Research Matters: Nicholas Stephanopoulos on "Partisan Gerrymandering and the Efficiency Gap"
Research Matters is a regular feature in which a member of the faculty talks about some of his or her latest work and its impact and relevance to law and society.
Professor Nicholas Stephanopoulos wrote “Partisan Gerrymandering and the Efficiency Gap” with Eric M. McGhee, a research fellow at the Public Policy Institute of California. The paper, which proposes a new method for measuring the fairness of partisan district plans, will be published in the University of Chicago Law Review in 2015.
Q. Why did you write this paper?
A. For years and years, the cause of action for partisan gerrymandering has been in a weird limbo. The U.S. Supreme Court, by tenuous majority, continues to think that it is a legitimate, justiciable theory for bringing a suit. But, at the same time, five justices keep rejecting every single standard that anybody comes up with to distinguish between what’s a lawful district plan, and what’s an unlawful district plan. So there’s a real need in this area of law for a test that courts can use. The idea in this paper was to propose a test — and not just any test, but one that responds to what I think is a narrow, but real, opening in the Court’s last decision, LULAC v. Perry, in 2006. In LULAC, for the first time in 20 years, the Court — with five justices signing on — expressed some openness to a test that is based on the notion of partisan symmetry. That means treating the political parties equally in terms of how their votes convert into seats. There could be a variety of a different kinds of tests. In the paper, my co-author and I argue for one based on a metric we call the efficiency gap.
Q. What is the efficiency gap?
A. It measures wasted votes. The two mechanisms that are part and parcel in gerrymandering are “packing” and “cracking” the other side’s votes. “Packing” means over-concentrating the other side’s votes so they win by huge margins in a small number of districts, and “cracking” means distributing their votes over lots of districts, so they lose each district by a relatively narrow margin. Packing results in surplus votes, and cracking results in lost votes; either way, they’re not contributing to a candidate’s victory in a district. The efficiency gap is a tally of all the packing and all the cracking that’s in a given district plan. We add up all the wasted votes for each party, and divide by all the votes cast. The idea is that in a perfectly fair plan, the parties would waste the same number of votes, so you would get an efficiency gap of zero. The bigger the difference between the parties’ wasted votes, the more one party has been either a victim or a beneficiary of gerrymandering.
Q. So if the gap is in favor of Party A, that means they essentially are able to win more seats with the same number of votes?
A. Yes. There’s a gap in the efficiency with which the parties’ votes turn into seats.
Q. You and your co-author computed the efficiency gap going back 40 years for congressional and state house races. What did you find?
A. Over the whole 40-year period, plans look pretty symmetric. There’s no systematic bias in favor of either the Democrats or the Republicans. But there have been serious trends toward favoring the Republicans. In the ‘70s and ‘80s, plans were tilted a little bit toward the Democrats. In the ‘90s, they were roughly fair. Then in the 2000s and 2010s, it moved toward the Republicans, more than it had ever been toward the Democrats. And not only are plans more pro-Republican on average, the typical plan benefits some party — whether it’s the Democrats or the Republicans — more so than in the past. Overall, the extent of gerrymandering is getting worse. But it’s Republicans who are the big winners.
Q. Why?
A. That’s the big question. Is it because they’re deliberately manipulating district lines more than in the past? I think they probably are, and they are probably aided in doing this by advances in technology. It’s much easier to produce a really good, really high efficiency gap in a district plan now. They might also be the beneficiaries of things other than deliberate efforts to screw over the other party. It might just be that Democrats are even more concentrated in cities than they used to be, and so there might be a natural gerrymandering effect against Democrats.
Q. Did the findings surprise you?
A. It was pretty obvious that Republicans were really benefitting from gerrymandering in 2012. You have a number of states — Wisconsin, Pennsylvania, Ohio, Michigan — where Republicans won less than the majority of the total vote in the state and then won big majorities of the seats. Even an eyeball test shows that’s extreme gerrymandering. But I was surprised at how across-the-board it is. I wouldn’t have known that Wyoming is really bad, or that all sorts of states that aren’t in the news are really bad. I was surprised that 2012 was so much worse than previous cycles. I was surprised that Democrats don’t seem able to gerrymander effectively when they’re in charge.
Q. Why should the average American care about district plans and the efficiency gap?
A. It affects the policies coming out of legislatures. At the federal level, if there had been an efficiency gap of zero in every state in 2012, Democrats would have controlled the U.S. House. The last two years would have been completely different with a Democratic Congress. Immigration reform would have passed, a climate-change action might have been taken. Gerrymandering is not just about seats and votes; if you affect who gets elected, you affect which policies come out. Gerrymandering makes it possible to have a big distortion in what the public wants versus what the public gets. There’s been a big uproar over voter ID laws, over campaign finance —and all those things are important, but neither a voter ID law nor campaign finance is going to give you a 15-seat swing in a state legislature. There’s no way Republicans won the U.S. House because of voter ID laws or campaign finance. This makes sense because those other laws directly affect the parties’ vote shares. The power of gerrymandering is that it keeps the votes constant and just totally changes how they translate into legislative power. You might fight and fight over the voter ID law to get a half-percent improvement in one party’s vote share — but if you give me the power to draw the lines, I can give you way more than a half-percent benefit.