The Formation Boom Is Real. The Pipeline Inside It Is Shrinking.

In June 2026, Americans filed 531,423 new-business applications — up 15.6% from a year earlier, and the highest June in the U.S. Census Bureau's record. If you follow small-business headlines, the story writes itself: entrepreneurship is booming, a new generation is starting companies, the "great small-business wave" has arrived.

All of that is true. And if your living depends on businesses actually coming to market — if you're a broker, a searcher, a lower-middle-market advisor, or an SBA lender — most of it is noise. Because underneath the record headline, the part of it that matters to you is doing the opposite of booming. This is a piece about the number under the number: what it is, why it's quietly deteriorating, where the real pipeline is forming, and why all of it collides with the largest ownership transfer in the history of American small business.

The number under the number

Every month, Census publishes two versions of the formation figure. There's the headline total — every application filed. And there's a quieter subset it flags as high-propensity: applications that carry the statistical fingerprints of a business that will actually hire employees and run payroll.

That definition isn't arbitrary. In Census's own methodology, a high-propensity application is one that comes from a corporate entity, indicates planned hiring or provides a wages-paid date, or falls in an employer-heavy industry — construction, healthcare, manufacturing, retail, professional services, food service. These are the markers that historically separate a business with a future payroll from an EIN attached to a side hustle.

And that distinction is the whole game for anyone downstream in the deal economy. High-propensity applications are the businesses that, three to ten years out, become the companies that get listed on BizBuySell, underwritten by an SBA lender, and sold to a searcher or a strategic acquirer. The rest — the majority — are sole proprietors, gig workers, and single-member LLCs that will never hire a second employee, never build transferable enterprise value, and never carry a sale price.

The headline grew 15.6%. High-propensity grew 5%.

So the number everyone is celebrating is climbing three times faster than the part of it that ever becomes an acquirable business. Fewer than three in ten June filings were even on the employer track. The boom is real. It is mostly not future deal supply.

Where the boom actually came from

To understand why the composition matters, it helps to remember where the surge began. For most of the 2010s, business applications ran a steady rhythm — around 290,000 to 300,000 a month, with high-propensity filings a stable ~38% of the total. Then 2020 happened, and the line broke upward: monthly applications jumped past 450,000 during 2021 and have stayed structurally elevated ever since.

That post-2020 surge was, in aggregate terms, real and durable — applications never returned to their pre-pandemic baseline. But it was disproportionately a surge in the low-propensity end: e-commerce sellers, independent contractors, creators, and solo operators formalizing into LLCs. The headline pile grew. The employer-track share began to slip. What looked like an entrepreneurial renaissance was, underneath, a change in the mix of what "starting a business" means — more people formally registering activity that used to be informal, and comparatively fewer building the kind of company that hires, scales, and eventually sells.

Five years on, that shift hasn't reversed. It has compounded.

The quality of the boom is falling — and that's the actual story

Here's the part almost no one reports, because it only appears when you track the ratio over time instead of the headline month to month.

The high-propensity share of applications has declined almost every year for the better part of a decade:

• 2018: 38.1% of applications were employer-track
• 2021: 34.2%
• 2023: 33.8%
• 2025: 29.9%
• 2026 (through June): 28.7%

Look at June specifically and the drift is even cleaner: 38.2% of June 2019's applications were high-propensity, versus 28.2% this June. Ten percentage points of "quality," gone in seven years.

And it isn't only the ratio. In raw counts, June 2026's total applications set an all-time June record — yet its high-propensity count of 149,714 actually came in below June 2021 (153,433) and June 2023 (156,205). The pile of applications keeps getting taller. The employer-track businesses inside it are no more numerous than they were three years ago, and proportionally they have never been scarcer in the series.

Say it plainly: the formation numbers look better every year, and the pipeline they represent looks worse. A record headline sitting on a record-low share is not a contradiction — it's the single most important thing in the data, and it's the thing the headline is structurally incapable of showing you.

The map: where the pipeline is actually forming

That 5% growth figure for high-propensity applications is itself an average, and the average buries the real movement — because the employer pipeline isn't slowing evenly. It's rotating. We took June's high-propensity applications, cut them by NAICS industry sector, and compared each to June of last year.

Two things stand out. First, the growth is concentrated: those five expanding sectors — Accommodation & Food, Construction, Retail, Professional & Technical services, and Health Care — account for roughly three-quarters of all employer-track formation. The next generation of acquirable businesses isn't forming across the economy; it's forming in a handful of consumer- and service-facing corners of it.

Second, the contraction is broad and, in places, steep. Finance & Insurance employer-track formation fell nearly a third year over year. Transportation & Warehousing — the sector that ballooned during the e-commerce and last-mile boom — is down more than 16%. Wholesale, administrative services, and manufacturing are all shrinking. These aren't rounding errors; they're the leading edge of a thinner bench of sellers in those industries a half-decade from now.

A formation map, read this way, is really a supply forecast. It doesn't tell you what's for sale today. It tells you where the raw material for tomorrow's deals is being created — and where it isn't.

Why this matters now: the exit wave is already here

None of this would be urgent if the deal market were quiet. It is anything but. It's staring down what may be the largest ownership transfer in the history of American small business.

Baby boomers own an estimated 12 million U.S. businesses, employing more than 25 million people. A widely cited McKinsey analysis projects that roughly six million small and mid-sized businesses will be involved in a "great ownership transfer" by 2035, with about a million of them expected to actually change hands in a sale — cumulatively worth on the order of $5 trillion. Other estimates put annual small-business exits as high as 665,000 a year by the mid-2030s. And across nearly every survey, the same gap appears: fewer than one in three of these owners has a formal succession or exit plan in place.

That's the demand side of the deal economy — or more precisely, the supply of sellers. A historic cohort of owners is aging toward the door over the next decade, most of them unprepared, all of them eventually needing a transaction.

Now set it against what the formation data just told us. The existing stock of businesses is about to turn over at unprecedented scale — while the pipeline replenishing that stock with new employer businesses is proportionally the thinnest it's been in a decade and concentrated in five sectors. A wave of sellers is arriving. Behind them, in most industries, comparatively little is forming to take their place.

That is not a boom for everyone in deal flow. It's a squeeze. And where you sit determines which side of it you feel.

Why it lands differently on every desk

The headline number is identical for everyone. The number that actually matters is different for every book of business.

If you broker or search in food service, construction, or healthcare, both currents run in your favor. You have a wave of owners aging toward exit and a thickening pipeline of new employer businesses forming behind them. Your funnel fills from both directions. The strategic move is to plant relationships now — with the owners nearing exit and with the operators just starting the ten-year climb toward becoming next decade's inventory.

If your book is finance, wholesale, manufacturing, or logistics, the exit wave still reaches you — those owners are retiring too — but there's measurably less forming behind them to replace what sells. Fewer new employer businesses today means a thinner bench of quality targets tomorrow, tighter supply of the good ones, and more competition for each. The strategic implication is the opposite: quality will be scarce, so the premium shifts to sourcing, relationships, and the ability to move on a good business before three other buyers do.

If you're an SBA lender, the composition shift is a credit-quality signal in slow motion. A formation environment that skews toward non-employer, thin-file, low-durability businesses is a different underwriting environment than one seeding genuine employer firms — and it varies sharply by sector. The industries building real employer pipelines are also the industries where, a few years out, you'll find bankable acquisition targets.

None of this changes what's on your desk this quarter. It changes what's on it in 2028 — and, read sector by sector, it tells you where to plant now.

How we read it — and why the distinction matters

Most coverage cites the total. We don't, because the total is precisely the figure that misleads. Business Formation Statistics exists in the first place because raw application counts overstate durable business creation. The high-propensity series is Census's own correction for that — and even it is an early signal, not an outcome. Census separately publishes how many applications actually transition into employer businesses over the following four and eight quarters, along with a measured delay between the EIN filing and the first payroll; the Small Business Administration's Office of Advocacy has documented the real, and widening, gap between employer-firm applications and actual employer starts.

Our discipline is three steps. First, use the high-propensity series, not the headline total. Second, use it not seasonally adjusted for the sector detail — the basis on which Census publishes industry breakdowns — and compare year over year to strip out seasonality. Third, and most important, watch the ratio, not just the level, because the level makes new records while the ratio quietly tells the truth.

That's the whole method: not the biggest number, the right one. It's the difference between reporting the news and reading it — and it's the layer a neutral, methodology-first read is built to add.

What we're watching next

Formation is a leading indicator of supply. On its own it says nothing about price. But the two are linked, and the link is exactly where this gets actionable: where a thickening employer pipeline meets a rising tide of retiring owners, competition to buy those businesses — and the multiples they command — builds with it. Conversely, where formation is hollowing out beneath the same exit wave, scarcity does the work on price. That interaction, sector by sector, is what turns a formation map into a valuation signal, and it's the layer we track next.

For now, the takeaway is simpler than the data behind it. The next time you see "business formation hits record high," ask the second question, and the third: High-propensity, or not? Building, or hollowing? In my sector, or someone else's?

The headline is written for everyone. The answer is written for your desk.

Sources: U.S. Census Bureau, Business Formation Statistics, June 2026 — business applications and high-propensity applications (seasonally adjusted); sector detail by NAICS sector (year over year, not seasonally adjusted); high-propensity-share trend computed from the 2018–2026 monthly series. High-propensity definition and 4-/8-quarter application-to-employer transition measures per Census BFS documentation. Employer-application-versus-actual-starts gap per the U.S. Small Business Administration, Office of Advocacy. Ownership-transfer ("silver tsunami") figures per McKinsey and related 2026 reporting; those figures are projections, not actuals, and are attributed as such.

FAQ

Disclaimer