# Counting Real Estate Commissions and Sales, from Kim Zussman

July 7, 2008 |

A real-estate agent shared some data on the local housing market over the past year, including listing price, sale price, days on market, and selling agent commission. He posed this question:

Commissions are paid by the seller to the listing agent and the selling agent (the agent that brings the buyer). Selling agent commissions range from 2.5% - 3%. The seller selects which commission rate he wishes to pay. Which yields the best result for the seller? My hunch is the higher commission (for many reasons), but I would like to see if the data supports my hunch.

To look at results for the seller, I was interested in DOM (Days on Market), and how much below the listing price the actual sale transacted - a variable I called "discount":

discount = [(listing price)-(sale price)] / (listing price)

In the data I noticed that there were other commissions besides 2.5 and 3.0%. When including these outliers, the analysis was messy - so since there weren't many I threw out all but exactly 2.5% and 3.0% selling agent commission data. Here is a test comparing the mean DOM for sales commission of 2.5 and 3.0%:

Two-Sample T-Test and CI: DOM_1, Commission 1_1

Two-sample T for DOM_1

`Commission 1_1 N Mean StDev SE Mean`
`0.025             398 90.4   73.3    3.7       T=2.08`
`0.030             164 78.5   56.3    4.4`

Those paying sales commission of 2.5% stayed on market an average of 90 days, whereas those who paid 3% stayed on 78, and the difference was statistically significant. Evidently the agent worked harder!

However the discount (drop in price from listing to selling) went the other way:

Two-Sample T-Test and CI: Disc_1, Commission 1_1

Two-sample T for Disc_1

`Commission N    Mean StDev SE Mean`
`0.025       398 0.0453 0.0467 0.0023 T=-1.34`
`0.030       164 0.0513 0.0488 0.0038`

The sellers paying the higher commission also had to take about 0.5% more discount from listing to selling price (though not quite statistically significant). This, of course, is made worse for the seller because the commission was also higher by 0.5% - so you could say that the net was really 1% worse including the higher commission.

Taken together it looks like the higher commission for selling agent incentivizes him to close the sale faster, though at a slightly worse price for the seller.

Capitalism works!