The geographic map of American job growth has always been mixed, and the present day is no exception. What we find interesting, especially ahead of the Midterm elections next week, is how the job growth of local labor markets varies by their political leanings.
Zooming into the nation’s metro areas, the only ones to have collectively lost jobs in the 20 months since Donald Trump took office are the most Blue. Redder parts of the country are doing better, and Purple Metros — metropolitan areas where nearly equal numbers of residents voted Republican and Democratic in the 2024 presidential election — have enjoyed the strongest job growth of all.
A quick methodological note before we dive into the analysis: We classify metropolitan areas by their political leanings based on a tally of votes cast for each candidate in the 2024 presidential election.1 (See the Appendix below for more detail.)
Purple metros are defined as those in which the Trump and Harris vote shares came within 3 percentage points of each other (a +/- 3 percentage point margin).
We also distinguish between metro areas that lean toward a particular party (+/- 3 to 10 percentage points) and those that strongly favor one (greater than 10 percent in either direction). In other words, a metro area is labeled as Very Red if Trump’s vote share exceeded Harris’ by at least 10 percentage points, and Very Blue in the other direction.
The battlegrounds
Ten percent of the country’s metropolitan areas are Purple. These political battlegrounds are home to 13 percent of the nation’s population. Together, the 38 Purple metros house 37.2 million Americans. Only the Lean Blue slice of the nation’s geography is smaller.
Very Blue metro areas house the most people (123.6 million, or 36 percent of the nation’s population), but Very Red metros are the most numerous, representing over half (54 percent) of all metro areas.
The two most Purple metropolitan areas in the United States are both in the Carolinas: Greensboro, North Carolina, and Columbia, South Carolina. The others are mostly scattered throughout the Southeast, Midwest, and West.
Deeper Red metropolitan areas tend to be smaller and concentrated in either the interior south or the central plains. The metropolitan scale reveals meaningful political differences across metros of similar sizes even within states and regions too, however. Metro Cincinnati appears much Redder than Cleveland or Columbus, while Tampa much Redder than Orlando and Miami. Tampa, Cincinnati, and Nashville are the three most populous Very Red metros.
Very Blue metro areas are fewer in number but greater in population. They tend to be either large coastal cities or smaller hubs of education, government, and knowledge economy jobs. Boulder, Colorado, is the most Blue metro area in the country, and West Coast tech hubs such as Silicon Valley and Seattle rank among the most Blue as well.
Resilience and stagnation amid slowing growth
Purple metros have demonstrated more resilience in terms of job growth over the past year and a half than the other political camps.
The figure below indexes employment levels aggregated across all metro areas in each group to 100 in January 2025, the month that President Trump took office. There was little difference in the employment trajectory across groupings in the runup to inauguration, but a distinct splintering set in after, with the gap widening at the start of 2026. Purple metros sustained their momentum longer after inauguration than other groups and also experienced a strong early 2026. Red metro areas followed similar general patterns, albeit with a more pronounced early slowdown.
By contrast, employment in Very Blue metro areas fell hard and fast after January 2025 and has only gradually recovered since hitting bottom later that fall. On net, Very Blue metropolitan areas still confront a jobs deficit relative to January 2025.
The next figure shows how the trend break for Very Blue metro areas was especially sharp. The rate of job growth slowed by more than half across Purple metros over the 20 months after Trump took office compared to the 20 months prior, but job growth collapsed completely in Very Blue ones. Growth has fallen sharply across Lean Red and Very Red metropolitan areas since inauguration, too.
The pervasiveness of growth follows a similar pattern. At the high end, 63 percent of Purple metro areas have registered job growth since inauguration, compared to roughly half (51 percent) of Very Blue metros at the low end.
Industries driving the trends and variation within Purple metros
Before expanding on the industries fueling each political group’s job trends, it is important to note some factors that are not responsible for the differences across metros.
Purple metros’ breakout job performance is not due to more favorable industry composition than other areas. A simple shift share (plotted below) shows that job growth between September 2024 and September 2026 would be near-identical in each metro type had it been driven by national industry performance trends.2 In fact, Very Blue metros, the worst performing group in reality, actually have a modest advantage over the four others in industry composition. The same industries are simply doing better in Purple and Red metros than in Blue ones.
Nor do some recent high-profile federal policy choices, like tariffs or federal government layoffs, fully explain the differences across the political group aggregates. Manufacturing jobs declined across all five groupings, ranging from -0.6 percent in Very Red to -2.2 percent Lean Blue metros, ruling out any broad-based tariff-induced manufacturing bump.
Similarly, federal government employment tanked across every political group, falling by between 7 percent in Lean Red metros and a whopping 12 percent (161,000 jobs) in Very Blue metros. Notably, Very Blue metro areas managed to largely offset those massive federal losses with growth of 125,000 in local and state government jobs over the same time period so that the net decrease in total government employment was less than 1 percent.
State and local employment differentiates Purple from Red metro areas, too. More new public sector jobs in Purple metros explains almost all of the difference in overall employment growth between them and Red metro areas.3
Some industry trends are shared by all political camps. Healthcare and education, for example, is growing everywhere — fastest in Purple metros but most voluminously in Very Blue ones, where healthcare and social assistance is one of the few expanding sectors. With all other sectors combined declining on balance, job growth in Very Blue metro areas would have been much more steeply negative without healthcare.
The map below of job growth since January 2025 shows how much variation our political groupings can still mask, however. The (Very Blue) DC metro area stands out as the most struggling large metro in the country, driven by disproportionately steep job losses in government and related sectors. A handful of Very Blue metros such as Charlottesville, Raleigh, and Asheville are thriving, by contrast — as are tech-heavy San Jose and Austin. Even among comparably Blue university towns, Champaign, Illinois, is adding jobs healthily while next-door Bloomington, Indiana, struggles.
Such local variation exists within the Purple group, as well. One fault line is visible around manufacturing: The fastest-growing Purple metros are much less exposed to manufacturing than the slowest-growing Purple metros on average. The exceptions are fast-growing Muskegon, Michigan, where more than one-fifth of the workforce is employed in manufacturing, and Dover and Des Moines, where slow growth stems from exposure to other sectors.
The three largest metros at the top of the Purple growth rankings — Charlotte, Kansas City, and Las Vegas — are very different places but, generally, have diversified economic structures and are less reliant on manufacturing than the national average.4
Some other of the fastest-growing Purple metros — State College, Pennsylvania, and Athens, Georgia — are college towns with major research universities. Notably, their strong performance contrasts with that of many of their peers similarly specialized in higher education including Ames, Iowa, another Purple metro with a university research hub.5 In general, Farm Belt metros across Iowa and Illinois and manufacturing-heavy metros like Toledo and Akron, Ohio, are among the most ailing Purple metros.
If top-line sector mixes offer only limited explanations for the different employment trends across the map, other factors must by definition be at play. Adverse tariff shocks may be much more localized than our broad geographic aggregates detect, for example. In addition, underlying population growth is strongly tied to job growth and helps partially explain the natural tailwinds enjoyed by booming Charlotte but not aging Cedar Rapids.
Furthermore, the deportation drive and slowdown in new immigrant arrivals might most-directly affect formal employment figures (i.e., the BLS jobs data analyzed here) for the metro areas where foreign-born workers have historically been concentrated. This could help explain the struggles of some Very Blue metro areas, though the strong job growth in Purple metro area El Centro (31.5 percent foreign-born) cautions that local nuance still holds here, too.
Conclusion
What implications might the political geography of job growth have on the 2026 election cycle? Your authors will refrain from potentially embarrassing prognostications in favor of informed observation.
This analysis of job growth suggests that the economy is doing worse on average in areas least likely to support the president or Republican party anyway, and the job market looks healthiest in the areas where Democrats need to rack up the most defections to change the balance of power, all else equal.
All else is not equal, of course. Metropolitan areas do not send representatives to Washington — states and congressional districts do. Economic logic does not feed into the latter’s boundaries. Voters do not cast their ballots with only economic considerations in mind, and gerrymandering will ensure that the relationship between local economic conditions and election results will be weaker than we geographers might otherwise anticipate.
Nevertheless, prevailing economic conditions remain electorally relevant. Job growth has shifted strongly downward across the map over the past twenty months. We’ll find out soon whether the downshift has been severe enough in the places that really matter in U.S. elections to contribute to a drubbing for the party in power on November 3rd.
Appendix: Metros and Methodology
Though political results are often mapped at the county level, this analysis is conducted at the metropolitan scale for several reasons. First and foremost, the Bureau of Labor Statistics provides data on employment levels by metropolitan area with only a one- to two-month lag (through September 2026, at the time of writing), allowing for a near-contemporaneous assessment of job trends. States have similar availability but less granularity (top-line growth in Florida says little about what is happening in Orlando, for example). Counties offer more granularity and cover the entire map, but the long time-lag (data is currently only available through Q1 2026) makes them less well-suited for assessing conditions in the months before the election.
Metropolitan areas are conceptually useful for economic analysis, too. In total, 294.5 million Americans reside in metro areas, or 86 percent of the population. Metro areas represent labor markets, making them well-suited for approximating Americans’ lived economic experience, such as whether they have access to good job opportunities and whether they see growth rising around them. Metropolitan areas also blend typically “Blue” central counties with more Purple suburbs and Redder exurbs. By spanning the full breadth of settlement patterns, metros offer much better microcosms of the nation than either counties or states. Metropolitan areas thus offer a novel, salient, and timely geography for political analysis.
Voting data was obtained at the county level from the U.S. Election Atlas and then aggregated to metros using the Census Bureau/OMB’s 2023 CBSA delineations.
SAE does not provide seasonally-adjusted figures at the metro-sector level, so September 2024 is chosen as the reference point for this section’s figures rather than January 2025 to account for seasonality. The total nonfarm employment figures used elsewhere are seasonally-adjusted, so they are reported with January 2025 as the reference.
Looking at only private sector employment, Very Red metros have a slight advantage, registering 1.08 percent growth over the period, compared to 1.03 percent for Purple metros and 0.95 percent for Lean Red metro areas, respectively. On the public sector front, Purple metros expanded state and local payrolls by 2.1 percent, compared to 0.7 percent in Very Red metros.
In September 2024, the Charlotte, Kansas City, and Las Vegas metros reported a manufacturing employment share of 8.0 percent, 7.8 percent, and 2.7 percent, respectively, compared to a national metro average of 9.7 percent. The average manufacturing share in the ten fastest-growing Purple metros was 8.0 percent (inflated significantly by Muskegon’s outlier share of 21.3 percent), compared to 10.8 percent among the ten slowest-growing Purple metros.
Both Very Blue Bloomington, IN (Indiana University), and Very Red Blacksburg, VA (Virginia Tech), count among the bottom five metro areas nationwide for job growth since January 2025.
I tend to like your articles. This one leaves me cold. In the past, employment and unemployment have been big players in voters sense of what is happening in their area. Now it is clearly inflation, corruption and terrible trade decisions. These are hard to analyze at the local level, I get this.
I am wondering why you choose to tackle this analysis at all. the differences are small, somewhat meaningless compared to inflation, policies and war, and tarrifs. As you point out MSA's are not perfect political boundaries. Perhaps looking for export/import effects of the tarrifs by MSA? You minimize the DOGE effects on the DC MSA. I live here and this has had a real profound effect on the region. Neighbors loosing jobs and homes. School systems looking at changing enrollments, funding , all of Virginia looking for ways to shore up money needs for the south and west of the state since NOVA has been hot so hard.
Please dont just make stuff up to fill the content rules.