A few months ago, the liberal blog Think Progress released an investigation into the costs of drug testing for welfare applicants, a popular conservative policy. Conservatives argue that this will root out spongers and junkies who mooch off the state. Think Progress’s report showed that in the states which do give drug tests to welfare applicants, the costs of administering the tests are quite high and the number of people who produce positive drug tests is quite low.
Liberals do not like the idea of drug testing welfare applicants. But arguments like Think Progress’s strike me as a very bad way of opposing the policy, if one actually seeks to disprove conservative reasoning. First, there’s the obvious point that any drug test supporter would instantly make: in order to evaluate the success of the tests in achieving their goals, you need to look at those who were deterred from applying and being tested (because they knew their test would be positive) not just those who applied and tested positive. I am sure Think Progress knows this, since it’s completely elementary: if drug tests are in place, surely a vast number of the people who would test positive aren’t going to go through with application and drug test if they know it’s a waste of time. But it looks more like a waste of money if you say “$300,000 was spent on drug tests with only 10 positive results” instead of factoring in cost savings from deterred applicants. Conservatives who support the policy would instantly point this out.
But evasive statistics by partisan think-tanks are no big news. The more important point is this: focusing on cost does not get you to a persuasive argument against drug-testing welfare applicants. Because a conservative could agree that the costs were not reaping sufficient benefits. All they’d have to say in response is “Well, if you’re right, and we’re wasting money, we should have applicants pay for their own tests instead of subsidizing them.” Problem solved, costs saved!
Cost-saving arguments are like this a lot of the time. Liberals make them because they know they have bipartisan appeal (nobody admits they like to waste government money), but they don’t actually create a case for liberal policy preferences. And in fact, if you rely solely on a cost-saving argument, you’re taking a huge risk. Because what if it turned out the numbers weren’t in your favor? What if it turned out drug tests did root out a lot of drug users? Liberals would still oppose drug-tests for welfare applicants, but their wastefulness argument would be devastated!
I feel the same way about arguments for prison reform that are cost-based. Often, advocates for different prison policies (like myself) slip into arguments about how our prisons waste money, how (for example) good post-prison re-entry and training programs will save states money by reducing future expenditures in re-housing recidivists. But if your main argument is a cost-saving one, you need to be prepared: if it turned out that prison reform cost states more money, what would you say? And if we really are focused on cost (rather than alleviating unfair human suffering), then what if someone proposed that you could save even more money by just slashing the prison healthcare budget?
It’s all a bit dishonest, for liberals to pretend they’re just about cost-saving, because they’re only about cost-saving if that argument can be mobilized in favor of their already-preferred program. If their program requires more money, then they’re no longer for cost-saving. That’s why ultimately, those (like me) who believe drug testing for welfare applicants is wrong and must be opposed, should state exactly what our reasons are. My reasons are that I think it’s an assault on human dignity, and that someone should have the right to both take drugs and receive welfare. It’s not that it’s too costly; I don’t really care if it is or isn’t. It’s that it serves no useful objective at any price. I think most Think Progress types ultimately feel the same way, and unless they’re up-front about it, when someone exposes their statistical legerdemain, they’ll be left without defensible arguments, despite being completely right.