__Commenting period has been reopened until 10/8/2020 11:59 pm EST __

Written comments should be submitted via the Federal eRulemaking portal at www.regulations.gov.
Comments may also be sent via email to farmbill.hemp@usda.gov


Hemp is as challenging to regulate as it is to produce, and USDA has boldly confronted these challenges in drafting an Interim Final Rule to implement the hemp production authorities enacted in the 2018 Farm Bill.

But in its rush to promulgate the IFR, USDA has issued a rule that endangers farmers and regulators with severe flaws that contradict available science, create an inherent presumption of guilt, misconstrue the legal definition of hemp, and overlook practical realities of hemp production.

We’d like to see USDA use its discretion it used to delay enforcement of the IFR’s provisions containing these errors, just as it did in February for the provisions requiring DEA certification of labs and destruction of ‘hot crops.’ This would relieve farmers and regulators of the imminent impacts of these errors, particularly as we are all contending with unprecedented wildfires and storms as well as COVID-19.

The Rush to Regulate – Haste Makes Waste

Despite the continued availability of 2014 Farm Bill authorities, the Department was pressed to rush a new rule with immediate effect, without adequately considering public input, available science, and potential impacts on the industry. In the rush, USDA had to cut corners on requisite public comment and economic analysis. The Department asserted “good cause” for exempting itself from these requirements, but these assertions would not be likely to survive judicial review¹.

  • The Administrative Procedures Act requires public notice and comment promulgating regulations like the IFR, but the Department did not carry out the required notice and comment procedure before making the rule effective.
  •  The Regulatory Flexibility Act requires analysis of economic impacts on small businesses, but USDA’s analysis for the IFR was cursory at best.

There has also been excessive influence by DEA. As Secretary Perdue testified before Senate Appropriations in February: “We had some pushback from DEA that really didn’t like the whole [domestic hemp production] program to begin with…our hands were constrained many times by the interagency process…” While it is true that the Farm Bill directed USDA to ‘consult’ with the Attorney General (who oversees DEA), it also


¹ The law provides for exemptions from these requirements if the agency demonstrates that they are “impracticable, unnecessary, or contrary to the public interest,” but USDA’s assertions that the IFR meets such a standard are dubious at best.

specified USDA as having the sole authority to regulate hemp production. Why, then, is the interagency task group tasked with steering hemp regulation led by DEA and the White House ONDCP rather than USDA, especially given the Secretary’s own acknowledgment that DEA is opposed to hemp production to begin with?

Secretary Perdue made it clear in his testimony to Senate Appropriations: “I would functionally say it [the IFR] is a draft plan…” No one at USDA wants to see farmers used as guinea pigs.

‘Guilty Until Proven Innocent’ Contrary to Law, Threatens Farms & Regulators

The rush to regulate led to technical errors that unlawfully create a presumption of guilt that misconstrues the statutory definition of hemp. In addition to putting farmers at risk of immense financial losses, it also leaves State and Federal authorities vulnerable to claims of unjust regulatory takings.

The central problem is the IFR’s incorrect application of the legal definition of hemp in the Controlled Substances Act, as amended by the 2018 Farm Bill: “The term ‘hemp’ means the plant Cannabis sativa L. and any part of that plant…with a delta-9 tetrahydrocannabinol concentration of not more than 0.3 percent on a dry weight basis.” By stating ‘not more than 0.3 percent,’ the statute requires authorities to prove that a crop is more than 0.3 percent in order to find it to be unlawful. But three of the IFR’s provisions make it impossible for authorities to prove this beyond a reasonable doubt:

• The definition of post-decarboxylation is unsupported by science and overestimates the amount of THC in hemp crops, endangering farms with false findings of non-compliance.

• Exclusion of sampling error from THC testing leads to false findings of non-compliance.

• Sampling of flowering tops contradicts the scientific basis for the 0.3% THC threshold.


Total THC Definition Contradicts Science and Overestimates THC

The rule’s definition of Total Theoretical THC and its use in determining post-decarboxylation levels of Δ9-THC assumes that decarboxylation fully converts all THCA molecules solely to Δ9-THC and to no other substances. However, the literature (Dussy et al 2005, Iffland et al 2016) found the conversion rate to be actually between as low as 30% or 40%, and under no circumstances greater than 70%, and that much of the THCA converts to other cannabinoids or remains in ash. In fact, 100% conversion cannot occur in a realistic setting².

By assuming a rate of conversion higher than anything realistically possible, the “Total Theoretical THC” method chosen by USDA to measure post-decarboxylated


²This is not to be confused with the 87.7% factor that we see in Total THC conversion formulas, as this is used to account for the difference in the mass of the two types of molecules. This calculation needs to be further multiplied by 40% to reflect real situations.

THC massively inflates the amount of Δ9-THC (as some marijuana producers certainly welcome). This theoretical maximum thus renders authorities unable to prove beyond a reasonable doubt that a sample contains more than 0.3% d9-THC as required by the legal definition of hemp, a reasonable lower end of the range must be used instead. This is a basic tenet of hypothesis testing, as any scientist would know, and forces us to prove ‘innocence’ when the law requires us to prove ‘guilt.’

This misinterpretation of law and hypothesis testing has devastated farms as crop purchasers are backing out of their contracts due to misperceptions of risk associated with theoretical estimates of THC rather than real data. It also renders useless about 82% of the hemp flower and 74% of hemp biomass produced in the U.S. (HempExchange, 2019), and it excludes 68% of all the varieties planted (Orser, 2018). This restricts competition by harming smaller farms that depend on higher-potency varieties for their economic viability. The definition of post-decarboxylation methods for measuring THC must be revised to use reasonable estimates of the actual conversion rate of THCA to Δ9-THC.

Exclusion of Sampling Error Will Lead to False Findings of Non-Compliance

Nowhere does the rule provide for reporting the error resulting from sampling of the crop in the field, which accounts for the greatest variability in estimating THC, by far. As anyone who’s seen a hemp field would know, there is usually great variation from one plant to the next. However, the IFR requires testing as few as one plant per acre, and using that result as the THC level of everything harvested from that acre without reporting a margin of error.

It’s like weighing a few people in a town—or maybe even just one person—and assuming that’s what everyone in the town weighs. Or it’s like asking 4 Americans whom they plan to vote for, and one of them said Daffy Duck: would that mean that 25% of Americans plan to vote for Daffy Duck?

The rule specifies that samples are to be taken ‘to produce 95% confidence that no more than 1% of plants will exceed the Acceptable Hemp THC Level.’ This may sound like good statistics, but those parameters are meant for reporting a confidence interval—or margin of error—and yet the rule makes no provision to report such a margin of error. Without reporting a margin of error, the 95% confidence is meaningless. The rule does account for “Measurement Uncertainty,” but the rule defines this only as the uncertainty resulting from accuracy of the devices used to test the samples, not the variability within the sample itself.

The effects of this oversight can devastate farms of all sizes, especially smaller farms or farms growing different varieties. Consider a realistic example where 5 plants from a 5- acre lot test for the following THC Levels: 0.20, 0.20, 0.20, 0.20, 0.80: The THC level reported for this crop would be the average of those numbers, which is 0.32% – thus the crop would be deemed non-compliant and condemned. But if this level were reported with a margin of error at 95% confidence that includes the variability of the sample, the result would be 0.32% ± 0.21%, meaning that the actual THC Level is somewhere between 0.11% and 0.53%, and thus the lot would be deemed compliant and allowed to go to market.

The margin of error for Acceptable Hemp THC Level must account for sampling uncertainty based on valid statistical methods.

Sampling of Flowering Tops Contradicts the Science and Overestimates THC

The IFR’s requirement to sample only flowering tops contradicts the scientific basis for the 0.3% Δ9-THC threshold, overestimating the amount and threatening false non-compliance findings. It is well understood that the highest concentrations of all cannabinoids, including THC, is found in the flowering tops. However, the scientific basis for the 0.3% Δ9-THC threshold was established by Small and Cronquist in their seminal 1976 publication for use in measuring THC in the leaves of young, vigorous plants—not flowers and is by their own acknowledgment arbitrary. Sampling must thus be based on leaves or whole plants, not flowering tops alone.


A Practical and Natural Taxonomy for Cannabis Author(s): Ernest Small and Arthur Cronquist Source: Taxon, Vol. 25, No. 4 (Aug., 1976), pp. 405-435 Published by: Wiley

Stable URL: https://www.jstor.org/stable/1220524 Accessed: 08-07-2019 21:03 UTC


Implementation Requirements Overlook Realities of Hemp Production

The 15-day window between testing and harvest is entirely impracticable given testing-lab capacity and variable farming conditions and logistics.  Imagine sending a test sample, getting a borderline result back a week later, and then having to contend with a rainy spell or broken-down tractor—let alone field staff suddenly sickened with COVID—with just a few days before the 15-day window expires. Much if not most hemp cultivated in 2019 was under regulations providing a 28- or 30-day testing window using samples taken from whole plants.  Farmers commonly had to wait more than 15 days just to receive results back from the lab.

Lab capacity remains in very short supply in 2020, even without requiring DEA certification (see below). Labs could be exposed to tort claims for causing farms to miss mandated deadlines, and this also raises anti-competition issues, as larger farms can more readily absorb weather and logistical issues, and are far more likely to have their own lab capacity to secure testing deadlines. The testing window must be revised to 30 days, and specified to begin on the date the test results are reported by the lab.

Requiring the use of DEA-certified labs is unpracticable given lab capacity and restricts competition. We welcome USDA’s decision to delay its enforcement, but it must be permanently removed. There are not enough DEA labs to test the amount of hemp being grown in the US. This will restrict market access for many farms, especially those independent of well-capitalized verticals that have their own DEA labs and offer services preferentially to their own subsidiaries and suppliers. The DEA requirement must be replaced with State accreditation of labs based on standards such as ISO or ASTM.