Why Two People With the Same Job Title Earn Very Different Amounts

Four things explain most of it

Two people carrying the same job title can be paid very differently, and the reasons are more structural than personal. Industry is usually the largest single factor: the same accounting work pays differently in investment banking than in local government, and the gap is about what the employer’s output is worth rather than what the accountant does each day. The output being priced is the employer’s, not yours.

Employer size is the second, and it is remarkably consistent across occupations. Larger organizations tend to pay more for the same nominal role, partly because they have more formal band structures, partly because scale makes the marginal cost of a higher band smaller, and partly because they compete for candidates against other large employers rather than against local ones. National competition sets a higher floor than regional competition does.

Geography is the third and it is enormous. Median metro variation across occupations runs around 2.3 times from the lowest-paying metro to the highest, which is a wider spread than almost any individual performance difference could produce. Geography is simply a bigger lever than merit.

The fourth is the date each person was hired, and it is the one nobody expects. It is not seniority in the sense of experience — it is the market conditions and band level in force on the day somebody’s salary was first set. Nothing in the ordinary pay cycle resets it afterwards.

Why the hire date matters so much

Salaries are set once, at hire, against the market as it stood that quarter. After that they move by percentage increases applied to whatever the starting figure was. Meanwhile the band for the role moves with the external market, which does not care what anybody already inside is paid.

So somebody hired during a tight labor market comes in at a number reflecting that competition, and somebody hired eighteen months later into a softer market comes in lower, doing identical work. Neither figure was wrong when it was set, and the difference persists for years because internal raises preserve proportions rather than resetting them. Percentage increases cannot close a gap expressed as a percentage.

Run the arithmetic and the durability becomes obvious. A four percent gap at hire, with both people receiving identical three percent raises, is still a four percent gap a decade later — and a larger one in dollars every single year. Nothing in the ordinary running of a pay system closes it.

Salary compression, and why employers tolerate it

The most visible version of this is compression, where new hires are brought in at or above the salaries of people who have been doing the job for years. It happens because external offers are priced against the current market while internal salaries are priced against their own history instead. That asymmetry between the two pricing methods is the entire mechanism.

Employers know this is happening and frequently do nothing, for reasons that are more about budgets than about fairness. Correcting compression means raising a group of existing salaries at once, which is a large one-off cost against a payroll budget, while hiring one person at market rate is a small marginal cost. The incentives on the employer point firmly in one direction.

The practical consequence for anybody who has stayed in a role several years is worth stating plainly. If your pay has moved by percentage increases while the market moved faster, you are probably below the band for your own job, and no amount of good performance will fix that — because the constraint is the mechanism rather than an assessment of you or your work. Performance arguments simply cannot reach a constraint of that kind.

What does not explain as much as people think

Individual skill and performance are real and they explain less of the variance than most people assume. Within a single employer and level, the range produced by performance ratings is usually a few percentage points, because merit budgets are small pools divided among many people. The spread between a good and an average rating is a point or two.

Negotiation matters at the point of hire and much less afterwards, which is why the starting number carries so much weight. A person who negotiated well once is ahead permanently; a person who negotiates well every year is fighting over a few points of a merit pool each year. One good conversation at hire is worth ten review conversations afterwards.

Tenure on its own turns out to do surprisingly little. Where the published data can speak to it, the step changes in pay come from changing level, employer or specialization rather than from accumulating years in the same seat. Time is a weak lever compared to any of those.

The title itself is doing less work than you think

Part of the apparent mystery is that the two jobs are not the same job. A title like Analyst, Engineer or Manager covers a wide range of scope, and two employers using the same word may mean very different things by it. Titles are internal labels rather than market definitions.

This shows up in the data as an unusually wide percentile spread within an occupation. Where the ratio between the tenth and ninetieth percentile is large, the title is covering several distinct jobs, and comparing yourself to a median inside it tells you very little about your own position. Compare the actual scope of both roles before comparing salaries.

Before concluding you are underpaid relative to somebody with your title, it is worth establishing whether you are doing the same work — the same scope, the same decisions, the same accountability. Frequently the answer is no in one direction or the other. Scope differences hide inside two identical job titles all the time.

How to find out where you stand

Work the question from the outside in rather than the reverse. Start with the published percentile range for your occupation in your metropolitan area, which tells you the shape of the market rather than one number. Then find posted ranges for comparable roles at comparable employers, weighting jurisdictions where ranges are legally required, because those are the most honest advertised figures available.

Then get the internal number. In several states you can request the pay scale for your own position and the employer must provide it, and even where there is no such right, asking often works. The band tells you the ceiling on what your current role can pay without a level change, which is the single most decision-relevant figure available to you.

Two readings — the external market and your internal band — answer different questions and you need both. The market tells you whether your employer is competitive; the band tells you whether you have room where you are now. You need both readings to act sensibly.

What to do about a gap you find

If you sit low in your own band, that is a raise conversation with a specific, checkable basis. Bring the band position and the market data rather than a performance argument, because the question you are raising is about placement rather than about merit. That is a much easier conversation to have.

If you are near the top of your band and below the market, the raise conversation cannot fix it and pursuing one will waste a year. The remedy at that point is a level change, a move to a different employer, or a shift into a specialization with scarcity behind it.

And if the gap is explained by industry or employer size, understand that it is not going to be argued away. Those differences are properties of where you work rather than of how you work, and the only lever that moves them is changing one of the two.