Redfin has one of the best corporate blogs not just in real estate, but in…well, business. Posts like this are the reason why. Glenn Kelman takes the Freakonomics guys to task (gently, as is his wont) for relying on bad sample set to conclude that brokers have little to no value.
Stephen Dubner and Steven Levitt renew their argument that real estate brokers aren’t worth 6%, citing a study (PDF) conducted by Stanford economist B. Douglas Bernheim and one of his graduate students, Jonathan Meer, which shows that using a broker has no effect on a home’s average selling price.
We are an (online) broker ourselves, but have argued that consumers should be able to choose the real estate services for which they pay, so I’m not sure we have a dog in this fight. In the past, we have welcomed studies showing that buyers and sellers can get along without a broker, and argued that a client working with an online broker negotiates a better price. But in this case I was surprised that the Freakonomics team didn’t evaluate the Stanford economists’ methodology.
The Stanford study only evaluated 800 homes sold on the Stanford campus, “the ownership of which is limited to Stanford faculty and a limited number of senior staff.” In such an environment, marketing is much easier because of the small number of potential buyers, trust is high because of the buyers’ affiliations with one another, and supply is extremely limited: many academics would kill, or even teach an extra freshman survey course, to live on the Stanford campus.
It’s worth reading the whole thing for Glenn’s take on it. He is, after all, the CEO of a fairly unique brokerage proposition.
As usual, Glenn is much nicer than I am. 🙂 One has to be willfully blind to the flaws of the Stanford study to make any conclusions about the value of real estate brokers based on that sample set. That’s like drawing conclusions about commercial real estate based on a study of Class-A buildings in midtown Manhattan — the city that breaks the laws of physics. (Did you know that Manhattan office buildings routinely have more rentable square footage than physical square footage?) Neither place has much resemblance to other places, like say… planet Earth.
One of the biggest flaws of the Stanford study is the severe restriction not just on supply, but on demand. These 800 houses are limited to Stanford faculty and limited staff (presumably, university brahmins). Maybe I’m a wealthy VC in the area and would like to own a house right on Stanford campus — sorry, no can do. And we’re going to make conclusions about pricing based on that? It’s a neat trick to make statements about the free market once you’ve eliminated the free market from the analysis, no?
Glenn thinks college towns may be a signpost to the future:
In real estate and in life, college is a smaller, more perfect vision of how the rest of the world could be. We thought it was interesting that the previous academic study on brokers’ effectiveness focused on Madison, Wisconsin, because this is also a small college community where alternative approaches to real estate have reached critical mass. Maybe these communities point the way to a post-brokerage world waiting for all us, where both sides abandon their brokers, where we can access information for ourselves online, where we can come to terms more easily and economically.
I seriously, seriously doubt it. For one thing, college housing may be immune to cyclical downturns. For another, I’d imagine only military bases have higher turnover in population than college towns, even amongst the faculty.
This is not to say that real estate brokerage will survive in its current form. There are enough broken things about how real estate works today that some change is inevitable. It remains to be seen how things will evolve, and what the value of brokerage services will be in the future.
It may not be, as the Freakonomics team tells us, based on the idea that using a broker can get you higher prices on your home. On the other hand, it may be that using a broker makes it easier to get you higher prices. Huh?
I mean that presumably there’s a set of activities that one has to do in order to maximize price on a house. Listing and marketing are a part of it, but so is something like staging a house. The homeowner himself can stage the house, but maybe he’s a Professor of Economics and knows as much about proper staging as a real estate agent knows about the Black-Scholes theorem. If there is some finite set of activities, then someone has to do them — otherwise, the price of the house will not be maximized. A buyer will offer less simply because the homeowner forgot to clean his bathroom.
Real Estate agents may end up becoming something like a specialist in house staging, and extract value for that service.
Or it may be (as the Stanford authors admit) that using an agent results in faster sales, although not necessarily higher price. There’s value in that.
Heck, there’s value in not having to be around to show my damn house to some damn couple. It’s a convenience thing.
Now, none of those things may be worth 6%. That’s a different kettle of fish.
For what it’s worth, I believe the future holds variable pricing for real estate brokerage services. Rather than trying to enforce a one-price-fits-all model on consumers who don’t want all those services, brokers may end up offering a buffet-style set of choices. Want me to show your house? That’ll be 1%. Want flyers made up? That’ll be 0.5%. Want me to make a website for your house? That’ll be $500. That provides flexibility and choice to both consumers and professionals, allowing both to feel that they’re getting what was bargained for.
But even that model may not work on the Stanford campus with its restriction on supply and demand.