We know that only a large minority of the homeless abuse drugs routinely and that a similarly large minority are homeless due to mental illness. One would expec

We know that only a large minority of the homeless abuse drugs routinely and that a similarly large minority are homeless due to mental illness. One would expect the two groups to overlap, so it seems reasonable to say that a large proportion of homeless people are neither mentally ill nor drug addicted, or at least that a large proportion aren’t homeless for this reason. Given that this is the case, you would expect a rise in the price of housing to correlate with a rise in homelessness. Since we don’t observe that, I’m guessing that there’s a variable we’re missing. Another potential explanation could be that people tend to move out of the high cost area before going homeless, so they don’t contribute to the homelessness statistics in the high cost area.