Report 39 — Gerrymandering, Safe Seats, and the Engineering of Electoral Choice

Full report

On this page

Research date: October 6, 2026 Scope: Partisan and racial redistricting, competitive representation, safe seats, federal and state legal constraints, independent commissions, and the effect of machine intelligence on map design.

Executive conclusion

Gerrymandering is one of the clearest examples of a democratic institution being formally preserved while its competitive substance can be weakened.

Voters still vote. Ballots are still counted. The manipulation occurs earlier: the political system chooses which voters will be grouped together before those voters choose their representative.

Partisan gerrymandering typically works through two basic techniques:

• packing opposition voters into a small number of districts they win overwhelmingly;
• cracking opposition voters across multiple districts so they remain minorities and win few seats.

The result can be a durable mismatch between statewide voter support and legislative seats, reduced competition, increased incumbent insulation, and a shift in accountability from general-election voters toward primary electorates.

In 2019, the Supreme Court held in Rucho v. Common Cause that claims of excessive partisan gerrymandering present political questions beyond the reach of federal courts because the Court found no judicially manageable federal constitutional standard for deciding how much partisanship is too much. The Court explicitly said that its ruling did not condone excessive partisan gerrymandering and noted that states and Congress possess tools to address the problem.

Rucho did not eliminate all redistricting law. Racial vote dilution remains prohibited by Section 2 of the Voting Rights Act, and the Supreme Court reaffirmed the established Section 2 framework in Allen v. Milligan (2023). State constitutions and state courts may also impose rules against partisan gerrymandering, and independent redistricting commissions remain constitutionally permissible. In Arizona State Legislature v. Arizona Independent Redistricting Commission (2015), the Supreme Court upheld Arizona voters’ use of an initiative to place congressional redistricting in an independent commission.

1. Why district boundaries matter

In a single-member district system, representation depends on geography.

Suppose a state has 10 congressional districts and voters divide 52% to 48% between two parties.

A neutral-looking distribution could produce something close to 5–5 or 6–4.

But if the 48% side is heavily packed into three districts that it wins 80–20 and cracked across the remaining seven, the 52% side might win seven seats despite only a modest statewide advantage.

Nothing about the ballots themselves needs to be fraudulent.

The distortion is produced by district design.

This is why gerrymandering can be understood as pre-election manipulation of representational structure.

2. Packing and cracking

Packing

Concentrate opposition voters into districts they win by overwhelming margins.

Votes beyond what is necessary to win do not produce additional seats.

Cracking

Split opposition voters across many districts so that they consistently lose by smaller margins.

Together, these techniques can convert a narrow popular-vote advantage into a large seat advantage or preserve legislative control even when statewide support shifts.

3. Safe seats change what “the election” means

When districts become strongly one-party, the decisive election may no longer be the November general election.

It may be:

• a low-turnout primary;
• a party convention;
• an endorsement process;
• candidate recruitment by donors and party officials.

This has several consequences.

General-election accountability declines

A representative in a district that is 70% aligned with one party may face little realistic threat from the opposing party.

Primary incentives increase

The official’s greatest electoral risk may come from a challenger within the same party.

This can reward candidates who appeal strongly to the most active primary electorate rather than the median voter of the district.

Voters may feel strategically irrelevant

Opposition-party voters in a safe district can rationally believe their general-election vote is unlikely to change representation.

Over time this can reduce participation and trust even when the election itself is honestly administered.

4. The Rucho decision and the federal judicial gap

In Rucho v. Common Cause (2019), the Supreme Court considered partisan gerrymandering challenges from North Carolina and Maryland.

The Court held that partisan-gerrymandering claims present political questions that federal courts cannot adjudicate because there is no judicially manageable constitutional standard for deciding when ordinary political districting becomes impermissibly partisan.

Crucially, the majority stated that the conclusion did not condone excessive partisan gerrymandering.

Instead, it pointed toward political remedies:

• state constitutional provisions;
• state statutes;
• state courts;
• independent commissions;
• congressional legislation under federal election powers.

The practical result is important:

there is no general federal constitutional lawsuit available merely because a congressional map is excessively partisan.

That places much greater weight on state law and political reform.

5. Racial gerrymandering and partisan gerrymandering are not the same legal claim

This distinction is essential.

Section 2 of the Voting Rights Act prohibits voting practices, including redistricting plans, that result in racial or language-minority vote dilution under the statutory framework.

The Department of Justice describes vote dilution as an electoral structure that minimizes or cancels out the voting strength of protected minority groups.

In Allen v. Milligan (2023), the Supreme Court reaffirmed the longstanding Thornburg v. Gingles framework for Section 2 vote-dilution cases.

Thus:

• partisan gerrymandering generally lacks a federal constitutional judicial remedy under Rucho;
• racial vote dilution remains subject to federal statutory review under the Voting Rights Act.

The two can overlap factually because race and partisanship can be correlated, but the legal doctrines are different.

6. Why both major parties have incentives to gerrymander

Gerrymandering should not be analyzed as a defect unique to one party.

When state political actors control mapmaking, the governing coalition has an incentive to convert temporary electoral control into durable structural advantage.

The mechanism is symmetric even when its use is not equal in every state or decade.

That is why reform should focus on who draws the map and what rules constrain them, not on assuming one party will always be the beneficiary.

A system that depends on the temporary majority voluntarily refusing available structural advantage is fragile.

7. Geographic sorting complicates the picture

Not every noncompetitive district is gerrymandered.

Americans are geographically sorted by:

• urban versus rural residence;
• education;
• industry;
• race and ethnicity;
• age;
• cultural preferences;
• housing costs;
• migration patterns.

Dense urban concentrations can create naturally lopsided districts even under neutral rules.

Therefore a serious analysis should not infer gerrymandering solely from unequal seat outcomes.

Instead, analysts examine:

• map geometry;
• alternative maps;
• partisan symmetry;
• compactness;
• county and municipal splits;
• communities of interest;
• demographic patterns;
• intent;
• historical election data.

8. Machine intelligence changes redistricting

AI and optimization software can intensify gerrymandering because mapmakers can evaluate millions of possible district configurations.

A political actor can optimize simultaneously for:

• expected seat count;
• incumbent protection;
• demographic composition;
• legal compliance;
• compactness appearance;
• minimizing visible anomalies;
• resilience under forecasted vote shifts.

The old stereotype of a grotesquely shaped district is therefore increasingly outdated.

A mathematically optimized gerrymander can look visually reasonable.

But the same computing power can be used defensively.

Researchers and regulators can generate large ensembles of legally valid maps and ask whether the enacted map is an extreme partisan outlier.

Machine intelligence is therefore dual-use:

it can optimize capture or expose capture.

9. Independent commissions

One major reform strategy is moving redistricting away from ordinary legislatures.

Arizona voters created an independent redistricting commission through a ballot initiative in 2000. The Supreme Court upheld the use of that commission for congressional redistricting in 2015.

Independent commissions vary widely.

Important design variables include:

• who appoints commissioners;
• whether elected officials can serve;
• partisan balance;
• role of independents;
• transparency of meetings;
• mapmaking criteria;
• public-comment requirements;
• judicial review;
• staff independence.

A commission can itself be captured if appointment rules are poorly designed. “Independent” is not a magical property; it must be engineered.

10. Reform options

Independent or citizen redistricting commissions

Remove direct control from incumbent legislators.

Benefit: weakens self-dealing.

Risk: appointment systems can reproduce elite influence in another form.

Transparent map criteria

Publish priorities such as:

• population equality;
• Voting Rights Act compliance;
• compactness;
• contiguity;
• preserving communities of interest;
• limiting unnecessary county splits;
• partisan fairness where state law permits.

Public map data and reproducibility

Release census data, election data, software, proposed maps, and scoring methods.

This allows citizens and researchers to reproduce the mapmaking process.

Ensemble analysis

Generate many legally compliant maps and compare the enacted map to the distribution.

If an enacted plan is an extreme partisan outlier, officials must explain why.

Multi-member or proportional systems

These can reduce the importance of district boundaries because representation depends less on winner-take-all single-member geography.

Tradeoff: federal House elections are currently governed by statutory single-member district requirements, so major change would require legislation and careful Voting Rights Act analysis.

11. How gerrymandering can produce democratic decay

The mechanism is cumulative:

1. temporary majority wins mapmaking authority;
2. majority draws districts maximizing future seat advantage;
3. many general elections become noncompetitive;
4. primary voters become more important than general-election swing voters;
5. incumbents become less responsive to opposition constituencies;
6. polarization can increase;
7. legislative majorities become less sensitive to statewide changes in public opinion;
8. the map helps reproduce the coalition that drew it.

This is one of the clearest examples of a system making itself harder to remove through ordinary voting.

12. AMIANISM interpretation

Within the Throne framework, gerrymandering matters because political power is allowed to redesign the arena in which that power will later be judged.

The democratic principle should be:

the officeholder should not choose the electorate more than the electorate chooses the officeholder.

Redistricting cannot be perfectly apolitical. Geography always requires choices.

The goal is therefore not to eliminate judgment but to force mapmaking power to answer through:

• transparent criteria;
• public data;
• independent review;
• reproducibility;
• legal challenge;
• citizen participation.

Core finding

Gerrymandering weakens democracy by changing the translation between votes and representation before Election Day. It can leave ballots formally free while making many districts structurally noncompetitive and insulating governing coalitions from changes in public preference. The most durable solution is not trusting a preferred party to draw fair maps, but designing institutions in which no temporary majority has unilateral power to engineer its own future electorate.

Primary and high-value sources

https://www.supremecourt.gov/opinions/18pdf/18-422_9ol1.Pdf

1. U.S. Supreme Court, Rucho v. Common Cause, 588 U.S. ___ (2019).

https://www.justice.gov/crt/redistricting-information

1. U.S. Department of Justice, Civil Rights Division, “Redistricting Information.”

https://www.justice.gov/d9/2024-04/Section%202%20Guidance%20on%20Redistricting.pdf

1. U.S. Department of Justice, Guidance under Section 2 of the Voting Rights Act for redistricting and methods of electing government bodies.

https://www.supremecourt.gov/opinions/boundvolumes/576BV.pdf

1. U.S. Supreme Court, Arizona State Legislature v. Arizona Independent Redistricting Commission, 576 U.S. 787 (2015).

1. Allen v. Milligan, 599 U.S. 1 (2023), discussed in DOJ Section 2 guidance above.