Richard Epstein and Glen Whitman on Libertarians and the Presidential Election

Well-known libertarian scholars Richard Epstein and Glen Whitman have recently weighed in on a question that has been much-debated in the blogosphere: Who, if anyone, should libertarians support in the presidential election. Epstein argues that we should support Romney as the lesser of the two available evils:

In the final countdown to what promises to be a close election, the libertarian finds himself without a comfortable home in either political party. Political parties and their presidential candidates offer market baskets of policy prescriptions on a large array of different issues. We do not have the option of picking out from each basket the policies that we like and rejecting the rest. Politics do not come served a la carte in our two-party system….

Though no libertarian can take comfort in the blurry Romney campaign, the scorecard does tip in his balance. The state of play nationwide on social issues is decidedly mixed, with too much intolerance on both sides. But on economic issues, the one confident point is that in an age of bloated government, the correct vote goes to the party, when the campaigning is mercifully done, that is more likely to limit the rate of government growth, if not shrink the size of government altogether. This election cycle, that party is the GOP. It is time for a change from Blue to Red, from Obama to Romney.

Epstein’s analysis of the Romney vs. Obama tradeoff is in many respects similar to mine, though I am less convinced about Romney’s superiority than he is. Epstein also makes an important point about social issues. While conservative Republicans are very bad in this area from a libertarian point of view, liberal Democrats also favor many types of social regulation, some of which are just as intrusive as those favored by Republicans. He mentions the HHS contraception mandate as an example. He might also have mentioned the much larger example of the Obama Administration’s expansion of the War on Drugs (which I noted in my own Romney vs. Obama post). If the Democratic Party really were there party of laissez-faire on social issues, it would be much more appealing to libertarians. But much of the time, it simply isn’t.

Glen Whitman argues that libertarians in swing states perhaps should support the lesser of the two major-party evils. But in the other 35 or so states, they should vote for Libertarian candidate Gary Johnson:

[I]f I were in a swing state where conceivably a group of libertarian-minded voters could affect the outcome if they all voted together, I would hold my nose and vote for one of the two major party candidates.

According to the New York Times electoral map, only 7 states are considered “toss-ups”: CO, FL, IA, NH, OH, VA, WI. To these, you might add the 8 “leaning” states: ME, MI, MN, NM, NV, PA (for Obama), AZ and NC (for Romney). If you’re a libertarian voter in one of these 15 states, then I have nothing useful to tell you.

But that leaves 35 states that are solidly in the Democratic or Republican camp, with a combined eligible-voter population of over 136 million (about half that number voted in 2008). None of these states would by any stretch of the imagination get tipped by your vote-of-exaggerated-size. In these states, there is no good reason to vote for Obama or Romney. You can vote your conscience with no fear that your conscience will have doomed our country to the greater of two evils.

And fortunately, there is an excellent vote-of-conscience choice available this year: Gary Johnson. Imagine if everyone like us (that is, libertarians in non-swing states) voted for Johnson. If even 1% of voters were in this category, Johnson would get over a million votes — which might actually be enough to get some attention, and maybe establish a beachhead for another run in 2016.

Whitman’s position is very reasonable. Although I disagree with Johnson on some issues (such as the Fair Tax and foreign policy), I think he’s much better than either Romney or Obama. In my view, the real problem with supporting Johnson, even in a non-swing state, is that third party politics is a poor strategy for promoting libertarianism. I worry that a relatively strong showing by Johnson would lead libertarians to invest additional resources in the Libertarian Party instead of in other efforts to promote liberty that are likely to be more effective. Obviously, this calculation would change if the LP does well enough that they have a serious chance of displacing one of the two major parties in the foreseeable future. But I see little if any chance of that happening.

Finally, it’s worth noting that, even if you live in a non-swing state, there is a very small chance that your vote will make a decisive difference in a presidential election. In this interesting article, Andrew Gelman, Nate Silver, and Aaron Edlin estimated that the chance of casting a decisive vote in 2008 was about 1 in 10 million for voters in several swing states, and an average of 1 in 60 million for the nation as a whole. In some non-swing states, it could be 1 in 100 million or 1 in 200 million. Those are extremely low odds. But, as I explained here, even those low probabilities might make it rational to vote for the lesser of evils if you think the difference between the two is great enough. Moreover, if the odds of affecting the outcome between the two major parties are extremely low, the odds of casting a decisive vote for Johnson in terms of “sending a message” are even lower. While there are situations where one additional vote might make the difference between victory and defeat for Obama or Romney, there probably isn’t any situation where adding one additional vote to Johnson’s count makes the difference between sending an effective libertarian message and not sending one. The only possible exception is that, if Johnson gets 5% of the national vote, the Libertarian Party would qualify for federal matching funds in the next presidential election.

UPDATE: I originally misidentified Glen Whitman as a legal scholar. In reality, he’s an economist. I have now corrected the mistake.