Arnold Kling  

Steven Levitt

China Menace?... Health Care Reform...

The New York Times profiles Steven Levitt, the recent Clark Medal winner.

Using data from more than 50,000 home sales in Cook County, Ill., he compared the figures for homes owned by real-estate agents with those for homes for which they acted only as agents. The agents' homes stayed on the market about 10 days longer and sold for 2 percent more.

For Discussion.
One proposed reform in real estate is to use marginal commissions rather than average commissions. Instead of earning, say, 6 percent of the total sales price of the house, the agent would earn, say, 50 percent of the difference between the sales price and a benchmark price. The benchmark price might be 90 percent of the appraised value of the house. How would that affect real estate agent behavior?

Comments and Sharing

CATEGORIES: Microeconomics

COMMENTS (17 to date)
Ted Harlan writes:

One proposed reform in real estate is to use marginal commissions rather than average commissions... How would that affect real estate agent behavior?

Drastically. Real estate agents are self-interested actors just like the rest of us. We could have problems, however, with self-interested appraisers receiving kickbacks for undervaluing a property.

Realtors have a pretty good racket...

In my neighborhood, I am noticing a very high percentage of houses being sold FSBO. Anyone else noticing the same?

Bernard Yomtov writes:

I don't see why real estate commission structures need to be completely standardized. Different sellers have different objectives. Why can't an agent offer a choice of contracts?

For example, maybe you are moving out of town to take a new job, and want to sell quickly. So the standard 6% fee will work. The incentives match pretty well.

But if you are just buying a larger home, and can afford to let the current one sit on the market while you wait for the best price some sort of sliding scale with more incentive to get a high price, rather than a quick sale, could be better. Even a flat hourly fee might work here.

When you offer a house for sale you are, roughly, trading off time, and holding costs, for price. Different sellers prefer different tradeoffs. why shouldn't the agent's contract reflect that?

Let me add that my observation, based on the behavior of friends, is that sellers tend to drastically underestimate holding costs, and would usually be well-advised to sell quickly at a somewhat lower price than hoped for.

Eric Krieg writes:

Real estate agents are for wusses. Take it from a Cook County, Illinois resident. The market is so good here that an agent is an unneccessary expense. You can get the few services that you actually need (like a MLS listing) a la carte.

This report is just one more reason not to use an agent.

Eric Krieg writes:

>>Let me add that my observation, based on the behavior of friends, is that sellers tend to drastically underestimate holding costs, and would usually be well-advised to sell quickly at a somewhat lower price than hoped for.

Chris writes:

Real estate contracts are negotiable now. Nothing says you have to pay them 6%, or whatever the standard is in your area. The good agents will be able to refuse to negotiate, the weaker or newer agents will be willing.

That said, there are a lot of not particularly bright people making six figures in the real estate market. Wait until rates pop up another point or two and we'll see who the good ones really are.

Micha Ghertner writes:

Here's an off-topic question about the Levitt article: one thing I found reassuring was Levitt's lack of experience in higher mathematics. I too have little experience in math, but an interest in economics. For those in the know, would I be able to survive Econ grad school? Would choosing a less math-based program like the Austrian programs at NYU or GMU be a wise choice, or do those still require too much math?

dsquared writes:

The article is way off base on that one. I've read Levitt's papers. He's not into all the analysis and higher math that marks out an MIT economist, but his econometrics chops are extremely good. The statistical tools needed to handle non-standard datasets are difficult to manage and Levitt does it better than anyone else.

Any economics program will teach you the maths you need to get through it, so I wouldn't be put off by that. For all the worst reasons, Austrian econ. is a bit out of favour at the moment, so unless you're doing the course purely recreationally, I suspect that taking an Austrian-heavy degree to avoid the mathematical heavy lifting would be a career-limiting move.

Micha Ghertner writes:

Thanks, Daniel. I'm interested in Austrian economics for ideological reasons as well as my mathematical ineptitude. Regardless, I'm planning on combining the econ with a law degree, so I'm not *too* worried about the career limitations.

By the way, is the term "maths" in general usage, or is it just a Briticism?

Matt Young writes:

I am in the process of selling a property in California, and I use a flat rate seller who will charge me $2,500 for a $400,000 valued property.

I will accept the best offer minus buyer commissions, so the buyer has an incentive to get a lower commissioned buying agent.

The real estate agents are suffering a bit of dis-intermediation because of the computer industry.

I am also buying a property, and I have access to the MLS listings on the Internet. I will investigate the properties I want on my own, write my own offer, and simply hire an agent with a flat fee to handle escrow.

Eric Krieg writes:

D^2 and Micha, I'm just curious, what is considered rigorous math in the economics profession.

ODE's? Some calculus?

As someone who was not strong in math before going to college, and who chose a major that required rigorous math, let me tell you that math ability CAN be developed, even later in life. I worked VERY hard in my first four math courses (2 calculus, one diff eq), essentially doing EVERY solved problem in the textbook. It was a couple of hours of work every night, but the payoff is that I am now "good" at math. Which essentially means that I have the ability to relean diff eqs every time they comes up in a subject! (and don't they come up in so many subjects: economics, heat transfer, shock and vibration, acoustics, analog circuits, etc.)

The only way to be strong in math is to put the time in. Too many people have the foolish notion that math ability is genetic. It is not, at least not at the level of college calculus.

Bob Dobalina writes:


Don't you think Calculus (at the college Calc II level) gets a little ridiculous? Especially the trigonometric stuff. Mountains of formulas to memorize.

Micha Ghertner writes:

If it's any indication, I don't even know what "ODE's" are. I took Calc I and did ok until the very end of the semester when we started getting into complex integrals. During the first week of Calc II, I thought that I had mistakenly enrolled in ancient Greek - I had no clue what the hell was going on.

Incidentally, I came to college as a math major, having performed very well in all my high-school algebra and pre-calculus courses. Calculus at a top-ranked engineering school (Georgia Tech) put an end to that idea.

Eric Krieg writes:

I have never memorized a formula in my life. OK, maybe that's a little of an overstatement, but with calc, you learn the basics and derive the rest.

Yeah, outside of Calc I (derivatives and integrals) and differential equations (ODE is ordinary differential equations), the rest of the Calc (Calc II and III) is not so important. I've been an engineer for 7 years, and I've never used a power series, for example.

But don't discount these classes. They're hard, but they also make you learn to think.

That's an important point. Today's primary schools don't teach kids to think, that's why the kids have such a difficult time with college calculus. It is a problem I had coming out of high school.

Another thing about calculus is that I didn't actually retain a lot of the information. But derivatives, integrals, and diff eqs are used throughout engineering. The concepts were reinforced with many different classes. I essentially relearned the topic every time I took them.

I think this would happen for you, Micha. If you are using the stuff in your economics studies, in real applications, you will learn it better than in an actual math class. That's my experience, anyway.

And once you have a mastery of these topics, it comes back to you quickly. I'm working on a masters degree in engineering, and I have to use calc all the time. I inevitably start out a little rusty each semester, but it comes back quickly.

Micha Ghertner writes:

I'm thinking of picking up "Calculus Made Easy," by Silvanus P. Thompson and Martin Gardner.

Has anyone read this book? Would you recommend it? Are there better books out there?

Any book that has recommendations by Julian Simon, Stephen Jay Gould, AND Noam Chomsky must be interesting.

dsquared writes:

Eric, Micha:

Basically, I would expect a graduate of a decent economics program to handle calculus up to real analysis, topology up to fixed point theorems, linear PDEs but not ODEs in any depth, stochastic calculus and dynamic programming. (Most of the mathematical element of an econ. degree is aimed at getting the little tykes to the level of handling Bellman's equation). I would also expect them to have extremely narrow knowledge of the topics in all these fields, because of the way in which it's taught. For example, I learned Lagrange's equation by rote as an undergrad and didn't understand why it worked for years afterward.

Most economists also have decent linear algebra and statistics (ranging from decent to excellent depending on the school), but bloody awful grasp of concepts in probability.

If you're doing econ. as part of a law degree, you should have no problem with the math if you could hack it in high school. "Maths" is the English term and only in general usage in the UK, btw.

Eric Krieg writes:

Linear PDE's? Ugh. When I get a PDE, I go right to either Excel or Mathcad and use a numerical method to solve it.

Good info, D^2. I only had one economics class that attempted to use calculus. It was a first order equation, and we integrated it.

But from the economic papers I've read on the web, I would think that statistics should be very important. In particular, I would think that a good class on design of experiments would be mandatory. Factor analysis would seem to me to be essential.

Eric Krieg writes:

Read it and weep, aspiring American economists:

Comments for this entry have been closed
Return to top