---

Summary

Key Takeaways from the Research

• District elections increase descriptive representation for racial/ethnic minorities when the group is moderately sized and segregated.
• District elections can allow neighborhoods to protect their interests.
• At-large elections encourage councils to consider a citywide perspective.

Important Questions We Need Better Answers To

• Are local legislative districts subject to the problems of gerrymandering that have been identified at higher levels of government?
• Does election type have policy consequences for local governments?
• Are marginalized communities’ substantive interests best served by at-large or district elections?

---

Introduction

The adoption of at-large systems typically flowed from efforts to dilute Black voters’ power in Southern cities after the Civil War and the municipal reform movement that sought to promote a citywide perspective in governance.

Local governments in the United States feature a wide variety of governing institutions. Some cities have elected mayors, others do not. Some cities hold elections concurrently with state and national elections, others do not. This brief explores another type of variation – the way in which local legislators are elected.

Cities in the United States tend to elect their city councils using two electoral system types: single member districts or at-large elections. When councilors are elected by district, the city is divided into geographic areas of roughly equal population size that elect a single member to the city council in a plurality or majoritarian contest. An at-large system is one in which members of the city council are selected by the entire city electorate. In most cities this means that voters are offered a slate of candidates and are allowed to vote for the same number of candidates as there are seats available. In some cases, cities allow casting multiple votes for the same candidate (e.g., if three council seats are up for election, the voter can cast three votes for whichever candidates they prefer). Cities often have majoritarian requirements such that if a candidate does not receive 50% of the vote, she is forced into a run-off election. Some cities designate seats or residency requirements for at-large positions turning the election into a series of single member contests, while other cities vote for only one at-large member in any given election. A small but growing number of cities use mixed-systems, electing some council members by district and others at-large.

Most research on the effects of district/at-large elections is focused on general purpose local governments, e.g., cities. Cities (along with towns, villages, and boroughs) are incorporated places that are governed by elected officials, and typically have the power to raise and spend revenue, regulate the uses of land, and enforce the law. Cities are chartered by the state in which they are located, and the rules governing incorporation vary from state to state. The determination of how legislators are elected occurs when cities are incorporated. Many smaller cities are general law cities. Their governance structure, powers, and functions are set by state statute. Most midsized and larger cities are governed by a constitution that is approved (and amended) by the voters of the city. These charter cities have home-rule. That is, charter cities are free to enact laws without attaining state permission. They have full authority to set their own tax rates and decide how and where to provide services. Importantly, they are able to determine various institutional features, such as how the city council is elected. Charter cities can and do change the way that their legislators are elected.

Up until the late 1800s, cities almost universally had district (or ward) elections.² The adoption of at-large systems typically flowed from two sources: efforts to dilute Black voters’ power in Southern cities after the Civil War and the municipal reform movement that sought to promote a citywide perspective in governance and limit the power of parochial neighborhood concerns.³ According to one commentator, “It is contended that at best [the district system] lends itself to the election of members who will pay more attention to the needs of their district than to the larger interests of the city as a whole; and that the concentration of the worst elements of the city’s population in some wards makes inevitable the election of a number of very objectionable members”.⁴ At-large systems spread rapidly throughout the United States in the first decades of the 20th century. The number of cities using pure at-large systems has declined over the past 20 years, but at-large elections remain a common feature in city politics.

The passage of the 1965 Voting Rights Act (and the 2002 California Voting Rights Act) laid the groundwork for legal contestation of at-large systems. Several court cases, including the landmark Thornburg v. Gingles,⁵ have found at-large systems to discriminate against Black voters by diluting their voting power under certain conditions. Voting must be racially polarized (where different racial groups vote cohesively for different candidates) and the racial group that is in the minority must be sufficiently large and compact to plausibly constitute a majority of voters in a single district. Over the last several decades, the court has required many legislative bodies to adopt district elections when these conditions are met. Nonetheless, today about 68% of cities elect their members at-large.⁶

The way in which members of the local legislature are elected affects who wins election to office, who is represented in the political system, and what policies are made. Generally, at-large elections shift representation toward voters rather than residents. In a districted system, regardless of the level of turnout in an area, the area receives representation on the council. In at-large systems, the voting majority selects the winners. Theoretical work by one group of researchers showed that at-large systems will be used as a disenfranchising mechanism when the minority group is small.⁷ When the minority group is closer to 50% of the population, the majority will prefer gerrymandered districts to maintain power.

---

One of the most persistent findings by scholars of urban politics is that single member district elections increase descriptive representation of historically underrepresented racial and ethnic groups on city councils.⁸ This effect has been found to be particularly strong for African Americans.⁹ Districts have also been found to be beneficial to Latinos.¹⁰ These statistical findings have been supported by extensive case study and historical research as well.¹¹ In sum, “the effect of …districts is unequivocally…greater equity”.¹² These consequences typically result when the minority group in question is large enough and sufficiently concentrated (e.g., segregated) to elect representatives in a districted system.¹³ However, one study also showed that at-large systems disadvantage Black candidates because they are more cognitively taxing for voters.¹⁴

The literature on the representation of women finds mixed effects for single member districts. Most research has concluded that districts are either meaningless¹⁵ or disadvantageous for women candidates.¹⁶ However, one paper found that women benefit from district elections. It argued that this is because at-large elections are typically more competitive and require more campaign funds and larger mobilization efforts for candidates.¹⁷

Single member district elections increase descriptive representation of historically underrepresented racial and ethnic groups on city councils.

Scholars have also sought to understand how electoral institutions affect women of color given conflicting institutional effects. Generally, scholars find that Black men, but not Black women are advantaged by districts.¹⁸

---

Because district systems tend to elect more officials of color, they also tend to make policy that is more responsive to minority interests. Higher proportions of minority councilors are associated with increases in minority appointments to the bureaucracy¹⁹ and increased focus and/or spending on minority policy issues.²⁰ Several scholars find that Black school board members are associated with more Black administrators and teachers; and that a higher proportion of Black teachers is associated with a higher proportion of Black students in gifted classes, as well as higher grades and test scores among Black students.²¹ Others produce similar findings for Latinos.²² One study found that increases in minority representation leads to higher test scores for minority students, likely because of increased funding and the hiring of more minority principals.²³

Furthermore, there is evidence that electing minorities in district elections produces better substantive outcomes for minority residents compared to legislatures in which minorities are elected to the council in at-large elections.²⁴ More specifically, a study found that in at-large systems increasing the number of Black or Latino school board members has no effect on the hiring of black administrators but find a powerful effect in districted systems. The paper argued that the explanation for the difference is that in the at-large setting Black and Latino administrators must attend to the preferences of the median (usually white) voter and as a result are more constrained in producing benefits for minority communities. In districted systems, where logrolling frequently prevails, minority communities appear to be better served. This is contrary to theoretical findings by another researcher, who demonstrated that policy outcomes are more likely to favor minority preferences under at-large systems, even though districts are more likely to produce descriptive representation.²⁵ Holding constant Black representation, another study found that at-large systems produce better representation for Black residents.²⁶

These results echo findings in the Congressional politics literature that explores the policy consequences of majority-minority districts and shows that such districts can undermine minority interests.²⁷ This kind of outcome is produced when the minority community is packed into a small number of districts such that their representatives are typically on the losing side of legislative votes. As a result, some scholars argue that minority interests will be better served when minority voters are able to impact the election of a larger number of legislators, even if this means they will not be descriptively represented.²⁸

We know virtually nothing about the present state of gerrymandering/district line drawing in local elections. Most municipalities hold non-partisan elections and those with partisan elections are frequently dominated by a single party. This means that it is unlikely that partisan gerrymandering shapes decisions about district lines. Given that district maps are adopted by the sitting legislature, it is likely that any bias in line drawing will tend to advantage incumbents and entrench existing hierarchies along race and class lines. So, it may be that the lessons learned from analyses of state and federal policy outcomes will be relevant at the local level. But new research is required to determine whether or not this is the case.

What we do have evidence of in local politics is the degree to which neighborhood versus citywide interests are represented under different election systems. At-large elections tend to promote a citywide perspective on policymaking, as opposed to a neighborhood perspective. One study found that district councils are more factionalized, with more geographically focused divisions and councilors playing a greater ombudsman role compared to at-large councils.²⁹ Another paper found that changing from at-large to district systems depresses the total amount of housing that cities build. Anti-development preferences are generally widely shared, but minority neighborhoods which tend to lack political power in at-large systems, are often saddled with new development. The implementation of districts, and thus a greater ability to dictate policy outcomes, means that minority neighborhoods are better able to resist development – and so, overall new housing declines.³⁰ Researchers found that districted cities tend to have larger budgets (spending, debt, and taxation), likely as a result of legislator deference to neighborhood priorities (e.g., log-rolling).

Electing minorities in district elections produces better substantive outcomes for minority residents compared to legislatures in which minorities are elected to the council in at-large elections.

However, another paper found that districts do not always have this effect. It showed that neighborhoods were routinely disregarded in favor of citywide needs in controversial decisions in one districted city (Los Angeles).³¹ Another piece of research also found that election type does not affect aggregate policy outcomes including per capita taxes and the share of taxes raised by sales tax.³² Thus, it remains something of an open question whether election type has a causal effect on policy outcomes, and for whom these outcomes are beneficial.

---

Conclusion

It is clear that district elections increase the likelihood that Black and Latino legislators will be elected when the population of Black and Latino residents is large enough to elect a councilor in a single member district and is residentially segregated.

The median city council has six members, and so, this means that Black and Latino populations must typically constitute at least 10% of the population and live in neighborhoods with heavy concentrations of Black/Latino residents to see an effect from districts.

Women appear to have a slight advantage in district elections as well, likely because of the lower cost of campaigning and lower levels of competitiveness in elections that cover a limited geographic area.

Scholars have also provided evidence of an association between descriptive representation and substantive representation. That is, having more Black and Latino legislators is correlated with city policy that reflects the interests of Black and Latino residents.

However, we do not have clear causal evidence that election type (district vs. at-large elections) generates policies that are preferred by marginalized communities.

Having more Black and Latino legislators is correlated with city policy that reflects the interests of Black and Latino residents.

We have stronger evidence that districts do offer the opportunity for neighborhoods to protect their interests. Generally though, more research is needed on the effects of election type on policy outcomes and whose interests are served by these outcomes.

Finally, it remains uncertain whether municipal legislative districts suffer from the problems of gerrymandering that plague district line-drawing at higher levels of government.

---

References

---