A recent analysis explores the relationship between objective, local price increases and shifting voter choices during the 2024 United States presidential election. The research indicates that the Republican candidate experienced slight electoral gains in counties with higher rates of inflation, particularly in lower-income areas. These findings were published in the journal Electoral Studies.
Political scientists often study a concept called economic voting. This is the idea that voters reward or punish politicians based on the financial health of the country. When the economy struggles, voters tend to cast their ballots against the political party currently in power.
Following the 2024 election, many journalists pointed to rising consumer prices as a primary reason for the outcome. Prior to this new analysis, researchers had only looked at survey data to explore this connection. Surveys ask individuals how they feel about the economy and who they plan to vote for.
Relying entirely on surveys presents a methodological challenge for researchers. A person’s political preference can influence their perception of the economy. For instance, someone planning to vote against the incumbent party might report feeling worse about the economy than someone supporting the incumbent.
Patrick Flavin, a political scientist at Baylor University, wanted to evaluate this economic dynamic using objective data. He sought to measure actual price increases across the country. He then wanted to see if those real-world price hikes correlated with changing vote totals.
Tracking inflation at the local level is surprisingly difficult in the United States. The federal government tracks national and regional price changes using the Consumer Price Index. This tool measures the average change over time in the prices paid by urban consumers for a market basket of consumer goods and services.
The government does not publish this specific inflation data for individual counties. Because no official county-level measure exists, researchers must find alternative ways to estimate local price shifts. Flavin needed a reliable dataset to see if local economic reality matched the national narrative.
To solve the data problem, Flavin turned to information provided by the Economic Policy Institute. This organization calculates a family budget metric to estimate the “income a family needs in order to attain a modest yet adequate standard of living” across the country. They break these estimated costs down for more than 3000 individual counties.
The institute calculates the cost of living across seven distinct spending categories. These categories include housing, food, transportation, healthcare, other necessities, childcare, and taxes. Flavin focused his analysis on the estimated costs for a household consisting of two adults and two children.
He compared the estimated local costs in 2023 to the estimated costs in 2024. By finding the difference between these two years, he created a working measure of county-level inflation. He calculated the percentage point change for each of the seven categories, alongside a measure of the total overall cost increase.
A percentage point is the simple numerical difference between two percentages. For example, moving from four percent to five percent is an increase of one percentage point. Flavin used these calculations to determine exactly how much prices rose in different parts of the country right before the election.
With the inflation data established, the researcher analyzed county-level voting records. He looked at the change in Donald Trump’s share of the two-party vote between the 2020 election and the 2024 election. A positive value meant the candidate gained ground, while a negative value meant he lost ground compared to his previous performance.
To isolate the specific impact of inflation, Flavin controlled for several other demographic variables. In statistics, controlling for a variable means using mathematical formulas to remove the influence of factors that might skew the results. For example, he accounted for the county’s unemployment rate, median age, and total population.
He also adjusted for the percentage of residents with a college degree and the percentage of white residents. Additionally, he included the rate of Evangelical Protestant residents in his statistical models. By neutralizing these factors, he could look more exclusively at the relationship between inflation and voting shifts.
The analysis revealed that overall inflation had a measurable connection to candidate performance. Across the country, Trump generally improved his vote share in counties that experienced higher overall cost increases. The researcher quantified this relationship using a statistical concept called a standard deviation.
A standard deviation is a number used to tell how measurements for a group are spread out from the average. A low standard deviation means the numbers are close to the average, while a high standard deviation means they are spread out over a wider range. By using this metric, statisticians can compare different types of data on a level playing field.
Flavin found that a one standard deviation increase in inflation predicted a 0.07 to 0.15 percentage point increase in vote share. This varied depending on the specific spending category being analyzed. Increases in the cost of childcare and taxes showed a positive correlation with improved performance for the Republican candidate.
However, the data revealed a different pattern for medical costs. Trump actually performed worse in counties that experienced higher inflation in the healthcare category. The researcher notes that citizens might simply trust the Democratic party more than the Republican party on the issue of healthcare.
Flavin also split the data into different groups to see if the trends held consistent across political boundaries. First, he looked at counties based on which candidate won them in the 2020 election. The results in this portion of the analysis were decidedly mixed.
In counties Trump won in 2020, he performed better where overall costs, childcare costs, and taxes were higher. In those same counties, he performed worse where transportation and healthcare costs saw higher inflation. In counties won by Joe Biden in 2020, Trump improved his margins where transportation and childcare costs rose, but lost ground where healthcare prices spiked.
The researcher then divided the counties based on their average income levels. He separated them into two groups, split by the median per capita income of roughly $54,000. Per capita income simply means the average amount of money earned per person in a specific geographic area.
In this income-based analysis, a very clear pattern emerged. The relationship between higher prices and an increased Republican vote share was much stronger in lower-income counties. In these areas, five of the eight inflation categories correlated with increased support for the challenger.
This pattern did not hold in wealthier areas. In higher-income counties, the Republican candidate actually lost ground in relation to three different inflation measures. The effect of inflation on voting behavior appeared heavily dependent on the local income bracket.
People in lower-income brackets often spend a larger percentage of their earnings on daily necessities. A sudden price increase for basic goods limits their purchasing power much faster than it does for wealthier individuals. The data suggests that this immediate financial pressure translated directly into shifting voter preferences.
In the United States, presidential elections are determined by the Electoral College. Because the national popular vote and several key swing states were decided by small margins, subtle local shifts hold great importance. A shift of a few percentage points in a handful of counties can alter the trajectory of a national election.
There are a few limits to keep in mind regarding this research. The study relied on estimates of a basic family budget rather than official federal inflation metrics. While this was a necessary workaround, it remains an approximation of the actual economic conditions in each county.
Additionally, the results for some specific categories were mixed and require further investigation. Future research could explore why inflation in certain sectors benefits one political party over another. Researchers might also examine how localized economic pressures interact with other voting issues, such as social policies.
Understanding these local dynamics could help analysts better predict how economic shifts influence close elections. The economy will likely remain a central focus for political scientists studying voter behavior in the coming years. Objective, local data provides a clearer picture of regional differences than national surveys.
The study, “Did Trump do better where inflation was worse? Evidence from county-level data,” was authored by Patrick Flavin.
Solicite contato
