Thanks for the thoughtful responses.

I believe that the narrative in mk's comment is incorrect, the idea that government welfare programs allow low-paying employers like Walmart and McDonald's to pay lower salaries, because their employees "don't need" as much income.

The idea wasn't immediately suspicious to me, but looking at it from the other point of view did make me wonder: "I was thinking about working for Walmart, but then I started receiving government welfare benefits; now I don't need the income from Walmart so much."

I believe that outside sources of wealth (be they welfare programs, lottery winnings, gifts, or something else) reduce a person's inclination to work, and employers must therefore "sweeten the deal" somehow to get potential employees on board.

If people who need income less worked for less, we would expect to see wealthier people accepting lower salaries. And very poor and hungry people who need income more would insist on high salaries. This is the reverse of what we actually see.

Bryan Caplan provides a perspective in How Welfare Hurts Walmart:

...Not convinced? Ask yourself: "If I ran Walmart, would I favor higher unemployment benefits?" Of course not. Why not? Because higher unemployment benefits make it easier to not apply for a job at Walmart. The same goes for any government program that makes idleness less unpalatable.

Elsewhere on the political spectrum, Arindrajit Dube (a supporter of minimum wage) argues in Public Assistance, Private Subsidies and Low Wage Jobs that "means tested public assistance programs are not tied to work, and we should not expect them to lower wages." He wonders whether such benefits actually raise wages, or merely cause people to work fewer hours, but both are consistent with reduced willingness of beneficiaries to work at a given salary, and "if they don’t lower wages, they can’t be thought of as subsidies to low wage employers."

We should remember that the decision to work or not does not depend solely on the salary; it depends on how attractive the next-best alternative is that the person will give up by going to work: the opportunity cost of employment. Thus we can't judge that $7.25 per hour is not "worth any adult's time" if we don't know what they will do otherwise.

"The most commonly cited reasons for using Mechanical Turk are to earn some extra spending money and productively use time that would otherwise be wasted watching television."