I’m always happy to see electoral reform brought up in the news because, well, our country really needs it. It’s for this reason that I’m glad former Congressman Mickey Edwards took to the Atlantic to address what he feels is a problem with our system – the lack of an absolute majority required for winning congressional races.

I do have a quarrel with a system that allows for the election of members of Congress (or governors or other officeholders) to whom most voters are opposed. Ben Quayle received 22.7 percent of the votes cast in his congressional primary; more than 77 percent of the Republicans who voted in that primary wanted somebody else to be their congressman. Quayle received just over 14,000 votes; more than 48,000 voted for somebody else, despite the fact that Quayle was the best known and most visible of the candidates. Running in a heavily Republican district, he will almost certainly become a member of Congress in January, representing a community that did not want him in that job.

While I agree that 22.7 percent isn’t a high mandate, I think his analysis is flawed. Edwards is referring to a primary election and primary elections are a different event from a general. Edwards claims that because Quayle was running in a heavily Republican district, he will almost certainly win, and therefore be “representing a community that did not want him in that job.” This is pretty misleading. Just because a majority of Republican primary voters didn’t vote for him in a (semi-closed) primary, doesn’t mean that they don’t want him in that job, it just means that in a multi-candidate primary, many had other first choices. If he really does win in November, it will most likely be because all of those other primary voters, plus many others, will eventually pull the lever for him. I’m willing to bet the vast majority of partisans, and a significant amount of soft-partisans, are pretty ready to vote on party ID and not care a great deal about the individual name behind it.  Just because the most active partisans’ did not get their first choice, does not mean the system isn’t fair.

There is one logical fix to what Edwards is concerned about, Instant Runoff Voting (IRV), but he dismisses it for a strange reason.

The problem with proposed “instant runoffs,” which would allow the reassigning of votes that had been given to “other” candidates, is that voters would be denied that opportunity to directly compare the policies and personas of the two remaining choices.

This is somewhat silly. The whole point of IRV – or Alternative Voting as it is known outside the US – is that a voter can rank their choices, so there should be no need to reevaluate from the two remaining candidates. Voters who chose from the top two candidates obviously supported one over the other, and voters who chose the eliminated candidates would have their backup votes go to the remaining two. In other words, they clearly already “compared” the two remaining choices in their initial ballot. Edwards displays an example of where IRV would have helped later on in his piece:

Supporters of Attorney General Bill McCollum have complained that it was the presence of a “minor” third candidate (who received only 10 percent of the vote) that allowed wealthy outsider Rick Scott to edge McCollum by three percentage points in the race for governor. Neither candidate got a majority (Scott received 46 percent of the vote, meaning he was opposed 54-46), and Florida has no runoff elections that would have permitted a Scott-McCollum faceoff.

If there were an IRV system, the minor candidate’s votes would have been reallocated to their supporters’ second choice, and we would have a definitive answer as to whether they were actually a spoiler or not.

Edwards goes on to correctly note that some states do have a majority runoff system, but fails to evaluate whether they are actually better.

With the exception of several states, mainly in the South and Southwest, majority support is usually not a factor in elections. In those states — often, amazingly, dismissed as “unsophisticated” — nobody can win an office of public trust absent that all-important ingredient: the support of the majority of those voters who took part in the elections.

I don’t dismiss these as “unsophisticated” but I do dismiss them as unproductive. For one, the record on runoffs does not seem to solve Edwards’ problem of lack of popular support. In my old state of North Carolina, Elaine Marshall and Cal Cunningham recently battled for the Democratic nomination for US Senate. Both competed with four other candidates in the first round of voting, before Marshall defeated Cunningham in the runoff. So what’s the problem with this? In the first round 425,343 voters turned out, while the runoff saw that number drop to 159,081. In other words, even less voters determined the nominee than before. This shouldn’t be surprising as primaries normally have pretty low turnout; why would we expect a second round to be any different?   It’s also important to note that runoffs discriminate against the numerical minority in any given district. So yes, there is a cost with this system, and that is that typically minority candidates in non-minority districts have a harder time winning primaries.

Edwards is right to notice something is wrong with the way we send people to Congress, but his suggestions, like most other US electoral reform ideas, are just cosmetic fixes. They may make things better on the margins, but fail to address the underlying problems with the system we use.