Hidden Fees, Drip Pricing, and Shrinkflation
Chicago Booth’s Jean-Pierre Dubé explains how retailers use hidden fees to obfuscate prices and avoid transparency.
Hidden Fees, Drip Pricing, and ShrinkflationMany communities in the United States face concerns about housing affordability. And while regulators and policy makers may consider a number of remedies, Chicago Booth’s Anthony Lee Zhang suggests that the conversation around affordable housing may be missing a key element: the role of the housing stock itself. Zhang explains evidence from his research with University of Southern California’s Erica Xuewei Jiang that indicates banks prefer to lend against certain types of houses—typically newer and more standardized housing—which can have big impacts on how easily different homes can be financed.
I find housing markets really interesting because they’re sort of simultaneously a combination of a real asset and a financial asset. So like from a theory perspective, you can think of them as you can apply all the cool tools of theoretical finance—risk, return, liquidity, market thickness, all of the things like that. Simultaneously, it’s also . . . it has this kind of tangibility. Everyone deals with this. Every household in the US has to deal with: How do I find a house to live in? Do I buy a house? Do I rent a house? Stuff like that.
A lot of people look at housing affordability. It’s a really important question. Can households in the US and elsewhere sort of afford to buy houses, right? Affordability tends to focus on prices, right? How do we keep house prices down so people can afford them? An important angle to whether people can buy houses other than prices is how much can you borrow against your house? So most people when they’re buying a house are gonna pay maybe like 20 percent of that house in a down payment and borrow something like 80 percent through a mortgage, right? And the question we wanted to ask in this paper was basically: What determines that 80 percent? So when are banks willing to lend sort of more against a house or less against a house? Are there certain kinds of houses that banks just don’t like lending against?
And why that’s important is because, take a house which costs, say, $200,000, right? Take a seemingly small change in basically how much a bank is willing to lend against it. Like say the bank’s willing to lend 80 percent of its value versus 75 percent, right? Houses are expensive, and this turns out to be a really big difference. You can borrow like 80 percent of the house’s value; you’re paying $40,000 for the house. You can borrow 75 percent; you’re paying $50,000 for the house. You’re literally paying an extra, like, $10,000 for the house if the bank changes its LTV by like 5 percent.
So the question we wanted to ask is: Are there differences in financeability of different houses? The main thing we found in the data, the sort of big thing that jumps out at us in this research is that some people in the US live in areas with houses that are old and nonstandardized, right? So think, for example, you could live in a suburb of San Diego. This is like one of the most liquid housing markets in the US. Why? Because the houses are relatively new and they’re very similar to each other. They’re very easy to value, right? Other people in the US live in areas with houses that are old and really nonstandardized, like massive mansions from like sort of the early 1900s or really small houses that are sort of in various states of decay.
The new houses, we find, are much easier to value. When we run a regression to predict the prices of these houses, the errors from that regression are pretty small. A house is gonna trade pretty close to what our model estimate for the house’s value is. Whereas if you think about it, like, a mansion, which is like really old, you can like try and predict its value, but you really can’t predict it that well. We fit this regression model and we find that for these kinds of houses—nonstandardized old houses—the model does a lot worse in predicting its price because there’s fewer substitutes available. The markets are thinner.
If you have a decaying mansion with like seven bedrooms out there somewhere, you might get really lucky. If you’re sitting there, and a buyer comes along, and this turns out to be their dream home, right? You might get a pretty good price for that house. But if you don’t get that dream buyer and you have someone who buys that seven-bedroom because they have no other choice, you’re gonna get a pretty bad price because they really didn’t want that house.
And what we find is lenders seem to care about this. So lenders lend a lot less against these old and nonstandardized houses than these new, standardized houses. Borrowers are able to borrow a larger fraction of the house value in places like San Diego with these nice cookie-cutter, similar houses than in these like places with decaying and very nonstandardized housing stock, right? So the lenders prefer lending against stuff like my condo, like these houses in San Diego, because of the fact that in the event of foreclosure, the lender has a much easier time making sure that basically the buyers that come along, there’s gonna be some buyer coming along that’s pretty interested in this house. And I can predict the price of this house upon foreclosure pretty well.
It turns out there’s another effect of nonstandardization, which is a bit unique to how housing markets work. Governments heavily regulate how banks and other lenders can make mortgage loans. One aspect of that regulation is the government doesn’t totally trust that when the buyer and the seller agree to a transaction price that they got that price exactly right. And so what the government does to make sure they’re not lending . . . banks are not lending a massive amount against overvalued collateral is the government orders an appraisal of the house. They order a mortgage appraiser to go along and try and figure out an independent assessment of what that house is worth.
So there’s a kind of interesting effect here, where the appraiser values the house simply by looking at sort of a bunch of recent sales of houses that were similar to my house and taking the average of those prices. Our seven-bedroom mansion, if the appraiser is gonna try and come up with the closest recent sales comparable to that, it’s not easy to find in anywhere near recent history stuff which is similar to that. The appraiser does their best, but sometimes they end up getting comparable sales of houses which are really different. And sometimes they come up with—just purely because of statistical error—they’re gonna come up with an appraisal value which is like way below the trade price of the house. And if that happens, government regulation implies that the person trying to buy that seven-bedroom can’t borrow as much against the house.
So that’s another kind of interesting way that sort of government regulation makes it so that nonstandardized houses have a tougher time getting through the appraisal process. The appraisers do their best, but the constraints that they have to really use comparable sales means that price noise enters into the appraisal process in this way.
Part of the reason we should care about this result is it turns out the people who live in the places with the hardest-to-finance houses tend to be poor. We find that the poor live in these neighborhoods where houses have high price dispersion and are hard to value, right? Precisely the people in the US who have the highest need for credit in order to get into homeownership have the hardest time getting financing because of the fact that they live in these areas where the houses are just not good collateral for mortgages.
So what might policy makers we do to kind of address this fact that sort of structurally there’s a structural friction that the poor are living in places where the houses are hard to finance? If you think that basically it’s a big policy goal to improve homeownership for the poor, and if you think that the market is just not going to . . . the free market is just not going to give financing at good terms to the poor because the collateral is not good enough, that might justify interventions where you go in and then you make government-sponsored mortgages at sort of unusually higher than market loan to values for poor people because the market is just not gonna step in here. And there are examples of programs that do things similar to this, like the FHA loan program.
Now, this is kind of an interesting question to think about, especially from a Chicago School kind of perspective, right? The lenders here are looking at the collateral. The collateral is worse, and the lenders are lending less against that collateral. And in some sense sort of, this reflects the invisible hand of the market, right? I think there’s a kind of interesting tension here because the way I think about it is: if a policy maker cares enough and thinks the outcome of the markets is just like sufficiently bad for already disadvantaged people, there might be some sense in which it makes sense to introduce policies that push against this invisible hand of the market, provide credit at unfairly and unusually high rates for these poor people. And sort of this is a way to justify some of these interventions that governments make in mortgage markets.
Another way to fix this problem, of course, is make the collateral better. And I think policy makers have not realized necessarily that things they do, which influence qualities of the housing stock—rebuilding policy, zoning policy, things like that—also affect the financeability of the housing stock. We started this paper with a goal of answering the question of showing that there’s large differences in how financeable different houses are, and these differences are big enough to matter for homeownership rates for affordability. And then our results then imply that if the regulators can actually do something to drive markets, to build housing that’s not only cheaper but also more standardized, newer, easier to finance, with easier to predict prices, that potentially matters a lot for homeownership. So our results basically point to if it’s possible for governments to regulate housing markets in a way which encourages building houses which are cheap and affordable for houses but also standardized, lenders are gonna be more willing to go in and provide financing against these houses. Low-income households in these areas are gonna have an easier time affording these houses because of the fact that the down payments are lower because they can borrow more against these houses, right?
So this is a very complicated debate, but one factor we really want to add is in thinking basically about affordability, governments should really think about financing. There’s things governments can do to encourage building houses that are not only cheap but also easy to lend against. And financing, our results show, is a really important part of the barriers to homeownership and something that governments perhaps haven’t been paying enough attention to until now.
Chicago Booth’s Jean-Pierre Dubé explains how retailers use hidden fees to obfuscate prices and avoid transparency.
Hidden Fees, Drip Pricing, and ShrinkflationNYU’s Marion Nestle joins the Capitalisn’t podcast to examine the role of Big Food in public health.
Capitalisn’t: The Money behind Ultraprocessed FoodsBethany McLean describes the factors that exacerbated the pandemic’s fallout for the most vulnerable in society.
Capitalisn’t: The Capitalisn’t of the US COVID ResponseYour Privacy
We want to demonstrate our commitment to your privacy. Please review Chicago Booth's privacy notice, which provides information explaining how and why we collect particular information when you visit our website.