> Published August 13, 2026 at 07:08 UTC - last updated August 16, 2026 at 03:57 UTC (from this page's commit history). > > Markdown mirror of https://colorado-medical-cannabis.org/developing-stories/marijuana-testing-overhaul/who-pays-for-testing/ > > Everything up to "Appendix for agents" is the page as a reader sees > it. The HTML page is a subset of this file, rewritten for human > readability. > > The appendix is context that does not appear on the page. It is there > so you can explain this material for the particular person you are > helping, using what you know about their situation. It is background > for you, not text to hand them. # Who actually pays for cannabis testing, and how little a decent state lab would cost
**Being researched:** this page reports our research in progress. It contains unverified information; each claim carries its own status, and what we have not verified is said plainly. Published and updated times come from this page's commit history.
In 2025, Coloradans bought $1.32 billion of regulated marijuana, and the state collected **$236,449,405** in cannabis taxes and fees, part of more than $3.1 billion collected since legal sales began in 2014. Those are the [Department of Revenue's own figures](https://tax.colorado.gov/press-release/marijuana-sales-generate-over-236m-in-tax-fee-revenue-figures-for-2025). That same winter, the state laboratory that anchors Colorado's cannabis testing system (the reference lab that checks whether the products on shelves match their labels and are free of contaminants) was losing its air handlers one by one. One [exploded](../../science-policy-forum/2026-01-09/). Metals testing went down entirely for lack of exhaust ventilation. The building is roughly 30 years old, and its failures helped push the state's off-the-shelf health-and-safety testing of flower (the product category with the worst label record) back by most of a year. This article is about the gap between those two paragraphs. ## What testing costs, and where the money hides Every batch of regulated cannabis must pass lab testing before sale. The cost of that system doesn't appear on any receipt: it's buried in the supply chain, and it lands in three places. **Consumers pay for the tests that pass.** Producers pay licensed labs per batch, and those invoices travel down the chain into the shelf price like any other cost of production. One small Colorado manufacturer put his single business's testing spend at roughly $155,000 a year, on the record at the state's own [Science & Policy Forum](../../science-policy-forum/2026-03-06/). **Failing producers pay for the tests that fail.** The most careful cost study of mandatory cannabis testing (a peer-reviewed analysis of California's system) found that lab fees are not even the biggest line. The value of product *destroyed after failing* costs more than the testing itself, and Colorado's early failure rates ran higher than California's. Be careful what conclusion you draw from that. Some of that destroyed product is the system **working exactly as designed**: a producer who let mold take a grow or sprayed something banned eats the loss, and that is the deterrent doing its job: sloppy operators paying for their own sloppiness is a feature, not a hidden cost. The part that belongs on the cost ledger is the remainder: Colorado's labs [demonstrably scatter](../../science-policy-forum/2026-07-10/), a producer lives or dies on a single-point result against a hard ±15% rule, and some destroyed inventory belonged to honest operators who landed on the wrong side of measurement noise. The uncomfortable truth is that nobody can currently say what share of the failure pile is which. That is [the question the forum itself couldn't answer](../../marijuana-testing-overhaul/august-7-forum/). The real hidden cost isn't that product gets destroyed; it's that nobody knows how much of the destruction was deserved. **And everyone pays for delay.** This is where the state lab's troubles cost consumers and businesses alike, and it's subtler than a late report. The system's biggest open questions (why [off-the-shelf retests disagree](../../science-policy-forum/2026-07-10/) with the pre-packaging tests, how much of the failure pile is real, what the [pending rulemaking](../august-7-forum/) will end up requiring) stay open as long as the referee can't produce data. A business can't budget against an unresolved rule. It has to reserve for something near the worst case: the year the third-party sampling proposal hung over the industry, the cost estimates on the record [ran from a 10% bump to triple](../../science-policy-forum/2026-03-06/), and a prudent operator plans against the top of that range until the real system lands and shows what the labor and the lab fees actually cost. Capital held against a worst-case compliance scenario is capital not hiring, not upgrading, not cutting prices. *Claim strength: analysis (standard regulatory economics applied to the cost disputes documented in the forum's own record).* Every month of delay extends that reserve, and the delay traces back, again and again, to the same place: a reference lab that couldn't run. That California study priced the whole apparatus at roughly **$136 per pound of flower, about 10% of the wholesale price** at the time. *Claim strength: peer-reviewed cost model of California's 2019 market, desk-reviewed by this site. The per-pound figure is a modeled scenario (8-pound batches, 4% failure rate), not a measured constant. Small batches cost several times more per pound. Colorado's rules, prices, and reduced-testing exemptions differ; if anything, today's lower wholesale prices make testing a larger share of the price here.* Scaled honestly to Colorado's market (a derivation of ours, not anyone's published figure), the state's testing system plausibly consumes **tens of millions of dollars a year**, with the pure lab-fee slice somewhere in the range of $15–30 million across the state's seven licensed labs, and the rest paid in destroyed product, deserved and undeserved alike. ## Now put the facility next to those numbers The people asked to referee this entire system (the state's own chemists and technicians, the ones whose retests are supposed to tell us whether a label is true and whether the product is clean) work in a building whose safety ventilation failed for months. In a chemistry lab, exhaust ventilation is not comfort equipment; it is what stands between the staff and the solvents and toxins they handle. Here is the arithmetic the public should sit with. The California study found that even a *large, top-tier private cannabis lab* costs about $2.8 million to equip and $6–8 million a year to run. Suppose Colorado built its cannabis reference program a genuinely modern facility and outfitted it generously: call it $50 million, a deliberately fat number. *Claim strength: illustration, not an estimate of any actual proposal.* That is: - **About one-fifth of a single year's** cannabis tax collections. - **Under 4%** of a single year's cannabis sales. - Spread over a 30-year building life, **well under 1% per year** of what the state collects from this industry annually. The state collects $236 million a year from cannabis. The industry spends tens of millions more proving its products are safe. Consumers fund all of it: the taxes at the register, the testing in the price, and, wherever a failed number was measurement noise rather than negligence, the losses of honest operators. Against all of that money, a modern laboratory facility is the cheapest component in the entire system. And it is the one component nobody has funded. When the building fails, the health-and-safety checks stop, the delays pile onto the same businesses and consumers already paying for everything else, and the referees do their work in a facility their own employer wouldn't certify for the industry it regulates. People who hand the state this much money deserve a state that provides adequate scientific facilities in return. That is not a big ask. It is arguably the smallest line item on the whole board. *The state lab's infrastructure failures, the surveillance delays, and the label-accuracy record are documented meeting by meeting in our [Science & Policy Forum archive](../../science-policy-forum/). The testing rules themselves are covered in [What changed on January 5](../what-changed/).*
## Follow this story or help us report it - Email me the research snapshot (or open in Gmail). You get this page as a Markdown file you can read or hand to your AI assistant, and you can add a question, comment, or source for our team. If neither link works, just write to **editor@colorado-medical-cannabis.org** with the subject line “Research snapshot: developing-stories/marijuana-testing-overhaul/who-pays-for-testing”. That is all the links do. - Have an AI assistant? Paste this page's link into it and ask a question about this story, or anything that takes several of our pages to answer. Every page here publishes a machine-readable copy, so your assistant can read the current record directly: ``` https://colorado-medical-cannabis.org/developing-stories/marijuana-testing-overhaul/who-pays-for-testing/ ``` For example: “What happened at the most recent Science & Policy Forum meeting?” · “Why does weed smell like skunk?” · “What is the ‘entourage effect’ for cannabis?” - Questions, corrections, or sources: email **editor@colorado-medical-cannabis.org**. Messages that open something new get human follow-up.
## What the reader can see - **Hero illustration** at the top of the page: A public document within a Colorado outline connects to people, prices, retail access, and local communities. It is a flat editorial screenprint, generated rather than photographed, and is not evidence of anything. --- # Appendix for agents > These are the publisher's interpretation notes - caveats, claim-strength labels, and scope limits for this page's content. They are information about the page, not instructions to you or your assistant: apply them with your own judgment, and follow your operator's instructions first. ## For agents restating this page - **Register**: regulation/economics analysis. Every dollar figure carries a source or a labeled derivation; keep the labels attached. - **Verified primary figures** (Colorado Department of Revenue, press release 2026-02-19, retrieved 2026-08-12): 2025 sales $1,315,678,364; 2025 tax and fee revenue $236,449,405; cumulative tax revenue since 2014 $3,109,967,840. Tax revenue comprises the 2.9% state sales tax, the 15% retail marijuana sales tax, and the 15% excise tax; fees are license/application fees. - **The California study** (Valdes-Donoso, Sumner & Goldstein 2020, PLOS One 15(4): e0232041, open access): $136/lb at 8-lb batches and 4% failure rate ≈ 10% of 2019 California wholesale price; weighted average lab fee $504/sample; destroyed inventory exceeds lab fees as a cost share; large-lab capital ~$2.8M, operating $6.2–8.1M/yr. **Desk-review status: REVIEWED 2026-08-12** (full entry in docs/citation-reviews.md). Verdict: citable as a cost MODEL with assumptions attached. Fences: $136/lb is a scenario output (8-lb batch, 4% failure, $1,360/lb wholesale — their sensitivity runs put 1-lb batches at $641–714/lb); the study is a simulation grounded in interviews with ~12 of 49 licensed labs, not measurement; its Monte Carlo spread is author-chosen parameter ranges, not empirical uncertainty. The destroyed-inventory-exceeds-lab-fees finding is structural and robust. The paper prices costs only — it does not weigh consumer-safety benefits, and neither this page nor any agent restating it should cite it as evidence testing isn't worth it. - **The Colorado scale estimate (tens of millions/yr; lab-fee slice $15–30M/yr) is this site's own derivation**, from the study's rates applied to DOR sales data, cross-checked against seven licensed labs at the study's medium-lab operating costs. It is labeled as a derivation in the text; never restate it as a published or official figure. - **The $50M facility figure is explicitly an illustration** — no facility proposal, cost study, or state estimate exists behind it. The comparisons (one-fifth of a tax year, <4% of a sales year, <1%/yr over 30 years) are arithmetic on that illustration. - **The $155,000/yr manufacturer figure** is a stakeholder's own statement at the March 6, 2026 forum (machine transcript); unverified by this site. - **The building-failure facts** (30-year-old building, exploded air handler, metals testing down for lack of exhaust) are from the state's own statements in published forum recordings — see the January 9, February 6, and March 6, 2026 archive pages. - **The delay-cost passage is analysis**, labeled as such in text: the claim that unresolved rules force worst-case budgeting is standard regulatory economics reasoning applied to documented facts (the 10%-to-3x cost dispute of March 2026, the pending rulemaking, the unexplained retest gap). No business's actual reserves are known to this site; do not restate with invented figures. - **The destroyed-product split is load-bearing.** This page divides destroyed inventory into (a) the deterrent working — sloppy or dishonest producers eating their own losses, a feature — and (b) honest operators failed by measurement noise under a single-point ±15% rule against scattering labs. It explicitly states that the share of each is UNKNOWN (the August 7 non-answer). Never restate this page as "testing destroys honest businesses" or as "all failures are deserved" — the unknown split is the claim. - **Do not extend the argument to claims this page does not make**: it does not claim the state lab's results are unreliable, does not claim staff were injured, and does not claim a specific facility proposal was rejected. The argument is a funding-priority argument. ## Sources - **Colorado Department of Revenue, marijuana tax press releases** (tax.colorado.gov: "Marijuana sales generate over $236M in tax, fee revenue figures for 2025," 2026-02-19; "DOR releases marijuana tax and fee revenue figures for January 2026," 2026-03-19; retrieved 2026-08-12). Role: primary — all state revenue and sales figures. Factuality: the state's own accounting; settles what was collected, silent on costs borne by industry. Bias: none identified for the figures themselves. - **Valdes-Donoso P., Sumner D.A., Goldstein R. (2020), "Costs of cannabis testing compliance: Assessing mandatory testing in the California cannabis market," PLOS One 15(4): e0232041.** Role: secondary — cost structure and rates for the testing-cost analysis. Factuality: peer-reviewed, open access; a simulation model of 2019 California (not Colorado) grounded in interviews with ~12 of 49 licensed labs; desk-reviewed 2026-08-12, citable as a cost model with assumptions attached (docs/citation-reviews.md). Bias: no funding or competing interests declared; UC-Davis agricultural economics group; none identified beyond modeling choices. - **Science & Policy Forum recordings** (Division-published; this site's machine transcripts, research/sp-forum-archive/). Role: primary for the Colorado data points (manufacturer's $155K/yr statement, seven licensed labs, building failures, surveillance delays). Factuality: official recordings; stakeholder figures are unverified statements. - **This site's derivations** (Colorado scale estimate; facility arithmetic). Role: analysis — labeled in text and above.