Managing Marijuana: The Role of Data-Driven Regulation

By Stephen Goldsmith • August 17, 2016

This post originally appeared on Stephen Goldsmith's Better, Faster, Cheaper blog on Governing.com.

When Colorado voters approved a ballot measure to legalize the sale of recreational marijuana in 2014, state officials knew they would have to quickly develop a robust system to safely and securely control the flow of the drug across the state, and they managed to do just that with the help of advanced tracking and data analytics. What Colorado is doing provides an impressive example of an emerging, more effective regulation model.

To deal with the consequences of marijuana's legalization, Gov. John Hickenlooper appointed Andrew Freedman as director of marijuana coordination, and the legislature created the Marijuana Enforcement Division (MED) in the state's Department of Revenue. MED drew on the state's experience with its preexisting medical marijuana regulations, examining what had worked previously to regulate and inventory controlled substances.

The widespread legal availability of marijuana and marijuana-infused products, however, presented a host of new concerns for Colorado, with those surrounding public health and safety first and foremost. The state established a comprehensive set of goals surrounding health and safety concerns, analyzed the gaps in its existing data, and devised a plan to better track and regulate marijuana.

To begin with, the state needed a way to track plants from seed to sale. MED contracted with Franwell, a Florida-based company, to create a Colorado-specific version of its Marijuana Enforcement Tracking Reporting Compliance (Metrc) system. The state requires growers to track each plant with a unique radio frequency identification tag. The RFID tags allow the plants to be inventoried more quickly without direct contact, and they create data at each step of the supply chain as plants move from growers to shippers to final packaging.

The RFID tags, like the sensors that are becoming ubiquitous in the emerging economy of connected devices, build regulatory intelligence directly into the regulated object, streamlining enforcement across the board. The tracking system, for example, can automatically flag facilities that are producing substantially less marijuana than expected based on the outputs of comparable growers, which allows state employees to more easily identify potential illegal diversion. And because the state is able to track the origins of any product, it can easily issue public recalls for specific batches or growers if regulators discover traces of potentially harmful pesticides.

This system of constant, real-time tracking allowed Colorado to shift away from an older regulatory model in which governments must depend on slow bureaucratic procedures for permitting and licensing. Rather than simply hoping that procedural factors would prevent noncompliance, the state can respond to problems as they arise, which increases accountability and allows for better-informed enforcement.

The Colorado Department of Public Health and Environment, for example, analyzed hospital-visit data and found that marijuana-related visits had tripled after commercialization and that poison control calls had doubled. Around the same time, MED noticed that marijuana-infused edibles were comprising significantly more of the market than expected -- fully 50 percent, according to Freeman. As MED discovered, the two were correlated: Many of the hospital visits and poison-control calls stemmed from a lack of dosing information or packaging safeguards.

MED quickly intervened, adopting new requirements that included childproofing edibles and clearly marking doses on the edibles' packaging. These changes have helped lower the instances of unintentional exposure to edible marijuana products. Colorado has hired a full-time analyst dedicated to deriving similar insights from these new flows of marijuana data.

Whatever one might think of the advisability of legalizing marijuana, there's much to be learned from Colorado's experience in data-driven regulation. While many of the practices Colorado is pioneering can be adopted by other states that may be considering legalizing marijuana, they also could help to broadly improve regulatory systems across all levels of government. Colorado is demonstrating how data-driven regulatory models can enable governments to quickly understand the realities of a market and hold businesses accountable -- ultimately resulting in stronger, more effective consumer protections.

About the Author

Stephen Goldsmith 

Stephen Goldsmith is the Derek Bok Professor of the Practice of Urban Policy at the Harvard Kennedy School and the director of Data-Smart City Solutions at the Bloomberg Center for Cities at Harvard University. He previously served as the mayor of Indianapolis and deputy major of New York City.

Read Professor Goldsmith's full bio here.