It's also relevant that SF is only 47 square miles, compared to the metro area's 3.5k square miles. In many of our cities, people commute to work in different c

It's also relevant that SF is only 47 square miles, compared to the metro area's 3.5k square miles. In many of our cities, people commute to work in different cities (hence the horrible traffic). You can end up with different salary:rent ratios comparing jobs in a city to rent in a city than you'd get looking at salary:rent ratios of jobs of people living in a city to rent in a city. Both are interesting statistics in their own right. I'd bet the difference in those two statistics indicates uneven housing supply (residential areas aren't the same as business areas) and small cities (residential areas are counted differently than business areas).