But it’s hard to know with any real certainty what the best long-term geographic investment for the company would be. The fact that tech companies tend to cluster doesn’t mean that a uniquely rich and powerful company couldn’t benefit from having a city of its own. The fact that bright young singletons gravitate toward coastal urbs right now doesn’t mean that you couldn’t attract talent — especially married-with-kids talent — to a heartland city whose Amazon District took advantage of sprawling housing stock left over from a prosperous past. Depending on which elements on Amazon’s wish list you weight most heavily, you can make a great variety of cities score impressively — from Bridgeport to Provo, Utah; Detroit to Rochester, N.Y.

Ultimately, as Lyman Stone, a prolific Department of Agriculture cotton economist, points out in a piece on Amazon’s decision, there is no city that comes close to meeting all of the company’s requirements: “No matter where Amazon goes, they will have to build their own fundamentals.” And no matter where the company goes, what it builds will change that city radically — so a static analysis of any destination will only take you so far.