Wednesday, April 06, 2005
And they are about to be told they can't trust their real estate agent.
Steven Levitt has been lauded by the American Economic Association as the most promising US economist under the age of 40. Ten of the economists it has previously so awarded have gone on to win the Nobel Prize. But Levitt seems to lack the necessary seriousness. Among his research topics are the voting behaviour of contestants on the game show The Weakest Link, the inner workings of the Ku Klux Klan, and the things people lie about at online dating sites.
His unifying theme is that people with access to information tend to abuse it. His hobby is catching them. Which brings us to real estate agents.
Levitt says he first became involved with agents when he went looking to buy a house. He felt they were subtly encouraging him to bid low. Which should have made no sense. Real estate agents are paid by commission. But then he took a closer look at the structure of those commissions.
Agents usually get a fixed proportion of the price the property sells for. (In Australia it is often 2 per cent.) That means that while the reward for actually getting a property sold is big, the extra reward for extracting a higher price can be tiny. An Australian agent who held out for an extra $50,000 for example might find herself earning only an extra $1000. But the cost to the agent of holding out might well be large: opening the house, accompanying visitors and negotiating contracts for weeks upon weeks until a higher price is reached. Holding out could destroy an agent's business.
But to the homeowner it's quite different. The owner usually wants the extra $50,000 and is prepared to wait. It is in the agent's financial interest to persuade the owner to sell more quickly than the owner should. And this is where what Levitt calls "information asymmetry" comes into play. The agent is an expert in the market. The owner is not. And they both know it. If the agent insists that a quick and cheap offer is actually the best offer the owner will feel obliged to agree.
Levitt figured that if this was happening, agents would act quite differently when it came to selling their own homes... They would hold out for longer and get higher prices. He examined the records of 100,000 sales in Cook County, Illinois, over a 10-year period and discovered that the agents did indeed get higher prices when they sold their own homes - about 3.7 per cent higher than the similarly featured and located homes they sold on behalf of their clients.
Of course, that could have been because they presented their homes better and tried harder to ensnare buyers. But if that was the case, Levitt reckoned that the agents' homes would have sold more quickly than those of their clients. In fact they sold a lot more slowly, staying on the market for 10 days, or 10 per cent longer. Agents appear to encourage owners to sell more quickly than they would do themselves.
And they appear to use a different language to describe those owners' homes. Certain words turn out to be associated with high prices when used in real estate ads. They include "granite", "gourmet" and "state-of-the-art". Each conveys specific useful information. Other words are associated with a lower price. They include "fantastic", "well maintained" and "charming". Levitt found that the agents tended to use the first group of words to describe their own properties, the second to describe their clients' homes.
He believes the words in the second group function as a sort of code. The phrase "well maintained" might signal to a buyer that a house is old but not quite falling down. A low offer might be accepted. "But to the 65-year-old retiree who is selling his house, 'well maintained' might sound like a compliment, which is just what the agent intends."
Levitt says it's difficult to function well as a real estate agent. You need to convince owners that you are acting in their interests while at the same time persuading them to settle for less than the best price and letting potential buyers know that a house can be bought for less than the best price.
It isn't that Levitt especially has it in for real estate agents. He believes that they are about as honourable as doctors, lawyers, car mechanics and teachers, all of whom can misuse their specialist knowledge to act against the interests of their customers. Doctors recommend and perform a higher rate of caesareans in regions where the birth rate is dropping. And teachers forge exam papers in order to push up pass rates.
In 2002, Levitt examined the eight years' worth of primary school exam papers submitted as part of the Iowa Test of Basic Skills. He looked for unusual patterns of answers that would suggest that teachers had "improved" the answers of their students (and their apparent performance as teachers) after the papers had been handed in. He found evidence suggesting that teachers had cheated in 4 to 5 per cent of the classes he studied. The school system retested 100 classes and fired the teachers it found guilty.
Which doesn't mean that we shouldn't put ourselves in the hands of experts. Levitt makes the point that it's often worth using experts even if you know they are trying to do you in. An untrustworthy real estate agent might still be better at selling your house than you would be yourself. A corrupt schoolteacher might still be better at teaching students than would their own parents.
But it pays to check up on experts. And that's getting easier. Levitt says before the arrival of the internet in the late 1990s Illinois real estate agents were able to sell their houses for about 5 per cent more than those of their clients. Afterwards the difference fell to 3 per cent.
It might also be a good idea to let the experts know you are on to them. I'd suggest buying two copies of Levitt's new book, keeping one for yourself and giving the other to your real estate agent. It's due out in the US next week. Here, you'll have to order it. Its disturbing title is Freakonomics: a Rogue Economist Explores the Hidden Side to Everything.