An oversight in your observation is failing to account for the fact that equilibriation processes are slower and less dramatic when there are smaller differences between two things. E.g. dropping ice onto a hot pan is quite dramatic. Dropping 60 degree water into 70 degree water not so much.

> Are you suggesting that there is a cutoff below which differences in income don't affect where people decide to live?

Of course there is, for plenty of reasons:

- Moving away from friends and family is not free, is risky financially (what if your new job does not pan out), and is often simply not something people want

- Population-wide differences do not guarantee that you personally will gain anything by moving (correlation vs causation)

- Small differences in average incomes are imperceptible to most people, and the quality of average person's financial analysis is lacking. For example, looking at just median income ignores differences in cost of living, quality of government services, lifestyle, and anything occupation-specific.

- There is often more variation in median income within a state (e.g. urban vs rural areas) than across states. Migration from rural to urban areas is a thing.

- etc.