In 2014, Washington State launched one of the nation’s first legal recreational marijuana markets, helping to usher in a wave of legalization that has since spread to 23 states and the District of Columbia. As Washington prepared to launch its pioneering market, it found that far more people wanted to sell cannabis than the state was willing to allow. In jurisdictions where applications outnumbered the available licenses, the state didn’t try to pick candidates based on merit, business plans, or capital. It simply held a lottery. Names were drawn, licenses were awarded to the winners, and the rest were turned away.
In solving an administrative headache, the regulators unwittingly designed a near-perfect natural experiment. The lottery left some of the new pot shops with more nearby competitors than others, providing an opportunity to understand how local competition—isolated from other economic factors—affects prices and product selection.
Economic theory says that businesses facing more local competition will try to get an edge through pricing, advertising, and what they choose to put on the shelf. But the stew of factors that shape their strategies makes it difficult to use real-world data to confirm that idea. A new working paper using data from the Washington cannabis shops by Yale SOM’s Kosuke Uetake, Brett Hollenbeck of UCLA and Sylvia Hristakeva of Cornell provides rare causal evidence that firms do lower prices and strategically differentiate what they sell in order to blunt competitive pressure.
“It’s not just theory,” Uetake says. “Firms are actually choosing these things—prices and assortment—as strategic tools.”
Let’s say that you compare prices at five stores in a busy commercial center and a single shop in a quiet strip mall, and find they are higher at the isolated store. That doesn’t prove that local competition is causing the price difference: The busy shopping hub could have attracted five stores because it’s a better location—more customers with more disposable income—and that demand could just as easily explain the difference.
“There is a reason why the firms chose those locations,” Uetake explains. “They expect higher demand; they expect people who want more variety.”
Washington’s lottery accidentally solved that problem. Applicants locked in their proposed store addresses before the drawing, so two businesses in equally promising neighborhoods could end up with different numbers of nearby rivals simply because of which names were drawn. That randomness let the researchers isolate competition’s effect from everything else bundled with a location’s appeal. And because Washington monitors and taxes every legal marijuana sale, Uetake and his colleagues were able to track nearly 400 retailers over time, mapping exact markups and how pricing and product mix shifted as competition changed.
They found that having a nearby competitor does have a measurable effect on prices, which fades quickly with distance. Retailers whose nearest rival was within 0.2 kilometers charged markups about 9% below retailers whose nearest competitor was more than 2.5 kilometers away. Push a competitor out past roughly a kilometer, and the price effect essentially disappears.
“I wouldn’t expect people to travel 100 miles to buy marijuana,” Uetake says. “People typically go to the local store. Competition here isn’t citywide; it’s block by block.”
Breaking competition down by headcount rather than distance revealed a similarly steep drop-off. Having one or two competitors accounted for nearly the entire decline in markups. A third, fourth, or fifth competitor barely moved prices further, a pattern that echoes classic theories in industrial organization.
When firms face more competition, they don’t just compete on price. They also look for ways to differentiate themselves. Our study shows that retailers do this by adjusting which products they offer, helping them soften price competition.
Retailers did not only cut prices when facing a nearby rival; they also expanded and reshaped their selection, carrying more products with less overlap with what that specific competitor sold. The closer two competitors were, the less alike their shelves looked. Differentiation mattered for pricing, too: when two nearby retailers carried the exact same product, that product’s markup came under significantly more pressure than a competitor’s mere presence would predict.
To understand the policy implications of the patterns they observed, Uetake and his colleagues simulated a simple change. What would happen if Washington allowed 10% more cannabis retail licenses statewide? Markups would fall by 1.4%, they found, translating into roughly $7.6 million in annual consumer savings, with the effect nearly doubling in Seattle, the state’s largest and densest market. In addition, existing stores’ assortments would grow by about 8%, meaning consumers would gain not just lower prices but more choice.
The findings could also have implications for other regulated markets. States use license caps to regulate liquor stores, pharmacies, and tobacco retailers, often with little hard evidence about what those caps cost consumers. Washington’s cannabis data is uniquely granular—tracking every transaction from grower to retailer—precisely because the state built its system from scratch as it launched a brand-new market. Most regulated industries are not tracked nearly as closely, Uetake notes, leaving regulators to guess at the tradeoffs of restricting competition—guesses the research suggests can carry real costs to consumers.
“They should have a decent monitoring and tracking system, like the one that the state of Washington implemented, Uetake says. “If they don’t have the data, they can’t estimate how consumers respond, or how firms respond. That data gives them insight into whether caps should be increased or decreased. Moreover, if the state cannot monitor and track the market, then there is more risk of a black market—which is exactly what happened in some states that have done a poor job of monitoring.”
For businesses, the lesson is that assortment is one of the levers they can use to ease competition; this study shows it actually happening in a real market, with retailers adjusting what they sell, not just what they charge, to reduce pressure from a nearby rival.
“When firms face more competition, they don’t just compete on price. They also look for ways to differentiate themselves. Our study shows that retailers do this by adjusting which products they offer, helping them soften price competition,” Uetake says. “Firms need to understand not only how consumers behave but how competitors do.”
“The Yale School of Management is the graduate business school of Yale University, a private research university in New Haven, Connecticut.”
Please visit the firm link to site
