The tech revolution that could fix America’s broken voting districts

abstract art of the white house

Illustration by Max Fleishman

Who’s behind the lines?

Barack Obama knows a thing or two about gerrymandering. 

Less than a year after his defeat to Bobby Rush in the 2000 Illinois race for the U.S. House of Representatives, Obama, then a state senator, found his home address cut out of the Chicago South Side district he tried and failed to represent in Congress. Rush denied moving the lines to intentionally box out his one-time opponent. But Obama harbored no illusions about what had happened. 

“What Rush did is not unique. Legislators choose boundaries that will advance themselves. … The system of redistricting in the U.S. tends to allow representatives to chose people, instead of people choosing,” Obama told the Hyde Park Herald, a community newspaper in Chicago. “It’s just politics.”

Of course, one does not occupy the Oval Office for eight years without being naturally talented at “just politics.”

The consultant, who had volunteered on Obama’s campaign against Rush, helped Obama literally map out his political future.

According to a fascinating history of Obama’s time in Chicago politics by the New Yorker‘s Ryan Lizza, Obama sat in a government office with John Corrigan, a Democratic consultant, who was charged with drawing districts in Chicago, and the pair remapped the swath of the Windy City Obama represented to help push him into mainstream politics. The consultant, who had volunteered on Obama’s campaign against Rush, helped Obama literally map out his political future—what Lizza calledthe most important event in Obama’s early political life.”

Fifteen years later, looking out over the U.S. House of Representatives while delivering his final State of the Union address, President Obama spoke of gerrymandering, the process of drawing political maps that are inherently unfair, with the full weight of his own—albeit unmentioned—history. “We have to end the practice of drawing our congressional districts so that politicians can pick their voters and not the other way around,” the president said. “If we want a better politics, it’s not enough to just change a congressman or a senator or even a president. We have to change the system to reflect our better selves.”

In the decade and a half since Obama tweaked his district to fit his ambitions, the technology that makes modern redistricting possible, graphical mapping software, has undergone a revolution. A new generation of computer programs are allowing anyone—not just professionals with years of training—to add their voice to the mix.

Laying down the line

As it turns out, drawing “perfect” districts is nearly impossible, but creating ones undeniably fairer than those concocted by career politicians isn’t all that difficult. With the help of new computer programs, it’s something even students can do, and it may be an answer to Obama’s question of how to make government more accountable.

Legislative districts are generally redrawn in the United States every 10 years, after the results of the decennial census provide information about how populations have shifted. The sheer number of choices for how to draw a voting district map is staggering. There are more ways to divide the state of Wisconsin’s thousands of equal-population census blocks than there are quarks in the universe, so it’s unsurprisingly that, for more than half a century, Americans have looked to computers to do much of the heavy lifting.

Well-drawn districts are compact, contiguous, and respectful of how communities are divided up in space. They should have relatively equal populations and not intentionally deprive one group a voice by splitting it up into a bunch of districts where it is overwhelmed by others. Drawing lines that respect these characteristics would be simple if people’s settlement patterns were simple. But every map, even those drawn with the most sincere of intentions, privileges one factor over another, leading to skewed results. 

If districts are drawn along municipal lines, the Democratic-leaning voters in an area’s urban center would get short shrift against a bevy of more Republican suburbs on the outskirts. Yet, if the districts are drawn to split up each suburb and match it with a swatch of the city, each district will be dominated by Democrats with Republican communities losing their voice. Intentionally drawing the lines to force one of these two outcomes—the thing Obama did years ago and condemns today—is a major problem good-government groups aim to solve.

The real fear about heavily gerrymandered districts is that they effectively lock legislators into safe districts with no fear of losing in a general election. As a result of gerrymandering, the only way to force lawmakers out of office is a successful primary challenge, typically coming from their party’s more extreme flank—exacerbating partisanship and pushing the parties to the extremes.

“Districting is just one of several arenas of conflict between the parties because the parties now differ from each other so significantly. … It further intensifies the partisan wars,” said Thomas Mann, a senior fellow at the Bookings Institution. “Now that elections are becoming more partisan, there’s less benefit to incumbency, in and of itself, outside of a partisan advantage. It really just protects them now in times when the national electoral tide is flowing against their party.”

There are more ways to divide the state of Wisconsin’s thousands of equal-population census blocks than there are quarks in the universe.

Starting in the 1960s, after the Supreme Court issued a series of decisions mandating all congressional districts have roughly equal population, there were a series of efforts to automate redistricting in a way that was blind to political concerns. Notably, in 1965, Stuart Nagel published an article in the Stanford Law Review entitled “Simplified Bipartisan Computer Redistricting,” which laid out “a simple and politically feasible computer program, which can reapportion a legislature or other body of people who represent geographical districts.”

Programs like Nagel’s were created and even distributed for free; however, not only were the computer systems required to run the programs far too expensive for most state governments, but the results they spit out wouldn’t satisfy legal requirements or even be plausible.

Instead, as it turns out, computers aren’t that great at drawing lines by themselves—they need some human help. As with anything potential substitution of human agency for algorithmic precision, there have been concerns that adding computing power to the equation could ultimately make gerrymandering worse. “A computer may grind out district lines which can totally frustrate the popular will on an overwhelming number of critical issues,” wrote Supreme Court Justice John Marshall Harlan II in a dissenting opinion on a 1969 redistricting case.

Those fears were echoed 40 years later by Justice Stephen Breyer, who opined, “The availability of enhanced computer technology allows the parties to redraw boundaries in ways that target individual neighborhoods and homes, carving out safe but slim victory margins in the maximum number of districts, with little risk of cutting their margins too thin.”

According to an analysis led by Micah Altman, director of research for the MIT Libraries’ Program on Information Science and a leading voice on redistricting technology, computers aren’t making politicians any better at gerrymandering than they were a generation ago. Instead, the effect of the technology is one that fundamentally democratizes the entire project.

“Our investigation shows that computers were adopted practically universally in the 1991 round of redistricting,” the study reads. “By the 2011 round of redistricting, mapping software has become substantially faster and cheaper, but it’s fundamental capabilities had not changed dramatically. The timing of almost universal adoption and the relative continuity of computer capabilities suggest that much of the blame assigned to computers for modern redistricting excess has been misplaced.”

For the people, by the people

In the early 1990s, Howard Slavin, founder and president of the software firm Caliper, noticed a disconnect between the rapidly increasingly processing power of desktop PCs and the expensive, specialized hardware being used for redistricting. “We observed that people who were doing redistricting were using very expensive mini-computers that filled a small room and cost up to a million dollars,” Slavin said. “The software for it cost hundreds of thousands of dollars.” 

Caliper had been producing mapping software for tasks like transportation planning and market analysis, but the company saw an opening. Slavin originally planned to market the company’s standard mapping software for redistricting, but he soon realized that a modified product tightly tailored to the needs of groups drawing lines on a regular basis would do even better, and Maptitude for Redistricting was born. At first, the company sold it to what Slavin called “financially disadvantaged” groups—such as the NAACP and Southern Christian Leadership Committee—that wanted to check if the lines drawn by politicians were biased against the constituencies they represented.

The program includes features like the Minority Balance Report, which can instantly compare the representation of a demographic group inside a given district to the group’s representation within a geographic area, like a city or county. If you want to see if boundaries are systematically disenfranchising African-Americans, just load up the shape of the district and click a button.

Over the ensuing years, Maptitude became the industry standard. It’s used by the national committees of the both the Democratic and Republican party, government organizations in over half of the 50 states, city and state governments from California to New Hampshire, as well has hundreds of educational institutions, nonprofit groups and corporations. It’s a powerful tool that brings in myriad data sets to show all the consequences of bunching homes into one group or another.

Even so, the process was slow to open up. “In 2000, you could count the number of redistricting plans submitted by good government groups and other members of the public on two hands,” said Mann. “A decade later, we saw hundreds of legal plans, some created by middle school students.”

What sparked this flood of plans were the development of new software—often distributed for free—designed to make drawing model districts as simple as possible. One such project, which Mann was involved in promoting, is called District Builder, the open-source product of a group founded by Altman called the Public Mapping Project. “The idea was to create tools to allow other guys to participate in the redistricting process,” Mann said. “We tried to bring publicity to the effort to put pressure on the people who actually drew the lines.”

Another effort, released in 2009, is called Dave’s Redistricting App. The work of a single programmer, it’s the first piece of redistricting software capable of being run in a standard Web browser for free. While it isn’t powerful enough to produce maps that meet the legal qualifications for official maps, it allows virtually anyone to try their hand at drawing lines.

Shortly thereafter, Caliper released its own program, Maptitude Online Redistricting, which allows governments to post plans on the Internet for the public to review and comment. States like New York, Idaho, and Arizona use the software in an effort to make the process more transparent.

“Technology has changed redistricting from something where the most you could do was complain about the results to something where you can actually go out and propose alternatives,” Slavin said. “In general, public plans are more competitive and balanced then legislative plans. … It’s clear the crowdsourced plans are better on balancing different criteria—from compactness to competitiveness—than ones drawn by politicians.”

“There’s too much at stake to let citizen’s groups really have a say when partisans control the process.”

Local and state governments in states like California and Florida have used tools like District Builder to allow members of the public to draw and then submit their own maps. Similar efforts have been made by good-government nonprofits and media organizations. Since the technology has such a shallow learning curve and intuitive interface, no prior experience or extensive training is necessary. Philadelphia even offered cash prizes for the best citizen-drawn maps as part of a program called Fix Philly Districts.

“I’d say our greatest success,” said Altman, “is in changing gerrymandering from a problem that people could only complain about afterwards to one in which they could to participate in solving,”

After former Virginia Gov. Bob McDonnell—who was later convicted on 11 counts of corruption—failed to follow through on his 2009 campaign promise to form a redistricting commission, the good-government groups that supported McDonnell approached Dr. Quentin Kidd, the director of the Judy Ford Wason Center for Public Policy at Christopher Newport University, who suggested organizing a redistricting competition among Virginia college students using District Builder.

As recounted in a paper by Altman and Public Mapping Project co-founder Michael McDonald, after the contest was announced, McDonnell issued an executive order establishing an independent, bipartisan commission that would provide recommendations to state lawmakers in charge of drawing districts. However, the governor neglected to give the commission any money and required it to present recommendations in two months.

“A knowledgeable redistricting expert might have concluded that … [the commission] was designed to fail,” Altman and McDonald wrote, “since organizing a commission alone might take two months and redistricting is a difficult task even without such restrictive time constraints.”

Luckily, there was already an infrastructure in place that could do much of the work the commission was tasked with accomplishing: the student competition, which was tweaked so that the plans generated by students could be submitted to the commission.

Around 150 students from a dozen colleges submitted plans for creating new congressional districts and each of Virginia’s two legislative bodies. “Redistricting plans were initially judged on districts’ contiguity, population equality, respect for the Voting Rights Act, respect for existing political boundaries and communities of interest, compactness, number of competitive districts, and the overall political fairness of the redistricting plans,” the paper reads.

Mann and American Enterprise Institute political scientist Norm Ornstein served on the panel of judges awarding cash prizes to the best plans. A plan submitted by law students at William & Mary University was adopted by redistricting commission as a centerpiece of its report to the legislature. Years later, redistricting in Virginia is a still a bitterly partisan mess, but the technology gave the public the rare opportunity to influence the discussion, at least a little.

The last mountain

The post-2010 census was a watershed for public participation in redistricting. Altman noted that the number of public plans submitted to government entities saw a 100-fold increase from one decade prior.

Nevertheless, simply expanding access to producing maps isn’t enough to get real change when the entire process is largely managed by politicians themselves. “If you had a nonpartisan commission drawing the lines, then I think they could have an effect,” Mann said with a sigh. “But there’s too much at stake to let citizen’s groups really have a say when partisans control the process.”

In most states, legislators are responsible for drawing and then approving congressional lines. However, a small handful—Alaska Arizona, California, Idaho, Montana, and Washington—employ independent commissions that typically exclude politicians or other elected officials. While Obama didn’t go into detail in his State of the Union about how he wanted to fix gerrymandered districts, a push toward more states adopting these independent commissions would go a long way toward that goal.

The hope is that commissions will be more responsive to the public than politicians primarily interested in keeping their jobs safe from serious competition. With the changing guard in redistricting software, that public could finally have a real voice.

Update 3:50pm CT, Jan. 25: This article has been updated to include information about Maptitude Online Redistricting. 

Illustration by Max Fleishman

Layer 8
The Islamic State is now broadcasting a satellite TV channel in Iraq
A new direction of the militants' media arm.
From Our VICE Partners

Pure, uncut internet. Straight to your inbox.