It turns out that as Americans collectively decided to buy more bulk toilet paper and rotisserie chickens, they also, with remarkable consistency, decided to drive while impaired at nearly identical rates—a relationship so tight it suggests either Costco membership somehow impairs judgment, or the universe is playing a joke on statisticians that would make a reasonable person question whether data means anything at all. One wonders if somewhere in a warehouse in Nevada, both trends are simply being pulled from the same cosmic vending machine.
The answer, almost certainly, is that both metrics are riding the same wave of economic expansion and population growth across the 2010s, before the pandemic briefly scrambled everything. As the U.S. economy improved and people had more disposable income, they bought more at Costco—and also, distressingly, drove more, which meant more opportunities for bad decisions behind the wheel. Think of it this way: Costco's revenue more than doubled from roughly $90 billion in 2010 to over $220 billion by 2022, while drunk driving deaths oscillated between roughly 9,000 and 13,000 annually. Both numbers are tracking the invisible current of American mobility and consumption; they're not causing each other so much as floating downstream together.
This is what happens when you have thirteen data points and a pattern-recognition instinct older than language itself—you find correlations that are technically real but fundamentally meaningless, which is either deeply reassuring or deeply troubling depending on your mood. The real story isn't that Costco somehow enables impaired driving, but that two entirely separate human behaviors can move in tandem for years without anyone being the wiser. Which should make us all slightly more humble about what our data is actually telling us.
As an Amazon Associate, getspurious.com earns from qualifying purchases. Learn more.
Want to learn more about why correlations like “Costco annual revenue” vs “Alcohol-impaired driving fatalities” don't prove causation? Read our guide to statistical thinking.