The Infrastructure of Unaccountability: Gerrymandering, Prison Gerrymandering, and the Communities They Silence

What Gerrymandering Is, and Why “Both Sides” Doesn’t End the Argument

Gerrymandering is as old as the republic. The term dates to 1812, when Massachusetts Governor Elbridge Gerry signed a redistricting bill creating a district so contorted it resembled a salamander — which a political cartoonist promptly named a “Gerry-mander” in his honor. The practice has never stopped. What has changed is the sophistication of the tools available to those who do it, the scale at which it can be executed, and the Supreme Court’s formal decision, in 2019, to look the other way.

It is true that both parties gerrymander when they have the power to do so. Democrats drew an aggressive map in Illinois that reduced Republicans to their fewest House seats since the Civil War. Republicans drew maps in North Carolina, Texas, Georgia, and Florida that entrenched structural majorities in states where the electorate is genuinely competitive. When partisans of either side respond to this criticism with “well, the other side does it too,” they are not wrong about the fact. They are wrong about what it proves. The fact that both parties engage in a corrupt practice does not make the practice acceptable. It makes it structural — and that is precisely why it requires a structural solution, not a partisan one.

The connection to criminal justice is not incidental. It is structural. The communities that bear the greatest burden of mass incarceration — over-policed, under-resourced, and subject to prosecutorial practices that have been repeatedly shown to produce racially disparate outcomes — are precisely the communities whose political power is most systematically suppressed by both forms of gerrymandering. They are packed into uncompetitive districts. Their incarcerated family members are counted as residents of rural prison towns. And the legislators who represent those prison towns gain political power from the census count of people who cannot vote for them, do not live there, and derive no benefit from their representation.

Prison Gerrymandering: The Mechanics of a Specific Injustice

Every ten years, the Census Bureau counts every person in America. The count determines how many congressional seats each state gets, how federal funding is allocated, and how state and local legislative districts are drawn. The Census Bureau counts incarcerated people where they are physically located on Census Day — in the prison, not at the home address they had before incarceration and will return to after release.

This seems like a bureaucratic detail. It is not. It is the mechanism by which political power is transferred, in bulk, from the urban communities that produce most incarcerated people to the rural communities that house the prisons. The Prison Policy Initiative, which has documented this phenomenon exhaustively, describes the practical consequence clearly: the miscount artificially inflates the population and therefore political representation of the predominantly white and rural districts that contain prisons, and the urban communities of color that most incarcerated people call home lose political power.

The racial dimension compounds the injustice. Black and Latino people are incarcerated at significantly higher rates than white people — a product of policing patterns, prosecutorial discretion, and sentencing disparities that are themselves the subject of criminal justice reform advocacy. Because incarcerated people are disproportionately Black and Latino, and because prisons are disproportionately located in rural, predominantly white areas, the census count transfers population — and therefore political representation — away from Black and Latino communities and toward white rural communities. The Brennan Center’s 2026 analysis of redistricting simulations found that counting prison populations at home could produce 6 additional Black-majority districts in Georgia alone.

Partisan Gerrymandering and the Criminal Justice Reform Pipeline

The connection between partisan gerrymandering and the failure of criminal justice reform is not theoretical. It is mechanical. When district maps are drawn to guarantee safe seats for one party’s incumbents, those incumbents become accountable to the party base rather than to the broader electorate. They have no incentive to respond to constituents who want reform, because their electoral survival depends on satisfying a narrower coalition — one that, in heavily gerrymandered districts, is often farther from the political center than the general population of that state.

The Prison Policy Initiative’s annual guide to winnable criminal justice reforms has documented for years how state legislatures that are products of extreme gerrymandering consistently block or roll back reforms that have majority public support. Polling consistently shows that Americans across party lines support sentencing reform, alternatives to incarceration for nonviolent offenses, and reentry programs. But in a gerrymandered state legislature, a senator whose constituents want reform and whose district is drawn to be safe can ignore those constituents with near-total impunity. The math doesn’t punish them.

The 2025 gerrymandering surge made this dynamic even more explicit. Following the 2024 election, President Trump publicly called on Texas Governor Greg Abbott to redraw the state’s congressional maps mid-decade to deliver additional Republican seats. Missouri followed. North Carolina drew yet another map targeting a seat won by a Democrat by fewer than two points in 2024. These are not organic democratic outcomes. They are engineered ones — and the communities whose reform legislation gets blocked by the legislators those maps protect have no mechanism for electoral accountability within the existing system.

What the Supreme Court Said — and What It Left Out

In June 2019, the U.S. Supreme Court issued its opinion in Rucho v. Common Cause, a consolidated case involving extreme partisan gerrymanders in North Carolina (favoring Republicans) and Maryland (favoring Democrats). Chief Justice John Roberts, writing for the 5–4 majority, acknowledged directly that the maps at issue were “highly partisan” and that excessive partisanship in districting “leads to results that reasonably seem unjust.” He then held that the federal courts had no power to do anything about it.

The majority’s reasoning was that partisan gerrymandering claims present “political questions” — disputes that lack “judicially discoverable and manageable standards” for resolution. Because there is no neutral principle for determining how much partisan advantage is too much, federal courts cannot police the line. The decision was not that gerrymandering is acceptable. It was that the federal judiciary is not the right institution to stop it.

Justice Elena Kagan’s dissent was a detailed accounting of what the majority’s reasoning left behind. She argued that the Constitution’s guarantee of free and fair elections, the First Amendment’s protection against viewpoint-based discrimination, and the Equal Protection Clause’s prohibition on intentional vote dilution all provided workable standards. She argued that the majority’s ruling left voters with no remedy against maps deliberately engineered to make their votes structurally irrelevant.

What Rucho did not foreclose was racial gerrymandering claims, state court litigation under state constitutional provisions, and political remedies — independent commissions, legislative reform, federal legislation. It handed the problem back to the institutions that created it and said: figure it out yourselves. Several states have. Most have not.

States That Have Acted — and States That Haven’t

Thirteen states have passed legislation to end prison gerrymandering, counting incarcerated people at their home addresses rather than their prison locations for redistricting purposes. The states that have acted represent a range of political compositions — this is not a partisan issue in state legislatures the way it has become in Congress.

The Prison Policy Initiative identified 14 states in November 2025 as most urgently in need of prison gerrymandering reform before the 2030 Census: Alaska, Florida, Georgia, Kansas, Kentucky, Louisiana, Michigan, North Carolina, Oklahoma, Oregon, Texas, Vermont, West Virginia, and Wyoming. These states have both significantly prison-gerrymandered districts and significant racial impact — disproportionate counts of Black, Native American, or Latino residents in the wrong place.

The window is not infinite. The Census Bureau is unlikely to change its counting policy before 2030. States that want to count incarcerated people at home for the next redistricting cycle need to pass legislation now to have the data infrastructure in place when the census arrives. Every year that passes without action is a year closer to another decade of maps drawn on distorted population data.

The Criminal Justice Reform Pipeline — What Gerrymandering Blocks

To understand what is at stake, it helps to be specific about what criminal justice reform legislation looks like in states where gerrymandered legislatures have blocked it, and what it has looked like in states where fairer representation has enabled it to pass.

The Prison Policy Initiative’s 2026 guide to winnable criminal justice reforms documents 34 high-impact policy changes that have broad public support and documented effectiveness at reducing incarceration without increasing crime: ending driver’s license suspensions for unpaid fines, decriminalizing traffic offenses, reforming pretrial detention, ending solitary confinement for extended periods, restoring jury eligibility to people with prior convictions, expanding family contact for incarcerated people, and reforming sentencing minimums. These are not fringe proposals. They have passed in states across the ideological spectrum when legislators face genuine electoral accountability.

The 2025 political environment made this worse. After the 2024 election results installed a federal government openly hostile to criminal justice reform, the Prison Policy Initiative noted that reform advocates were largely “playing defense” — trying to preserve gains already made rather than advancing new legislation. California voters passed Proposition 36, rolling back reform policies. Mid-decade partisan remaps in Texas and North Carolina were explicitly designed to entrench majorities capable of blocking reform for the rest of the decade. The momentum that had produced genuine bipartisan reform under different political conditions was running directly into structural walls.

Solutions That Exist and Why They Haven’t Passed

The solutions to gerrymandering are not mysterious. They are documented, tested, and in operation in states that have adopted them. What they are not is easy to pass — for the obvious reason that gerrymandering is enacted by legislators who benefit from it, and reform requires those same legislators to voluntarily dismantle their own structural advantage.

Independent redistricting commissions are the most widely adopted reform. Commissions composed of citizens rather than partisan legislators — with explicit restrictions on who can serve and transparent processes for map-drawing — have produced demonstrably more competitive maps in California, Colorado, Michigan, and Virginia. The Brennan Center’s analysis of the 2024 election found that 74% of the seats that flipped between parties were either in states using fairer map-drawing processes or in states without the need to redistrict. Competitive maps produce competitive outcomes. That is not an accident.

Federal legislation could solve the problem uniformly across all 50 states. The Freedom to Vote Act, which passed the House in 2022 and had majority support in the Senate, would have banned partisan gerrymandering in congressional redistricting and required independent commissions in every state. It failed when the Senate fell two votes short of modifying filibuster rules to allow an up-or-down vote. The bill has been reintroduced in subsequent Congresses. It has not passed.

Prison gerrymandering reform at the federal level would require the Census Bureau to count incarcerated people at their home addresses — a methodology change the Bureau has declined to make despite sustained advocacy. Absent federal action, the solution is state legislation, which 13 states have now passed. The model bill developed by the Prison Policy Initiative provides a template. The remaining 37 states require only the political will to use it.

Why This Needs to End — Not Be Managed

The argument for tolerating gerrymandering, when it is made explicitly, usually sounds like this: it is a tool both parties use; the alternative — courts drawing maps — is its own form of undemocratic intervention; the solution is political, not judicial; and voters can always pass reform initiatives if they care enough. Each of these arguments contains a thread of truth. None of them add up to a case for the status quo.

Both parties gerrymander. Both parties’ gerrymandering is wrong. The solution to mutual corruption is not to accept it as inevitable — it is to remove it from both parties’ hands simultaneously, through independent commissions or federal standards. The argument that courts should not draw maps is correct and is the argument for independent commissions, not for leaving the process with partisan legislators. The argument that voters can fix it through initiatives is undermined by the Utah case, where voters did exactly that and the legislature overrode them.

For criminal justice reform specifically, the stakes are not abstract. They are measured in the lives of the 2 million people incarcerated in the United States on any given day, the millions more who are on probation or parole, and the communities — disproportionately Black and Latino — that have absorbed decades of over-policing, over-prosecution, and the cascading economic and social consequences of mass incarceration. These communities have been systematically denied the political power to demand accountability from the system that has harmed them. That denial is not an accident. It is the designed output of maps drawn to protect the legislators who will never have to answer for it.