The USDA Rules in Detail: Testing Requirements

Published on Nov 7, 2019

As we discussed last week, the United States Department of Agriculture (“USDA”) released their interim final rules for hemp manufacturing, and those rules were published in the Federal Register on Thursday, October 31. Hemp growers should note the rules on testing within the 0.3% limit that dictate whether the produced crop is classified as marijuana or hemp. These rules provide for potentially onerous situations for growers, though they are not unexpected.

Pre-Harvest Testing

Within 15 days prior to the anticipated harvest date, a designated person (from a governmental or law enforcement agency) must collect samples of the crop. While this requirement is simple enough, it could end up having important ramifications if the state agency doing the sampling takes longer than 15 days. We are familiar with multiple experiences of Washington growers having to wait many weeks for the Washington State Department of Agriculture (“WSDA”) to take samples. The USDA itself notes that sampling beyond the planned harvest date could result in tests higher in delta-9 tetrahydrocannabinol (“THC”) than they otherwise would. This occurs because cannabis plants convert cannabinoids to delta-9 THC as they approach maturity and harvest. This could mean the difference between whether an entire crop is destroyed or not, so the stakes are high.

Acceptable Hemp THC Level

The interim rules provide for some interesting tolerances on the 0.3% delta-9 THC limit, and explicitly states that product could test slightly over that limit but still qualify as hemp. The relevant rule states, “The method used for sampling from the flower material of the cannabis plant must be sufficient at a confidence level of 95 percent that no more than one percent (1%) of the plants in the lot would exceed the acceptable hemp THC level.” “Acceptable hemp THC level” means,

Acceptable hemp THC level. When a laboratory tests a sample, it must report the delta-9 tetrahydrocannabinol content concentration level on a dry weight basis and the measurement of uncertainty. The acceptable hemp THC level for the purpose of compliance with the requirements of State, Tribal, or USDA hemp plans is when the application of the measurement of uncertainty to the reported delta-9 tetrahydrocannabinol content concentration level on a dry weight basis produces a distribution or range that includes 0.3% or less.

The rule then provides an example for how this could occur. When cannabis is tested, there is a margin of error of the precision of the result. This margin of error, or “confidence level,” must be reported in addition to the actual result. For example, if a sample is tested as being 0.33% delta-9 THC, but the confidence level was 95% (meaning a 5% margin of error), that means that the sample could actually contain anywhere between 0.28% and 0.38% delta-9 THC. Therefore, because the sample could be under the 0.3% limit, it qualifies as hemp.

That same sample could fail if the test had only a 2% margin of error, because the sample then could contain only between 0.31% and 0.35% delta-9 THC.

Confidence Level

As you can imagine, this creates an interesting moral hazard on the testing laboratories, who could then be incentivized to provide as imprecise results as possible in order to appease their customers. That is why the rule requires a 95% confidence level, or no more than a 5% margin of error. Even with that margin, testing labs will still likely be incentivized to stay at that margin of error rather than trying to minimize the margin, as one might expect a scientist would hope for.


The rules state that the testing method must include a validated testing methodology that uses postdecarboxlyation “or other similarly reliable methods.” The rule states that other methods that are currently acceptable include gas or liquid chromatography. Postdecarboxlyation is the value determined that determines the delta-9 THC content in a sample after decarboxylation, which is defined in the rules as “the removal or elimination of carboxyl group from a molecule or organic compound.” In laymen’s terms, the sample is burned or heated to convert the non-intoxicating cannabinoid THCA to the intoxicating delta-9 THC cannabinoid.

Test Failures

The stakes for growers are quite high because if the samples test above 0.3% delta-9 THC, then legally they are marijuana and not hemp. Marijuana is still a Schedule I narcotic under the Controlled Substances Act and thus regulated extremely tightly. This means that the entire lot will have to be destroyed. And it doesn’t stop there: the party destroying the product must be licensed by the Drug Enforcement Agency to handle Schedule I drugs. The sample applies for the testing laboratories, because if a sample fails the test, these laboratories will then be in possession of Schedule I drugs. The rules currently don’t provide for any mechanism to cure defective lots, such as removing delta-9 THC through post-harvest extraction. It is possible that, after receiving comments from stakeholders, the USDA will provide for such mechanisms.