Bernie Sanders and Alexandra Ocasio-Cortez have proposed capping credit card interest rates at 15%. It sounds appealing and benevolent, of course: wouldn’t people be better off if the interest rates on their credit card debt were reduced and literally nothing else changed? Of course they would be. What decent person would object to something that helps poor people? Clearly, anyone who would oppose a benevolent measure like this must have a heart of stone, a seared conscience, and perhaps a pocket lined with secret payments from sinister interests.

Decency, I would argue, is about a lot more than the intuitive fuzzies that come from imagining a world in which poor people are better off. Real decency requires us to think through the intended and unintended consequences of the policies we are proposing. In a competitive market for loanable funds, price ceilings—and since the interest rate is the price of borrowing a dollar, a cap on interest rates is a price ceiling—create shortages and induce people to waste resources competing for artificially-scarce loanable funds.

How does this happen? First, a price ceiling raises the quantity of loanable funds that borrowers demand. People want to borrow more at 15% than 25%. A cap would mean people seeking more loans. That looks like a good thing as it means more people seeking credit cards and, therefore, the opportunity to build a credit history. Think about arguments in favor of minimum wages that emphasize the increase in the number of people who might want jobs.

A price ceiling affects the supply side of the market, too, and an interest rate cap of 15% is going to reduce the number of dollars banks are willing to lend. The increase in quantity demanded combined with the reduction in quantity supplied means a shortage in the market for loanable funds. Holding everything else constant, the interest rate cap reduces people’s access to credit.

This is where a lot of the analysis stops, but it’s also where things get most interesting. It’s entirely possible that the benefits to people who are now paying lower rates are greater than the costs to people who are no longer able to get credit, and maybe there are good reasons to support that trade-off. We need to carry things a few steps further, though, before we arrive at a conclusion.

With a lower quantity of loans supplied, the value of the last loan made is now higher. The market-clearing interest rate might be 25%, rates might be capped at 15%, but with fewer loanable funds supplied people might be willing to pay 30% for these now-harder-to-come-by loans. How do they get the loans if they’re only allowed to pay 15%?

They can do a lot of things, just like they might do when competing for artificially-scarce apartments under rent control or artificially-scarce gas and other supplies when “price gouging” laws kick in after natural disasters. They over-invest in signals of their creditworthiness (relative to what they would have done without the interest rate control). They compete harder—and wastefully—for credit that is scarcer than it would have been without the control. Ultimately, the benefits of lower rates might evaporate.

And so here we are, yet again, discussing the unintended negative consequences of a policy that is ostensibly there to help the poor but that will actually make them worse off. No doubt, there are winners large and small from price ceilings: the politicians who propose them get elected and re-elected, and the voters who support them get to feel like they are helping people who are less fortunate. Reality, though, is something else entirely, and if we really cared about improving credit conditions for the poor, we would think twice about policies that make credit harder for them to obtain.