Question: One common argument against public assistance taking the form of direct cash handouts is that the recipients will use the money to buy things that taxpayers find objectionable, e.g., illicit drugs, gambling, etc. To avoid this outcome, the argument goes, public assistance should take the form of in-kind transfers, e.g., food, housing, medical care, etc. What does this argument assume about the income elasticities of objectionable goods? Suppose the recipients could costlessly resell the in-kind transfers. In this case, is there any difference between direct cash handouts and in-kind transfers?
Solution: A common argument against giving people cash instead of in-kind assistance—like food, housing, or medical care—is that cash might be spent on things taxpayers find objectionable: illicit drugs, gambling, or other “vices.” The idea is that if we hand out groceries or rent vouchers instead of money, we can prevent recipients from using the aid to fund consumption that we believe to be harmful or immoral.
But this argument rests on an assumption that doesn’t hold up under closer scrutiny.
At its core, the argument assumes that the demand for objectionable goods rises with income—that is, that these goods have a positive income elasticity. If you give someone more money, they’re more likely to spend more on drugs or gambling. That may very well be true.
The problem is that the argument simultaneously assumes something quite different about in-kind transfers: that giving people food, housing, or medical care will not lead to more consumption of objectionable goods. This is only possible if objectionable goods somehow become immune to income changes when the income comes in the form of in-kind support.
Even if someone can’t directly sell the food or housing they receive, getting those goods for free frees up money they would have spent on them. That extra money can then be used for anything—including objectionable goods. Unless we believe people will consume more of the in-kind good and nothing else with the money they save, we should expect some of that income to be reallocated toward whatever they value at the margin.
In other words, the logic of the in-kind transfer argument contradicts itself. It claims that cash causes bad behavior because income matters—but that in-kind transfers don’t because income suddenly doesn’t matter.
Now, let’s suppose recipients can resell the in-kind goods. In that case, the transfer becomes equivalent to cash in every meaningful way. They can turn the food or housing voucher into money and spend it however they like. Economically speaking, resale makes the in-kind transfer function exactly like a cash transfer.
But even if resale isn’t possible, the basic conclusion still holds. The key idea is fungibility: money is interchangeable, and so is the value of money saved. If a recipient was already buying food before the government gave them food, then the food transfer simply frees up their existing money to spend elsewhere.
Whether objectionable goods consumption rises as a result depends on one thing: whether those goods are normal goods, i.e., goods that people consume more of as their effective income rises. If they are—and the argument assumes they are when criticizing cash handouts—then any transfer that increases effective income, whether in-kind or in cash, will have the same effect.
