ISSUE 009
Fifty-eight versus forty-six
Everyone in economic development says the words skills gap. Almost nobody says which skills, in which county, against which jobs. I went looking for work that actually does the arithmetic, and found one study that did it for 565 local labor markets and reached a conclusion about rural places that is worth more than any amount of regional boosterism. Then I went looking for Missouri's own numbers, which turn out to be published four times a year by a board most people have never heard of.
An idea worth thinking about
Fifty-three percent of Missouri jobs are mid-skill and forty-six percent of workers have the skills for them
The Missouri Chamber Foundation published an update to Workforce 2030 in October 2024, with research by Economic Leadership LLC. The headline mismatch is a seven-point gap: 53 percent of Missouri jobs are classified as mid-skill, against 46 percent of workers holding the corresponding skills. Alongside it, Missouri's share of adults with a bachelor's degree or higher sits below the national average, at 33 percent for women and 31 percent for men, 33 percent for white adults and 27 percent for people of colour.
The geography in the same report is the part that should change how a rural business reads the labour market. Missouri's population grew 1.1 percent between 2019 and 2023 against 2.2 percent nationally, and the state's own projections to 2033 have thirteen counties growing more than ten percent while more than fifty lose population. On jobs, between 2018 and 2023, 29 counties grew more than five percent and 36 counties lost jobs outright. Employment is concentrating toward the centre of the state. The report lists thirteen high-demand occupations requiring no bachelor's degree, with between 1,100 and 6,828 average annual openings; heavy and tractor-trailer truck driving is the largest, at 6,828 openings and average earnings of $49,525.
Two disclosures. This is a business association reporting that employers cannot find skilled workers, which is a finding it has an interest in reaching, and the mid-skill definitions it uses are not universal. Its underlying labour-market data comes from Lightcast, a commercial provider whose estimates I have leaned on elsewhere through other publishers; the figures are not independently derived. I am reporting the gap because it is specific and because the county-level job and population counts behind it are checkable, not because the framing is neutral.
A tool I disappeared into
The state keeps one page that holds every number it has on Missouri colleges
The Department of Higher Education and Workforce Development maintains a research and data repository that is easy to miss and hard to exhaust. The Statistical Summary of Missouri Higher Education sits at the centre of it — a document the state has published since the mid-1980s, running in roughly its current form since the early 1990s, covering every Missouri public and independent institution across seven sections and around 145 numbered tables. Student preparation, persistence and graduation, assessment, tuition and financial aid, enrollment and demographics, institutional finance, and student transfer. Degrees conferred are broken out by level and discipline area; enrollment is reported by Missouri county of origin; graduation is reported at the 150 percent of normal time standard.
Alongside it on the same page: the 2026 Higher Education Factbook, an in- and out-migration report tracking students leaving and entering the state, a high school graduates report, dual credit and age demographics, and MoSCORES, the state's credential scorecard search. The repository also links the federal Post-Secondary Employment Outcomes data, which reports earnings for graduates of specific programmes at specific institutions — the closest thing available to a demand-side check on a credential before anyone enrols in it.
That last link is the one worth the detour. The Statistical Summary counts supply. It has no demand side at all, which is not a flaw — it is a reporting document for a coordinating board, doing exactly its job. But it means a reader who wants to know whether a programme leads anywhere has to bring the earnings data and do the matching themselves. The repository at least puts both on the same page. Nobody does the match for you. That is, in a sentence, the entire problem this entry is about.
Learn something
Somebody tried to count every credential in the United States and got to 1,076,358
Credential Engine has been attempting a census of American credentials since 2018. Its fourth count identified 1,076,358 unique credentials across 18 categories and four provider types. Postsecondary institutions accounted for 350,412 of them, in degrees and certificates. Secondary schools accounted for 56,179, in diplomas, alternative completion certificates and equivalency awards. MOOC providers accounted for 13,014.
Non-academic providers — industry bodies, software vendors, training companies, licensing boards — accounted for 656,505. That is roughly sixty percent of every credential in the country, issued by organisations that are not schools, in badges, course completion certificates, licences, certifications and apprenticeships. A fifth count was published in 2025.
Sit with what that means for a hiring manager in a town of twelve thousand. When a candidate presents a certificate, there are over a million things it could be, most of them issued by bodies with no common standard, no shared transcript and no way to verify what was actually demonstrated. The supply of credentials is not scarce. Legibility is. That is why this entry's hard number is about matching rather than counting — and it is worth noticing that the four provider-type figures in the published count sum to 1,076,110, which is 248 short of the stated total. Even the census of credentials does not quite reconcile.
What we’re seeing
Rural labor markets would have to change what 58 percent of their credentials are in
Jeff Strohl, Zachary Mabel and Kathryn Peltier Campbell at the Georgetown University Center on Education and the Workforce published The Great Misalignment in 2024. It compares the supply of certificates and associate's degrees against projected occupational demand in 565 local labor markets, defined by commuting zones rather than by metro boundaries — which means the unit is roughly the area people actually drive to work in, not a city limit.
In 283 of those 565 markets, at least half of all middle-skills credentials would have to be conferred in different fields for the mix to match projected demand. In the best-aligned quarter of markets, full alignment would need fewer than 42 percent of credentials redistributed. In the worst quarter, more than 60 percent. Even the most aligned markets in the country — Americus, Georgia at 18 percent, Manhattan, Kansas at 20, Paris, Tennessee at 21 — are not aligned.
The rural finding is the one to carry. Providers in very rural labor markets, defined as those where at least three-quarters of residents live in rural parts of the commuting zone, would need to redistribute 58 percent of the credentials they confer. In very urban markets the figure is 46 percent. Somewhat rural sits at 54 and somewhat urban at 53, so the gap is at the extremes rather than along a smooth gradient. The mechanism the authors identify is not rural incompetence; it is the number of institutions. Every ten additional providers in a market is associated with three to four percentage points less misalignment, and very urban markets have on average thirteen times as many middle-skills providers as very rural ones. Their analysis points to complementarity rather than competition as the reason — providers in a crowded market end up covering different niches, and the community-level mix lands closer to demand than any single institution does.
They are careful about what misalignment is not. Twenty-eight percent of all middle-skills credentials nationally are granted in programmes with no direct occupational match, mostly liberal arts, general studies and humanities, and the authors state plainly that this does not make them worthless — transfer, personal interest and general education are valid reasons to enrol. But only 21 percent of people who earn a middle-skills credential ever transfer to a four-year institution, so most of these students enter the labour market with that credential as their highest attainment. The report was funded by Ascendium Education Group, a philanthropy with a stake in postsecondary workforce policy; that is disclosed on its acknowledgements page and I am repeating it here.
Missouri's own completion numbers, published by the Coordinating Board for Higher Education on 4 March 2026, put a floor under all of this. Of 20,850 Missouri public high school graduates who enrolled full time in fall 2019, 57.8 percent had completed something within six years — 12.3 percent a certificate, 21.3 percent an associate's, 39.7 percent at least a bachelor's. Split by institution type, the divergence is severe: 77.0 percent completion at highly selective institutions, 73.1 at selective, 58.1 at moderately selective, and 47.1 percent at open-admission institutions, which is where most rural Missouri students go. Enrollment of recent public high school graduates is down 9.9 percent over ten years, and that decline is entirely at four-year institutions, down 18.5 percent, while two-year enrollment is up 1.5 percent.
So the Missouri picture, stated without decoration: fewer graduates enrolling, a growing share of them at the institutions with the lowest completion rates, in a state whose rural labor markets have the fewest providers and therefore, by Georgetown's finding, the worst match between what gets taught and what gets hired. The response that follows from the evidence is not more credentials. It is better-aimed ones, and somebody local doing the aiming.
One question
What is the one credential you would hire for tomorrow that nobody nearby is granting?
Georgetown's whole argument is that somebody has to notice the gap locally, because no state agency is going to notice it for you. Reply and name it; I am keeping a list.
More in this pillar as it's written.
— Shana
Sources link to the original publisher, never to a summary.