The Map Beneath the Career
A career is often described as a sequence of individual decisions: what to study, which job to accept, when to change employers, and how much effort to invest. Those decisions matter, but they are made inside a geography. The local labor market determines which employers are nearby, which industries are expanding, how much a worker can learn by changing jobs, and whether housing, transport, care, and public services make those opportunities usable. This series examines how geography shapes career opportunity—and how to judge the evidence without turning a region into a destiny.
The practical question is not simply whether one place is “good” or “bad” for work. It is: which part of a person’s opportunity set is being shaped by place, through what mechanism, and with what consequences over time? That question requires holding several outcomes apart. A region may have high economic output but low disposable income for its residents. A city may offer high wages but make homeownership inaccessible. A place may recover its employment rate while many incumbent workers never recover their former earnings. Geography matters, but it matters differently for different people and different definitions of success.
Why National Averages Miss the Career Map
National statistics are useful for describing the broad condition of a labor market. They can tell us whether employment is rising, unemployment is falling, or wages are recovering. But an average can conceal the local conditions through which work is actually found. Two people in the same country may face very different probabilities of finding a job, changing occupations, gaining experience, or moving into a better-paid role because they live in different local labor markets.
The OECD’s 2026 Employment Outlook makes this difference visible. Across comparable small regions, the average gap between regions in the highest and lowest employment quintiles is 11.4 percentage points. In more than half of OECD economies, the range between small regions exceeds 20 percentage points. Unemployment in the lowest-performing regional quintile is, on average, more than twice the rate in the highest-performing quintile. These figures are descriptive rather than a universal causal estimate, and the size of a gap depends on the country, year, and regional definition. They nevertheless establish the scale of the problem: a national employment rate does not describe a single job market.
The geography of income follows the same logic, but not in a simple one-to-one way. Employment is the main route through which most households receive labor income, so regions with higher employment rates generally have higher median disposable incomes. Yet GDP per capita is not the same as the income available to a typical resident. Output can be attributed to a headquarters while the workers commute from elsewhere. Capital-intensive industries can generate substantial production without creating proportionate household earnings. Taxes and transfers also compress differences between places. In some countries, regional GDP per capita and median household income are only weakly related.
This distinction changes how a worker should read a location’s apparent prosperity. A place with impressive output may have a less impressive resident opportunity set if jobs are inaccessible, housing absorbs the wage premium, or services are difficult to reach. Conversely, a place with lower measured output may offer a reasonable living standard when commuting, redistribution, and lower costs narrow the difference. The question is not which number is “real.” It is which number answers the question being asked.
The Local System Behind a Career Path
Geography shapes opportunity through a system rather than a single variable. One part of that system is the local concentration of firms and industries. Productive firms tend to cluster in particular regions, especially where they can draw on skilled workers, specialized suppliers, research institutions, infrastructure, and dense networks of customers. The resulting agglomeration can make it easier for employers and workers to find one another, spread knowledge between firms, and create a larger menu of roles. A worker in such a market may have more ways to enter, learn, specialize, and change employers without leaving the region.
The OECD reports that high-value-added services are disproportionately concentrated in capital-city regions. Across the countries with available data, these services account for an average of 26.6% of employment in capital regions, compared with 17.8% in other metropolitan regions and 12.5% in non-metropolitan regions. The figures do not mean that every job in a capital is better or that every rural job is a dead end. They show that the local mix of opportunities differs systematically. A region’s industry structure affects the kinds of experience that are available and the pathways by which one job can lead to another.
The system is reinforced by both sides of the labor market. Skilled workers may be attracted to places with more productive firms, while productive firms may choose places where specialized skills are already concentrated. This creates a feedback loop: more firms can create more learning and matching opportunities, which attract more workers, which make the region more attractive to additional firms. Lagging regions can face the reverse pattern as productive firms leave, job options narrow, and younger or better-educated residents become more likely to depart.
But worker sorting does not explain everything. In the OECD’s selected microdata comparisons, differences in observable characteristics such as age, gender, education, migration background, household composition, and disability explain only about one-third to one-half of regional employment gaps. Residents of low-employment regions remain less likely to be employed after those characteristics are taken into account. That result does not prove that living in a particular region causes a specific individual’s outcome. Unobserved ability, motivation, risk preferences, commuting, and selective migration can still matter. It does show why a purely individual explanation is incomplete: the same formal qualification can have different value when local employers, institutions, and networks differ.
The local system also includes the conditions that determine whether a job can be reached and sustained. The OECD finds that access to public employment services, early childhood education, and primary schools is generally worse in lower-income regions. In its European comparison, almost all residents of the highest-GDP-per-capita regions in France and Germany can reach a public employment service within 15 minutes by car, compared with fewer than half in the countries’ lowest-GDP-per-capita regions. This is a physical-access measure, not a measure of service quality or a causal estimate. It nevertheless illustrates an important mechanism: opportunity is not just the existence of a vacancy. It is the practical ability to search, travel, care for dependents, acquire skills, and remain employed.
Housing makes the system more difficult still. A high-wage region may offer a larger career ladder while pricing out the workers who could use it. A lower-cost region may make ownership or family stability easier while offering fewer high-value jobs. These are not contradictions to be resolved by choosing one universal ranking. They are trade-offs within the opportunity set.
The arrows run in both directions over time. A place shapes which opportunities are reachable, but the people who can move, the firms that invest, and the outcomes that accumulate also reshape the place. That is why geographic opportunity is neither a fixed property of the map nor a simple reflection of the people who live there.
How to Read a Geography-of-Opportunity Claim
The first discipline is to name the outcome. “Opportunity” can mean the chance of employment, the level of labor income, the median disposable income of a household, the ability to move up the income distribution, the probability of owning a home, or the accumulation of housing or total wealth. These outcomes are related, but they are not interchangeable.
The U.S. Census working paper in this series makes that point especially clearly. Using linked records for more than 3.4 million parent-child pairs, it estimates county-level mobility for total income, labor income, gross housing wealth, homeownership, and gross total wealth. Its rank-based framework distinguishes absolute mobility—the average adult outcome for children who start near the bottom—from relative mobility, which captures how strongly parental position predicts the child’s position. A place can offer better average outcomes to children from low-resource families without greatly weakening the relationship between parental and child position. Those are different kinds of progress.
The same distinction applies to geography. A map of income mobility is not a map of wealth mobility. The Census study finds familiar low relative and upward income mobility across much of the Southeast and higher income mobility across the Great Plains and parts of the upper Midwest. Wealth measures show less geographic clustering and more concentration by density. Many large urban areas display greater persistence in housing wealth than in income. Homeownership has another pattern again: high home prices can raise the value of homes owned by successful entrants while making ownership harder for children of renters.
The second discipline is to ask whether a statistic describes a place, a person, or the relationship between them. Employment rates in the OECD analysis are generally measured at the place of residence, while productivity and sectoral employment may be measured at the place of work. Commuting can therefore make a region look less or more prosperous than the daily flow of work would suggest. County-level mobility estimates describe associations across populations, not a forecast of what will happen to one reader who moves there.
The third discipline is to distinguish evidence of a pattern from evidence of a mechanism. The sources support a strong conclusion that geography is associated with career and economic trajectories. They also identify plausible mechanisms—firm concentration, industry mix, services, housing, selective migration, and shocks. But an association between a region’s home prices and wealth mobility does not by itself prove that prices caused the mobility pattern. The Census paper explicitly treats its local-correlate regressions and Great Recession results as descriptive associations, even when the timing makes a long-run interpretation informative.
This series therefore asks a narrower and more useful question than “Where is the best place to live?” It asks how a place changes the reachable set of jobs and life chances, which groups can use those opportunities, which outcomes improve, and who bears the cost when the local economy changes. The next step is to examine the mechanisms inside the local job market: why opportunity clusters, how that clustering creates advantages, and where its apparent benefits stop reaching residents.