other interesting studies of both the U.S. and Switzerland showing, among other things:

1.    Direct democracy leads to greater responsiveness to voter preferences.
2.    Direct democracy leads to lower levels of government expenditure, except perhaps on education.
3.    Direct democracy results in public services being provided at lower costs.
4.    Direct democracy leads to higher per capita incomes.

I have noticed that intellectuals are fond of the maxim that “democracy is the worst system, except for all the others.”  I gather that intellectuals who say this mean that we would be better off being ruled by a “philosopher king” rather than a bunch of mostly ignorant voters, but alas “power corrupts” (another overused cliché) and thus we need democracy to insure that the government is responsive to the wishes of the voters, and doesn’t become highly repressive.  I think this view is wrong.  Democracy is the best system, period.  It is the best way of deciding what to do, and it is the best way of insuring that the government actually does those things.  This seems to defy common sense, which is why I included it in my list of economistic views.  Indeed it is so difficult for intellectuals to forgo their strong prior belief that they know best, that even economists who believe that financial markets aggregate economic data efficiently, and can forecast better than experts, are often reluctant to extend that hypothesis to political markets.  In the US economists vote 3 to 1 for Democrats.  Rule by economists would push us toward the high-tax European model.  Fortunately, voters don’t allow this to happen.

Yes, there are many examples of where democracy failed to produce the optimal result, but these failures are anecdotal.  I know of no systematic study that shows experts make, on average, better public policy decisions than voters.  In the one example of somewhat direct democracy during the 20th century, we see voters who remained skeptical of highly statist economic policies, even when they were in vogue among policy experts and intellectuals.  Of course Switzerland has many flaws.  Its behavior toward Jews  during WWII was in some respects disgraceful, as was true of virtually all European countries, and the US.   But consider one other huge success””Switzerland remained neutral in 1914.  In contrast, many other European armies were marching off to senseless slaughter with the enthusiastic support of the overwhelming majority of their public intellectuals.  It also has lenient drug laws, which casts doubt on the view that direct democracy would lead to a nightmarish regime of socially repressive polices.

The concept of democratic accountability goes beyond voting.  It is almost impossible for governments ruling over a very large population to be responsive to voters’ needs in any sort of efficient way.  This problem of “diseconomies of scale” in governance has been well hidden because of two historical facts.  First, prior to WWII larger countries did have two important advantages, a large domestic market, and a greater ability to deter invaders.  Neither advantage is important for modern day members of the EU and NATO (or other developed country treaty organizations.)  Second, the extraordinary success of the U.S. has led to complacency about the ability of large countries to be successful.  (I say “extraordinary” because one would expect a large, ethnically diverse country to be closer to the middle of the pack in per capita GDP.  Also note that the U.S is a very happy country, despite press reports to the contrary.)  In recent years, however, there has been a little-noticed trend for the smaller countries in Europe and Asia to rise up the charts and surpass the bigger countries like Japan, Germany, France, Britain and Italy in per capita GDP.  As for why the U.S. has been so successful thus far, it first achieved its great success with a more decentralized framework of federalism, has been relatively open to referenda, and had a large economy when large domestic markets were still very advantageous.  We may now be coasting on our past success, and may have missed out on important parts of the neoliberal revolution.   Earlier I argued that the U.S. will eventually need to break up into an EU-type arrangement.  Once again, Frey and Stutzer (2002, p. 426) have something interesting to say about decentralization:

“The study on Switzerland (Frey and Stutzer 2000) measured the extent of local autonomy by an index based on survey results.  Chief local administrators in 1,856 Swiss municipalities reported on how they perceived their local autonomy using a 10 point scale.  The estimate reveals a statistically significant positive effect of decentralization on subjective well-being.  For local autonomy, the proportion of people who indicate being completely satisfied with life increases by 2.6 percentage points, compared to a situation in which the communes are one standard deviation less autonomous vis-à-vis their canton.”

It sounds good, although the thought of living with a bunch of Swiss who are “completely satisfied with life” makes me very depressed for some reason.

I am not saying that direct democracy is perfect.  I recall reading that Switzerland has a lot of petty paternalistic regulations.  Perhaps this is why it looks so pretty.  But this should come as no surprise as we observe the same thing in American gated communities—perhaps the closest American analog to direct democracy.  I find towns that ban purple houses and pick-up trucks in the front yard to be boring and sterile.  But I suppose people would tend to sort into the types of communities they liked best.  In any case there are worse things in the world than gated communities, much worse.

One final comment.  Although we disagree about democracy, I actually agree with 90% of Caplan’s book.  I just interpret some of the evidence differently.  One area I didn’t discuss was the trendy “self-interested voter hypothesis.”  We both agree it is nonsense.


Tags:

 
 
 

32 Responses to “Why are the Swiss so happy?”

  1. Gravatar of Richard A. Richard A.
    20. September 2009 at 10:40

    If my memory is not playing tricks on me, Pat Buchanan was a free trader in the 80s. He seems to have done a flip flop on this issue about a couple of decades ago.

    I’m am a pessimist when it comes to the future of free trade in the US. The Japanese electorate is much better educated than the US electorate and this gap seems to be widening. This will make it easier for the Japanese government to move closer to pure free trade.

  2. Gravatar of Byung Kyu Park Byung Kyu Park
    20. September 2009 at 11:45

    If my memory is not playing tricks on me, Pat Buchanan was a free trader in the 80s. He seems to have done a flip flop on this issue about a couple of decades ago.

    It may have something to do with the fact that “free trade agreements” such as NAFTA have little to do with laissez-faire free trade. It has more to do with propping up global corporations by using the power of the state. In an ideal world, such “free trade agreements” shouldn’t even be necessary as protectionist tariffs should have been an affront to our liberty and property—the way some of our Founding Fathers thought Bill of Rights shouldn’t have been necessary because they thought it would have been clear to any thinking person that the federal government was limited to the powers enumerated in the Constitution, which did not include trampling on these natural rights.

    By the way, I’m not understanding the “xenophobe like Pat Buchanan” bit. How is Pat Buchanan a xenophobe? I’ve skimmed through his Wikipedia page looking for anything that suggests that Pat Buchanan is a xenophobe or racist set in his ways, but it seems like the only thing constant in his view is “America first” (for one, he supports the U.S. support of Israel as long as it is consistent with our best interests), and really, shouldn’t that be the view of every American politician? (And given our relatively open immigration policy, members of any ethnicity is welcome here—as long as they become an American who stands for America first, not their old allegiances.)

  3. Gravatar of Thorfinn Thorfinn
    20. September 2009 at 12:09

    The Swiss do have a good record on direct democracy. However, California, India, and Indonesia have a much more mixed record. The Indonesia example is especially instructive, as there authorities implemented serious decentralization not too long ago, so you can compare things before and after the change. The record isn’t great, which doesn’t bode well for your thesis.

    When you move power away from centralized authorities to local powers, it’s much harder to get information. We all have a reasonably good idea of what Obama is up to. I think few of us know what our local county government is doing; this allows for sizable corruption by local power-brokers (this is what happened in Indonesia). There are economies of scale in terms of accountability and monitoring that go against the problems you raise.

    On the other hand, India has a fairly good experience decentralizing (expanding authority to village and municipal governments, increasing the number of states). But you also see greater variation; before, the country rose or fell on the quality of national politicians. Now, different areas have different quality politicians, and some areas lag behind others. Not that this is a bad thing; it’s resulting in a ‘race for the top’ in terms of improving governance.

    Rather than taking n = 1, attributing all success of Switzerland to its political system, and declaring decentralization a panacea (as in the study you cite); I think it’s more reasonable to say “it depends.” The Swiss model works because people are very local minded, generally honest, and are much more homogenous at the local level than nationally. India, though not homogenous at any level, at least had robust local institutions and was at such a centralizing extreme that moving the other direction could yield results. The example most similar to the US as a whole, California, is very discouraging.

    Any reasonable empirical analysis of decentralization needs to compare differences in outcomes after changes in decentralization status across a panel of countries. Note that such analyses haven’t even shown any positive impact of democracy; let alone the finer measures you advocate (democracy does seem to be useful for lowering the variance of growth, if not raising the average rate).

    Making the US go “EU” style seems like a solution without a problem. What problems, if any, will that solve? On many issues, the federal government is simply more competent, accountable, and monitored than state governments. Is there a trend for “smaller countries” to gain in GDP? Really? The top three have generally been China, India, and Indonesia. If the US is ‘coasting on its past success’; you have one hell of a counterexample to deal with.

    NCLB seems to have improved things. State politicians are, basically, part-timers with little ability or interest, who actions go practically unmonitored. The death of local journalism (Schulhofer-Wohl has a good case study on this) will make it even tougher to see what local governments are doing. The regional differences across Europe that make national governments make sense are nowhere near present in America.

    As a sidenote, I’m always shocked by economists who think that anti-trade positions can be equated with “irrational.” I think the evidence is clear that free trade has both aggregate and distributional outcomes–it raises the growth of everyone, while hurting particular groups. In theory, we can tax and redistribute to that everyone, including those groups, are better off. In reality, we don’t do that, and many people are worse off as a result of free trade. It’s entirely rational for them to be anti-free trade.

  4. Gravatar of azmyth azmyth
    20. September 2009 at 13:45

    I believe that the more important job of economists, even more than advancing the field through research, is to teach. I’m not confident in the ability of a republican form of government to seriously constrain the popular will. As worrying as the tyranny of the majority is, the best defense isn’t a constitution, it’s building broad support for a pluralist society. If people didn’t really want freedom of speech, etc, it would be gone within a couple of decades. Constitutions are vitally important, but they depend on the spontaneously generated rules and cultural norms of society to be effective. Some countries stick with a constitution for hundreds of years and some change them frequently. The reason the U.S. Constitution is so effective is that Americans practically worship the thing.

    I think the median voter hypothesis is about 80% as true as the efficient market hypothesis. Politicians can exploit rules to seek rents when the stakes are low. When the big stuff rolls around, society is controlled by the masses (or those with the most capacity for violence if one is in such a society).

    So what do we economists do? Teach the populace the counter intuitive way economists think. Try to promote liberal values in society. Hand out copies of Hayek and J.S. Mill and hope for the best. One thing I try to teach is trust in bottom up orders instead of relying on rules imposed from the top. Why should politics be any different than economics?

  5. Gravatar of ssumner ssumner
    20. September 2009 at 13:47

    Richard, The Japanese may be better educated, but what kind of education? Are they taught that free trade is good? I doubt it.

    Pyung, A lot of conservatives have argued that Buchanan is against free trade with non-white countries, and anti-immigration. I haven’t studied the issue in depth, so perhaps that perception is wrong. I will say, however, that almost everything I hear Buchanan say tends to fit well with this view. I seem to recall the National Review did a piece on him a few years back, which argued he was anti-semitic as well.

    Thorfinn, I agree that India has about the worst form of democracy possible. I recall reading that the national government picks textbooks that must be used in every district–in a country with 1.1 billion people! They have one railroad company (government owned) with more than a million employees. India is as far from Switzerland as possible.

    California is also too big. Because it has a nice climate politicians are able to exploit the fact that people want to live there even if the taxes are much higher than Texas. They would be better off with small local governments, rather than monstrosities like the LA school district.

    I don’t know much about Indonesia. It is also far too big, and its democracy is very young.

    Regarding local corruption, we don’t pay attention because counties have no power in the US. In Switzerland cantons set the income tax rate. If that was done in the US you can be sure that people would pay attention. But I would never deny that culture also plays a role in corruption–compare Minnesota and Louisiana.

    You said;

    “Rather than taking n = 1, attributing all success of Switzerland to its political system, and declaring decentralization a panacea (as in the study you cite);”

    The whole point of this overly long essay was to show it’s not a n = 1 situation. There are cross-sectional studies of Swiss cantons that show the effects of greater democracy and decentralization. And the evidence is statistically significant. So people can’t say “the Swiss are special” because these studies compare Swiss against other Swiss. But you are right that time series evidence would also help.
    You said;

    “The Swiss model works because people are very local minded, generally honest, and are much more homogenous at the local level than nationally.”

    How did the Swiss develop these good characteristics? Perhaps it was because they were given responsibility, rather that simply told to follow the orders of the central government.

    You said;

    “Making the US go “EU” style seems like a solution without a problem. What problems, if any, will that solve?”

    Perhaps the problem that we have MTRs that exceed 40%, and will soon exceed 55%. They may soon even exceed Sweden’s top rate. Conservative Texas and liberal Washington State and moderate New Hampshire all have zero state income taxes. That’s a choice I’d like to have.

    You said;

    “Is there a trend for “smaller countries” to gain in GDP? Really? The top three have generally been China, India, and Indonesia.”

    People of those ethnicities live in Singapore, and are 5 to 10 times as rich as the others. If you look at Europe and East Asia you will see that many smaller countries now exceed the big countries in per capita GDP. Remember when Britain was richer than Ireland? Remember when Germany was one of Europe’s richest countries? Now it is the poorest of the three German majority countries of Europe. Admittedly the crisis has hit some high-flyers like Iceland. So perhaps we need more time to really tell if the trend is real.

    You said;

    “NCLB seems to have improved things. State politicians are, basically, part-timers with little ability or interest, who actions go practically unmonitored. The death of local journalism (Schulhofer-Wohl has a good case study on this) will make it even tougher to see what local governments are doing.”

    I’m not surprised, as they have little power. Your observation doesn’t affect my argument at all, as I propose turning states into national governments. Many European countries are the size of American states, and voters do pay attention when they must give 50% of their money to the government.

    You said;

    “As a sidenote, I’m always shocked by economists who think that anti-trade positions can be equated with “irrational.” I think the evidence is clear that free trade has both aggregate and distributional outcomes-it raises the growth of everyone, while hurting particular groups. In theory, we can tax and redistribute to that everyone, including those groups, are better off. In reality, we don’t do that, and many people are worse off as a result of free trade. It’s entirely rational for them to be anti-free trade.”

    What you say is true, but read Pop Internationalism. 95% of anti-free trade arguments are not rational, they show the speaker does not even understand Ricardo’s comparative advantage theory. It’s fine to understand the argument but then reject it, but few non-economists understand the argument. They think trade is only beneficial if there is a “level playing field,” which is a silly argument.

  6. Gravatar of ssumner ssumner
    20. September 2009 at 13:54

    azmyth, I mostly agree. A few comments on constitutions:

    1. Latin American countries often have constitutions. These are only as effective as the underlying political system (which means they often are not very effective.

    2. The US constitution does guarantee free speech, but there are many cases were we do not adhere to the 1st amendment. I think it is possible that political speech is a bit freer here than in a European country (people don’t go to jail for denying the holocaust) but overall our record inst that good. There are massive limitations on commercial speech in all sorts of areas, and even political speech is regulated by campaign finance laws.

    I think you probably agree with both of these points (as you mentioned the importance of cultural norms), but I just wanted to emphasize them. It is one reason why I don’t fear direct democracy. But I also think constitutional protections of a few basic liberties can go hand in hand with direct democracy in other areas. I am not a dogmatist.

  7. Gravatar of Thorfinn Thorfinn
    20. September 2009 at 14:54

    Sumner,

    Your points are all well-taken, but you seem to be rather aggressive in excluding cases of decentralization that don’t work out. Once you take out everything that’s “too big” or “too diverse”; you’re left with only Switzerland. And I’m not sure that variations within Switzerland explain very much. Nor that Singapore shows conclusively that small governments are the best. Note that India is both the biggest and best performer out of all South Asian countries.

    Applying to the US; if you want to turn states into nations, California is very much the model. The flip side of low taxes in Texas is high taxes in New York and New Jersey. Cities aren’t going anywhere, and states which control big cities tend to tax like crazy.

    It seems odd to radically alter American political structure to save a few bucks in taxes, especially when similar reform has been tried and found wanting virtually everywhere but a small European mountain state. Alternatively, we could just cut the entire military budget.

  8. Gravatar of TGGP TGGP
    20. September 2009 at 16:15

    It isn’t just economists who are more rational: Caplan’s book shows that education is one of the best predictors of “thinking like an economist”. At his own blog he showed that including IQ bumps education down to the second most important factor.

    I don’t consider your claim that voters were less statist than experts in the past unless you provide the sort of evidence Caplan has. It is easy to say “Look how many smart people were communists/fascists!”, but that doesn’t show whether the masses a whole were less statist than the elite as a whole. I agree with you that lawyers would probably outcompete economists (and Caplan does have a great lawyers vs economists anecdote relevant to that), but we don’t know if lawyers think less like economists than the general public. The GSS probably has an occupational code for them that could be checked out.

    Speaking of Switzerland and gender equality, they were one of the last first world nations to grant women suffrage.

  9. Gravatar of q q
    20. September 2009 at 16:16

    before you laud direct democracy too much, look at how the referendum process has worked in washington state. read a bit about tim eyman. specifically look at initiative 695, which set state vehicle excise taxes at $35 statewide. (i left washington state in 2000, so i haven’t been keeping close tabs on him.)

  10. Gravatar of StatsGuy StatsGuy
    20. September 2009 at 16:40

    Generalizing from the Swiss case requires some care. All that a cross-canton relationship means (even if the causality is really real) is that in the context of 600-700 years of Swiss history, with a population that is highly homogenous (with the only major split being between protestants and catholics), with long-standing institutions, with a small military, with an economy in which there is no major racial group that entered with a disadvantage (which we know can become self-reinforcing due to many social/informational dynamics), there is at this moment in time a relationship. Maybe. If you believe the polling methods. At 95% confidence…

    In a sense, it’s n=26, but in another sense it’s still n=1.

    But does that mean that direct democracy is a better option in Rwanda? In Iraq? In Lebanon? It is easy to practice tolerance when that tolerance is not severely tested. In all of these societies, it is striking how rapidly a seemingly tolerant society destroyed itself when a minority ceded power to a violent majority.

    Even in the US, one of the things that allows direct democracy to work in the only real place in which it is widely practiced and might be said to work (local government) is the exit option (combined with federally enforced individual liberties and funding). It’s trivial to write a simulation that shows that even very slight preferences to dwell in the vicinity of similar people (or people with similar cultures) can lead to sharp divisions over time. And, indeed, we see such self-selection playing itself out in media fragmentation and the increasing polarization of political districts (aided, of course, by gerrymandering).

    In the international setting, the exit option is diminished – if not by actual barriers to emigration or immigration to desirable locations, then by the high costs associated with leaving one’s own physical assets, language, family, and friends behind. This does put a hamper on one of the primary release valves for living in a world where being a minority is unpleasant, or even deadly.

    Historically, btw, one of the greatest examples of the exit option being necessary for democracy is Pakistan/India, and the massive self-segregation of muslims and hindus. Consider the level of violence today… And imagine what it might have been with greater cultural “integration”.

    As to the (lack of) success of referenda in the US, others have addressed that…

  11. Gravatar of bil. A. bil. A.
    20. September 2009 at 18:54

    Thorfinn writes:
    “The flip side of low taxes in Texas is high taxes in New York and New Jersey. Cities aren’t going anywhere, and states which control big cities tend to tax like crazy.”

    I would point out here that Houston is the 4th largest city in the U.S. The Dallas-Fort Worth-Arlington MSA is the 4th largest MSA in the country at 6.3 million, and the Houston-Sugar Land-Baytown MSA is the 6th largest at 5.7 million. Having 2 of the top 10 MSAs would seem to suggest that Texas doesn’t support your hypothesis.

  12. Gravatar of bil. A. bil. A.
    20. September 2009 at 19:11

    Scott,

    I share your support for decentralization and interest in the Swiss model, however recently I read Forrest McDonald’s _Novus Ordo Seclorum: The Intellectual Origins of the Constitution_, and in one part of it he offers some historical support for the idea that it was the failings of unconstrained democracy on the State level* under the Articles of Confederation that led many to support the call for the constitutional convention leading to the U.S. Constitution giving more power to the (generally less democratic at the time) central government.

    *such as debtors using the State governments to regulate in their favor against creditors, etc…

    Also, I understand that it is relatively difficult to become a Swiss citizen, do you think that this “anti-immigration” stance is an important aspect of their success as a highly democratized confederacy?

    P.S. If you have time, and haven’t already, you should read about Robin Hanson’s idea of futarchy before you go to GMU:
    http://hanson.gmu.edu/futarchy.html
    You might like it, it is all about using betting markets to determine all of government policy, not just the monetary base!

  13. Gravatar of mbk mbk
    20. September 2009 at 19:14

    The EMH and the wisom of the crowds hypothesis are of the same kind. Both predict that the aggregate “opinion” is more correct that any randomly chosen individual opinion, be it expert or otherwise. If the wisdom of the crowds model is wrong, so is the EMH.

    Or put in a more subtle way, the EMH may fail occasionally for the same reasons the wisom of the crowds may sometimes fail to materialize. Both EMH and democracy require individual decisions where the deviations from the “truth” by the individual actors are uncorrelated amongst themselves.

    In situations where all information comes from one or few sources which are highly correlated (ratings agencies, group think amongst expert elite …) there is no reason to expect either EMH or democratic decisions to give “true” results. In addition to this, the system may fail for reasons of resonance (by slight abuse of language) , key words chaos theory and synchronization (see Steven Strogatz’s nice book “Sync”).

    The key to democracy is not to expect the voter to be rational (even the EMH does not assume rational market participants – it relies on the aggregate to be correct). The key is to get the voter to have unsynchronized, randomized if you will, opinions. This means, diverse, different, and therefore “random” sources of information and biases. By this logic one must presume that the worst biases be the ones from experts, because they definitely are more correlated as a group: sometimes very right, sometimes very wrong.

    Finally, both EMH and democracy are part of a wider issue which is how to build optimal systems from faulty parts. It is a standard fact of engineering that near perfect systems (say laptops) can be built from parts with dreadful tolerance ranges (ask anyone who ever saw a transistor data sheet). One of the most curious studies I have come across was about, how many of a bunch of faulty parts you havew at hand, should you average to get the best (“truest”) end result on average? Answer, about half of all parts you have at hand should be averaged (if you averaged _all_ parts, you would include all extremes as well. So you want to get the highest achievable smoothing by numbers, with a lowish risk of including biased extreme values). See here:

    http://focus.aps.org/story/v9/st32

    Now by this logic the older democracies seem to almost have self-optimized: we usually have about 50% voter participation, and ruling parties making up ca. 50% of the votes… Yes, it’s not a random selection, but a curious outcome nonetheless, would one not expect positive returns and the perfect party emerging after a series of elections?

    The elephant in the room which has not been discussed at all here is, one key question is on scope of government vs. method of choosing government – the range of things a ruler is allowed to decide on. What should be the scope of public decisions? What should be the scope of individual decisions? If the public scope is low then autocracy may not feel too obtrusive. If the public scope is high then even democracy may feel oppressive. etc etc. Next in line is the confusion of “What” with “How” . No question, experts are usually the best to tell you “how” to achieve what you want. But this “what” people want, is a question of discretionary volition, not amenable to improvement by expert opinion. If say people truly want an insular protectionist island, then this act of volition is on a different plane from any expert judgment – the expert can only predict what would then happen economically.

    To put it to the extreme: Experts have no place as political decision makers, only as tools to achieve said discretionary political will (and by implication as educators on the various trade-offs).

  14. Gravatar of rob rob
    20. September 2009 at 20:08

    I think the creation of more prediction markets would complement more direct democracy. For instance, if there were a betting market on what % of GDP will go toward healthcare expenditures in 10 years. One could use such a market to help predict and monitor the consequences of any changes in the law. More importantly, people might be driven to be more aware and more intellectually honest with themselves about the consequences of any new legislation. People might really start to vote their pocket-books a bit more than their systematic biases. Maybe each initiative should be required to spell out metrics for the objectives hoped to be achieved by the initiative. Those metrics would then form the basis for that prediction market. Perhaps if the actual metrics fell short of the claimed objectives by some degree it would be automatic grounds for a new initiative on the issue. I suppose for social issues like gay marriage, it would nonsensical to have metrics attached. Or maybe not. If people against the measure held the position that gay marriage would undermine family values, a metric such as increase/decrease in the number of straight marriages or divorces over the next ten, twenty years would be a reasonable metric to correspond with that initiative. So the proponents and detractors of each measure should get to propose metrics to measure the consequences of each initiative.

    The question is how do we get voters to actually vote in their own self interest?

    I know you don’t like this sort of thinking, but it reminds me of Socrates’s (in favor of philosopher-kings) quote:

    “The problem with letting the people vote is that either:

    1) People won’t know what they really want; or worse:
    2) People will know what they want and vote for it.”

    (Paraphrasing from distant memory)

  15. Gravatar of rob rob
    20. September 2009 at 20:28

    mbk,

    The big difference between the EMH and wisdom of crowds is that the EMH deals with markets, where each individual is forced to put their money and thus their self-interest on the line. In voting, most people realize their vote doesn’t really matter statistically. You can simply vote your emotions with nothing to lose. I voted for Kinky
    Freidman for Gov of Texas last time around simply because he was cooler than the rest, and it made me feel better about voting for him than the other dillweeds running. But would I have voted for Kinky if I thought my vote would push him over the top and we would really end up with Willie Nelson as energy-csar for the state? Well, possibly: but not if I weren’t insane.

    I voted for Obama because I thought, even if he turns out to be a poor president, it felt like the right thing to do. Anyway, he wasn’t going to win Texas, so my vote didn’t matter. But I’ve never invested in the market without tremendous concern over what my individual ROI will be.

  16. Gravatar of mbk mbk
    20. September 2009 at 21:10

    Rob,

    somehow I can’t believe you do not want a positive political outcome for yourself (rational expectations). At the same time your actual behavior may not be rational (bias etc). And as far as I follow the EMH, it does only require rational expectations, but not rational behavior. That means specifically the EMH, as one of a class of neutral models, does _not_ require that anyone be “good” by default, it only requires that all individual nutcase decisions, ideally drawn by chance, cancel out to reveal the collective “truth”. The EMH indeed woks best if deviations from the “truth” by individuals are truly random. Translated to the case of democracy, that allows for the occasional protest vote, disillusionment etc as fringe of a distribution.

    I see the prime problem with our current implementations of democracy in the West in 1- the scope of government (regardless of how we get the particular government) and 2- confounding of the supposedly separated powers, following the critiques by (again) Hayek. The executive (CEO) is not supposed to have influence over the legislative that makes the rules, and in addition the legislative has two separate jobs, one to make laws, the other to act as a board of directors to supervise the management (executive), i.e. the current executive’s administration of the state apparatus. All in the interest of the shareholders (voters). Say in the US the current president is widely expected to be able to solve healthcare legislation etc. Well in principle the US president is supposed to be the executive. He is not supposed to be a legislator. He is not even supposed to do much domestic politics.

  17. Gravatar of Current Current
    20. September 2009 at 23:59

    I think Statsguy is right to point out that Scott is still making claims from a single sample.

    It doesn’t matter if the other examples of direct and decentralized democracy that Thorfinn gives are not as he thinks. The point is that Scott is still relying on one example.

    Scott discusses the various Canton’s within a successful democracy that uses decentralization and direct democracy. But, this is begging the question. It presumes that such a state will be successful overall.

    Suppose we have a state that implements direct democracy and decentralization but becomes less successful. It may still be the case that the parts of it that have more decentralized power are more successful, and the people there happier.

    Trade demonstrates the problem. If a canton depends on a large percentage of external trade then it isn’t really like a small centralized country. So, the problem is how a federation of such cantons can be constructed that is successful?

    I’m not arguing here for rule by elites. I’m arguing for constitutional representative democracy and the rule of law. The idea that “Democracy is the best system, period.” is ridiculous. As many have pointed out it has been the rule of law that has been far more important in history. Law, Constitutions and division of powers are anti-democratic, but they have all being very successful.

  18. Gravatar of Current Current
    21. September 2009 at 02:01

    Thorfinn: “As a sidenote, I’m always shocked by economists who think that anti-trade positions can be equated with ‘irrational.’ I think the evidence is clear that free trade has both aggregate and distributional outcomes-it raises the growth of everyone, while hurting particular groups. In theory, we can tax and redistribute to that everyone, including those groups, are better off. In reality, we don’t do that, and many people are worse off as a result of free trade. It’s entirely rational for them to be anti-free trade.”
    Scott: “What you say is true, but read Pop Internationalism. 95% of anti-free trade arguments are not rational, they show the speaker does not even understand Ricardo’s comparative advantage theory. It’s fine to understand the argument but then reject it, but few non-economists understand the argument. They think trade is only beneficial if there is a ‘level playing field,’ which is a silly argument.”

    You’re missing Thorfinn’s point, which is a good one.

    As Jeffrey Friedman would say discussion about trade involve understanding the complex and counter-intuitive principles of economics. They involve counter-factual reasoning. Most people and I’d say most politicians can’t do this. Inevitably most arguments against free trade are ill-founded.

    However, Thorfinn’s point is about interest. It may be in the interest of someone who understands the argument for free-trade well to oppose it. If you are a person whose job is likely to be lost due to foreign competition for example.

    Even if everyone were to understand the comparative advantage argument – which is hopelessly unrealistic. Then there would still be a sizable portion of the population who would be against free trade because it would disadvantage them.

  19. Gravatar of johnleemk johnleemk
    21. September 2009 at 09:09

    “Then there would still be a sizable portion of the population who would be against free trade because it would disadvantage them.”

    Scott mentioned this in his post — Bryan has pointed out that this notion of the purely self-interested voter is bunk. If we can convince a majority of the populace that trade is in almost everyone’s best interest overall, they will vote for it.

    To the argument that California and New York are counterexamples against Scott’s thesis, I think Scott (or at least I) would say that these states are themselves too centralised. The primary issue is not devolving power to the states; it is devolving power as much as possible. (Hence Scott’s specific reference to county governments in one of the comments here.)

  20. Gravatar of StatsGuy StatsGuy
    21. September 2009 at 09:46

    Actually, Thorfinn’s argument is one of the reasons democracy can be hostile to free trade. Consider the Frieden/Rogowski type arguments.

    Here’s an example:

    Capital is relatively more mobile than labor. Opening up trade and investment allows capital to equalize across nations faster than labor.

    In countries with relatively less labor and more capital, capital outflows imply a distributional loss to labor.

    However, capital ownership is more concentrated than labor. Let’s say 10% of people own 90% of capital. In a pure democracy, with 50%+1 rules, the majority has strong incentives to raise trade barriers to prevent distributional losses.

    The concentrated minority (capital owners) have a couple options. One is to accede (ha ha, fat chance). One is to try to push through free trade against the wishes of the majority (btw, free trade is consistently not popular in polls) by using anti-democratic techniques, like interest groups and campaign financing and lobbying etc. One is to negotiate with labor and offer a bargain (e.g. attain free trade, but offer to help pay for adjustment/retraining costs).

    Now, here’s the counter-intuitive part.

    In the US, which is NOT a democracy, capital owners have chosen option 2: use special interest groups, lobbying, concentrated power. This is possible due to the scale of government/obfuscation/high cost of organizing for non-concentrated interests.

    In Switzerland, one could argue that the small scale and direct democracy make it much harder to use political maneuvering and lobbying, etc. Thus, capital owners have achieved a compromise with labor owners – free trade _in exchange for_ a stronger social safety net (e.g. covering retraining, adjustment costs, and basic needs). In essence, the low cost of organizing labor forced capital owners to engage in Coasian bargaining in Switzerland, while in the US capital owners opted for a confrontational approach (because the structural factors operated in their favor).

    But, as others have noted, Switzerland is an exclusive club. It’s hard to join. The US is easy to join (but getting harder), and make no mistake – the fact that the US is easy to join and that capital ownership is concentrated and powerful are related (good luck getting the causation right). Only recently has the cost of training/social care/education begun to eat away at the obvious gains from immigration (and even then, there’s a strong argument that at the current equillibrium, immigration is highly beneficial).

  21. Gravatar of Current Current
    21. September 2009 at 09:46

    johnleemk: “Scott mentioned this in his post “” Bryan has pointed out that this notion of the purely self-interested voter is bunk. If we can convince a majority of the populace that trade is in almost everyone’s best interest overall, they will vote for it.”

    I certainly agree that voter self-interest is not the only relevant aspect of democracy. The point Thorfinn and I are making though is that it is incorrect to claim that a particular voter is necessarily “irrational” because they don’t agree with free trade.

  22. Gravatar of Current Current
    21. September 2009 at 10:55

    Statsguy,

    You write: “Capital is relatively more mobile than labor. Opening up trade and investment allows capital to equalize across nations faster than labor.

    In countries with relatively less labor and more capital, capital outflows imply a distributional loss to labor.”

    That’s certainly correct. But you ignore the possibility that tariffs are a distributional loss to labour. In many cases they are because they reduce competition and allow producers to charge higher prices. Or because they put consumers at the wrong end of a comparative advantage issue.

    To give an example of the first. I visited England a couple of weeks ago. A relative of mine picked me up from the airport, he worked his whole career in the business of equipment for electrical distribution substations. He pointed out a hotel outside the airport where the bosses of those businesses would meet every year to fix the prices. That was before privatization when the state owned electricity industry had to buy British. After privatization the price-fixing had to end because of foreign competition. Many tariffs are similar to this, the group they benefit is the owners or managers of a particular industry.

    The same sort of thing is true of comparative advantage. The Corn Laws were repealed in Britain because of pressure from industrialists and workers. The farmers and land owners benefitted from the Corn Laws because it spared them from competing with more efficient American producers. This was of great concern to the lower classes because they spent much more of their income on food.

    You also ignore the possibility that labour have an interest in the gains from comparative advantage in the longer term.

    I think you are right about many modern free-trade laws though, which are only beneficial to the masses in the longer term. The masses though don’t support them for this reason. Rather governments support them because of corruption or for ideological reasons.

  23. Gravatar of Bababooey Bababooey
    22. September 2009 at 08:11

    Unlike Switzerland and unlike Caplan’s issue surveys, U.S. voters rarely vote on single “issues”, they vote on people. And voting puts you in a different mindset than casually answering a survey while watching T.V.

    Politicians are an amalgam of beliefs & personal characteristics that they try to hide or fake, so voters need their instincts to key off hints and tells. Worse, most of our political candidates come from a pool of people whose highest desire in life is to boss other people around, so there isn’t much to choose from.

    This is why Caplan’s book seemed, to me, to cover a fantasy world and not U.S. representative democracy. Caplan should have used Swiss referenda instead of surveys to prove that biases sustain themselves through the selective process of voting.

  24. Gravatar of Current Current
    22. September 2009 at 23:54

    Bababooey, I think your right about that. Much of the political science stuff that comes from economics seems to forget about the things you’re mentioning.

    I don’t know if normal political scientists are any better because I haven’t read that much of them.

  25. Gravatar of ssumner ssumner
    26. September 2009 at 08:37

    Thorfinn, You said;

    “Your points are all well-taken, but you seem to be rather aggressive in excluding cases of decentralization that don’t work out. Once you take out everything that’s “too big” or “too diverse”; you’re left with only Switzerland. And I’m not sure that variations within Switzerland explain very much. Nor that Singapore shows conclusively that small governments are the best. Note that India is both the biggest and best performer out of all South Asian countries.”

    I don’t follow your “too diverse” point. Switzerland is diverse in terms of language.

    Some commenters use India against me, saying “look, India is democratic and its a mess.” You are saying “look, India’s big and it’s a success.” I think the world is complicated, and many factors explain a country’s success. Yes, I have oversimplified in my discussion, but that’s hard to avoid. And I think those who disagree are equally guilty. If we were to do some sort of systematic survey–say comparing every single country in the world–then I think democracy and decentralization would come out looking pretty good, but of course I can’t be sure.

    You said;

    “It seems odd to radically alter American political structure to save a few bucks in taxes, especially when similar reform has been tried and found wanting virtually everywhere but a small European mountain state. Alternatively, we could just cut the entire military budget.”

    I don’t see where the Swiss system has been tried, except in Switzerland. I would say that the last 50 years show we can’t just cut the military budget. But if we were 50 nation states we might be able to, as they usually spend about 1/2 as much as we do as a share of GDP. So there is your best way to cut military spending in half.

    Regarding big cities, I predict that with 50 independent nation states the NYC and California top income tax rates would rise from 10% to about 25%. But their combined federal and state top rates would fall from about 50% to 25%. And the tax forms would be simpler.

    TGGP, You said;

    “I don’t consider your claim that voters were less statist than experts in the past unless you provide the sort of evidence Caplan has. It is easy to say “Look how many smart people were communists/fascists!”, but that doesn’t show whether the masses a whole were less statist than the elite as a whole.”

    Here is my evidence. At a time when it was fashionable among intellectuals to be communist (the 1930s and 1940s), the communist party got a tiny percentage of the votes in the US. Even the socialist party didn’t do well. Regarding your previous comment about empirical evidence, Caplan cites studies showing cognitive biases. But that doesn’t prove democracy works more poorly than rule by elite. I like to look at the real world performance of countries:

    Direct Democracy: Switzerland
    Representative Democracy: Rest of Europe
    Quasi-democracy; Singapore, Russia, Venezuela
    Dictatorship; China, Iran, Egypt,
    Totalitarian state: North Korea, Burma.

    Yes, Singapore is an exception, but to me it looks like countries tend to do better when they have a more democratic government. I would be stunned if a complete survey of all 200 countries showed Bryan was right.

    q, I agree, but can also cite cases where the voters are smarter than the experts, such as with tax cuts.

    Statsguy, You said;

    “Generalizing from the Swiss case requires some care. All that a cross-canton relationship means (even if the causality is really real) is that in the context of 600-700 years of Swiss history, with a population that is highly homogenous (with the only major split being between protestants and catholics),”

    Do you have a theory as to why the language groups get along better in Switzerland than in other democratic countries like Spain, Belgium, and Canada? Could it be because they each have a right to control their own affairs? Isn’t that logical?

    You said;

    “But does that mean that direct democracy is a better option in Rwanda? In Iraq? In Lebanon? It is easy to practice tolerance when that tolerance is not severely tested. In all of these societies, it is striking how rapidly a seemingly tolerant society destroyed itself when a minority ceded power to a violent majority.”

    I think Switzerland is a great model for these societies. I have advocated that Iraq adopt the Swiss system, giving the Kurds control over their area, and the same for the other groups. Of course these societies are troubled in all sorts of ways, but a fundamental problem is the tyranny of the majority. That is the beauty of decentralization.

    I agree with you about the exit option and local governments. That is why the suburbs have better schools than NYC and LA. They have less monopoly power than big cities, and must compete for residents. I’d like to see big cities broken up into much smaller units, with county governments handling regional issues like major infrastructure. Let local governments handle schools, or better yet go with vouchers.

    Regarding the media, when I was young all we had was the 3 big networks. And they were awful. Now through the internet we have access to all sorts of unconventional ideas, like TheMoneyIllusion. I think this is a huge advantage.

    You said;

    “In the international setting, the exit option is diminished – if not by actual barriers to emigration or immigration to desirable locations, then by the high costs associated with leaving one’s own physical assets, language, family, and friends behind. This does put a hamper on one of the primary release valves for living in a world where being a minority is unpleasant, or even deadly.”

    That is precisely why I like the Swiss model. If a country is well governed, and minorities are given more local autonomy, then there is less chance that they will be repressed by a overbearing centralized national government. But I do favor a few basic human rights being enshrined in a national constitution.

    I don’t follow your last comment. Are you saying the breakup of British India into several different countries is good? Or bad?

  26. Gravatar of ssumner ssumner
    26. September 2009 at 10:13

    bil A. I agree with you about Texas, it seems better run than a lot of other big states. One reason might be geography. It doesn’t have the physical or cultural amenities of NYC, SF, Miami, and LA, and thus must compete for migrants by have middle class–friendly policies on taxes and housing.

    I’m afraid I don’t know much about colonial history, but am skeptical of historians. Recall that they rate Wilson a great president and Harding one of the worst. They like centralization and big government. You may be right about the debtors, but perhaps there were offsetting advantages.

    I have read Hanson’s work on futarchy, and like it a lot.

    mbk, Those are very good points, and very interesting. I had never thought about the 50% turnout in that way. I also agree that there are similarities between democracy and the EMH, but at the same time I think Bryan Caplan presents some good evidence that voters may have consistent biases about some issues that are counterintuitive, like free trade.

    The question of the proper scope of government activity is also a good one. I think many things now done by government should be privatized.

    rob, I agree with you about prediction markets. Indeed they might even be interesting for moral issues. When someone says “I believe eventually society will come to recognize that X is immoral” it is a sort of prediction about future moral beliefs. Even if I opposed gay marriage, in a prediction market I would bet that the number of states allowing gay marriage in 50 years will be much more than today.

    Current; You said;

    “Scott discusses the various Canton’s within a successful democracy that uses decentralization and direct democracy. But, this is begging the question. It presumes that such a state will be successful overall.”

    I disagree strongly. I think you guys just keep moving the goal posts. First I was told that it is just one country, one observation. I was told that Switzerland was successful for cultural reasons that had nothing to do with direct democracy and decentralization. That explanation would predict that the most democratic and decentralized parts of Switzerland should be no more successful than the others. And yet they are more successful. Now I am told that all these new observations are also worthless.

    You said;

    “Law, Constitutions and division of powers are anti-democratic, but they have all being very successful.”

    I don’t agree. I think laws can be consistent with a democracy. I didn’t say all decisions should be decided by referenda, just lot’s more than right now. I’d like to see income tax rates set by referenda, but not freedom of speech. We have a split under representative democracy, with the legislature only having the right to decide some issues. Why couldn’t we have the same split under direct democracy?

    I now see what Thorfinn intended, but what are the implications of this? Most people benefit from trade.

    johnleemk, Yes, I can’t understand how people were using NY and California as counterexamples. I am saying “big is bad.” So people should be using Vermont and Wyoming as counterexamples. Except that they aren’t as screwed up. Why not divide NYC up into 5 separate governments? Or 50? You can still have transit authorities at the county or state level.

    Statsguy. You said;

    “Capital is relatively more mobile than labor. Opening up trade and investment allows capital to equalize across nations faster than labor.

    In countries with relatively less labor and more capital, capital outflows imply a distributional loss to labor.

    However, capital ownership is more concentrated than labor. Let’s say 10% of people own 90% of capital. In a pure democracy, with 50%+1 rules, the majority has strong incentives to raise trade barriers to prevent distributional losses.”

    I completely disagree. The average worker is better off in open economies than closed economies, by a wide margin. It is not a zero-sum game. Trade makes income far higher than autarchy. Most people benefit from trade. If the average country closed its borders it would go into a deep depression. Look at the countries that don’t trade, how well are they doing? You might say you just want to restrict trade, not end it. But that just makes the damage smaller, it doesn’t go away.

    You said;

    “btw, free trade is consistently not popular in polls)”

    I disagree. Caplan’s book has polling data that shows most Americans support free trade agreements. And after the Al Gore–Ross Perot debate opinion swung in favor of NAFTA, and Congress responded. It shows you how rational voters are when they focus. In any case, the relevant polling data does not come from the US, but from small European nations where free trade is much more popular. Remember that the gains of trade are much higher in small countries, and remember that I believe all countries should be small countries. And finally, remember that by far the most democratic country in the world has pretty free trade, except in agriculture.

    I also don’t agree with your version of American history. The relatively free trade regime set up in the US between WWII and the 1970s was strongly supported by labor, it certainly wasn’t forced down their throats by capital. And capital often supports protectionism in their own industry just as much as labor.

    The relative influence of labor in politics was near a peak in 1965 when our relatively open immigration laws were passed, so I don’t buy that argument either. The Democrats traditionally support more immigration, and they also represent labor.

    bababooey, Caplan’s book just looks at the issue of cognitive bias. But it doesn’t consider how voting might improve outcomes. I agree that empirical studies of real world democracies are the best way of evaluating democracy. What Caplan’s book showed is that there might be problems with democracy, not that those problems are worse than the alternative.

  27. Gravatar of StatsGuy StatsGuy
    27. September 2009 at 05:10

    ssumner:

    “I don’t follow your last comment. Are you saying the breakup of British India into several different countries is good? Or bad?”

    Morally bad, but pragmatically good. And I count myself a pragmatist. The end – two countries that are now more religiously homogeneous – is almost certainly more peaceful than what we would have had otherwise, and the resultant societies (at least, India’s) more stable. However the means were pretty awful.

    I agree that there are many countries that could benefit from the Swiss model. Getting it into place is a challenge; much is dependent on whether you believe deliberate social construction can approximate the effects of centuries of social evolution. To date, our experience with “nation-building” has been mixed. We can count some significant successes – Japan, Germany, S. Korea. But some serious failures as well, particularly recently. S. Vietnam, Iraq, Afghanistan.

  28. Gravatar of StatsGuy StatsGuy
    27. September 2009 at 05:32

    ssumner:

    “The average worker is better off in open economies than closed economies, by a wide margin. It is not a zero-sum game.”

    You are comparing complete openness to complete closure. This is a straw man of the argument above, and not entirely fair. The essential argument presented is one in which optimal readjustment costs to an open economy are partially paid for by winners. My argument (and it’s not really mine, since it’s borrowed) is that the distribution of who pays the adjustment costs depends on the relative bargaining power of domestic actors, and that in SOME situations democracy CAN BE hostile to free trade.

    You note that unions were pro-trade in the immediate post-war era. YES, because at that time the US controlled half of the entire world’s manufacturing base (making out workers more productive and yielding very desirable terms of trade), but ALSO because the US had higher tariffs across the board. As tariffs were dismantled and lowered, and other areas of the world caught up in manufacturing, labor opinion rapidly shifted. Much of this shift came when the US first started to feel the impact of real competition in heavy industries during the oil shocks (Toyota brought this to a head). The story I’m telling doesn’t begin until late 70s through now, and that coincides with the drop in the US manufacturing base, the increase in the size of the finance sector (as well as other service sectors), deregulation, and a slight long term downtrend in capacity utilization rates in general. It also coincides with massive fiscal deficits (during the Reagan years especially) that helped preserve an overvalued dollar by subsizing trade deficits.

    On a relative basis, the US population has been far more protectionist than elites. Your argument is that this is because they are simply unenlightened. My argument is that even if they were enlightened, there are situations when a majority of people lose out in free trade, that distributional losses can for for over 50% of people result in a net loss (even after accounting for overall gains on average), and that a simple-majority-rule system can in those cases be hostile to free trade. My secondary argument is that in such situations, bargaining can occur – indeed, Coasian bargaining is almost certain to occur. The results of this bargaining depend on the structural, cultural, and institutional (aka, political) factors. Outcomes can include repression, burden-sharing, and almost anything in between. Moreover, in some cases, “second best” burden sharing agreements can – in the long run – yield a more open domestic market by decreasing majoritarian opposition to free trade in specific circumstances. I believ ethis is what happened in Switzerland.

    BTW, you are correct that much of the opposition to free trade comes from specific losing industries. In practice, these industries serve as catalysts to motivate mass opinion against free trade – consider the textile industry, the auto industry, and various “buy american” campaigns.

  29. Gravatar of mbk mbk
    27. September 2009 at 07:57

    Scott, thanks for the roses. The scope of government is really the key, in my mind: the decentralization that you mention, so that “minorities can decide on their own affairs”, when taken to the extreme, would then of course be a situation where every _individual_ can freely decide on their own affairs: decentralization down to the the minority of the individual. One does not have to take it to such literal extreme, but battles over public policy usually ignore the first decision that has been already made and take it for granted: that the thing in question must be decided “publicly”, which is really to say, in a decision made binding for all (“public decision” just means, some body constituded by whatever means and now considered “public”, decides what everyone now has to do). So the first step for a liberalization of the mind would be to create a clear process for deciiding which decisions really _must_ be made by the so called public. Here, from hearsay, Switzerland often falters as well: the joke goes that in Switzerland all things are either illegal, … … or mandatory.

    I have a qualification to add re: the EMH and the wisdom of the crowds. I just read an extensive 1970 Fama review paper on the EMH written after its first few years of existence, to make sure I really understood how it was meant in detail:
    Efficient Capital Markets: A Review of Theory and Empirical Work, Eugene F. Fama, The Journal of Finance, Vol. 25, No. 2, (1970), pp. 383-417
    http://www.jstor.org/pss/2325486
    As it turns out the actual EMH does not glorify the markets nearly as much as its reputation says. Maybe this is just news to myself and all others here knew that already but here we go.

    Above I erroneously assumed the EMH stated that markets would reveal some kind of “true price” by aggregation of individual assessments, themselves ideally randomly distributed. Well at least in this paper Fama goes to great lengths to claim that the EMH does no such thing. It just predicts that markets will reflect all available information _including investors’ tastes_ (Fama’s words) , for risk for instance but for any other possible preference (now we’d call it “biases”) as well. Fama keeps on repeating how the EMH says really nothing about whether markets will reveal any underlying truth, or correctly predict what is actually about to happen – the EMH just states that efficient markets will reflect all available information and by extension, investor beliefs, whatever they may be. Fama says right at the beginning of the paper that even the usual practical statement that prices will reflect expected returns (discounted by risk) is already an assumption by induction for the sole purpose of making testable predictions, but not a necessary assumption of the EMH. You could test the EMH with an assumption that investors want to lose money (my words). Fama also insiste heavily that the EMH does not require randomness, nor any other specific generator of the input distribution, or any specific distribution to begin with. He cites his own work and Mandelbrot, as far back as 1970, on fat tailed price change distributions, and states that this is not an argument against the EMH because the EMH does not require a random walk, or frictionless markets, or markets with good information, or low transaction costs, or investors without biases. It only requires that markets reflect all of that, good bad and ugly.

    In essence the original EMH seems to make much weaker predictions than say the wisdom of the crowds argument – it does not make assumptions about correctness. As far as I understand Fama there, the EMH does not preclude irrational exuberance or any other kind of collective bias.

    In economics this means that an efficient market that collapses does not need to mean that anything at all has happened. All it means is that sentiment is now different, and reflected in prices. Whether sentiment is justified or not is not in the EMH.

    Related to democracy this means that the point Caplan appears to be making in his book, which I have not read, i.e. that voters have pervasive biases that are reflected in their voting patterns, would simply follow from the EMH applied to democracy (call it an analogous “efficient democracy hypothesis”). Markets are efficient if all available information cum biases are reflected in prices, and democracies are efficient if all available information cum biases, is reflected in the outcomes of the votes. That means two things to me: 1- democracies are already analogous to a market system 2- applying “more market” to political decision making, as a supposed improvement over democracy, is unlikely to change the quality of political decision making, because any bias in politics will also show up in markets.

  30. Gravatar of Current Current
    27. September 2009 at 10:05

    Current: “Scott discusses the various Canton’s within a successful democracy that uses decentralization and direct democracy. But, this is begging the question. It presumes that such a state will be successful overall.”

    Scott: “I disagree strongly. I think you guys just keep moving the goal posts. First I was told that it is just one country, one observation. I was told that Switzerland was successful for cultural reasons that had nothing to do with direct democracy and decentralization. That explanation would predict that the most democratic and decentralized parts of Switzerland should be no more successful than the others. And yet they are more successful. Now I am told that all these new observations are also worthless.”

    I think it’s likely that the same cultural reasons that make Switzerland successful also make direct democracy successful there. I think that is the simplest explanation of what we see.

    In businesses it generally works like this. If a department has competent and motivated employees then little oversight is needed. A great many decisions can be devolved to those employees. But, if a business has poor employees then such devolution of power is not effective. (I’m not saying citizens are employees of a country, I’m just drawing an analogy).

    Current: “Law, Constitutions and division of powers are anti-democratic, but they have all being very successful.”

    Scott: “I don’t agree. I think laws can be consistent with a democracy. I didn’t say all decisions should be decided by referenda, just lot’s more than right now. I’d like to see income tax rates set by referenda, but not freedom of speech. We have a split under representative democracy, with the legislature only having the right to decide some issues. Why couldn’t we have the same split under direct democracy?”

    I fairly much agree with you here, constitutions and laws *can* be consistent with Democracy. But nothing says that they must be consistent with Democracy. The electorate may not agree with them.

    What I’m trying to show you here is that you are not really advocating “democracy”. Rather you are advocating a specific flavour of decentralization, Dan Hannan and British conservatives would call it “localism”.

    To me this all seems very devious and underhand to call what you describe “Direct Democracy”. It’s more like “Democracy that Classical Liberals like”, there’s nothing wrong with that, just we should misrepresent it.

    Statsguy, I agree with you in principle now, though I don’t know about the particulars of the history you mention. The distributional outcomes of introducing free trade may favour capital or labour, it depends on the circumstances.

    Mbk, thanks for the info on the EMH, I didn’t know that. It seems to me that Fama’s view is smarter than I previously thought.

  31. Gravatar of ssumner ssumner
    27. September 2009 at 10:07

    Statsguy, In principle I don’t think there is a “morally bad but practically good” option, as I am a pragmatic utilitarian in my moral views. But I actually see your point about the India break-up. I favor smaller countries, but also wish different kinds of people could live together. When they can’t, then hard choices must be made. I don’t know enough history to know what options were available in 1948. I doubt any of the options were good. Even (mostly) Pakistan couldn’t stick together, and that war in 1970 was also bloody. BTW, if it had stayed together it would now have the world’s 3rd largest population.

    On trade, I agree my no trade scenario was extreme, but I think it is a continuous loss function. I really think the average person does best with free trade (although some manufacturing workers do lose jobs.) And I think losses get continuously larger as trade barriers get higher, even starting from free trade. The vast majority of low paid workers are in agriculture, services or export industries–only a relatively small proportion are in import competing industries.

    mbk; You said;

    “Scott, thanks for the roses. The scope of government is really the key, in my mind: the decentralization that you mention, so that “minorities can decide on their own affairs”, when taken to the extreme, would then of course be a situation where every _individual_ can freely decide on their own affairs: decentralization down to the the minority of the individual. One does not have to take it to such literal extreme, but battles over public policy usually ignore the first decision that has been already made and take it for granted: that the thing in question must be decided “publicly”, which is really to say, in a decision made binding for all (“public decision” just means, some body constituded by whatever means and now considered “public”, decides what everyone now has to do). So the first step for a liberalization of the mind would be to create a clear process for deciding which decisions really _must_ be made by the so called public. Here, from hearsay, Switzerland often falters as well: the joke goes that in Switzerland all things are either illegal, … … or mandatory.”

    Again, good points. Everything should be decided at the lowest level possible. For schools, that is the free market, through vouchers. For many things now done by the Federal government, the states would be better. The many successful European social welfare states with 5 to 10 million people gives lie to the fear of liberals that decentralization makes the welfare state impossible. It’s possible, but they would have to convince voters state by state.

    As I’ve said, I think Switzerland may be too paternalistic for my taste. But my taste doesn’t matter. If most Swiss like things neat and tidy, the government should reflect their wishes, not mine.

    You said;

    “Above I erroneously assumed the EMH stated that markets would reveal some kind of “true price” by aggregation of individual assessments, themselves ideally randomly distributed. Well at least in this paper Fama goes to great lengths to claim that the EMH does no such thing. It just predicts that markets will reflect all available information _including investors’ tastes_ (Fama’s words) , for risk for instance but for any other possible preference (now we’d call it “biases”) as well. Fama keeps on repeating how the EMH says really nothing about whether markets will reveal any underlying truth, or correctly predict what is actually about to happen – the EMH just states that efficient markets will reflect all available information and by extension, investor beliefs, whatever they may be.”

    I think this is really important. But I would make one additional point that you may disagree with. This bias includes “smart investors” as well. So when he says investors may be biased, there is no easy way to take advantage of those biases, because he assume you and I are biased too! If we weren’t we’d arbitrage away any inefficiencies

    Now I know that when we get to the real world the smart money may not have enough wealth to arbitrage away inefficiencies produced by stupidity. But the point is that they can do a lot. In contrast, there is no way for a smart voter to arbitrage a stupid vote. So I think financial markets are a bit more efficient than elections. Nevertheless, I agree with your general point.

  32. Gravatar of Current Current
    28. September 2009 at 04:10

    Statsguy: “In Switzerland, one could argue that the small scale and direct democracy make it much harder to use political maneuvering and lobbying, etc. Thus, capital owners have achieved a compromise with labor owners – free trade _in exchange for_ a stronger social safety net (e.g. covering retraining, adjustment costs, and basic needs). In essence, the low cost of organizing labor forced capital owners to engage in Coasian bargaining in Switzerland, while in the US capital owners opted for a confrontational approach (because the structural factors operated in their favor).”

    I know what you mean here, but it’s not Coasian bargaining. Coase’s theory is about externalities, these are the subject here. A change in the market that comes about from a change in supply is not an externality. I see how it is like an externality, but it isn’t an externality. There are no poorly defined property rights here.

Leave a Reply