Realty Times, along with a number of other online publications reported that the percentage of people who sell their homes themselves has declined from 1997 to 1999, based on the NAR's homebuyer/seller surveys from 1997 and 1999. But were we right or wrong?

Other reports suggest that FSBO numbers are on the rise as much as 30 percent to 40 percent. Could the NAR just be putting a positive spin on what is actually data disasterous to the industry? Or could the discrepancy lay in what the other reports call FSBOs?

It's an intriguing question, and one that casts confusion on any "research" that depends on the definition of the FSBO.

While Realty Times may be good with words, we admit our math could use a little work, so when an eagle-eyed reader Keith McLane, a Realtor with the Greater San Francisco Association of Realtors, pointed out a possible problem in the reported results between the NAR's 1997 and 1999 surveys, we decided to investigate. If McLane were correct, the surveys could point to a rise in FSBOs, not a decrease, as Realty Times and others previously reported. The news site could be facing a possible retraction.

According to the 1997 survey, FSBO sellers have a number of reasons for selling a home themselves, including not wanting to deal with an agent, having an available buyer already, and not wanting to pay a commission among others. This information was displayed in Table 48. The summary below the table clearly states that 13 percent of sellers sold their homes without using an agent.

McLane says that the 13 percent "doesn't appear to break out the percentage of FSBOs who sell to a friend, relative or relocation company. The 1999 survey, however, does break out this subset of the FSBO population on page 51 in the following way:

"There were a small percentage of homeowners in 1999--around four percent--who sold their previous homes either to friends or family or to a relocation company. The remaining 16 percent of homes were sold by the previous owners. In 1999, these "For Sale By Owner" transactions, or FSBOs, consisted of two types: those in which the seller knew the buyer and those where the seller wanted to sell their home without the assistance of a real estate professional."

"Hence, the reality of the situation appears to be that, per NAR's own research, 13% of sellers sold their homes themselves in 1997 as compared to 20% (16% + 4%) of sellers who sold their homes themselves in 1999--a 7 percent increasein FSBO sellers," writes McLane.

So why don't the numbers add up? Realty Times asked NAR research director Dr. Fred Flick and survey author Kevin Roth for help.

According to Flick, the problem in the numbers comes from the definition of what is and isn't a FSBO. "If you cut a deal to sell your home to your brother, you didn't really sell it on the market," explains Flick.

The FSBO numbers in the surveys, says Flick, refer to people who put their homes on the market and sell them to someone they don't know.

He also says that the 1997 figures are no longer relevant to the 1999 findings because, as McLane discovered, the questions were asked in more detail in 1997, revealing different results.

The four percent of sellers who didn't use agents weren't true FSBOs, according to Flick. "We stand by the 16 percent number of FSBOs," says Flick.

Roth adds that the respondents to the survey are not associated with the NAR but are found in public records by independent research firm, Experian.

Roth explains that the 1997 version of the book had a different author and presented information differently. It all starts with the question, "What did you do with your previous home?" found in the homebuyer section of the survey, not the section on FSBOs.

"In 1997, the author printed the entire results which drives down the FSBO number," says Roth. "In the 1999 survey we presented the equivalent data but broke it down in two categories - those who did not sell their homes due to hanging on them as an investment or were unable to sell them."

If you take out that number you get 18 percent. "We had extracted out those non-sellers previously, but we didn't do that in 1997," says Roth.

"In 1999, the researchers presented Tables 6-4 and 6-5 is all repeat homeubyers who sold their previous homes. "We removed the people who did not sell their previous home and just analyzed those repeat buyers who did sell their home and in that situation that is where you get the 16% of FSBOs for 1999."

All clear now?

"It wasn't meant to be confusing," says Roth. "I thought the 1997 survey was confusing. It didn't seem appropriate to be counting the people who didn't sell their homes in the mix."

So McLane was right about finding the discrepancy, and the researchers understand why he drew the conclusions he did. And Realty Times has no retraction to print.

But the moral of this story is that while it is common journalistic practice to report the findings of a survey at face value, sometimes with a little digging, other facts can come to light.

Now we know what is and isn't a FSBO and how that definition can make a difference in the interpretation of data related to homebuying and selling.

The 1999 and 1997 surveys are both available for viewing at http://nar.realtor.com/research .

Published: May 15, 2000

Use of this article without permission is a violation of federal copyright laws.