Reihan rightly worries about New York mayor Bill de Blasio’s plan to ease the strict work requirements Michael Bloomberg had placed on the city’s food stamp program:
Is this a badly needed correction from the bad old days of Bloomberg? It is important to understand that, for better or for worse, the Bloomberg administration was very accommodating when working poor applicants sought to enroll in the food stamp and Medicaid programs. A big part of the reason was simply that the city government didn’t set the eligibility rules for these programs, and the federal and state governments had grown more permissive over time. But it also reflected a public philosophy that the billionaire mayor was never very good at articulating—that those who can work and choose not to do so are different from those who do not.
This is a distinction that advocates of an unconditional basic income see as pernicious and that those who want to ease up on work requirements see as needlessly punitive. But it is a distinction that makes eminent public policy sense.
Dylan Matthews tries to push back by defending the very “just give poor people money” approach that Reihan decries in his article, citing evidence from the field of international development:
A recent randomized trial found that Kenyans who received no-strings attached cash from the charity GiveDirectly built more assets, bought more goods, were less hungry, and were all-around happier than those who didn’t get cash. But voters and politicians generally prefer giving people specific goods — like housing, food, or health care — rather than plain old cash, for fear that the cash might get misused by unscrupulous poor people. Maybe the recipients will just blow the cash drinking! …
Now, the World Bank’s David Evans and Anna Popova are out with a new paper reviewing what evidence is out there about aid to the global poor and alcohol/tobacco consumption.* They found 19 studies which attempted to measure the effect of cash transfers — both no strings attached ones and ones families receive if they fulfill certain conditions, like school attendance — on the purchase and consumption of “temptation goods”; the studies contained a total of 44 estimates of cash’s effect in various contexts. 82 percent of those estimates showed that the transfers reduced consumption of or spending on alcohol and tobacco. The vast majority of those weren’t statistically significant, so the best conclusion is that there’s no evidence transfers affect drinking or smoking behavior.
