the seats. In four other states that year, the party that won a majority of the votes in congressional races got fewer than half the seats.1 In North Carolina in 2016, Republican House candidates received 53 percent of the vote but 10 of 13, or 77 percent, of North Carolina’s House seats.2 How can these results have happened? Perhaps more importantly, is there any way in which these outcomes can be considered democratic?
This book addresses both of those questions. The short answer to the first of them is that the congressional district boundaries in Pennsylvania, like legislative district lines in many states, were gerrymandered. Gerrymandering is the process of drawing legislative district boundaries to give one party (or group of voters) an electoral advantage over others.
Gerrymandering in the United States is quite unpopular with the public. According to a bipartisan poll conducted in December 2018, 63 percent of all likely 2020 presidential voters had an unfavorable view of partisan gerrymandering. Another 32 percent had no opinion while just 5 percent had a favorable view.3 Those views were shared, with only slight variation in the percentages, by Democrats, Independents, and Republicans alike. When respondents were asked if they would prefer districts with no partisan bias, even if it meant fewer seats for their own party, or districts with partisan bias, assuming that their own party would win more seats, only 15 percent chose biased districts while 65 percent preferred unbiased districts.4
Nevertheless, when legislators have the opportunity to gerrymander district lines, many – perhaps most – of them will seize the opportunity. Voters are unlikely to punish their own party for doing so (despite their stated preference for unbiased districts) and legislators can enhance their party’s power by creating additional districts in which they have an electoral edge. With little downside and the potential for gaining seats in the state legislature or in Congress, gerrymandering is hard for politicians to resist.
The second of our questions is the more difficult one. How one answers it will depend on what one means by ‘democracy’ and whether one thinks the redistricting process should be a normal part of politics. Though democratic elections are expected to be free and fair, it’s not immediately clear what would constitute a violation of this expectation.
The rest of this chapter will introduce gerrymandering by explaining, in a bit more detail, what it is and why it occurs. Gerrymandering is not unique to the United States but its practice here is in many ways exceptional. The chapter will then address the reasons that gerrymandering stirs so much controversy. Beyond the obvious power struggle that gerrymandering initiates, there are competing visions of how democracy ought to operate that are at play.
The Need to Draw District Boundaries
In any political system with meaningful legislative elections that take place in districts not demarcated by otherwise permanent boundaries (e.g., state or national boundaries), the lines around legislative districts will have to be drawn. In most places, these lines will be redrawn periodically to account for population shifts. This process of redrawing district lines is called redistricting or boundary delimitation.5
In the United States, redistricting typically takes place every ten years, following the constitutionally mandated national census. For congressional representation, census data is used for reapportionment, or the process of adjusting the number of members of the House of Representatives from each state based on changes in population. For example, as a result of the 2010 Census, Texas gained four seats in the House while New York and Ohio each lost two.6 District lines in states that gain or lose seats will obviously have to be redrawn. However, they’ll also be redrawn, even if only slightly, in states that did not gain or lose seats.7 That’s because, as we’ll see later in the book, it is now a legal requirement that legislative districts within a state have equal population sizes. This applies to state legislative districts as well, so census data will be used to redraw state House and Senate districts to ensure equal population sizes in those districts.
The states are responsible for drawing state legislative and congressional district boundaries. In most states, the state legislature draws district lines and adopts the new maps as they would any normal piece of legislation. Some states, however, let commissions established for this purpose draw the lines for state legislative and/or congressional districts. Regardless of the model a state employs, the process is virtually always political.
These two facts – that district boundaries must be redrawn regularly and that the redistricting process is political – create opportunities for those who wish to gerrymander districts. In countries where districts correspond to pre-existing administrative units, there is no opportunity to gerrymander because there is no need to redraw district boundaries. In Israel, for example, all 120 members of the Knesset (the national legislature) are elected nationally by proportional representation. In other words, the national border serves as the district boundary for the one (nationwide) legislative district. Few countries that do not redraw district lines are that extreme. There are a number in which electoral districts correspond to predetermined sub-national governmental jurisdictions (e.g., provinces or states).8 In the remaining countries, where districts don’t correspond to permanent territorial units, gerrymandering becomes a possibility.
Even in countries that utilize redistricting commissions, gerrymandering can occur. Of course, the extent to which this is possible depends on how the commission is constructed and how it operates. Nevertheless, even in places where redistricting commissions are supposed to be neutral, forms of gerrymandering can take place. The United Kingdom, for example, uses Boundary Commissions (one each for England, Scotland, Wales, and Northern Ireland) to establish parliamentary constituencies. Despite the fact that the Commissions are designed to be independent and non-partisan, Ron Johnston, Charles Pattie, and David Rossiter have found significant bias (in favor of the Labour Party) in the results of UK general elections through 2005. In part, this bias was the result of “increased efficiency of Labour’s votes.”9 This increased efficiency of the vote, in turn, is explained (at least in part) by the Labour Party’s efforts to influence boundaries during the Public Inquiries conducted by the Commissions.10 Thus, even when partisan operatives don’t control the redistricting process, they may nonetheless influence it.
It should be noted that electoral systems using single-member legislative districts (i.e., one representative per district) almost inevitably produce disproportional results. The percentage of seats won by the victorious party will usually be larger than the percentage of votes they received because of the winner-take-all nature of these districts. Winning districts with 75, 60, or even 51 percent of the vote results in 100 percent of the representation for those districts.11 However, this disproportionality is not the same as bias, as Johnston and his colleagues point out.12 We’ll discuss various definitions, and measures, of partisan bias later in the book. For now, it’s worth noting that gerrymandering, by definition, results in biased electoral results and bias is the chief problem with gerrymandering.
Of course, gerrymandering can sometimes be used for purposes other than maximizing the number of seats for a party. The protection of incumbents is another, quite common, use of gerrymandering (sometimes referred to as “bipartisan gerrymandering”). Though it is possible to protect incumbents while also maximizing seats for a party, it is generally thought to be difficult to do both effectively. To protect an incumbent in one place often means giving up a seat to the other party elsewhere. Regardless of why it’s being done, however, the root problem with gerrymandering is the same – it creates an unlevel playing field in a given district.
The need to redraw district lines, in and of itself, doesn’t create biased maps.13 Indeed, unbiased districts can be drawn, as the experience of many countries, and even many American states, demonstrates. Instead, it’s the political nature of the redistricting process that makes gerrymandering so hard to avoid.
To say that the redistricting process
