[giantg2]

"but only to be high enough relative to wages elsewhere to compensate for the cost of commuting."

True, but then the implied issues with not living in the city you work in become moot if truly compensating for the commute and cost of living outside the city.

[abeppu]

I think maybe you mean something by "truly compensating" which isn't actually being achieved. But the lesser compensation we are achieving is still sufficient to impact people's behavior.

Suppose you're able to get two part time jobs with relatively little predictability in your shift schedule. Both are minimum wage, but one is in SF where the minimum is $17/hr and another is in Concord at $14/hr (the state minimum for a company with <25 employees). Say you live in Antioch, and can get to the Concord job in around 35 minutes for $4 with BART, and get to the SF job with more than 1h and $7.30 on BART (one way). So for every shift in SF you pick up, you get a $3 premium per hour relative to the Concord job, but for the first 2 hours that just goes to cover your extra BART fare, _and_ you lose an extra hour in transit. In dollars you still come out ahead on an 8 hour shift in SF relative to in Concord, but I think it would be unreasonable to say that "the implied issues ... become moot".

[giantg2]

I say theoretically, because in a theoretical world the market would be efficient and the employee would be analyzing the income and costs. Your examples are using locale in a HCOL state. Now image if the workers there actually did analysis on moving to lower cost states.

Yes, the moot part is based on them being truly compensatedfor the extra commute, in which case it would be moot. It's much more interesting to me to look at the income inequality and market side than the "solution" of creating subsidized housing. I'd rather go after the true root problem than cover up one symptom and allowa the problem to persist.

You're looking at it as it exists today. I'm saying theoretically this other stuff should work (but may not in practice).

[abeppu]

I think this line of thinking is just sloppy. "Theoretically" the efficient market only gets to pareto optimality, i.e. we can't make the commuting low-wager worker's situation better without someone else being at least slightly worse off, but there's no lower bound on how bad things can be for that low-wage worker. Nothing assures us that we converge to a state where at the low end "the wages should come up to make it possible to live [in SF]" or anywhere particularly close.

[giantg2]

I find your thinking sloppy. There, was that helpful?

"better without someone else being at least slightly worse off"

Exactly. The point (if you go back through my comments) is that there seems to be a demand and supply imbalance based on the skill/type of labor. We should be looking at ways to slow higher end wage growth and promote it in the lower end. One way to do this could be around education. Essentially, increase the number of developers on the market while decreasing the number of people who are only qualified for the lower wage work. We've kind of seen this with the pandemic shrinking the labor force and driving those lower end wages higher.