Daily AI-surfaced causal investment ideas from the news.

Every lead comes with a deep research prompt. Paste both into ChatGPT or Claude and go find the trade.

78Strong lead
Sep 17, 2026Comment period on both proposed rules closes September 21, 2026policy · defense · technology

Washington is redrawing the line between a small company and a big one, and $71 billion of federal contracts moves with it

The Small Business Administration proposed on August 20, 2026 to lift the revenue ceiling for federal IT-services contractors from $34 million to $531 million and to measure size by headcount rather than revenue almost everywhere. By its own estimate, 37,002 firms already holding $71 billion of federal contracts would newly count as small — and the set-aside pool most government-services revenue flows through gets re-sorted.

The US Small Business Administration published two linked proposed rules in the Federal Register on August 20, 2026: a Revised Size Standards Methodology at 91 FR 54096, and a Small Business Size Standards rule at 91 FR 53741 (RIN 3245-AI67). Together they would replace the current table of 995 size standards, spread across 978 NAICS industries and 18 contracting exceptions, with 338 standards set at the four- and five-digit NAICS level. Comments on both close on September 21, 2026.

The numbers inside the rule are not incremental. NAICS 5415, computer systems design and related services, would move from a $34 million receipts ceiling to $531 million. NAICS 541330, engineering services, would move from $25.5 million to $252 million. All explicit maximums are eliminated; new minimums are set at 500 employees or $30.6 million in receipts; all 18 existing federal-contracting exceptions are eliminated; and SBA proposes not to lower any standard, even in the 45 industries where its own analytics support a cut.

The methodology also flips the default. Where SBA has discretion, size would be measured by employee headcount rather than average annual receipts. That removes what the industry calls the benefit cliff — the mechanic by which winning a large contract pushes a firm’s five-year average receipts over the threshold and costs it its small-business status.

SBA’s own regulatory analysis puts the effect at a net 114,541 additional firms classified as small, taking the national total from 6,344,967 to 6,459,508. Of those, it estimates 37,002 unique firms held FY2025 federal contracts — roughly 105,655 contracts worth more than $71 billion — and would be newly eligible as small businesses.

Opportunity

The obvious reading is that this is administrative housekeeping at an agency that does small-business lending, of interest to compliance lawyers and nobody else. That reading holds right up until you notice what the small-business designation actually controls. Federal agencies carry a statutory goal of directing 23% of prime contract dollars to small businesses, and the Rule of Two requires a procurement to be set aside whenever two capable small firms are expected to bid. Size status is therefore not a label. It is the gate on which contracts may be competed at all, and by whom.

Raising the ceiling on that gate does two opposite things at once. It hands a much larger addressable market to firms that were previously just over the line, including firms that graduated out of small status and watched their set-aside revenue run off. It simultaneously floods the set-aside pool with bigger, better-credentialed competitors, which is bad for the genuinely tiny incumbents who have been winning restricted competitions against thin fields.

The M&A consequence is separate and arguably larger. SBA does not treat a portfolio company owned in substantial part by a licensed Small Business Investment Company as affiliated with that SBIC for size purposes. An SBIC-backed government IT services platform today has to stay under $34 million of receipts. Under the proposal it could run to $531 million. That is roughly a fifteenfold expansion of the box a sponsor can build inside while keeping set-aside eligibility, and it makes small-to-small roll-ups and all-small joint ventures viable where they previously were not.

Hypothesis: nothing in public-market GovCon valuation currently contains an SBA size table, which is why a rule that re-sorts $71 billion of contract eligibility has moved through a comment period with essentially no coverage outside law-firm alerts and trade press — while the sub-scale end of the listed sector trades at its lows. This is an inference about what is and is not in prices, not an established fact about anyone’s positioning.

Share prices verified September 17, 2026. DLH Holdings $3.90, 40.6% below its one-year high and down 29.1% in three months. Castellum $0.62, 52.9% below its one-year high. Parsons $46.61, 47.8% below. Booz Allen Hamilton $78.89, 24.2% below. Leidos $131.00, 34.4% below. The scaled survivors are not where the damage is: CACI $635.26 sits 6.6% off its high and SAIC $134.17 closed at its one-year high.

How it could play out

SBA takes comments through September 21, 2026 and finalises some version of the rule, having stated it intends to proceed. Contracting officers re-run market research against the new NAICS groupings and discover that procurements previously too large to set aside now have two or more capable small bidders, so more work moves inside the set-aside fence and away from full-and-open competition. Firms that had aged out of small status re-qualify and bid work they had written off; the smallest incumbents lose win rates to them; a wave of size protests accompanies the transition. Sponsors with SBIC structures find their permitted build size multiplied and start buying, which turns a regulatory change into deal flow in a sector where the sub-scale listed names are trading near multi-year lows.

Questions worth asking

  • Which listed government-services companies would actually fall under the proposed thresholds in the NAICS codes where their revenue sits? That is a mechanical, checkable exercise against 91 FR 53741, and it decides whether this is an investable idea or merely an interesting one.
  • Is the real trade the newly-small companies, or the large primes losing addressable work as more procurements become set-asides? Those are opposite positions and the rule supports both.
  • Does the shift from receipts to headcount change what kind of company wins? A revenue test punishes growth; a headcount test punishes labour intensity and rewards software leverage, which is a different selection pressure on the whole sector.
  • Who owns the private side of this? SBIC-backed and PE-backed GovCon platforms get a fifteenfold larger box in IT services — does that show up first as takeout demand for small listed contractors rather than as organic revenue?
  • Eliminating all 18 contracting exceptions is buried in the methodology rule. Which specific programs did those exceptions protect, and is anyone listed dependent on one?
  • The proposal explicitly declines to cut any standard even where SBA’s analytics support a cut. What does that asymmetry say about how the final rule is likely to land relative to the proposal?

Where to look

  • DLH Holdings — small-cap federal health and IT services whose history runs through small-business set-asides, trading 40.6% below its one-year high
  • Castellum — micro-cap federal IT and cyber services, 52.9% below its one-year high, the size band where re-eligibility would matter most
  • ICF International and VSE — mid-caps close enough to the boundary that the NAICS mapping is worth doing rather than assuming
  • Booz Allen Hamilton, Leidos and Parsons — the potential losers if more work moves inside the set-aside fence, all well below their one-year highs already
  • SBIC-backed and private-equity-owned government services platforms — the non-listed side where the fifteenfold expansion of the permitted build size lands first

Thesis check

The strength is that this is primary-source, quantified and dated: SBA has published the thresholds, published its own estimate that 37,002 firms and $71 billion of contracts change classification, and set a September 21, 2026 comment deadline, and the government-services small-cap complex is trading at or near multi-year lows while none of it is discussed outside government-contracting law practices. The weakness is that a proposed rule is not a rule — SBA could retreat on individual thresholds under comment pressure from incumbent small businesses who lose the most, the final rule’s effective date is unknown, and size status alone does not confer eligibility for the 8(a), HUBZone, SDVOSB or WOSB programs, each of which imposes its own criteria. There is also a real chance that the clean expression here is private rather than public, in which case the correct output is the NAICS-mapping exercise rather than a trade.

Sources

Federal Register, Small Business Size Standards, 91 FR 53741, Aug 20 2026 · Federal Register, Revised Size Standards Methodology, 91 FR 54096, Aug 20 2026 · Holland & Knight, SBA Proposes Sweeping Overhaul of Small Business Size Standards, Aug 20 2026 · Washington Technology, SBA's size standards overhaul means real winners and real losers, Sep 2026 · Mayer Brown, SBA Proposes Major Overhaul of Small Business Size Standards for Federal Contracts, Aug 2026

what this copies· Post it:XRedditHN

Subscribe to get daily leads by email.