Communities of Disinterest: How an Algorithm can put an End to Gerrymandering

April 15, 2016

For over 200 years, the devious practice known as “gerrymandering” has undermined our democratic process and effectively instilled a sense of political despondency among our electorate in the United States. What is gerrymandering? The term is derived from the word “salamander” and the name of Elbridge Gerry, a former governor of Massachusetts in the early 19^th century. Gerry presided over the drawing of a new state senate voting district in 1812, and its creation was largely believed to favor his party, the Democratic-Republicans. The map was said to resemble the shape of a salamander, complete with various appendages and depicted in the Boston Weekly Messenger with claws and fangs. Simply put, the term “gerrymandering” has come to mean the intentional manipulation of electoral maps to favor a particular party or interest.  Many solutions have been proposed to stop this practice, but one of the most promising ones may turn out to be a computer program.

One can usually spot a gerrymandered electoral district by its obtrusive shape. This graphic illustrates how the manipulation of electoral boundaries can give a disproportionate advantage to a given party. For example, here are Tennessee’s nine congressional districts as they appeared in 2004. Notice the distinct shapes of districts seven and three. Those shapes are not arbitrary. They were drawn up specifically to favor house Republicans. In fact, not a single House member in Tennessee lost a bid for re-election from 1980 to 2005. This is gerrymandering in full effect. Whenever a politician at the state level has remained in office for multiple terms, it’s not unreasonable to suspect that gerrymandering has played a role in some way. Job security is usually a good thing—but when it comes to politicians, accountability should be part of the job description. Electoral districts are said to be based on “communities of interest,” defined by Washington Post reporter Christopher Ingraham as “people who share a common demography, culture, class, etc.” That’s a pretty broad term. How do we decide what a community consists of? The law doesn’t define it. Who gets to decide?

The United States has become increasingly polarized in the past few decades. Thomas Mann from the Brookings Institution writes, “The parties in Congress are as polarized—internally unified and distinctive from one another—as any time in history.” The increased polarization, both in rhetoric and in actual representation within Congress, is perhaps a symptom of a larger systemic problem: Americans by and large feel they are alienated from their representation in government. This dramatic presidential race is perhaps a manifestation of these trends. And it certainly makes for a lot of great headlines. At least presidential elections are generally competitive.  But what happens in Congress is arguably more consequential for the lives of average Americans. And it’s directly related to the heart of the matter. Congress currently has an approval rating of 11 percent. That’s only 2 points higher than the all-time low of 9 percent in November 2013, directly after the government shutdown. A whopping 86 percent of the public says they disapprove of Congress. Yet in 2012, the re-election rate for incumbents in Congress was 90 percent for House members, and 91 percent for the Senate. Now at this point, you might be wondering how it is that representatives in Congress remain in office when the public disapproves of them in record numbers. That’s a great question.

Part of the problem is that most folks tend to like their own elected representatives, but think that Congress as a whole is ineffective. That explanation holds water with Senate elections, which are not subject to the vices of gerrymandering. But what about the House of Representatives? Take a look for yourself. Ingraham concludes that Democrats are underrepresented by at least 18 seats in the House. It would be irresponsible to blame only one party when it comes to gerrymandering, but the Republicans have benefitted disproportionately in recent elections. So it isn’t particularly surprising that Democrats as of late have widely attempted to pass legislation to end the practice; one wonders if the shoe were on the other foot, if they would be as eager to do so.

So gerrymandering is a big problem. What can we do to fix it? Is it fixable? The drawing of electoral boundaries has been controversial for as long as electoral boundaries have existed in the US. The formation of independent, nonpartisan redistricting commissions is a potential solution. California’s map in the 2000s is a prime example of the influence of gerrymandering. The bipartisan effort in 2000 virtually guaranteed incumbent victories. In 2002, no seat changed party hands in California state legislature. Districts were dominated by either the Democrats or the Republicans, with very few competitive districts. The state’s solution is often touted as an ideal model. In 2008, the state voted to pass Proposition 11, which removed the responsibility of drawing the California’s congressional districts from the state legislature. Surely that’s a step in the right direction. The power to redraw was given to the California Citizens Redistribution Commission, consisting of 14 members. As a result, the redistricting commission created a map that gave rise to the most competitive congressional districts in the country. But redistricting commissions, as good as they may sound, are still subject to corruption. It shouldn’t be a surprise that as soon as the new commission in California was created, politicians were already trying to figure out how to exert influence over the process. Maybe it’s time to take humans out of the equation.

A recent study conducted by marketing research firm Intentions Consulting and Nikolas Badminton shows that some Canadians have a distinct lack of trust in, well, humanity. Roughly a quarter of participants thought than an unbiased computer program would be more trustworthy and even more ethical than their own managers. They preferred a computer program to be responsible for screening, hiring and assessing job performance. Younger adults shared these beliefs in higher numbers, with about a third of them preferring a computer program over an actual person. It’s safe to say that at least some Americans share this preference in some capacity. What if, as a solution to gerrymandering, we let a computer program redraw electoral districts?

It’s true that there will always be losers when it comes to drawing electoral boundaries, which makes it all the more important that it’s done in an impartial way. Maybe, in this instance, simple is best. The “shortest split-line” method of redrawing districts is a clear, straight forward solution, touted for its impartial nature. Essentially, shortest split-line is based on an algorithm with clear, transparent parameters. Creating a mathematical definition of an electoral district inherently removes the bias of human influence. In reference to the algorithm, The Center for Range Voting (CRV) states, “Which of those people are liberal, conservative, Republican, Democrat, black, white, Christian, Jewish, polka-dotted, or whatever has absolutely zero effect on the district shapes that come out.” CRV created a detailed comparative analysis of what states electoral boundaries would look like using shortest split-line. One of the major hurdles when it comes to the method is that its indiscriminant nature can actually be illegal in some circumstances, specifically with respect to the Voting Rights Act. The Equal Protection Clause expressly prohibits the formation of electoral districts that dilute the votes of racial groups. This law espouses a nice sentiment. But consequentially, it mandates affirmative racial gerrymandering, sometimes with unintuitive results; it lumps minorities together, wherever possible, into one single district. As evidenced by the situation with Florida’s fifth congressional district, this actually diminishes the influence of minorities, because it gave them less influence in surrounding areas. Wouldn’t it be best for all voters if it could be ensured that the redrawing process was impartial?

Gerrymandering isn’t a problem that makes headlines often. Both parties do it. It’s an arbitrarily complicated problem, and it’s not a sexy or even a particularly interesting problem. And that is part of the reason we allow it to persist. But gerrymandering gives rise to a polarized, corrupt and complacent legislature. In previous essays, we’ve seen multiple instances in which the U.S. government has been slow to adapt to emerging technologies. Copyright laws have been slow to adapt to the emergence of digital media. Transparency, once praised as a virtue in government by candidate Obama, is all but a hopeless dream as far President Obama is concerned. Ironically, that’s one thing the executive and legislative branches seem to agree on, despite the fact that transparency is now easier than ever with the development of new methods of electronic communication. This distinct inability for our legislative body to keep up with the advancement of technology is largely attributable to the checks and balances inherent to the way our government functions. Passing legislation isn’t an easy thing to do. The more comprehensive the legislation is, the more difficult it is to get through Congress, and rightfully so. But redistricting based on a shortest split-line algorithm is actually an instance in which technology can directly aid the democratic process, thereby potentially making our government more efficient and fair. It shouldn’t be controversial to suggest that politicians ought not be given the authority to draw their own electoral boundaries. It’s a clear conflict of interest. In principle, gerrymandering is patently undemocratic, regardless of which party is drawing the map. And rest assured, both parties are complicit in this practice. Voters should choose their representatives, not the other way around.

David Stockdale is a freelance writer from the Chicagoland area. His political columns and book reviews have been featured in AND Magazine. His fictional work has appeared in Electric Rather, The Commonline Journal, Midwest Literary Magazine and Go Read Your Lunch.  Two of his essays are featured in A Practical Guide to Digital Journalism Ethics. David can be reached at dstock3@gmail.com or via his website.