Redistricting

Introduction

Congressional redistricting, the focus of this explainer, occurs every ten years after the release of the decennial U.S. Census. Based on these data, state officials redraw boundaries so each district has approximately the same population, consistent with the principle of one person, one vote established in Reynolds v. Sims (1964) and Wesberry v. Sanders (1964)—around 760,000 people as of the 2020 redistricting cycle.

Redistricting occurs alongside reapportionment, the process of determining the number of seats each state has in Congress. There are 435 voting seats in the House of Representatives, and a state may gain or lose seats every decade depending on whether the state’s population has increased or decreased relative to other states. While the Constitution mandates that reapportionment happen every decade, it does not offer specific guidance on how to carry it out. Instead, federal law (Title 2 of the U.S. Code) dictates the reapportionment process, while Article I, Section 4 of the Constitution grants state legislatures the power to determine the “Times, Places and Manner of holding Elections” and Congress the ability to “make or alter such Regulations.”

Individual states require that the redistricting process follows certain criteria, which commonly include contiguity (the district is a single, unbroken shape, not scattered pieces), compactness (can be measured in many different ways, for example by minimizing the distance between all the parts of a constituency), and the preservation of communities of interest (areas with common social, economic, or political interests). While these criteria promote geographically coherent districts, some states additionally require competitiveness or partisan fairness. When districts are drawn strategically to favor a particular party or group, it is called gerrymandering.

Gerrymandering, explained

Gerrymandering is the intentional manipulation of district boundaries to unfairly advantage a party or group. The term was coined in 1812 after Elbridge Gerry, then governor of Massachusetts, signed a bill that created a salamander-shaped district to advantage the Democratic-Republican Party. Though states outline criteria like compactness and contiguity to curb gerrymandering, there remains a strong incentive to gerrymander in states where partisan actors like state legislators or political commissions oversee redistricting.

A grid-based visualization showing three scenarios of political redistricting on a 10x10 grid of 100 voters, split 50-50 between Blue/Democrat and Red/Republican parties. Individual voters are represented as dots organized in structured 5x5 blocks (each containing 12 Blue and 13 Red dots, with the middle dot changed to Blue in the bottom-left and top-right blocks).

Competitive: The grid is divided into four 5x5 square districts. Two districts (bottom-left and top-right) contain 13 Blue and 12 Red cells (won by Blue), and two districts (top-left and bottom-right) contain 12 Blue and 13 Red cells (won by Red), yielding an even 2-2 seat split where all districts are highly competitive toss-ups.
Packing: Democrats are packed as extremely as possible into a single safe district on the left (25 Blue, 0 Red) which they win 25–0. Because Democratic voting power is concentrated entirely here, Republicans win the other three districts (containing 14, 15, and 21 Red cells, respectively), resulting in a 3–1 seat split for Republicans.
Cracking: Republicans are packed into a single safe district in the top-right (20 Red, 5 Blue) which they win 20–5. The remaining Republican voters are spread out ("cracked") across the other three districts, allowing Democrats to win them by narrow margins (each having exactly 15 Blue and 10 Red cells), resulting in a 3–1 seat split for Democrats.

The two most common methods of gerrymandering, called packing and cracking, are visualized in Figure 1 above. The leftmost panel shows a competitive baseline with the same voters divided into four compact districts, Democrats (blue) and Republicans (red) each carrying two seats. The other two panels redraw the lines around those same voters to produce a different result.

Packing refers to the drawing of district lines to crowd as many voters of a particular party or group into as few districts as possible. Because their voting power is concentrated in so few districts, they win fewer seats overall. In the middle panel of Figure 1, Democratic voters are packed into the leftmost district, allowing Republicans to make up the majority in the other three districts. Cracking works by drawing district lines so that voters of a particular party or group are spread out across many districts, limiting their political influence. In the rightmost panel of Figure 1, Republican voters have been “cracked” such that they form a majority in just one out of four districts. These gerrymandering strategies can give a particular party an electoral advantage even when districts contain roughly the same number of voters.

How and when congressional districts are drawn

Who draws congressional districts

Two maps of the United States, side by side, comparing how each state drew its congressional districts in the 2010 and 2020 redistricting cycles. States are shaded by how much authority the mapmaking body holds: legislature, commission, advisory commission, backup commission, nonpartisan staff, and at-large district. Legislatures drew the maps in most states in both cycles, falling from 29 states to 26. Commissions with primary responsibility grew from six states to ten, adding Colorado, Michigan, Montana, and Virginia. Advisory commissions fell from five states to four, while backup commissions rose from two to three as Ohio shifted from advisory to backup. Iowa is the only state where nonpartisan legislative staff draw the maps. Six states had a single at-large district in the 2020 cycle, down from seven.

States distribute redistricting authority across a spectrum rather than in clean categories, and commissions in particular can vary widely state to state. The categories shown in Figure 2 group states by how much authority the mapmaking body holds, not by who sits on it. The National Conference of State Legislatures, the source for the figure, applies the term commission to any non-legislative body given a role in redistricting and does not separately designate independent commissions.

Legislature: In most states, the state legislature draws congressional maps as ordinary legislation subject to the governor’s veto. This concentrates mapmaking power with the same partisan actors who stand to benefit from it.

Commission: A body outside the legislature holds primary responsibility for drawing the maps. These bodies vary widely in composition. Some, like Michigan's, are built to exclude sitting officials, candidates, and lobbyists and to seat a mix of partisan and unaffiliated members; others, like New Jersey's, are made up of elected officials and their appointees, so politicians retain control. Figure 2 does not distinguish between the two.

Advisory commission: A commission drafts or recommends maps, but the legislature retains the authority to adopt them. The commission's role is to assist, not to decide.

Backup commission: The legislature draws the maps in the first instance, and a commission takes over only if the legislature misses its deadline or cannot agree.

Nonpartisan staff: Iowa alone assigns map drawing to nonpartisan legislative staff, who work without access to political or election data, including the addresses of incumbents. The legislature still votes the maps up or down.

Since the 2010s, a wave of reform, mostly enacted through ballot measures, has changed congressional redistricting processes and moved a number of states away from legislative control:

Colorado (2018): Voters approved Amendments Y and Z, removing mapmaking power from the legislature and creating two independent 12-member commissions—one for congressional districts (Amendment Y) and one for state legislative districts (Amendment Z). Each seats four Democrats, four Republicans, and four unaffiliated voters, with any map requiring at least eight votes, including two from the unaffiliated members.

Michigan (2018): Voters passed Proposal 2, an initiative that took authority from the legislature and established the Michigan Independent Citizens Redistricting Commission: a 13-member panel of randomly selected applicants (four Democrats, four Republicans, five independents), with sitting politicians, recent candidates, and lobbyists barred from serving.

New York (2014): Voters approved Proposal 1, creating a 10-member redistricting commission (each of the four legislative leaders appoints two members, who then choose two unaffiliated members). In practice, the commission only advises. The legislature can reject its proposals and, on a second failure, amend and adopt its own map, which is why Figure 2 shows New York as an advisory commission.

Ohio (2018): Voters passed Issue 1, a constitutional amendment creating a tiered process for congressional maps: the legislature first attempts a plan requiring a three-fifths supermajority that includes at least half of each of the two largest parties; failing that, a seven-member bipartisan commission takes over; and if the commission cannot agree, the task returns to the legislature under relaxed rules.

Utah (2018): Voters passed Proposition 4, creating an independent advisory redistricting commission and banning partisan gerrymandering for both congressional and state legislative maps. In 2020, the legislature repealed much of it (via SB 200), effectively regaining full control of mapmaking. In July 2024, the Utah Supreme Court held that voter-approved government-reform initiatives are constitutionally protected from unfettered legislative repeal. Based on this ruling, a state district court struck down SB 200 in August 2025, reinstating Proposition 4 and forcing a new congressional map.

Virginia (2020): Voters approved Question 1, a constitutional amendment transferring mapmaking authority from the legislature to a 16-member bipartisan commission—eight legislators (four Democrats, four Republicans, split evenly across the two chambers) and eight citizens. The commission’s maps must still be approved by the General Assembly. Note that Figure 2 above shows Virginia as a commission state, since the National Conference of State Legislatures classifies the commission as holding primary responsibility for drawing the plan.

Note: Not every change visible in Figure 2 reflects a deliberate reform. Montana, for instance, shifted from a single at-large seat to a commission only because it gained a second congressional district after the 2020 Census, giving its existing districting commission a congressional map to draw for the first time in decades.

When Congressional districts are drawn

Redistricting typically occurs once per decade, but nothing in federal law confines it to that schedule. In League of United Latin American Citizens v. Perry (2006), the Supreme Court held that neither the Constitution nor Congress prohibits a state from redrawing districts mid-decade. A number of states bar or constrain mid-decade redistricting under their own constitutions or statutes. Only two states had voluntarily redistricted for partisan advantage mid-decade between 1970 and 2025, but a cascade began in 2025 when Texas, pressured by the Trump administration, enacted a new congressional map intended to allow Republicans to gain up to five seats. This prompted California voters to approve a countervailing map (Proposition 50) adding about five Democratic-leaning seats. Missouri, North Carolina, and others soon followed.

Recent legal controversy

Allegations of partisan and racial gerrymandering are pervasive, with legal cases frequently brought before state and federal courts. These cases have forced states to redraw district boundaries mid-cycle and set important precedents for what forms of gerrymandering are permissible. High-profile Supreme Court cases in recent years have upended the status quo and ushered in a new era of redistricting battles.

Partisan gerrymandering

In recent years, two cases brought before the Supreme Court have considered the legality of mapmakers drawing lines to advantage their party: Gill v. Whitford (2018) was brought by Wisconsin voters challenging a Republican-drawn state assembly map they characterized as extreme partisan gerrymandering. The Court unanimously held that the plaintiffs failed to demonstrate a “concrete and particularized” injury caused by the gerrymander and so did not rule on the question of partisan gerrymandering itself.

The following year, North Carolina voters, the North Carolina Democratic Party, and democracy watchdog Common Cause brought a similar case before the Court regarding the state’s remedial 2016 congressional map in Rucho v. Common Cause (2019). In a 5-4 ruling, the Court declared partisan gerrymandering a “political question” outside federal jurisdiction, blocking future federal lawsuits on the issue. However, legal battles continue to play out in state legislatures, courts, and ballot measures. Critics of the decision argue it opens the door for mapmakers to engage in more extreme partisan gerrymandering, as well as racial discrimination masquerading as partisan gerrymandering.

Racial gerrymandering

Perhaps the most heavily litigated area of redistricting is the drawing of districts based on their racial composition. Under the Equal Protection Clause of the Fourteenth Amendment, the Supreme Court recognized in Shaw v. Reno (1993) that a district drawn to separate voters by race can be unconstitutional. Miller v. Johnson (1995) then set the operative test: a map becomes constitutionally suspect when race, rather than criteria like compactness and contiguity, is the predominant factor used to draw a district. On the other hand, Section 2 of the Voting Rights Act (VRA), as interpreted in Thornburg v. Gingles (1986), forbids maps that deny minority voters a realistic opportunity to participate equally in the political process and to elect candidates of their choice. In short, the Constitution restricts the use of race in redistricting, while the Voting Rights Act can require mapmakers to consider it. This tension drives recent landmark redistricting litigation with significant consequences for both map design and political representation.

After the 2010 Census, the North Carolina legislature raised the Black share of two districts (NC-1 and NC-12) above 50 percent. The heads of the House and Senate redistricting committees offered a different justification for each district. They said NC-1 was drawn to comply with Section 2 of the Voting Rights Act, but denied using race for NC-12, defending it as a partisan gerrymander. North Carolina voters David Harris and Christine Bowser were registered to vote in the new districts but alleged that the legislature had packed Black voters into the districts on the basis of their race. The voters’ case eventually reached the Supreme Court in Cooper v. Harris (2017). In a 5-3 decision, the Court ruled that the North Carolina General Assembly relied on race too heavily. In the majority opinion, the Court argued that race was the predominant criterion used to create the new map and that North Carolina had not given a substantial reason for giving it such weight. However, legal scholars noted that the decision did not resolve the underlying tension. For example, the Harvard Law Review commented:

“Given the independent harms of partisan gerrymandering and the unique entanglement of race and party, it is both unnecessary and futile for courts to tackle the impossible challenge of distinguishing between racial and partisan gerrymandering. Rather, the fact that Harris could easily have gone the other way suggests the need for an additional standard that addresses partisan gerrymandering directly and thus allows courts to police more accurately the affront to democracy that both racial and partisan gerrymandering represent."

Following the 2020 Census, Black voters in Alabama challenged the state's new congressional map in Allen v. Milligan (2023). They argued that drawing only one majority-Black district unfairly diluted their political power, given that Black residents make up about a quarter of the state's population. Many observers expected the Court's conservative majority to use the case to weaken Section 2. Instead, in a 5-4 decision, it declined to do so, reaffirmed the Thornburg v. Gingles (1986) decision, and ordered Alabama to draw a second district in which Black voters could elect their candidate of choice.

Just a year later, in Alexander v. South Carolina State Conference of the NAACP (2024), the NAACP argued that South Carolina had moved tens of thousands of Black voters out of its 1st Congressional District to make the seat safer for Republicans, creating a racial gerrymander. In a 6-3 decision, the Court disagreed. It called a lower court's finding that race had predominated a “clear error.” More consequential than the result was the reasoning: the Court strengthened the presumption that legislatures act in good faith. It held that when a state offers a partisan explanation for its map, challengers generally must produce an alternative map showing the same partisan goal could be achieved without the racial effect. A state can therefore defend a map that disadvantages minority voters by attributing it to party, a motive Rucho v. Common Cause (2019) placed beyond federal review.

The same year, Louisiana redistricted to create a second majority-Black district in accordance with Allen v. Milligan (2023). A group of non-Black voters then challenged that new district as an unconstitutional racial gerrymander. Their case, Louisiana v. Callais (2026), was heard by the Supreme Court twice: first in March 2025, then again that October after ordering a rare reargument to address whether the intentional creation of a majority-minority district (where more than half of the population is a racial or ethnic minority) under Section 2 of the VRA violates the Fourteenth or Fifteenth Amendments. On April 29, 2026, the Court ruled in a 6-3 decision that the new district could not stand. Because the VRA had not, in the Court's view, required a second majority-Black district in the first place, Louisiana had no compelling interest to justify its reliance on race. The decision also added new requirements for proving a Section 2 claim—including that challengers show racially polarized voting cannot be explained by partisanship—making claims of vote dilution, the harm Section 2 of the VRA is meant to remedy, substantially harder to bring going forward. Together with Alexander v. South Carolina State Conference of the NAACP (2024), this decision left both the constitutional and the statutory routes for challenging maps that weaken minority voting power considerably narrower than before.

Redistricting literature

Gerrymandering and redistricting research is exceptionally broad, with a vast literature exploring how the redistricting process affects party dynamics, electoral politics, and representation. While partisan gerrymandering is often seen primarily as a tool for securing partisan advantage, research suggests that its aggregate partisan effects may be limited: Kenny et al. (2023) find that widespread partisan gerrymandering largely cancels out at the national level because both parties benefit from favorable district maps. Where a single party does control the process, however, the advantage is real—Jeong and Shenoy (2024) estimate that the party drawing the map gains roughly 11 percentage points in its probability of winning a given House seat. Gerrymandering also substantially reduces the number of competitive districts by creating safer seats (Carson, Cottrell, and Williamson 2025), though whether this deepens political polarization is contested (McCarty, Poole, and Rosenthal 2009). These changes have downstream impacts on elections: competitive districts modestly increase voter turnout (Ainsworth et al. 2024), while partisan gerrymandered districts can depress turnout (Jones, Silveus, and Urban 2023) and reduce trust in government and support for institutions (Williamson and Justwan 2025). The effect on turnout is not uniform, however. Voters who feel disadvantaged by a gerrymander may themselves be mobilized to vote (Izatt 2025).

Redistricting has implications for political representation. Moving voters into districts with unfamiliar representatives can weaken the legislator-constituent relationship and reduce voters’ familiarity with the officials who represent them (Hood and McKee 2010). How representatives interact with constituents also changes. Following redistricting, legislators become more reliant on campaign contributions from donors outside their districts, weakening their linkage with and accountability to constituents (Crespin and Edwards 2016). At the same time, representatives adapt their communication strategies to the partisan composition of their newly drawn districts (Kaslovsky and Kistner 2024), indicating that legislators can modify their behavior in response to the changes in the constituencies they serve.

These representational concerns are particularly salient for racial and ethnic minority communities, which have been the focus of much redistricting scholarship. A substantial body of work examines how creating majority-minority districts (in which more than half of the population is a racial or ethnic minority) affects that group’s political representation in state legislatures. In his review article titled “Race and Redistricting” (2022), David Canon of the University of Wisconsin–Madison defines the practice of racial redistricting as “redrawing district lines to advantage one racial group of voters over another.” Canon explores the scholarly debate on this issue, citing arguments of authors who advocate for racial redistricting and those who oppose the practice. Opponents of racial redistricting favor a “colorblind” approach to representation, arguing that race is no longer as central to politics as it was in previous decades (Thernstrom and Thernstrom 1997, Thernstrom 2009). Some also argue that drawing districts on the basis of race segregates minority voters and amounts to a “triumph of tokenism,” or the appearance of representation without the substance. (Guinier 1991). Proponents believe that descriptive representation—having elected representatives accurately mirror the demographics of their constituents (Guinier 1995, Phillips 1995, Hawkesworth 2003, Young 2000)—is essential for cultivating trust among the electorate (Mansbridge 1999). From this perspective, representatives who share the racial or ethnic backgrounds of their constituents may be better equipped to understand and represent the political preferences of historically marginalized communities. More recent work has expanded this conversation beyond the election of minority representatives to consider how the redistricting process itself disproportionately affects minority communities. For example, Gaynor and Hayes (2026) find that demographic minorities are more likely than the general population to experience district boundary changes and are more likely to be moved collectively during redistricting.

Redistricting can also influence public policy and the distribution of government resources. For example, Ansolabehere, Gerber, and Snyder (2002) analyze the influence of redistricting on public finance. They find that counties with more legislative seats per person before a court-ordered redistricting process received more funds from the state afterward, while counties that lost seats received less. They also find that redistricting and population equalization diverted approximately $7 billion annually from overrepresented to underrepresented counties.

A growing body of work approaches redistricting computationally, using algorithms to draw and evaluate districts. By generating large numbers of alternative maps that follow a state’s own rules, these methods can reveal whether an enacted plan is a statistical outlier—strong evidence of a gerrymander—and such analyses increasingly appear in redistricting litigation. Approaches range from Markov chain and sequential Monte Carlo sampling (DeFord, Duchin, and Solomon 2021; McCartan and Imai 2023) to mathematical optimization; notably, Chuang, Hanguir, and Stein (2024) show that drawing maximally competitive districts is computationally hard, which helps explain why fair maps can be difficult to produce even in principle.

Redistricting’s financial, social, and electoral consequences are well documented at the federal and state levels. In contrast, local redistricting processes are harder to audit and quantify. Brian Amos, a professor at Wichita State University, chronicled these difficulties in a piece for MEDSL. He claims many of these challenges stem from a lack of accessible data. While some states have their redistricting data available digitally (on a website or by request) in a clean, interactive format, it is more common for this data to be presented as a printable PDF. These PDFs are sometimes low-quality and difficult to reproduce. Furthermore, Amos has found that some of these local district maps do not meet the constitutional requirement of near-even population numbers (such as Floyd County, Indiana).

Search