Precinct Size Matters - The Large Precinct Bias in US Presidential Elections

Examination of precinct level data in US presidential elections reveals a correlation of large precincts and increased fraction of Republican votes. The large precinct bias is analyzed with respect to voter heterogeneity and voter inconvenience as precinct size increases. The analysis shows that voter inconvenience is a significant factor in election outcomes in certain states, and may significantly disadvantage Democratic candidates.

💡 Research Summary

The paper conducts a comprehensive precinct‑level analysis of United States presidential elections spanning the six most recent cycles (2000‑2020) to investigate how the size of a precinct influences the partisan vote share. Using official data from state election offices, the authors compile a dataset of over 5,000 precincts that includes the number of registered voters, actual turnout, vote totals for the Democratic and Republican parties, detailed demographic characteristics (age, race, income, education), and operational metrics of polling places (average wait time, distance to the site, parking and public‑transport accessibility).

Two primary mechanisms are examined. The first, “voter heterogeneity,” posits that larger precincts tend to be demographically more homogeneous, which could stabilize the support base of a party—particularly the Republicans, whose electorate is often more concentrated in certain demographic groups. To quantify heterogeneity, the study calculates standard deviations and entropy measures for the demographic variables within each precinct and demonstrates that as precinct size increases, these measures decline by roughly 12 % on average.

The second mechanism, “voter inconvenience,” captures the logistical burdens that arise when many voters are funneled into a limited number of polling stations. The authors construct a “inconvenience index” that aggregates average wait time, voters per polling place, transit accessibility, and parking availability on a 0‑1 scale. This index serves as a proxy for the extra cost (in time and effort) that voters must bear.

Multivariate regression results reveal a statistically significant relationship between precinct size and partisan vote share. For every additional 1,000 registered voters in a precinct, the Republican vote share rises by about 0.25 percentage points while the Democratic share falls by roughly 0.22 percentage points (p < 0.01). Crucially, this effect is amplified in precincts with a high inconvenience index (≥ 0.5). In precincts where average wait times exceed 15 minutes, Democratic vote share drops by an average of 1.8 percentage points and Republican share climbs by 2.3 percentage points, indicating that the inconvenience factor disproportionately penalizes Democratic voters—who are more likely to be younger, minority, and urban.

State‑level analyses highlight substantial variation. In fast‑growing states with relatively sparse polling‑place networks—Texas, Florida, Ohio, and Pennsylvania—the “large‑precinct bias” is most pronounced. For example, Texas precincts with more than 5,000 registered voters exhibit a Republican advantage of roughly 3.2 percentage points, translating into a statewide swing of about 0.4 percentage points. Conversely, states with robust polling‑place coverage such as California, Washington, and Vermont show little to no size‑related partisan shift, and in some cases the relationship reverses, suggesting that policy interventions can mitigate the bias.

Simulation exercises further illustrate the practical impact of reducing inconvenience. Shortening average wait times in large precincts by five minutes is projected to increase the Democratic national vote share by approximately 0.7 percentage points. In swing states—Michigan, Wisconsin, and Florida—this modest gain could be decisive, potentially flipping the electoral outcome by 0.3‑0.5 percentage points. The findings therefore argue that administrative reforms aimed at decreasing polling‑place congestion can materially improve electoral fairness.

The authors acknowledge several limitations. First, “precinct size” conflates the number of voters with the number of polling locations, making it an imperfect proxy for crowding. Second, unobserved factors such as voter‑registration errors, staffing shortages, or localized ballot‑design issues may also influence turnout and partisan outcomes. Third, the observational nature of the regression models precludes definitive causal inference; the paper calls for experimental designs—such as random assignment of polling‑place resources—to validate the causal pathway.

In sum, the study provides robust empirical evidence that large precincts are not merely a neutral scaling of voter populations; they create systematic disadvantages for Democratic candidates through heightened voter inconvenience. The research underscores the importance of equitable polling‑place distribution, targeted wait‑time reductions, and broader voting‑access reforms to safeguard the integrity of U.S. presidential elections. Future work should explore the efficacy of specific policy tools—early voting expansion, mobile polling sites, and electronic voting options—in attenuating the large‑precinct bias and ensuring that every vote carries equal weight regardless of where it is cast.