Agricultural Biotech and the Cracks in the Coordinated Framework

The Coordinated Framework Applied to Agriculture

BP1 introduced the Coordinated Framework. Agricultural biotech is where its assumptions are most strained.

The 1986 framework assigned agricultural biotech to three agencies:

USDA APHIS regulates plants under the Plant Protection Act. Its authority covers organisms that could pose a "plant pest risk" — meaning they could damage agriculture or natural plants. Plants engineered with genes from known plant pests (most early GMOs used genes from Agrobacterium tumefaciens, a plant pathogen) triggered review under this authority.

EPA regulates pesticides under FIFRA, including plant-incorporated protectants (PIPs) — proteins genetically engineered into plants to kill or repel pests. The most famous example is Bt crops, which express insecticidal proteins from Bacillus thuringiensis. Each new Bt event requires EPA registration.

FDA regulates food safety for genetically engineered foods. In practice, FDA review of GMO foods has been voluntary — through a "consultation" process where companies submit safety data and FDA confirms it doesn't have concerns. This was set in a 1992 FDA policy statement that has remained largely unchanged.

This three-agency setup made some sense for the first generation of GMOs. Roundup Ready soybeans (engineered to tolerate glyphosate herbicide), Bt corn (engineered to kill corn borers), and similar products clearly involved foreign DNA insertion using methods (like Agrobacterium transformation) that triggered all three agencies' authority.

It made much less sense for what came next.

---

The Gene-Edited Crop Question

Modern gene editing — CRISPR especially — can produce changes to crop genomes that are indistinguishable from natural mutations. A targeted deletion of a few base pairs. A single nucleotide change. The introduction of an existing gene variant from one part of the species into another.

Are these GMOs? Legally? Practically?

The traditional definition of "genetically modified organism" assumed transgenic technology — the insertion of foreign DNA from one species into another. Gene editing doesn't necessarily involve any foreign DNA. After the editing process, the final organism may have only minor deletions, single base changes, or rearrangements that could have arisen through natural variation.

The USDA confronted this question directly with the SECURE rule (Sustainable, Ecological, Consistent, Uniform, Responsible, Efficient), finalized in May 2020.

SECURE narrowed USDA's regulatory authority over plant biotech in a major way. Under SECURE:

• Plants with edits that could have been achieved through conventional breeding are not subject to USDA regulation
• Plants with edits that involve a single deletion, single base substitution, or introduction of a gene from a sexually compatible species are exempt from review
• Only plants with transgenic insertions or more substantial modifications require the full regulatory process

The CRISPR mushroom from this module's opening was an early example of what SECURE would later formalize: small CRISPR edits would essentially fall outside USDA regulation entirely.

This was a major win for plant biotech companies. It was also a major loss for the precautionary regulation advocated by many environmental groups, who argued that genetic engineering — regardless of method — produces changes that wouldn't occur in nature and should be assessed for ecological and health risks. The debate over SECURE was furious and continues.

The EU went the opposite direction. In 2018, the European Court of Justice ruled that gene-edited crops are GMOs under EU law and require full regulatory review. The result: a CRISPR-edited crop can go to market in the US in a matter of months, but face years of regulatory review (or outright prohibition) in the EU. The same crop, the same gene edit, different legal status.

Whether the US or EU approach is better depends on what you value. The US approach favors innovation and speed. The EU approach favors precaution and consumer choice. Both are defensible positions. They cannot both be implemented in the same country.

---

GMO Labeling: The National Bioengineered Food Disclosure Standard

For decades, GMO labeling was one of the most contested food policy issues in the United States. Consumer groups wanted mandatory labels. Industry groups argued labels would mislead consumers into thinking GMOs were dangerous. Various states started passing their own labeling laws — Vermont's 2014 law was the first to require mandatory GMO labels on food sold in the state.

The patchwork of state laws was untenable for national food companies. So Congress preempted state action with the National Bioengineered Food Disclosure Standard (NBFDS), signed in 2016 and implemented in 2022.

The NBFDS requires disclosure for foods containing "bioengineered" ingredients. But the implementation involves multiple compromises that critics argue undercut the law's purpose:

• The term "bioengineered" was chosen instead of "GMO" or "genetically engineered," because USDA argued the public was less familiar with the term — reducing the chance consumers would avoid labeled products
• Disclosure can be made via text on the package, a symbol (the "BE" logo), a QR code, or a phone number — meaning consumers may need a smartphone to actually know whether food is GMO
• The threshold for triggering disclosure is high: foods with less than 5% bioengineered material may not require disclosure
• Highly refined ingredients (like sugar from GMO sugar beets, or oil from GMO soybeans) are exempt because they don't contain detectable modified DNA, even though they came from GMO crops
• Gene-edited foods that fall outside USDA regulation under SECURE may not be considered "bioengineered" under the NBFDS at all

The result is a labeling system that technically exists but in practice provides limited transparency. Industry groups consider this appropriate (since they view GMO labeling as scientifically unjustified). Consumer groups view it as a deliberate undermining of meaningful disclosure.

International comparison sharpens this. The EU requires clear "Genetically Modified" labels on the front of packages. Many countries that require labeling have lower thresholds, fewer exemptions, and clearer language. The US disclosure standard is among the weakest in the developed world.

---

Pesticides, Plant-Incorporated Protectants, and Ecological Risk

EPA's role in agricultural biotech is less visible than USDA's but arguably more consequential for ecological policy.

Plant-Incorporated Protectants (PIPs) are proteins engineered into plants that confer pest resistance. The dominant examples are Bt crops — corn, cotton, soy, and other crops engineered to express insecticidal proteins from Bacillus thuringiensis.

EPA regulates these proteins as pesticides under FIFRA. Each Bt protein must be registered, with safety data on human health, non-target organism effects, and resistance management. EPA registrations are typically time-limited and require renewal — providing periodic opportunities to update requirements based on emerging data.

The major ongoing policy questions:

Resistance management. Like any pest control method, Bt crops can select for resistant pest populations. EPA requires resistance management plans — typically requiring farmers to plant non-Bt "refuge" acres to maintain a population of pests that haven't been selected for resistance. Refuge compliance is variable, and resistance has emerged in multiple pest species over the past two decades. The fundamental issue: every Bt crop deployed at large scale will eventually face evolved resistance. This is biology, not policy failure.

Non-target effects. Whether Bt crops affect beneficial insects, soil microbes, or non-target species has been studied extensively. The general scientific consensus is that effects on non-target species are limited and significantly smaller than the alternative (spraying broad-spectrum insecticides). But the consensus is contested, especially for novel applications like RNA-interference traits or stacked multi-toxin crops.

Synthetic biology applications. Newer agricultural biotech products go well beyond Bt — engineered nitrogen-fixing bacteria, synthetic biological controls, gene drives for pest suppression. EPA's regulatory framework for these is still developing. Some of these technologies, especially gene drives, raise issues that traditional pesticide regulation isn't designed to address (covered in BP8).

The bigger picture: EPA's role in agricultural biotech is fundamentally about ecological risk assessment. As biotech moves into more ambitious ecological applications — synthetic biology, gene drives, engineered microbes deployed at scale — EPA's frameworks are increasingly stretched. This is one of the cracks in the Coordinated Framework: it has neither the staff nor the statutory authority to regulate the kinds of ambitious environmental biotechnology that's now on the horizon.

---

Wait, Actually...

The EU's "process-based" regulatory framework for GMOs has produced an interesting paradox.

Because the EU regulates based on whether a crop was produced through genetic engineering (process), rather than what the final product actually is (product), it ends up treating new technologies very differently from old ones — even when the outcomes are essentially identical.

A crop developed through mutagenesis breeding (irradiating seeds with gamma rays or treating them with chemical mutagens to induce random mutations, then selecting useful variants) is not regulated as a GMO in the EU. Despite producing many more genetic changes than CRISPR — typically thousands of random mutations across the genome, most of unknown function — mutagenesis breeding has been used commercially for nearly 100 years and is grandfathered as "conventional."

A crop developed through CRISPR editing to make a single, precisely targeted change is regulated as a GMO in the EU, requiring years of review and labeling.

The same outcome — say, a herbicide-tolerant wheat — could face zero EU regulation if developed through mutagenesis or extensive EU regulation if developed through CRISPR. From a risk perspective, the CRISPR version is almost certainly safer (fewer off-target changes, more characterized). From a regulatory perspective, the CRISPR version is more burdensome.

This is the cost of process-based regulation. It can systematically penalize newer, more precise methods because they're new — not because they're riskier. The US's product-based approach avoids this paradox but creates others (gaps where novel products fall outside any agency's authority).

There's no neutral regulatory choice. Every framework has structural biases. Knowing them is the start of policy literacy.

---