Stephen Ansolabehere is the Frank G. Thompson Professor of Government at Harvard University. He is an expert in public opinion and elections and has published extensively on elections, mass media, and representation, political economy, and public opinion, especially concerning energy and the environment. He is the director of the Center for American Political Studies at Harvard, as well as principal investigator of the Cooperative Congressional Election Study.
Christopher T. Kenny is a Ph.D. Candidate in theDepartment of Government studying AmericanPolitics and Political Methodology at HarvardUniversity. He focuses on redistricting and the USCensus, at the intersection of American politics andmethods. He is the author of over a dozen R packagesfor working with data and methods for politicalscience and voting rights research. He is a memberof the Algorithm-Assisted Redistricting MethodologyProject and is also affiliated with the Center forAmerican Political Studies at Harvard University.
Summary
Promise of the Reform
"Far too often, redistricting plans are gerrymandered, leaving politicians less accountable to the citizens they represent. Independent redistricting commissions are frequently proposed as a way to draw fair district lines that don’t favor either party or any candidate."
Key Takeaways From Existing Research
Gerrymandering occurs in many states and municipalities, leading to less responsive legislatures.
Redistricting is a game of tradeoffs, complicated by geographical limits.
Independent commissions have likely contributed to the creation of fair plans, but their impact is hard to measure properly, and it is likely far less than the effect of court interventions.
Important Questions that the Research Does Not Answer
How do each state’s redistricting statutes and constitutional provisions impact how commissions draw plans?
Introduction
“How can we ensure that maps are drawn fairly and that districts reflect what people want? How can we minimize or even eliminate the manipulation of districts for political ends?”
The US House of Representatives is one of the great innovations of modern representative government. It was designed to represent the people, not places or the will of local elites. The nation would conduct a census of people every decade, following which the states would draw new districts in line with the changing population. This was a wholly new and enduring way to ensure popular representation. Yet, 250 years later the United States continues to struggle with the most fundamental problem raised by an electoral system that relies on districts: how can the districts be defined to ensure that they reflect the population fairly and equitably?
It did not take much time before the fundamental problem with the Founders’ innovation reared its head. Following the 1810 Census, the leaders of the Democrat-Republican party in Massachusetts, crafted a congressional district snaking from the New Hampshire border to the towns north of Boston, a configuration that stacked the deck in their favor. The district was pilloried in the Whig newspapers and nicknamed the “Gerry-mander,” after the Democratic-Republican governor Elbridge Gerry and its supposed resemblance to a salamander. That term has become synonymous with the entire American redistricting process.
The grant of this power to the state legislature exposes a deep-seated political contradiction in the idea of popular representation in an electoral system that relies on districts. Either the districts become outdated quickly as populations shift, or the political system adjusts the districts to take account of population shifts. Allowing the politicians to perform this task gives them the power to shape the coming decade’s electoral outcomes. That seems fundamentally unjust, as parties, individual legislators, or dominant interests in the legislature can skew the districts in their favor.
Over the past two centuries, as representative democracy has spread globally as a political institution, dozens of nations, including the United States, have grappled with the question of how to solve this conundrum. There are two general approaches: constraints and processes. Constraints are rules and standards, such as requiring that all districts have equal populations. Processes are procedures for drafting and enacting new maps, such as allowing the legislature to do it or empowering an independent commission with that responsibility. There has been both extensive theorizing about these approaches and empirical analysis of the performance of these approaches to redistricting.
Constraints on Redistricting
How can we ensure that maps are drawn fairly and that districts reflect what people want? How can we minimize or even eliminate the manipulation of districts for political ends?
One approach is to impose constraints on the map-maker.1 The state might require that any legal map must have certain characteristics, otherwise it will not be used in elections. The US Constitution places constraints on map-makers in its guarantee of equal protection under the laws (the 14th Amendment) and its prohibition on the denial of the right to vote on the basis of race or color (the 15th Amendment). Federal laws, such as the Voting Rights Act and periodic apportionment acts, place further constraints, including setting the number of congressional districts to be drawn and assigning the number of districts to each state. State constitutions and laws impose still further constraints on what is acceptable for a legal map.
Each of these constraints reflects an underlying value of the society or an idea about how representation can be best achieved. For example, equal population requirements reflect the values of popular control of government and equality. And, each of these values is formalized into a standard. Famously, the standard of equality was formalized by the Supreme Court of the United States to mean One Person, One Vote.2 To implement the standard, legislation and court decisions often develop specific metrics or criteria.
The standard of One Person, One Vote, for example, has been formalized by the Supreme Court of the United States to require exact population equality of congressional districts. The populations of districts for the US House of Representatives within any state must not deviate by more than one person.
Of the many values and standards at play in drawing districts, six are central.
First, equal representation of population.
The Supreme Court allows deviations of up to 10 percent in state legislative and municipal districts but requires exact equality in the US House districts. Even still, the standard of One Person, One Vote is not absolute. The Court has allowed unequal population representation in special districts that do not have legislative powers.3
Second, geography. Districts must be contiguous and, in many states, they must have some degree of compactness. Districts cannot have too many jagged edges, be very long and narrow, or have elbows and bends. Some states prohibit donut-shaped districts, such as a district that covers a city and a second district that encircles that city.
Third, race. Although the 15th Amendment, enacted in 1870, prohibits abridging the right to vote on the basis of race, it wasn’t until the 1960s that the federal government intervened to enforce that right. In 2023, the Supreme Court reinforced the status of the Voting Rights Act as “settled law,” but it still requires careful interpretation of the law to implement it in each jurisdiction. Some states have adopted their own versions of federal laws, such as California’s Voting Rights Act.
Fourth, party. There is no federal standard that prohibits partisan gerrymandering, but most states have imposed some constraints to prevent it, either in the form of statutory prohibitions or court interpretations on the states’ constitutions.
Fifth, incumbency. Some states, like Florida, expressly forbid drawing districts to favor incumbents. These rules are less common than prohibitions against partisan favoritism.
Sixth, political boundaries and communities.
Nearly all states have laws that discourage the division of counties and municipal boundaries. Beyond administrative practicalities, respect for municipal boundaries and even neighborhoods are thought to improve the representation of communities contained in those areas. It is an open question, however, whether a community has more influence or gains better representation if it is entirely within one district or divided between two.
Balancing so many criteria complicates the task of the map-drawer, and, ultimately, tradeoffs must be made. Which of the values embodied in these criteria are primary and which are secondary? Many states’ laws provide a hierarchy of these criteria. The Fair Districts amendments to the Florida constitution, for example, states that prohibitions on intentional racial, partisan, and incumbency gerrymandering are “tier 1” criteria and take primacy over concerns about the geographic configuration of districts or respect for political boundaries, which are considered “tier 2” criteria. Ultimately, though, it is the political process that determines how to balance competing criteria.
Constraints, thus, are only part of the story. Their effectiveness depends on what happens if there is a violation of the rule and on who enforces these constraints. These are questions of process.
Theoretical Research on Redistricting Processes and Standards
“By the early 1960s states such as Alabama and Tennessee had not been redistricted in half a century.”
The process embedded in the US Constitution leaves it to the states to determine the rules governing elections, including the configuration of districts. Whatever the legislature and the executive negotiated became the law defining congressional and state legislative districts. The courts were called on in the relatively rare cases in which no deal could be reached through the normal legislative process.
Unfortunately, the exclusive reliance on the legislative process, without a check by the courts, led to considerable dysfunction. The state legislatures in essence got to draw their own districts, as well as the districts of their members of Congress. The process was even more tightly controlled as the responsibility to draw the map usually fell to the committee within the legislature whose duty this was and then only to the majority party members of that committee or the chair. If no deal could be reached, as was often the case, the old boundaries from the prior decade were left in place. By the early 1960s states such as Alabama and Tennessee had not been redistricted in half a century. The result was malapportionment, districts that heavily favored the majority party, and thoroughly entrenched incumbents.
How Gerrymandering Works
The challenge for the gerrymander-er is to solve the question of, for any distribution across geography of different types of people or different voters within a state, what strategy maximizes the probability that the interest or party that the gerrymanderer represents wins the majority of the seats in the legislature? What strategy maximizes the total number of seats? The analytical and mathematical formulation of this problem is called the theory of the optimal gerrymander.4 It boils down to a couple of key insights.
The optimal strategy depends on the underlying balance of the two competing interests or parties. If the opposing group is relatively small, say one fifth of the population, then the way to minimize its influence and maximize yours is to divide that group across multiple districts so that it does not constitute a majority of any district. This strategy is called cracking. If the opposing group is relatively large, say half the population, then the optimal strategy is to put as many people from that group into as few districts as possible. This strategy is called packing.5
Creating Fair Districts
This raises the question of what fair districts would look like. To answer that question, political analysts have considered what approach could be used to derive the “neutral” plan, a counterfactual to the optimally gerrymandered plan. Such a counterfactual must embody legal requirements, such as contiguity of all parts of each district, while eschewing partisan-motivated decisions. This turns out to be an exceedingly complicated object to define given the practical realities of the distribution of populations in actual states and cities.
“Legislatures, left to their own devices, cannot be trusted to create a legal district map that will respect the rights of all voters.”
Three solutions are frequently proposed. First, a neutral plan may be one than adheres to principles of neutral redistricting, such as partisan fairness. Second, a neutral plan may be one that is drawn without using specific information, such as ignoring partisanship or even drawn districts at random.6 Third, neutrality might be relative to an existing map, and we may ask whether a map improves on the prior map. Whichever definition of a neutral map is used, the history of redistricting in the United States amply demonstrates that unconstrained legislatures will pack and crack their way to plans that advantage the majority party. To attain something approaching a neutral map, one still needs a process that will yield such a map.
Redistricting Commissions
The first historical crack in the state legislatures’ monopolies over redistricting came with the interventions of the federal and state courts in the 1960s. More recently, reformers have turned to commissions as a way to further separate the drawing of maps from the interests of individual legislators and parties.
An independent redistricting commission, by virtue of being separate from and independent of the legislature, will not be subject to the same political forces. Its independence will produce fairness. That at least is the ideal. Although the institution may be independent of the legislature, it is not necessarily independent of the political forces that shape the legislature. Commissions, inevitably, are comprised of people who are embedded in the social, economic, and political fabric of a society. It is an empirical question whether such an independent body indeed acts independently of those broader political forces. In the next section, we will review the empirical research on the adoption of commissions and the fairness of maps.
The intervention of the courts and the creation of commissions are themselves recognition of the failure of the legislative process. Over the past two centuries, Americans have come to the realization that legislatures, left to their own devices, cannot be trusted to create a legal district map that will respect the rights of all voters. Nonetheless, the courts still defer to the legislature whenever possible. The US process, then, is one of respecting the intentions of the legislature, so long as its actions stay within legal bounds.
Empirical Research on Redistricting Processes and Standards
“Empirical research has further found that commissions improve partisan and racial fairness somewhat.”
The central question for institutional design is which processes and standards work. The problems of partisan or racial gerrymandering, it is hoped by many reformers, can be remedied with a shift away from legislatures to independent commissions or special masters, or to clear standards and criteria with strong enforcement. Measuring the effectiveness of different processes and standards has been a substantial area of empirical research.
Drawing Maps
While the composition of the US population has changed, so too has the institutional design of map drawing. Over a dozen states have adopted commissions to draw their state’s district lines.7
Four primary classes of commission have been identified.8 Advisory commissions (ACs) make proposals but do not have the final power to pass maps. Backup commissions (BCs) step in only if the primary method fails. Politician commissions (PCs) are composed of politicians and appointees, typically with some level of partisan balance. Independent Citizen commissions (ICs) are composed of citizens with the power to pass a plan without legislative approval.
Clearly, some of these are stricter than others. Legislatures can completely ignore Advisory commissions or may never need backup commissions. Politician commissions hold more power than Advisory or Backup ones, but may be more akin to insurance: they attempt to limit the worst-case impulses of a single party in power. However, when political power is not balanced within the commission, this is weaker. Independent Citizen commissions are clearly the biggest limit on direct politician influence, so we may expect them to have the largest effect on electoral outcomes.
In practice, most commissions are Advisory, Backup, or Politician Commissions, but there are now a handful of truly Independent Commissions, such as in Arizona, California, and New York. The National Conference of State Legislatures offers excellent resources on redistricting.
Academic research on the effectiveness of commissions has reached many different—and often conflicting—conclusions. Some commissions have succeeded in the most basic task of proposing a map in a timely fashion (e.g., California in 2012 and 2022 and Arizona in 2022), but others ground to a halt and failed to reach a decision on a map (e.g., New York 2022). Some commissions have avoided lengthy legal conflicts (e.g., California in 2012), while others have become mired in the courts (e.g., Arizona in 2012). The same decision-making challenges that havevexed legislatures potentially also affect commissions.
Empirical research has further found that commissions improve partisan and racial fairness somewhat. The effects are not substantial and often not statistically distinguishable from no effect at all. One study found that commissions tend to better preserve administrative units, like towns and counties, while also drawing more compact districts.9 Another, however, finds that commissions often protect incumbents.10
“The sweeping changes that many reform advocates anticipated have not yet materialized in many of the states that adopted commissions.”
Research has also shown that commissions produce more competitive plans and plans with better opportunities for minority representation.11 While, on average, commissions tend to draw plans with less partisan bias, large partisan bias does still appear in some commissions.12 Taken together, it is very hard to say if process reforms, such as commissions, have had a substantial impact. In general, it seems that they may perform slightly better on average than legislatures, but the sweeping changes that many reform advocates anticipated have not yet materialized in many of the states that adopted commissions.
Assessing the Work of Commissions
There are several reasons why it is hard to discern the effects of commissions. First, reforms are rarely a singular change, say from legislative-drawn mapping to commission-drawn mapping. Many other aspects of the redistricting law change at the same time, making it hard to isolate the impact of specific reforms.
Second, constraints and process reforms go hand-in-hand. When a state adopts a commission, it usually also adopts a set of constraints that did not exist before, such as specific rules for compactness and partisan fairness. The correlation between less “political” processes and stricter constraints is quite high. Very few states have strict constraints without some form of commission. (Florida is an example of such a state.) Third, clarity is vitally important. When the process is changed, not only are constraints adopted, but the clarity of the rules is improved. The shift to equal population districts under the One-Person, One-Vote precedents of the 1960s was extremely swift.13 One possible reason is that it is clear and simple to implement. Courts adopted a bright- line rule that congressional districts should be nearly perfectly equal in size, as a function of equal protection law. For state legislative and other sub-state districts, any deviation over plus-or-minus 5% is assumed unconstitutional.As such, drawing equal population districts requiresthe mapmaker only to be able to tally people.
A good example of the lack of clarity and the inability to enforce improvements is partisan gerrymandering. Dozens of measures have been proposed.14 Perhaps most notable are the deviation from partisan symmetry15 and the efficiency gap.16 State redistricting rules in Michigan and Ohio have gone as far as requiring consideration of the partisan bias within a plan. Yet, federally, partisan gerrymandering has been determined non-justiciable (i.e., the federal courts cannot hear partisan gerrymandering cases) precisely because there is not a singular partisan fairness criterion that follows from the principle of equal representation.
Fourth, it is hard to gauge what the effects of a single reform might be because redistricting involves tradeoffs among multiple values and criteria. A new process or a new constraint will affect how the legislature or commission makes tradeoffs among all the values at stake. Tightly limiting population deviations may require splitting more municipalities or worsening compactness. Increasing compactness might decrease minority representation. How trade- offs are made depends on every other factor or value under consideration as well as the values of those people involved in the process.
“It is unclear that specific changes in processes or constraints have definitive consequences, especially for partisan fairness.”
Finally, redistricting occurs in the broader political context, one that invariably involves checks and balances. Since the 1960s, there has been a long-term trend toward fairness and equity in redistricting. The movement to equal population districts equalized political power across populations, drastically shifting power in state legislatures. Much of this can be attributed to court involvement and the creation of clear standards for mapmakers to follow. Federal and state laws, including the VRA and Equal Protection, have clearly supported the creation of minority- opportunity districts. State legislatures do not act without a judicial check. The possibility of legal challenges has changed the way legislators think about the process. The effects of changing the process from the legislature to a commission should really be thought of as the effect of changing from a legislature plus the courts to a process of a commission plus the courts. The legal process has had a clear, substantive effect. Beyond that, it is unclear that specific changes in processes or constraints have definitive consequences, especially for partisan fairness.
The Role of the Executive
One other institutional check cannot be ignored, and that is the executive. The long-term trend toward partisan fairness of redistricting maps also coincides with long-term trends in electoral competitiveness in states and divided control of government. Several studies have found that divided control of state government at the time of redistricting results in maps that are much more fairly balanced. This is a simple result of institutional bargaining that comes with any piece of legislation. When the Republicans hold the governorship and the Democrats hold the legislature, or vice versa, the parties must come to a deal. As a result, the interests of the majority party in the legislature will be traded off against the interests of the governor’s party. From the 1990s to the 2010s most of the reduction in partisan biases came because of a rise in divided government in the states. Much as political commentators decried the gridlock and acrimony in such states, one important result was an increase in partisan fairness in the district maps.
Legislatures also act in a broader cultural context, one in which the values of what is acceptable or ethical action have shifted considerably. Many states, such as Massachusetts in 2011, without the impetus of reform or a requirement of partisan fairness, have taken apart long-standing partisan gerrymanders. Not all legislatures have taken that step, but many have. Given the complexity of the political process that creates the rules themselves, it is difficult to determine what has driven the shift toward partisan fairness in many states and why others have insisted on perpetuating practices of partisan gerrymandering. Process reforms and new constraints have likely contributed, but their effects are on the margins.
Most likely a broader shift in our legal and political system dating back to the early 1960s changed both the process and the political culture in subtle ways that have led the states to make redistricting, over the long run, fairer.
The trend toward fairness does not mean, however, that the desire to manipulate the boundaries for political gain has dissipated. Quite the contrary. They remain a constant feature of our politics, one that is rooted in the very idea of representation through political districts. Over the past half century, there has been a continued political effort to remove the effects of gerrymandering. These battles have been fought in the legislatures, the commissions, and, perhaps most importantly, the courts. Such disputes are costly and time-consuming, but they have been highly consequential, applying pressure on the states to cure the mischiefs of partisan and personal political ambition.
Conclusion
“The partisan biases in the US Congressional map nationally are negligible.”
The politics of redistricting cut to the heart of one of the fundamental problems of democratic politics. Elections and representation do not “just happen.” They require rules, laws, and management. Writing the rules, in turn, allows the elected officials and political parties to put their thumbs on the scale. Redistricting, then, is never without conflict, drama, and bad behavior. The sea change in the politics of redistricting came in the 1960s when the federal and state courts asserted their power to check the political process and to protect individual voters’ rights. The courts have not solved those problems, though their intervention significantly improved them.
The first two decades of the 21st century have seen the states embrace new processes and constraints on the redistricters—most notably, the adoption of commissions in several US states, an approach that is common elsewhere in the world, including the UK. The states have also imposed constraints such as requirements for compactness, respect for communities, and partisan fairness.
It must be acknowledged that the partisan biases in the US Congressional map nationally are negligible.17 Some states have significant biases, but those balance out, and the overall bias appears to be only a handful of seats in the Republican direction.
Commissions surely contributed to some of those improvements, but their independent impact is hard to distill. The decades since the Supreme Court intervened had seen the steady erosion of partisan biases in the states and nationwide, as well as improvements in racial representation. It is nearly impossible to separate the effect of the commissions from the national trends or from other institutional reforms, such as the imposition of constraints on the maps, or from changing norms that have led some legislatures to do the right thing.
Endnotes
1 Cain, B. E. and Butler, D., “Redrawing District Lines: What's Going On and What's at Stake,” The American Enterprise 1991, 2 (4): 28-40
Cain, B. E., “Redistricting Commissions: A Better Political Buffer?” Yale Law Journal (2012) 121 (7): 1808-1844
2 Ansolabehere, Stephen and Snyder, James, The End of Inequality: One Person, One Vote and the Remaking of American Democracy (WW Norton, 2008)
3 See for example, Salyer Land Co. v. Tulare Lake Basin Water Storage District, 410 U.S. 719 (1973)
4 Friedman, John N and Holden, Richard T., “Optimal Gerrymandering: Sometimes Pack, but Never Crack,” The American Economic Review 98 (1) (2008): 113–44
Grofman, Bernard, Owen, Guillermo, and Feld, Scott L., “Thirteen Theorems in Search of the Truth,” Theory and Decision 15 (3) (1983): 261–78
5 Grofman, Owen and Scott, Thirteen Theorems, 1983
6 Chen, Jowei, and Rodden, Jonathan, “Unintentional Gerrymandering: Political Geography and Electoral Bias in Legislatures,” Quarterly Journal of Political Science 8 (3) (2013): 239–69
Fifield, Benjamin, Higgins, Michael, Imai, Kosuke, and Tarr, Alexander, “Automated Redistricting Simulation Using Markov Chain Monte Carlo,” Journal of Computational and Graphical Statistics 29 (4) (2020): 715–28
McCartan, Cory, and Imai, Kosuke, “Sequential Monte Carlo for Sampling Balanced and Compact Redistricting Plans,” Annals of Applied Statistics (Forthcoming 2023)
7 Warshaw, Christopher, McGhee, Eric, and Migurski, Michal, “Districts for a New Decade—Partisan Outcomes and Racial Representation in the 2021– 22 Redistricting Cycle,” Publius: The Journal of Federalism 52 (3) (2022): 428–51 https://doi.org/10.1093/publius/pjac020
8 Cain, Redistricting Commissions, 2012
9 Edwards, Barry, Crespin, Michael, Williamson, Ryan D., and Palmer, Maxwell, “Institutional Control of Redistricting and the Geography of Representation,” The Journal of Politics 79 (2) 2017: 722–26 https://doi.org/10.1086/690633
10 Henderson, John A., Hamel, Brian T., and Goldzimer, Aaron M., “Gerrymandering Incumbency: Does Nonpartisan Redistricting Increase Electoral Competition?” The Journal of Politics 80 (3) (2018): 1011–16 https://doi.org/10.1086/697120
11 Warshaw, McGhee, and Migurski, Districts for a New Decade, 2022
12 Best, Robin E., Lem, Steve B., Magleby, Daniel B., and McDonald, Michael D., “Do Redistricting Commissions Avoid Partisan Gerrymanders?” American Politics Research, (December, 2021): 1532673X211053216 https://doi.org/10.1177/1532673X211053216
13 Ansolabehere and Snyder, The End of Inequality, 2008
14 Katz, Jonathan, King, Gary, and Rosenblatt, Elizabeth, “Theoretical Foundations and Empirical Evaluations of Partisan Fairness in District-Based Democracies,” American Political Science Review 114 (1) (2020): 164–78
15 King, Gary, and Browning, Robert X., “Democratic Representation and Partisan Bias in Congressional Elections,” American Political Science Review 81 (4) (1987): 1251–73
16 Stephanopoulos, Nicholas O., and McGhee, Eric M., “Partisan Gerrymandering and the Efficiency Gap,” University of Chicago Law Review 82(2015): 831
17 Kenny, Christopher T., McCartan, Cory, Simko, Tyler, Kuriwaki, Shiro, and Imai, Kosuke, "Widespread partisan gerrymandering mostly cancels nationally, but reduces electoral competition." Proceedings of the National Academy of Sciences 120, no. 25 (2023): e2217322120
About the Democracy Reform Primer Series
Narrowing the gap between research and public dialogue, the University of Chicago Center for Effective Government's Democracy Reform Primers responsibly advance conversations and strategy about proposed changes to our political institutions. Each Primer focuses on a particular reform, clarifies its intended purposes, and critically evaluates what the best available research has to say about it. The Primers do not serve as a platform for either authors or the Center to advance their own independent views about the reform; to the contrary, they serve as an objective and authoritative guide about what we actually know—and what we still don’t know—about the likely effects of adopting prominent reforms to our political institutions.
In some instances, the available evidence may clearly support the claims of a reform’s advocates. In other instances, it may cut against them. And in still others, the scholarly literature may be mixed, indeterminate, or altogether silent. Without partisan judgment or ideological pretense, and grounded in objective scholarship, these Primers set the record straight by clarifying what can be said about democracy reforms with confidence and what requires further study.
About the Center for Effective Government
The University of Chicago Center for Effective Government was founded in 2019 at the University of Chicago’s Harris School of Public Policy to help solve the problems of government ineffectiveness with a multi-faceted theory of action. The Center organizes its work and activities around three key areas—ideas, education, and eangagement—and builds bridges across differences between scholars, students, practitioners, leaders, journalists, and advocates. Through robust, innovative programming, the Center works to strengthen institutions of democracy and improve government’s capacity to solve public problems.