Redefining “Small”: The SBA’s Proposed Overhaul of Size Standards
On August 20, 2026, the Small Business Administration (“SBA”) proposed a comprehensive overhaul of its small business size standards for the first time in decades. The proposal would purportedly simplify industry categorization and raise applicable size thresholds, which the SBA says would allow “small but rapidly growing firms to continue qualifying and [add] over 110,000 firms to the 36 million small businesses in America.”
The SBA and its programs (government contracts, loans, and technical assistance) sit at the center of a tension that runs through most small business policy: balancing the duty to use public resources efficiently against the goal of actually reaching the people the program was meant to serve. Is the goal to support the most vulnerable businesses or to allocate resources to high-growth firms so they can grow into larger ones? Is it to secure the best deal for taxpayers or to direct capital toward underserved communities? The current proposal promises simpler and more efficient size standards, but does it account for the shifting of resources away from the smallest businesses toward larger, better resourced businesses?
To understand what this looks like in practice, we must first understand how the current size classification system came to exist.
How we got here
The Small Business Act of 1953 authorized the SBA to establish size standards for determining eligibility for federal small business assistance; the agency established its initial standards on January 1, 1957 and size was measured in two different ways:
Manufacturing industries were considered to be “small” if they had no more than 500 employees. This figure became the long-standing “anchor” for employee-based measurement standards to come.
Other industries (such as construction, trucking and warehousing, retail, and wholesale trade) used a revenue-based standard, ranging from $1 million to $5 million.
Since then, the SBA has made numerous changes to the size standards:
In 1959, it split the 500-employee anchor into different tiers: financial assistance programs used 250-, 500-, and 1,000-employee standards depending on industry, while federal contracting programs retained the flat 500-employee standard.
The revenue measurement was also expanded several times. By 1963, SBA revenue-based size standards were: $1 million for retail trade industries; $1 million for services industries; $5 million for wholesale trade industries; and $7.5 million for construction industries.
In 2000, it scrapped the use of the Standard Industrial Classification it had used since inception in favor of the North American Industrial Classification System (NAICS), which it still uses today.
In 2009, the agency moved to calculating a distinct standard for each individual NAICS industry, replacing the older “anchor” approach with what it now calls a “percentile” approach which calculates the applicable standard based on a particular industry’s own firm size distribution, concentration, and competitiveness.
These shifts produced the current system which comprises 1,037 separate size standards. Employee-based thresholds now range from 100 to 1,500 depending on the specific industry and receipts-based thresholds now range from roughly $2.25 million to $47 million.
The other "small business" definition
To complicate matters even further, the SBA’s Office of Advocacy (“SBA Advocacy”) — a separate, independent unit within the SBA created by Congress in 1976 to research small business issues and represent small business interests in the regulatory process — uses a related but distinct, and more generalized, definition: businesses with up to 500 employees. That’s the number behind most of the headline statistics we see because it’s simple, memorable, and consistent across industries. SBA Advocacy uses this for more effective communication of its research and national statistics.
This is a research convention, not the operative legal standard, and the distinction is important. Eligibility for programs such as SBA loans and government procurement opportunities strictly apply the SBA’s current patchwork of size standards.
Proposed changes to the legal framework
The first substantive change would revise the entire methodology SBA uses to set size standards by moving away from the 6 digit NAICS industry codes currently used. Instead, the SBA would set certain standards at the broader 4- or 5-digit NAICS Industry Group level instead. This would reduce the number of size standards from 995 to 338 (a reduction of around 66%). The stated rationale is twofold: it’s easier for small businesses to identify where they fit within those broader categories and therefore to pursue federal contracting opportunities, and it reduces the risk of contracting officers assigning the wrong NAICS code and therefore the wrong size standard to a contract, excluding eligible small businesses from contention.
There is also a proposal to make employee count, rather than revenue, the default measure wherever appropriate, representing a meaningful departure from a system long tethered to revenue-based measurements for services industries. The SBA estimates that this would bring a further 114,541 additional firms under the threshold, including about 37,002 existing federal contractors that combined held more than 105,655 contracts worth over $71 billion in FY2025. Conversely, it estimates that fewer than 200 businesses nationwide, out of more than 6.3 million, could lose their small business designation.
What these changes could mean for small businesses
This isn’t the SBA’s first attempt at broadening eligibility, and there’s new empirical evidence on what happens when it does. A 2026 NBER working paper studied the last major wave of SBA size standard increases following the Small Business Jobs Act of 2010. By exploiting quasi-random variation in the timing of these expansions, and using confidential Census Bureau firm-level data from 2008 to 2017, the authors isolate the causal effect of raising a size standard.
The findings are clear: when an industry’s size standard increases, revenue for that industry’s smallest firms (those below 50% of the old threshold) falls by 6.8%, an average loss of more than $90,000 in annual revenue per firm. The decline shows up immediately, holds for at least two years afterward, and doesn't appear before the change, confirming that the size standard is the cause of the decline. Firms that become newly eligible for small-business support because of the higher threshold see the opposite effect: their revenue goes up by 15.4%.
The reason for this is the increased competition for the same fixed pool of support. The clearest example is seen in federal contracting numbers: the authors find that although total dollar amount of contracts flowing to an industry doesn’t change after a size standard increase, the share of contract dollars going to the smallest firms drops by 11.3 percentage points, and the share of individual contracts they win drops by 13 points.
Who absorbs these losses isn’t evenly distributed either. The revenue losses are worse for younger firms (8.1% vs. 6.7% for older ones) and for more productive firms (9.9% vs. no measurable effect for low-productivity firms). The effect is also sharply worse for firms that depend more on external financing or SBA lending (9.1% revenue decline vs. 2.3% for less-constrained firms).
The negative effects of this extend beyond revenue. Wages at the smallest firms fall by 2.8%, driven entirely by reduced payroll rather than layoffs. At the industry level, firm exits increase 9.1% overall and 12% among the smallest firms specifically, with no offsetting increase in new firm entry, and patent applications (a proxy for innovation) fall 14% the year after a size standard increase, 15% the year after that, and 31% within three years, consistent with resources shifting away from the small, young, high-productivity firms.
There's also the likely effect on the already problematic heterogeneity within the SBA’s definition of small business. As we’ve discussed in earlier blogs (see: here and here), the definition used by the SBA (and consequently by many government agencies and other organizations) is inherently flawed. First, employees are only counted if they are W2, excluding 1099 and off-books workers which are a core supplier of labor to small businesses. Second, it conflates businesses that are fundamentally distinct in size, capital, workers, resources, revenue, and many other factors. A 2-person coffee shop and a 400-person, multi-site medical business are not the same thing and shouldn’t be treated as such. Third, it misses the nuance needed to understand the different needs of different sized businesses and how those needs can be better served by the SBA and other organizations. Broadening the pool of “small businesses” means that even more firms with meaningfully different capital needs, cash-flow cycles, and financing constraints are measured by the same yardstick.
Underneath this proposal is a tension that runs through most policy: the duty to use public resources efficiently against the goal of actually reaching the people the program was meant to serve. Broadening the definition of who counts as “small” widens the pool of firms eligible to compete for the same fixed resources. We know that these resources are more likely to be directed to larger firms that can a) deliver economies of scale, or b) have the knowledge and resources to navigate government procurement, so changes like this (much like the change in 2010) actually harm some of the small businesses that the SBA was set up to serve. That tension is why size standards have been revised so many times over the past seventy years; each revision is an attempt to strike a balance that the last one didn’t quite get right. This doesn’t mean that policy changes are the problem – there is an inherent need for policy to continue to adapt and evolve with changing conditions. But good policy only works if it accounts for the unintended consequences a change like this can produce.
The SBA is currently seeking comments on the proposed rule and following consultation will issue any necessary revisions.



Comments