National park permits that sell out in minutes skew toward richer ZIP codes

In the Nature Communications paper, the two groups in the main comparison booked through the same reservation systems. What separated them was how far ahead they got their permits ^[1][3]. That keeps the destination fixed, so the income gap cannot be put down to richer households preferring some parks over others. The skew shows up in the most competitive releases, where most tickets are gone within five minutes of going on sale ^[4].

The effect is modest, and it is measured indirectly. A ZIP code whose median household income is 3% to 6% higher ^[4] is a somewhat richer neighborhood. The sample behind the figure is large: more than 5 million reservations, or more than 450,000 per system on average ^[2][15]. But the income number describes where bookers live, so the study cannot say how much richer the individual winners are than the people who missed out ^[3].

The thing this doesn't tell you is why the early bookers win. Hughes said it could be because higher-income visitors have more flexibility to plan trips months ahead, know the reservation systems better and tend to be more tech-savvy ^[5]. He offered those as possible explanations for a pattern in ZIP codes and booking times. His own routine fits them. He recalls coming to campus early for the faster Wi-Fi and booking on two computers ^[7]. "It made me think about who has the ability to participate in such a process. Not everyone can plan months ahead, is comfortable with technology or has access to a fast internet connection," he said ^[6].

I think the case for lotteries at parks rests mostly on Hughes's earlier work. In a study of Western river permits, handed out through a mix of lotteries and advance reservations, he found that people who got permits by advance booking tended to have higher incomes than lottery winners ^[8]. That compares allocation rules head to head. The reporting on the park study describes early and late bookers within reservation systems and does not mention a lottery comparison ^[3]. Hughes said parks could experiment with how many permits are released months ahead versus close to the visit date, and that in some cases random lotteries may provide a more equitable alternative ^[11].

Rocky Mountain closes its timed-entry season on Oct. 18 ^[10]. Hughes put the stakes in the park system's founding language. "But if you look at the founding document of the Yellowstone National Park, our first national park, it is to be 'set apart as a public park ... for the benefit and enjoyment of the people,' not just rich people," he said ^[12]. "Income is just one indicator," he said, adding that education, age and race could also play a role ^[13].