Evan Soltas
Feb 26, 2014

But Somebody Pays That

Mike Konczal has asked me to revisit the underlying question in the piece Mankiw's textbook republished. (You can see a post he wrote in response to my old one here.)

The piece argues that universities are raising sticker prices far more than they are sticker prices as a tool of price discrimination -- and, in particular, that the discrepancy is used so that high-income families cross-subsidize low-income families.

Here's how it works, intuitively. Say the sticker-price tuition at a school is $40,000 a year. But very few people actually pay $40,000 a year. Most students get large amounts of financial aid that reduces the net cost of college to, say, $15,000. But not everybody. The high sticker price allows for a high top price, and then the school's financial-aid office can calibrate prices below that according to their aid formula. By raising the sticker price, but not the net price, schools create more of a range over which this price discrimination can occur.

What would we see in the data if this was actually going on? We'd see the difference between sticker price and net price grow for rich families more than poor families. And that is what we see. The data comes from the College Board's "Trends in College Pricing" report.

I've zeroed out the 1999-2000 academic year because poor students have always received some discount on their education relative to rich students -- what we care about is has that cross-subsidy grown.

What the numbers show is that you now get a $2000 greater discount at public four-year universities if you're in the bottom quartile than if you're in the top quartile versus 1999-2000. The discount is $4500 greater at private four-year universities now than in 1999-2000.

There are a few other ways you can look at the data. Konczal and Matt Bruenig both seem to focus on the net cost of college, inclusive of room and board. These are important numbers, but they're removed from what I'm looking at. Since we don't have the full sticker price, just the tuition sticker price, and what I'm curious about is the pricing strategy, that's why I focused on the difference (in this post as in the original).

So, yes, as I see it, the pricing strategy is redistributive: Raise the sticker price and give bigger discounts to the poor.

I'm not saying that there's no public-policy influence here. I'd agree with Konczal that the rise in federal money is helping to hold down net tuitions for the poor. The price-discrimination part comes in when sticker tuitions rise across the board, and the grants offsets the increase only for the poor.

And I should have been a bit more careful on a line Konczal quotes. The precise wording of my point about subsidization isn't that the rich are paying above and beyond the cost of their education. It's that the increases in net cost have mostly been borne by people in the higher reaches of income.

When we look at just changes in net cost, what we see is that net costs have grown for the rich, not for the poor, at private four-year universities. However, the increases have been broad-based at public universities -- I imagine that's so because of cutbacks at the state level.

I don't think I'm missing something -- if I am, I'd love to know. Leave me a comment. From my view in college, the price-discrimination story seems pretty obvious.