I think this is an excellent solution, but it is one that won't be replicable by cities like NY, Seattle, or San Francisco until they fix their housing cost problems. If you think about it from a systems perspective, with inflows and outflows, you will begin to see the problem.

1) Inflows: higher housing costs increase the inflow into homelessness. Whether it is job loss, drug addiction, mental illness, or whatever...if you have higher housing costs, you are more likely to end up homeless.

2) Outflows: higher housing costs become a higher bar to leaving homelessness. Let's say someone can address, over time, their underlying cause to homelessness. That person will take longer to overcome the bar of housing costs in a high cost area than an equivalent person with lower housing cost bars to overcome.

Both of these arguments seem obvious, almost tautological, but combined what you have is a higher rate of inflows, and a lower rate of outflows. This means two things: ceteris paribus, you have more homeless people, and the homeless people that you do have take longer to leave homelessness.

Now take into account the fact that in high housing cost areas, housing costs more to build because land prices and construction prices are higher (even taking into account Eminent Domain), and you have: 1) More demand (in terms of quantity of people) for relief housing 2) Higher expected time of use of housing. 3) Higher cost of building housing.

These factors are multiplicative, not additive. Combined, I wouldn't be surprised if the cost of attaining the same success rate is two orders of magnitude higher in San Francisco than it is in Salt Lake City. Construction costs alone take care of your first order of magnitude...now multiply those by the higher inflow rates and outflow rates.

This solution is obvious, but you can't put the cart before the horse. Salt Lake City never had to worry about it...San Francisco does.