Perfect voting systems don't exist, but let's try some better ones

Ed Lotterman l Bismark Tribune l November 17, 2013

My home state of Minnesota had its first experiment with allowing local governments to use ranked-choice voting and it seems the system passed its first test. At least there were no glitches or unforeseen outcomes.

RCV is a variation in instant-runoff voting used by a few cities in several states, including California, Maine and North Carolina. It allows voters to choose second and third preferred candidates if no single candidate gets at least 50 percent of votes case.

In this case, all but the top two candidates are eliminated and votes cast for others are determined by their second choice. People can vote for smaller-party candidates knowing that doing so will not have the effect of electing someone they dont want to see in office, as may have happened in the past with George Wallace, John Anderson, H. Ross Perot and Ralph Nader.

Successive small successes like this at local levels increase awareness of the advantages of the system and show us how it might break some of the gridlock at the national level.

The question of devising voting systems that result in the fullest expression of the desires of the largest majority of the population is a knotty one. It has challenged economists as well as political scientists and law scholars. At a highly theoretical level of maximizing the total satisfaction of society for the nation as a whole, economists tend to be pessimists.

Kenneth Arrow, one of the greatest U.S. economists of the 20th century (and uncle of one-time Treasury Secretary Lawrence Summers, a recent competitor for chairing the Federal Reserve Board), won a Nobel prize in part for his 1951 Ph.D. thesis that is the base of his impossibility theorem.

Briefly, it argues that when there are three or more candidates, no voting system, including any ranked-choice one, can reliably end up representing the true preferences of voters. To the extent Arrow is right, moves to let local governments use ranked-choice voting is futile.

However, like much economic theorizing, Arrow's work is highly abstract. While Wikipedia is not always reliable, its entry on Arrow's Impossibility Theorem gives a good explanation of his ideas to those with the grit to plow through it, and an illustration of its level of abstraction to the rest of us.

His work is part of a broader area known as social choice theory that deals with many challenges for complex modern societies.

Economists have good tools for examining how decisions get made when individuals make choices about purely private goods. Do we want a chocolate or strawberry shake? Should we re-do the bathroom or take a vacation in Cancun? Pay for a transmission job and try to drive the old buggy a few more years or take on a big loan to buy a new car instead?

Evaluating how such decisions get made so as to maximize satisfaction is easy, however, compared to the question of how to do so for collective goods and public goods that we pay for and consume as groups.

Should people pay more in taxes to have better roads? How many post-secondary campuses should a small state keep open? Does the United States need a new, multibillion-dollar set of littoral combat ships? Should illegal immigrants be turned away from emergency rooms at county hospitals? Should people in their 20s and 30s pay higher taxes, and have less to spend on themselves, so that people in their 70s can continue to get three times as much in Medicare benefits as they ever paid for in taxes?

Those are societal choices, ultimately settled through political processes, not private ones made by any individual. However, as Arrow and many others before him demonstrated, expressing the aggregate preferences of society is knotty.

The only way for such questions to be answered with a measure of both fairness and economic efficiency is through an effective democratic system. Effective democracy depends greatly on election systems that deliver leaders who best represent the majority wishes of voters.

That, however, is not as easy as it might seem. U.S. citizens, particularly those who have not lived abroad or had some political science in college, tend to assume that the way we do it here is the only way to do it. But there are many ways to skin the cat of democracy.

We have a system of congressional representatives based on districts, designated by population, with the winner in each district being the first past the post. Under this system, 49 percent of the people in a state could have no representative from their party in Congress.

Many other nations use some form of proportional representation, where each party get seats in the legislative branch based on the party's proportion of the total vote. Both approaches have advantages and disadvantages.

We had a system where candidates at state and local levels were chosen by party leaders or conventions, which often meant the same thing. In a move intended to increase representation, many states instituted primary elections.

Now these are used by activists in the wings of each party to effectively hijack control away from more moderate majorities. The result is political deadlock at the national level, and that fosters economic sluggishness.

It is difficult to start a new party, whether on the left or right. Voters dont want to throw their votes away by voting for someone they prefer but who is doomed under the current winner-take-all system.

Ranked-choice voting is designed to take power away from the wings of parties and to give voters assurance that the process will be more likely to produce a result that best represents majority views.

Arrow proved that achieving this perfectly is impossible, at least in the conditions of abstract theory.

But there is much evidence that improvements can be real in practice. Better functioning democracy often means a better functioning economy. Perhaps Minnesota and the few other areas that have ranked-choice voting will lead the way.

(Economist Edward Lotterman teaches and writes in St. Paul, Minn. Write him at