Liberal Democracy is Viable, But Can We Do Better?

Note: This post is part of our League Symposium on Democracy. You can read the introductory post for the Symposium here. To see a list of all posts in the Symposium so far, click here.

~ By James Hanley

Synopsis: Liberal democracy is viable, but always threatened, both in its liberal aspect and its democratic aspect, by the inevitable fact that it creates losers. An alternative approach that I call Tiebout democracy creates more winners and fewer losers.  Is it a better approach?

This isn’t the post I wanted to write, but writing a popular version of Arrow’s Theorem still has me stumped.  And this isn’t a completely developed post, but more of an off-the-cuff assemblage of thoughts that have been rolling around in my skull, with lots of echo but not necessarily much coherent pattern.  Take it for what it’s worth; i.e., my measly two cents.

Definition: Liberal Democracy
To define liberal democracy, I’m going to cadge a bit from Jaybird, and say that we have liberal democracy when we select those who govern us through some method of “counting heads” while ensuring some sufficient amount of “freedom, individualism, equality, fraternity, and, yes, liberty.” What counts as that “sufficient amount” need not be precisely defined here, except that I would say to the extent possible with maintaining public safety, more is better.

The Basic Ground: There is No Group Preference
With that definition out of the way, let’s move on to what I will call the basic ground, which is that there is no method of counting heads that can satisfactorily aggregate our individual preferences into a rational group preference order (or as the economists like to say, a “social welfare function”).  This is the point of Arrow’s theorem, and why it’s such an important, albeit little known, concept.  As I said, popularizing it is at present beyond my ability (and as far as I can tell so far, beyond anyone else’s too), but the key point is that any method of counting heads must either deny some people the right to hold their own preference order, or result in an outcome based on the rank ordering of some irrelevant alternatives, or choose an alternative that is less preferred by everyone over an alternative that is more preferred by everyone, or select an alternative determined by a dictator (someone whose preference trumps everyone else’s).

All that is to say, the idea that democracy reveals what “the public wants” is utter bunkum. There is no collective will or public preference.  Democracy doesn’t reveal what the public wants; it reveals whatever result the particular head counting method happens to produce.  One of the assignments I give my students is a voting assignment, in which they tally the votes in a hypothetical election using different head counting methods (plurality, plurality run-off, sequential run-off, and Borda count), with each head counting method producing a different outcome, even though none of the voters changed their preferences.¹

This is why I have no patience with the term “the public good.” It’s just another term for public preference, and there is no such thing.  What I think is the public good is different from what you think is the public good, and so there is often no public agreement on what constitutes the “public good.”²

What It Means: The Structural Problems of Liberal Democracy
What this means is that democracy (head counting) will always and forever be not about finding out what the public wants, but about determining which group gets to impose its will on others. And that has two other implications: 1) The winning group’s will won’t always be liberal; 2) There will always be losers, and losers will always be resentful.

Illiberal Democracy
The illiberality in democracy has been handled by others here, so I won’t go into it in depth.  I only want to add one bit of empirical evidence to support it.  A study of citizen initiatives in the U.S. revealed that while only 1/3 of all initiatives and popular referenda are approved by the voters, over 3/4 of those that restrict the civil rights of a particular group are approved.³ It appears that direct democracy not only may result in illiberal majority tyranny, but that it is more likely to do that when given the opportunity than it is to do anything else.

The Resentment of Losers
Even when the results of democratic head counting aren’t particularly illiberal, the process produces losers. And losers tend to be resentful, not just because no-one likes to lose, but because they have been taught that the process is supposed to represent their interests, and instead they see their interests being overridden and shoved aside.

Because democracy inevitably breeds resentment, it is always breeding a population that may work to undermine it. I don’t laugh and shrug it off when I hear yahoos talking about it being time for a 2nd Amendment solution, or armed revolution, not in response to government acting undemocratically, but in response to a democratic outcome they don’t like.  I’d never thought deeply about the issue until I was in Beirut last year, observing the consequences of the Lebanese Civil War.  Twenty years after it ended, you can still see the shells of bombed out buildings, apartment buildings with shell holes in the wall, and street corners with soldiers standing behind sandbags.  A place where everyone I talked to thought renewed civil war was a very real possibility.  A place where the social ties were absolutely shattered by the war.

The number one virtue in democracy, like the number one virtue in sports, is the ability to lose gracefully.  We see how often people fail to display that virtue in sports; why would we expect it to be any better when we’re dealing with issues that matter even more to people?

Tiebout Democracy: An Alternative to Liberal Democracy?
But authoritarianism breeds losers, too!  Indeed, which is why liberal democracy is viable.  Most people realize the traditional alternative is much worse.  But our biggest danger isn’t really armed insurrection. A Timothy McVeigh here and there is a tragic nuisance, but not a precursor to sustained revolt.  The real danger is that disgruntled losers, when they finally win, will rig the system to minimize their future chances of losing.  See: Gerrymandering; Texas.  See: Voter ID Laws; Wisconsin, Indiana. And it’s not just conservatives.  I know left-wing academics who ask seriously whether democracy is satisfactory, given that it doesn’t produce the “right” outcomes.

Ultimately it is the winners who are afraid of becoming losers who will undermine democratic structures, making them not quite truly authoritarian–because that would really set the new losers off–but, back to Arrow’s Theorem, effectively playing the dictator role, or effectively denying participation to those holding opposing preference orders.

So liberal democracy, even if in a degraded form, is viable, and both more viable and more desirable than a true dictatorship. But particularly in that degraded form, and to some extent even in its ideal form, it is a soft dictatorship.  Those who control the agenda and determine the voting procedure will dominate–there is no public will.

Can we do better? I don’t know, but let me throw out an alternative that is currently being bandied about, what I will call Tiebout democracy. Charles Tiebout was an economist who argued that different municipalities compete for citizens by offering different mixes of taxes and services. On one level, this is just federalism–Mississippi attracts one kind of person, while Massachusetts attracts another.  On another level, it is an explicitly consumerist model; polities as businesses chasing customers. Phrased in the latter way, some will see it as very undemocratic.  After all, democracy is about voting, about counting heads, right?  But isn’t the market in large part about counting heads?  How many people are willing to buy a Ford, how many are willing to buy a Chevy?  We tend to call that voting with our pocketbook, or voting with dollars, but the amount we pay merely expresses the intensity of our preference; each purchase is still a single choice, a single head count.  But to the extent that we are enabled to express not only our preference but the intensity of it, this system is actually more democratic than a simple head-count, because it more accurately represents the real preferences of the whole public.

Federalism is a democratic form, one that, like markets, allows people to vote with their feet.  What’s more, this type of democratic form reduces the number of losers.  When I and 95% of all my neighbors vote to buy Ford, the handful of Chevy preferrers in the neighborhood don’t lose.  Tiebout democracy increases the number of satisfied citizens.

How far should we take this?  I’m a staunch federalist that thinks we’ve moved too far away from it in the U.S., nationalizing too many issues.⁴  But am I on-board with the charter cities model?  I’m not yet sure. But I am sure that a truly liberal person wouldn’t reject them out-of-hand. As David Ellerman writes:

For classical liberalism, the basic necessary condition for a system of governance is consent. Consent could be to a non-democratic constitution which alienates the right of self-governance to some sovereign–which in the case of a charter city would be the technocratic viceroys, or their principals such as some well-meaning foreign governments. Consent plus free entry and exit suffice to satisfy the governance requirements of classical liberalism. Classical liberalism per se sees no moral necessity in democratic self-governance at all…

He is right. Traditional forms of democracy are just tools used by liberalism, and if another tool achieves the liberal end, then traditional democracy can be shunted aside in favor of the other tool.

Ellerman is wrong, though, on two counts. First, he argues that “The theory of inalienable rights… shows why such contracts to voluntarily give up the rights to self-governance are inherently invalid.”   This is wrong, because as long as you have free entry and exit–particularly free exit–you have not actually alienated any of your rights.  You may have chosen not to exercise them because the city you chose doesn’t allow that exercise, but you have not alienated them because if you change your mind and prefer to begin exercising those rights you can exit that city and choose another.  Second, he claims that this approach is non-democratic.  But how impoverished is our understanding of democracy if it does not include voting with one’s feet? As with any candidate for our political allegiance, a charter city’ success is ultimately determined by the number of heads they can count in their favor.  I can either vote between two candidates to govern the place where
I live, or I can vote between two places to live.

And is there not something perverse about favoring a democratic method that produces more losers, rather than one that produces more winners?

No, I am not proposing that this is a perfect system. I’m not even sure I would prefer to live in a world of nothing charter cities rather than one where charter cities aren’t even an option.  But of course that is true of our current systems as well.  To point only to the imperfections of Tiebout democracy is to implicitly engage in the nirvana fallacy. And realistically, just as politicians have to pander to us, and corporations have to pander to us, wouldn’t charter cities have to pander to us if there is free entry and exit?

In my experience, libertarians are apt to incline in favor of charter cities, while liberals are apt to recoil from the idea (I have no idea where conservatives stand on the issue, probably in favor).  What does that say about the two group’s respective stances on both liberalism and democracy?

_______________________________________
¹ The assignment is cadged from Ken Shepsle and Mark Boncheck’s Analyzing Politics. In their version, a 5th voting procedure, the Condorcet method, produces yet another winner, again without any voters changing their preferences.

² There are, of course, public goods, but that’s a technical term that applies to a very limited class of goods.  The term is frequently misapplied by people who prefer ideology to analysis.

³ Barbara S. Gamble. “Putting Civil Rights to a Popular Vote.” American Journal of Political Science, 41(1): 245-269. 1997.

⁴ I’m appalled at the degree to which we Americans nationalize issues not because we’re dissatisfied with our state’s policy, but because we’re dissatisfied with other states’ policies.  It’s not enough to win locally and adopt a live-and-let-live attitude toward other places; we must force those others to succumb to our will.  We insist on turning winners into losers, and then we say that is more democratic.