A Texas group health premium is quoted by an issuer for a particular business. What gets published is a federal survey average: AHRQ’s Medical Expenditure Panel Survey–Insurance Component reports an average total single-coverage premium of $8,737 per enrolled employee in 2025 at Texas private-sector establishments with fewer than 50 employees that offer health insurance (MEPS-IC 2025, Table II, row II.C.1).
Where the one published Texas figure comes from
AHRQ publishes a Texas-specific workbook, and the figure above is a single cell in it. The table is titled “Table II. State of Texas, private-sector data by firm size, 2025” and the column is headed “Less than 50 employees”; the workbook itself is Texas2025.xlsx. Its data year is 2025 and it was published in July 2026, which is the newest Texas edition there is.
The survey is not a collection of quotes. Its own methodology page says who gathers it and where the sample comes from:
The U.S. Census Bureau serves as the data collection agent for the MEPS-IC survey. The List sample is drawn from the Business Register, a confidential list of establishments in the United States maintained by the Census Bureau.
So the figure is a weighted estimate built from a federal register of workplaces. That is a genuinely useful thing to know about the Texas market. It is not a rate anyone offered to anyone.
The figure is conditioned on already offering coverage
Every premium row in that table is measured, in AHRQ’s own words, at establishments that offer health insurance. The denominator excludes every Texas workplace offering nothing at all, which means the average does not describe Texas small employers generally. It describes the ones already buying.
That is the first reason the number cannot travel to a particular business. It is an average over a population an employer may or may not be in, in a year that has already closed, and it carries no information about any one census, industry or plan design. We publish no premium of our own, in digits or in words, and we do not restate AHRQ’s annual figure as a monthly one — a published annual average that has been divided is no longer the published figure.
It is worth being blunt about why that restraint matters more here than it looks. A survey average is exactly the kind of number that reads like guidance: it has a source, it has a year, it has a state attached to it, and it is easy to repeat. What it does not have is any relationship to the business reading it. An employer who anchors a budget on it has anchored on the outcome of other employers’ plan choices, staffing and tier mix, none of which are theirs. The figure is worth publishing because it is the only Texas measurement anyone has; it is worth framing carefully for the same reason.
“Establishments” is not another word for “businesses”
MEPS-IC counts establishments and sorts them by the size of the firm that owns them. Both words are defined terms on the survey’s glossary page, and the difference between them is the difference between reading the column correctly and reading it wrong:
ESTABLISHMENT - A particular workplace or physical location where business is conducted, or services or industrial operations are performed. Also known as a site. The sample is chosen at the establishment level, and the MEPS-IC data are collected at the establishment level whenever possible.
FIRM - A business entity consisting of one or more establishments under common ownership or control. Also known as an enterprise. A firm represents the entire organization, including the company headquarters and all divisions, subsidiaries and branches. A firm may consist of a single-location establishment or multiple establishments. In the case of a single-location firm, the firm and establishment are identical. Firm size is the total number of employees for the entire firm as reported on the sample frame.
Read together, the column is narrower than its label suggests. A Houston distribution site belonging to a national company is an establishment, but its firm is not small, so the site is not in the small-firm column at all. A Lubbock shop that is the whole of its own company is in it. The unit being counted is the workplace; the size test is applied to the whole organization behind it.
The survey’s size column is not the Texas legal definition
This is where a Texas reader is most likely to be misled, because the survey’s column boundary and the state’s statutory ceiling happen to share a number while describing different things.
Texas writes its own definition. Tex. Ins. Code § 1501.002(14) defines a small employer as a person who employed an average of at least two employees on business days during the preceding calendar year, subject to a ceiling of 50 employees, and who also employs at least two on the first day of the plan year. The survey’s column has no floor at all, applies no first-day test, and counts the employees of the entire firm as the sample frame reports them. The two populations overlap. They are not the same set, and a figure drawn from one is not a statement about the other.
If the definition itself is the question you came with, the Texas state page is the better starting point than any premium figure.
Family coverage is published separately, never derived
MEPS-IC also publishes a family row for the same column. It reports an average total family premium of $21,949 per enrolled employee in 2025 at Texas private-sector establishments with fewer than 50 employees that offer health insurance (MEPS-IC 2025, Table II, row II.D.1).
Those are two separately published cells, and they stay separate here. We do not subtract one from the other, express one as a multiple of the other, or describe a family-versus-single gap, because every one of those operations produces a figure AHRQ never published. The honest use of the pair is that family coverage is measured and reported on its own terms, and that a plan’s tier mix is one of the things an issuer prices — which is a reason to get a real quote rather than to do arithmetic on a survey.
How often Texas establishments in that column offer coverage at all
MEPS-IC reports that in 2025, 28.2% of Texas private-sector establishments at firms with fewer than 50 employees offered health insurance (MEPS-IC 2025, Table II, row II.A.2).
That is a statement about the survey’s sample in a single year, and it is worth reading for exactly what it is. It is not a benchmark, not a norm, and not evidence about any particular employer’s obligation, eligibility or standing against neighbors. It is also not a trend: one edition is being read here, and no earlier edition is used, so nothing on this page says the Texas offer rate is rising or falling.
What AHRQ says about the precision of its own estimates
AHRQ publishes a rule for when one of its estimates should not be reported at all, and it publishes it in the open:
RSE > .50: Estimate should not be reported or displayed in tables due to extremely large sampling error. .30 ≤ RSE ≤ .50: Estimate can be reported but flagged with an * to indicate that its precision is questionable. RSE < .30: Estimate is deemed precise and can be reported with no *.
None of the Texas cells behind the figures on this page carries that asterisk, on the value or on its published standard error. That is worth saying plainly, because it is the only reliability signal the survey offers at state level, and because an author who does not check it has no idea whether the number they are repeating is one AHRQ would stand behind.
What actually sets a Texas group’s rate
A survey cannot tell a business what it will pay, because the survey is not what sets the rate. In the small group market the federal rule at 45 CFR 147.102(a)(1)(i)–(iv) closes the list of things a premium may vary on: whether the coverage is for an individual or a family, the rating area, age, and tobacco use. Nothing outside that list may move the rate.
That closed list is why a real Texas number is cheap to obtain and impossible to guess. The inputs are your own — your census, your address, the plan you pick — and the arithmetic belongs to the issuer. Getting a Texas quote puts the actual factors in front of an actual carrier; reading an average puts none of them anywhere. If you already have a plan and the question is really about next year, shopping the renewal is a different exercise from pricing a first plan, and group health coverage is where the plan mechanics sit.
What this page will not put a number on
MEPS-IC publishes the employee side of the premium split and publishes no employer share. Subtracting the employee figure from the total would produce a number the survey never reported, dressed in a citation it does not support, so the employer’s portion is described here in words and nowhere in digits.
The same rule governs everything else on this page. There is no monthly figure, no per-employee division, no year-over-year comparison, no family-to-single multiple, and no Texas county or rating area described as cheaper or dearer than another — not because those questions are uninteresting, but because no source read for this page publishes an answer to them, and a figure nobody published is a figure nobody can defend.
The bottom line
The one published Texas figure is a survey average across establishments that already offer coverage. It is not a rate, not a starting point, and not yours. A Texas number comes from an issuer, on a closed federal list of rating factors, and we publish no premium of our own.
Frequently asked questions
Is there a published average premium for group health insurance in Texas?
There is one. AHRQ’s Medical Expenditure Panel Survey–Insurance Component reports an average total single-coverage premium of $8,737 per enrolled employee in 2025 at Texas private-sector establishments with fewer than 50 employees that offer health insurance. It is a survey average across establishments that already offer coverage, published in July 2026 for data year 2025, and it is not a price anyone was quoted.
Does the survey say what a Texas employer pays toward the premium?
No. MEPS-IC publishes the employee side of the split and publishes no employer share at all. The employer pays the rest. How much the rest comes to is not a number the survey reports, and it is not a number that can be worked out from the ones it does report, so we do not print one.
Is the survey’s small-firm column the same thing as a Texas small employer?
No. The survey column counts establishments sorted by the size of the firm that owns them, with no floor and no plan-year test. Texas defines a small employer by statute at Tex. Ins. Code § 1501.002(14), which requires an average of at least two employees across the preceding calendar year and at least two on the first day of the plan year. The two populations overlap; they are not the same set.
Can the annual figure be restated as a monthly cost per employee?
Not honestly. AHRQ publishes an annual average per enrolled employee and nothing else. Dividing it into months, or into head-count slices, produces a figure the survey never published and nobody has to stand behind. This site does no arithmetic on a published figure.
Why can’t you quote my business from that number?
Because the survey is not what sets your rate. In the small group market the federal rule closes the list of things a premium may vary on, and a survey average of other people’s establishments is not on it. A Texas number comes from an issuer that has your census, your address and your plan selection in front of it.
About the author
Nate Jones, CPCU, is the founder of Wexford Insurance and Wexford Health Insurance, an independent agency placing small-business group benefits in 9 states. Nate reads the MEPS-IC Texas workbook against the Texas Insurance Code every year so that the one published Texas average is never handed to an employer as though it were a quote. Start with the Wexford Health Insurance quote form.