The Geography of Pay

When the Bureau of Labor Statistics surveys employers across the country each May, it produces what is effectively an annual census of American compensation — not sampled from tax returns or household surveys, but reported directly by the establishments that sign the checks. The 2024 data paints a strikingly clear picture: the United States isn’t one labor market with regional wrinkles, but a collection of vastly different economies that happen to share a currency and a flag.

At the summit sits the District of Columbia at $88,000 — a median wage that reflects the unusual nature of the capital’s economy, where federal agencies, lobbying firms, law offices, and think tanks crowd out the low-wage retail and food service jobs that dominate most states. DC’s median is 25% higher than the next state, Massachusetts ($62,270), and 78% above the national figure of $49,500. It is, in effect, a city-state with an economy structured more like Zurich than like the rest of America.

Below DC, a tier of high-wage coastal states clusters between $56,000 and $63,000: Massachusetts ($62,270), Washington ($61,590), New York ($58,560), Connecticut ($58,400), Colorado ($58,210), Maryland ($58,050), New Jersey ($57,230), and California ($56,940). Alaska ($59,400) breaks the coastal pattern — its high wages reflect the cost premium of operating in a frontier economy where a gallon of milk can cost $8 and a plumber charges $150 an hour simply because there aren’t enough plumbers willing to live in Fairbanks.

Then comes a dense middle band where 20 states crowd between $46,000 and $53,000 — from Rhode Island ($54,040) down through Florida ($46,860). This is where most of America actually lives and works: the Great Lakes manufacturing belt, the Sun Belt growth corridors, the agricultural Midwest. The typical worker in Ohio ($48,060) earns almost exactly the national median. The typical worker in Texas ($47,500) — the state that has added more jobs than any other in the past decade — earns slightly less.

At the bottom, a cluster of Southern and border states falls below $44,000: Oklahoma ($43,950), Alabama ($43,830), Louisiana ($43,770), West Virginia ($43,320), Arkansas ($41,020), and Mississippi ($39,070). Mississippi’s median of $39,070 is lower than the national median was in 2019 ($39,810). In wage terms, Mississippi in 2024 is where the nation as a whole was five years ago.

Group the 51 jurisdictions into the four Census regions and the geography of pay snaps into focus. The Northeast — nine states from Maine to Pennsylvania — has an employment-weighted median of $56,117 and 27.2 million workers. The West — thirteen states from Alaska to Hawaii — comes in at $55,011 with 36.3 million workers. The Midwest — twelve states from Ohio to the Dakotas — drops to $48,704 with 32.8 million workers. And the South — seventeen states plus DC, from Delaware to Texas — pays the least at $47,749, but employs the most people: 57.8 million workers, or 37.5% of the national total.

This is the central irony of American wage geography: the region that pays the least employs the most. The South’s low wages are not a small-sample curiosity confined to a few rural states. They are the daily reality for nearly two in five American workers. When politicians talk about “the American worker,” they’re statistically most likely talking about someone in Georgia or Texas or North Carolina earning less than $48,000 a year.

The gap between the Northeast and the South — roughly $8,400 in employment-weighted median terms — is large enough to buy a used car, cover six months of groceries for a family of four, or pay a year’s worth of health insurance premiums on the ACA marketplace. It is the difference between accumulating savings and running deficits each month. And it persists not because Southern workers are less skilled or less productive, but because the industries that dominate the Southern economy — logistics, healthcare support, food processing, retail distribution — are structurally lower-paying than the knowledge-economy sectors that anchor the coasts.

Perhaps the starkest way to frame the divide: only 20 of 51 jurisdictions have a median wage at or above the national figure of $49,500. Those 20 jurisdictions employ 65.4 million workers. The remaining 31 — home to 88.8 million workers — pay below the national median. The “typical” American wage is, in fact, earned by a minority of Americans. The majority work in states where the median falls short.

Aggregate medians tell a story about economic structure — which industries are present, how many managers versus cashiers populate a given state. But the more revealing question is: what happens when you hold the occupation constant and change only the geography? The answer, for most occupations, is dramatic.

Consider four occupations that collectively employ tens of millions of Americans: registered nurses, software developers, elementary school teachers, and cashiers. Pick five states that span the wage spectrum — California, New York, Texas, Ohio, Mississippi — and watch what happens to the same job title.

A registered nurse in California earns a median of $140,330. The same credential, the same scope of practice, the same rotating shifts in a hospital ward — in Mississippi, that nurse earns $74,470. That’s a spread of $65,860, or 1.88 to 1. California’s nurses benefit from a powerful state nurses’ union, a high cost of living that pushes all wages up, and a chronic shortage that gives them leverage. Mississippi’s nurses work in a state where hospital margins are thin, many facilities are rural and publicly funded, and the cost of living is low enough that $74,000 goes further than it sounds — though not twice as far.

A software developer in California earns $170,910; in Mississippi, $86,460 — a ratio of 1.98 to 1. In New York, that developer earns $161,260; in Ohio, $107,690. The geographic spread for software developers is the widest in dollar terms because the top-paying states are bidding for talent that can build products worth billions, while the bottom-paying states employ developers who maintain internal systems, run government IT, or work for smaller firms without venture capital’s wage inflation.

An elementary school teacher in California earns $98,190; in Mississippi, $48,440 — a ratio of 2.03 to 1. This is remarkable because teaching is the most geographically standardized occupation on the list: every state requires certification, every teacher follows a curriculum, every classroom has roughly the same student-to-teacher ratio. Yet the pay gap is wider than for nurses. The explanation is almost entirely about state and local funding formulas. California’s per-pupil expenditure exceeds $21,000; Mississippi’s hovers around $10,000. Teachers are paid by tax revenues, and tax revenues follow property values and income levels — which brings us back to the same geographic divide.

Even cashiers — the most standardized, lowest-skilled occupation on the list — show a meaningful geographic spread. A cashier in California earns $36,270; in Mississippi, $22,410 — a ratio of 1.62 to 1. The gap is smaller in percentage terms than for nurses or developers, but in human terms it may be larger: the Mississippi cashier’s $22,410 is below the federal poverty line for a family of four ($31,200 in 2024). That cashier in California, even at $36,270, is above it — though still far below California’s own cost of living.

The geographic wage divide is most consequential not at the median but at the 10th percentile — the wage below which 10% of workers in a state fall. This is the floor of the labor market, the rate paid to the least-experienced, least-credentialed, least-leveraged workers in each state. And the floor varies enormously.

In Louisiana, the 10th percentile wage is $22,030 — roughly $10.59 per hour for a full-time worker. In Mississippi, it’s $22,110. In Alabama, $23,520. These figures are below the poverty threshold for a family of three. A worker at the 10th percentile in any of these states, working full-time year-round, cannot afford the average rent for a one-bedroom apartment in their own state without spending more than 50% of gross income on housing.

Compare that floor to Washington state, where the 10th percentile is $37,490 — 70% higher than Louisiana’s. Or DC at $38,480. The difference at the bottom of the distribution is actually larger than the difference at the median, in proportional terms. The lowest-paid workers in Washington earn 1.70 times what the lowest-paid workers in Louisiana earn; at the median, the ratio is only 1.41 times. Geography punishes the poor more than the middle class.

Flip to the other end and the pattern inverts. The 90th percentile in DC is $191,880; in Mississippi, $83,400 — a ratio of 2.30x. The 90th percentile in California ($157,210) is 1.89 times Mississippi’s. High earners also benefit from geography, but they have something low earners don’t: mobility. A software engineer in Mississippi can relocate to Seattle for a $60,000 raise; a cashier in Mississippi cannot easily move to Washington for a $15,000 raise, because the cost of the move itself — security deposits, moving expenses, lost social networks — is a larger fraction of their savings.

Not all states are equally unequal. The P90/P10 ratio — how much more the 90th percentile earns than the 10th — varies from a low of 3.00 in Maine to a high of 4.99 in DC. This ratio captures something that median wages alone cannot: how wide the internal spread is within a single state. A state can have a high median but extreme inequality (DC), a low median but moderate inequality (South Dakota at 3.09), or a mid-range median with alarming inequality (Texas at 4.65).

The most unequal jurisdictions — DC (4.99x), Texas (4.65x), Virginia (4.58x), California (4.46x), Louisiana (4.46x), Maryland (4.38x), Georgia (4.35x) — share a common feature: they are economies where a small but significant knowledge-economy sector (government, tech, finance) coexists with a large low-wage service sector. Virginia’s P90/P10 ratio of 4.58 reflects the chasm between a Booz Allen consultant in Tysons Corner earning $150,000 and a hotel housekeeper in Virginia Beach earning $29,000. Texas’s 4.65 ratio captures the distance between an oil & gas engineer in Houston and a warehouse worker in El Paso.

The most equal states — Maine (3.00x), South Dakota (3.09x), Puerto Rico (3.09x), Vermont (3.13x), North Dakota (3.30x) — tend to be smaller, more homogeneous economies without extreme high-wage sectors. Vermont doesn’t have Wall Street; Maine doesn’t have Silicon Valley. Their wage distributions are compressed because there simply aren’t many $150,000+ jobs to pull the top end up.

The mean-median gap tells a related story. In California, the mean wage ($79,900) exceeds the median ($56,940) by $22,960 — a ratio of 1.40x. This means a relatively small number of very high earners pull the average substantially above what the typical worker makes. In South Dakota, the mean ($55,480) exceeds the median ($45,620) by only $9,860 — a ratio of 1.22x. The bigger the mean-median gap, the more top-heavy the wage distribution. California, Illinois (1.38x), New York (1.38x), Georgia (1.37x), and Virginia (1.36x) have the biggest gaps. Iowa (1.22x), South Dakota (1.22x), and Alaska (1.23x) have the smallest.

State-level wages are not random. They follow industry composition with remarkable fidelity. The single best predictor of a state’s median wage is the share of its workforce in computer and mathematical occupations — the tech sector, broadly defined. DC has the highest tech concentration at 6.70% of employment, and the highest median. Washington (6.02%), Virginia (5.86%), Maryland (5.08%), and Colorado (4.63%) follow — and every single one of these states ranks in the top 10 for wages. California, despite its enormous size, has 4.10% tech workers and 739,550 of them in absolute terms — more than the total workforce of Wyoming, Vermont, and North Dakota combined.

At the other end of the spectrum, production and manufacturing concentration correlates with middle-of-the-pack wages. Wisconsin leads the nation with 8.90% of its workforce in production occupations, followed by Indiana (8.23%), Arkansas (8.06%), and Iowa (7.95%). These states cluster in the $41,000–$49,000 median range — above the bottom but well below the tech-driven coasts. Manufacturing pays better than retail or food service, but it doesn’t generate the $150,000+ salaries that pull state medians into the upper tier. Michigan, the historic heart of American manufacturing, has 7.30% production workers and a median of $48,300 — almost exactly the national figure.

Perhaps the most counterintuitive finding: healthcare concentration does not predict high wages. West Virginia leads the nation with 11.60% of its workforce in healthcare occupations (practitioners plus support combined), yet its median wage of $43,320 places it 49th among states. Louisiana (10.10% healthcare, $43,770 median) and South Dakota (10.01%, $45,620) tell the same story. The reason is structural: healthcare employment in these states is dominated by support roles — nursing assistants, home health aides, medical assistants — that pay $28,000–$38,000 rather than the practitioner roles that pay $80,000+. Having a large healthcare sector means having a large number of modestly paid workers, not a large number of well-paid ones.

The lesson is that not all jobs are created equal, and neither are the industries they cluster in. A state’s wage level is largely determined by whether its dominant employers are in knowledge-intensive sectors (tech, finance, government consulting) or labor-intensive sectors (retail, healthcare support, agriculture, warehousing). Policy, unionization, cost of living, and minimum wage laws all matter at the margin — but the primary driver is what the state’s economy actually makes and sells.

The state wage ladder tells you what the typical worker earns. The payroll map tells you where the money actually flows. Multiply each state’s total employment by its mean wage and you get an estimate of total annual wage expenditure — the sum of every paycheck signed in that state in a year. The results reveal just how concentrated America’s economic gravity really is.

California dominates with an estimated payroll of $1.44 trillion — 18.1 million workers earning an average of $79,900 each. That single state’s payroll exceeds the GDP of all but about a dozen countries. Texas follows at $881 billion, then New York ($769 billion), Florida ($619 billion), and Illinois ($419 billion). The top five states alone account for roughly $4.1 trillion in wages — approximately 39% of the national total.

The concentration sharpens further as you move down the list. California’s payroll is larger than the combined payroll of the bottom 25 states — from New Hampshire ($47B) down through Wyoming ($16.8B). New York’s payroll ($769B) is larger than that of 40 states. Even within the Sun Belt growth story, the payroll numbers reveal that growth in bodies hasn’t translated into growth in per-worker spending: Florida has 9.8 million workers but a mean wage of only $62,990, giving it a payroll-per-worker figure 21% below California’s.

The smallest payroll belongs to Wyoming at $16.8 billion — 278,500 workers earning an average of $60,200. That is roughly what Apple pays its 160,000 employees in total compensation. Vermont ($20.2B), Alaska ($23.4B), and South Dakota ($25.1B) are scarcely larger. These are not poor states in per-capita terms — Alaska and Wyoming both have above-median wages — but their small populations mean their total economic footprint is a rounding error in the national accounts.

The most revealing way to see the geographic divide is not through medians or means but through the full wage distribution — the P10, P25, median, P75, and P90 — across the highest and lowest-paying states. This reveals that the gap between states doesn’t just exist at the middle of the distribution; it runs from floor to ceiling.

In DC, the wage distribution runs from $38,480 (P10) to $53,310 (P25) to $88,000 (median) to $141,300 (P75) to $191,880 (P90). The distance from P10 to P90 is $153,400 — the widest internal spread of any jurisdiction. A DC worker at the 25th percentile ($53,310) earns more than the median worker in 36 states. The capital isn’t just wealthy; it is spectacularly unequal.

In Mississippi, the distribution runs from $22,110 (P10) to $28,980 (P25) to $39,070 (median) to $58,860 (P75) to $83,400 (P90). The P10-to-P90 spread is $61,290 — less than half of DC’s. Mississippi’s 90th percentile ($83,400) is lower than DC’s median ($88,000). The highest-earning workers in America’s poorest state earn less than the typical worker in its richest jurisdiction.

Between these extremes, states like Massachusetts and California combine high medians with wide spreads: Massachusetts ranges from $35,800 to $157,220 (P10 to P90), and California from $35,260 to $157,210. The surprise is that their P10 figures — $35,800 and $35,260 — are not dramatically higher than the national P10 of $29,990. Even in wealthy states, the bottom tenth of workers earns only modestly more than the national floor. The geographic advantage is concentrated from the 25th percentile upward.

States like Vermont and South Dakota present a different model: relatively compressed distributions. Vermont runs from $34,480 to $107,920 — a P90/P10 ratio of just 3.13x, the lowest of any state. South Dakota: $28,880 to $89,190, a ratio of 3.09x. These states lack the very high-wage industries that create extreme top-end skew, but they also lack the extreme low-end depression that characterizes the deep South. They are middle-wage, middle-inequality economies — unremarkable on any single measure but perhaps the closest approximation to what a “typical” American wage structure looks like before coastal and Southern distortions pull it apart.