Should the Victor Share the Spoils?
America's winner-take-all electoral system has its problems. Proportional representation might fix them.
“Left Party whip Keith Ellison spoke in Washington today in an attempt to rally centrist support for tighter financial regulation—his liberal coalition has support on the issue from Tea Party leader Steve King, but without more Democrats and Republicans the bill is doomed to fail. Leaders of the Green Party have yet to take a stance on the bill but …”
Wait, what?
This might sound absurd in the United States, but it’s not as crazy elsewhere in the world. The American system of government is stable, popular, and backed by the Constitution—and dominated by two political parties. A political system comprised of multiple, smaller parties and shifting coalitions may be unimaginable in America, but it’s the norm in most other democracies. While the United States is one of the world’s oldest democracies, and spreading democracy is a central tenet of the country’s foreign policy, our winner-take-all system itself is among our least-popular exports. In Western Europe, 21 of 28 countries use a form of proportional representation in at least one type of election.
What is proportional representation, or PR? It’s a system that aims to gives parties the same percentage of seats as the percentage of votes they receive—and it might be able to end our gerrymandering wars.
Every ten years, state officials are charged with redrawing district maps to account for population shifts in the Census. In practice, incumbent lawmakers often turn into cartographers with the power to change maps to suit their needs. The problem is epigrammatic: Rather than voters choosing their legislators, legislators are choosing their voters. That’s how you get districts that look like Maryland’s third congressional district, pictured below:
The word gerrymandering is a blend of Massachusetts Governor Elbridge Gerry’s name and his Democratic-Republican Party’s 1812 creation of a salamander-shaped district. Experts, like my colleague Garrett Epps, blame the distorting process for part of the current disconnect in some states between a state’s voters and their representatives. After the 2010 Census, a bevy of Republican state legislatures packed Democratic voters into lopsided throwaway districts, usually in urban centers. The problem has been especially acute in swing states like Pennsylvania and North Carolina. Democrats have replied by splitting liberal suburbs like Montgomery County, Maryland, into several smaller districts where they hold a slight advantage, but Republicans are still winning this political game at the moment: In 2012, Democratic House candidates received more votes nationwide than Republicans—and Republicans won a 33-seat majority.
The most commonly prescribed fix for gerrymandering is using a bipartisan commission to draw congressional districts. Six states currently use commissions whose members are appointed by political leaders from both parties, while California goes a step further, selecting commission members from a pool of bipartisan voters. Iowa uses a non-partisan, independent agency and has achieved some of the most competitive races in the country.
Do these commissions work? Yes, to some extent. A New York Times analysis found that “states where courts, commissions or divided governments drew the maps found a much smaller disparity between the share of the popular vote and the number of seats won in Congress.” But there are limitations to their effectiveness. In 2010, California Democrats were able to hijack the 2010 redistricting process to protect their party’s incumbents. Arizona and New Jersey don’t limit committee members’ future political activities, and the Supreme Court may soon decide whether Arizona citizens ever actually had the right to take the map-drawing process out of the legislature’s hands. What’s more, Americans increasingly sort themselves by political affiliation: Liberals tend to live in cities, while conservatives are most prevalent in rural areas. Even with good, non-partisan intentions, it’s getting harder to draw single-member districts that get a party’s seat share to approximate its statewide vote share.
That’s where PR could come in. In PR, each party wins seats in proportion to its support. Israel elects all 120 members of its national legislature from a single multi-member district that encompasses the entire country, and the Netherlands does the same with its lower house. But districts that large can lead to over-representation of fringe parties who receive just a small percentage of the vote, as well as giving numerous tiny parties the ability to make outsized demands from big parties if they lack a majority. So larger countries often break themselves down into smaller districts to ensure legislators have some connection to a particular geographic area.
In the United States, those geographical areas could be the states. Imagine Oregon sent five members of the House. Under PR, if Democrats got 60 percent of the statewide vote and Republicans got 40 percent, three Democrats and two Republicans would be elected to the House. Or if the two big parties got 40 percent each, and the Green Party won 20 percent of the vote, the Greens would send a representative to Congress. The largest states could use several smaller electoral districts, so that, for example, someone from San Diego isn’t represented only by northern Californians.
There are different ways of determining which candidates from the parties make it in. Most European democracies use what's called an open-list PR system, where each party nominates (at most) as many candidates as that district sends to its legislature. Voters get a single vote for a candidate that also counts for that candidate’s party.
Think again of five-member Oregon. One popular local Republican wins 40 percent of the vote, and two other Republicans win 10 percent each. All three go to Congress because the party won three-fifths of the state’s votes. The top Libertarian candidate receives 20 percent. He or she, too, goes to Congress. No Democratic candidate gets more than 8 percent of the vote, but because the total number of votes cast for Democrats adds up to 20 percent, the last congressional seat goes to the first-place winner among them. (Results that don’t break down so cleanly are decided by seat-allocation formulas like the “largest remainder method.”)
The voting reform group FairVote advocates a PR method called “instant-runoff” voting, where citizens rank the candidates on a ballot into first place, second place, and so on. Once a candidate reaches a threshold, votes are reapportioned to second- and third-place finishers until all seats are filled. Here’s an explainer video from Minnesota Public Radio:
In another variation, Germany mixes PR and winner-take-all single district elections by allowing citizens to vote twice, once for a party and once for a local representative.
To be sure, these various PR systems are more complicated than the winner-take-all method. Simplicity for voters is not to be disregarded. Plus, there are some advantages to the current system. When voters must make a clear choice between two big parties, those parties must present themselves as big tents open to different groups. (The Senate, which allocates only two seats to each state, isn’t as well-suited to PR and would continue to privilege the big parties if the House reformed its rules.)
But all PR systems accomplish one crucial feat: cutting out wasted votes. Right now, American elections produce a lot of them. Under the current system, a candidate who receives 49 percent of the vote and a candidate who receives 5 percent of it in a two-way House election receive the same reward: none. This provides little incentive for Democrats in rural Georgia to vote in House elections at all, for example. Conversely, it’s easier to stay home on Election Day if you’re confident your liberal Manhattan neighbors will carry the day—Reihan Salam calls the experience of voting Republican in hyper-liberal Manhattan “a socially approved form of venting.”
Americans, understandably, are committed to local representation as historically practiced, where each House member has a tight connection to a small constituency. The lower house is the people’s house, and communities want their own representative. Perhaps, absent the desire for a massive change in election rules, PR is best used as a mirror to reflect the problems—wasted votes, safe-seat extremism, minority underrepresentation—of winner-take-all.
The problem of wasted votes is tough to address through minor changes, but looking at PR changes the way we think about single-seat districts. France is the only other country that allows self-interested legislators to draw their own maps without incorporating PR to balance that power. At the least, truly non-partisan commissions must be established to end gerrymandering—and with it, the creation of too many artificially safe House districts in which politicians must demonstrate the firmness of their ideological positions, not the rationale for them.
Minorities and women are still underrepresented in Congress. In 2013, the Supreme Court tossed out a provision of the Voting Rights Act of 1965 that had prevented some states with a history of discrimination from changing their election laws without federal approval. That ruling led to a debate on whether “majority-minority” districts, whose creation the VRA had encouraged, substantively increased minority representation or were merely a form of tokenism. And as Katrina vanden Heuvel writes in The Washington Post, “women win about 10 percent more seats in multi-seat state-legislative and city-council elections than they do in congressional districts.” Would drawing fairer maps be enough to achieve greater equality? What other steps can be taken to give voice to voters of all backgrounds?
A PR system would also give third parties a higher chance at holding office, at least in the House. Instead of dragging the Republican Party to the right, a Tea Party politician could run on his or her own, aiming for the 20 percent of the vote he or she would need in a five-member district, and the Republican candidate wouldn’t need to pander to the loudest voices. That may not be a good thing—again, the stability of two big, theoretically centrist parties can be an advantage. That said, if America wants its two dominant parties to be more moderate, it will have to try methods other than the “top-two primary.” In 2012, California changed the rules to make candidates all of parties face off in state congressional primaries, with the top two finishers advancing to the general. The state hoped the move would increase independent turnout and boost moderate candidates. It hasn’t worked as advertised.
If PR is not the answer, we need to start asking more questions.