The Federal Trade Commission and Department of Justice held a joint public workshop in Washington, DC on June 5, 2018, to explore competition issues in the residential real estate brokerage industry.
What’s New in Residential Real Estate Brokerage Competition – An FTC-DOJ Workshop
June 5, 2018
BILL ADKINSON: If you could please take your seats, we’ll start the last panel.
Thanks very much. Welcome to the final panel of the day—Regulatory and Industry Factors
Affecting Residential Real Estate Competition. We’ll cover a variety of policy issues over the
next 90 minutes. So I’ll get right to brief introductions.
My name is Bill Adkinson. I’m an attorney advisor in the Office of Policy Planning at the FTC.
The panelists’ impressive bios are online and also have been distributed outside. The panelists
are Stephen Brobeck, who is Executive Director of the Consumer Federation of America; Katie
Johnson, General Counsel and Senior Vice President, National Association of Realtors; Dr.
Panle Barwick, Associate Professor of Economics, Cornell University; and Brian Larson,
Associate Professor of Law, Texas A&M University School of Law, and Of Counsel, Larson
The questioning will be led by my colleague Matthew Mandelberg, who is an attorney in the
Competition Policy and Advocacy Section of the Antitrust Division of the DOJ. We’ll start by
asking each of the panelists to take two or three minutes to convey their general perspectives on
these issues. After that, Matthew will lead the conversation. Since our time is limited, we’ll ask
the panelists to keep responses to questions to a minute or two.
MATTHEW MANDELBERG: So with that, we have the all-star panelists that Bill introduced.
And so, without further ado, I’ll ask each of my panelists to spend a couple of minutes just
providing their perspectives. So our role here, in this panel, is taking a step back after hearing
about developments in technology and hearing about some of the new and innovative business
models and some of the traditional business models that remain relevant in the industry.
And here, we’re going to look for, more broadly, at the state of competition and what regulatory
or industry factors may be facilitating or inhibiting competition. So I’ll turn it over to each of my
panelists to spend a couple of minutes providing their general overview and perspective on the
BILL ADKINSON: So Stephen, can you start?
STEPHEN BROBECK: Sure.
BILL ADKINSON: Thanks.
STEPHEN BROBECK: Thank you for this opportunity. I’ve submitted written comments late
yesterday. These can be found on the CFA website in the Real Estate Brokerage subsection in
the Housing Issue area. My comments are going to really echo, to a large extent, much of what
was said in the last panel.
CFA believes that this brokerage marketplace provides good information to consumers about
home listings and facilitates relatively quick sales. However, we also believe that there’s
insufficient price competition. Commission rates are relatively high and they are uniform.
In a number of developed countries, these rates are between 1% and 4% in part because a single
estate agent is involved. In the US, the average rate appears to be over 5%. And what struck me,
as I was researching my comments, was that it’s no lower than it was in 2001. This may appear
puzzling because of supposed efficiency gains through technology. But it really doesn’t surprise
us. Traditional agents and brokers practically never advertise and are even reluctant to publicly
disclose base commission rates.
The coupling of listing and buyer agent commissions gives buyers the impression that they are
not paying commissions, and also allows steering of these buyers to properties with the highest
commission splits. The glut of agents, about 2 million for 4.5 million homes sales annually,
creates agent and broker social pressures to maintain high rates. And consumers, who are
focused on the sale price and a timely sale, know relatively little about brokerage services, which
are complex and rarely purchased but depend on personal services of agents.
For all of these reasons, nontraditional brokers have had a very tough time competing with
traditional agents and brokers. Exclusive buyer brokers, who I worked with in the 1990s when I
first engaged these issues, have almost disappeared. And nontraditional agencies have small
market shares that are growing slowly or not at all.
Because of these market imperfections, the FTC and DOJ need to continue their vigorous efforts
to ensure greater price competition. They should continue striving to eliminate state anti-rebate
and minimum service laws. These statutes not only restrict price competition in a certain number
of states, but may well encourage informal discrimination against rebaters in other states. The
culture is very important here of the traditional brokerage industry.
Second, investigate proposals for more effective price disclosures. I don’t have time to outline
some ideas. I hope to have that opportunity later. And then finally, closely examine the price
impacts of large real estate portals like Zillow. Now Zillow does make available—and the other
large portals—valuable information to consumers about listings. But the fees these portals charge
agents and brokers increase acquisition costs, encouraging high commission levels—that’s a real
problem in the industry. There are too many agents. I mean, they compete too vigorously. They
have to spend a lot of money in order to acquire clients. And that keeps the commission levels
And then also the information published by Zillow on service quality is deceptive bordering on
the fraudulent. I may have an opportunity to comment on that later.
BILL ADKINSON: Thanks very much, Stephen. Katie?
KATIE JOHNSON: Good afternoon, everyone. First, thank you to the FTC and the DOJ, and all
of the attorneys that we’ve been able to work with and talk to, for putting on this presentation and
all the panels today that have been a really good discussion on the state of competition in our
industry today. So my opening remarks are—I’m going to try to summarize what I think we’ve all
gleaned throughout the panels today. I think, first, that means that technology has definitely had
a positive impact on the consumer experience and it has added costs to the costs of doing
So search is a great example. Because there are so many websites, thousands of websites and
apps, that deliver property information to consumers in an instant. But also, technology has
enabled brokers to deliver varied business models. We saw four different ones and heard about
many others in the second panel today.
They have enhanced digital marketing capabilities, professional photography, CRM tools that
just help them deliver a better consumer experience. And then, of course, the transaction itself
has become almost entirely electronic with the advent of transaction management tools,
electronic signatures, and now eNotarization.
So that has all enhanced the consumer experience, but it costs something. Brokers and agents are
spending money to have websites and apps and digital marketing and professional
photographers. And I didn’t even mention lead generation. So that’s been discussed throughout
this day too. Lead generation has significantly increased in the last 10 years. I think, even, just
five years ago, like in 2013, the average agent spend on Zillow was about $3,000. And now, five
years later, Zillow and Trulia have combined and it’s double or almost even triple that. So there
are costs to the technology.
But also like many, many other industries, while technology has enhanced the consumer
experience, it has not displaced the worker. Real estate agents and brokers are valuable
resources. We’ve heard that. I think we’ll consistently hear that today too. A home is often the
most complex purchase of any person’s life. They want somebody to help them through that
process, through that complex, that emotional process. They need a local, trusted real estate
advisor who understands the market, the neighborhoods, the communities because they actually
live there and they understand what’s going on.
So that leads us to the second theme I think we’ve heard and is true—and that’s that even with
these great advancements in technology consumers still choose real estate professionals to help
them through the buying and selling process.
And then third, the real estate industry is robustly competitive. The market is working and
government intervention could stifle this competitive environment.
So what we heard is that the real estate is unlike any other industry in the world, including the
real estate industries in other developed countries. Gross differences, not just geographically—
although that matters—but also in the way they do business, the way they represent the
consumers, and the way their consumers are protected in real estate transactions in other
So here we are unique because our brokers agree to cooperate with their competitors. And they
do that for the benefit of the consumers. And the way that happens is through the MLS. And
they’re incentivized to participate in the MLS because it helps consumers, and it helps them do
their business and get paid for it.
MATTHEW MANDELBERG: Could you just—
KATIE JOHNSON: Wrap it up? Destroying incentives to belong to the MLS would harm
consumers, not help them. Free market, competitive—no need for government intervention.
BILL ADKINSON: Thanks, Katie—a little intervention. So Panle, could you speak please?
PANLE JIA BARWICK: I’d like to thank the workshop organizers you for giving me this
opportunity to share some insights that we have learned from economic research. And over the
past couple of decades, we have seen a proliferation of technology innovations and many new
business models. However, much of the potential of those innovations has yet to be realized in
the real estate sector, partially due to some structural hurdles. And here, I’ll focus on one, which
is the seller pays for the commission of both the listing agent and the cooperative agent. I’m
going to use a word—buying agent.
Historically, both agents worked for the seller. So it’s natural that the seller paid for both agents.
A couple of decades ago, many states changed the regulation so that the buying agents are
answerable to the buyers, but they didn’t require the buyers to pay for the service and instead
kept the traditional model. Why might this become a hurdle to competition or reaping the benefit
of innovation? Because first, the seller pays a fraction of the transaction price as commission to
the buying agent. And that has the—potentially can create a conflict of interest because the
buying agent prefers a high transaction price, but the buyer prefers a lower transaction price.
Second—and here the literature has documented, provided evidence of steering and suggests that
this could limit price competition. The researchers have documented two identical properties
with different commissions to the buying agent have different outcomes. The property with the
higher commission is more likely to sell and sells faster. As a result, paying a lower commission
to the buying agent, a form of price competition, is not a winning strategy.
Third, when the seller pays for both commissions, the buyer ends up paying for the commission
through increased transaction price—despite that the fact the buyer doesn’t write a check to the
buying agent. And we have heard much of this. The confusion that the buyers don’t know what
the role, who is paying for the commission? And that, in some cases, the buying agent gives
some rebate to the consumer, but this is not very effective and limits the choice the consumer
Lastly, it’s difficult to get data from other countries. But according to research that we have, the
US commission rate is on the high end of the spectrum compared to other countries. In the UK,
for example—I’m not necessarily advocating this country as a model—but there, the seller pays
for the leasing agent and the buyer doesn’t pay the agent. The realtor fee varies from 1% to 3%
and seems to be decreasing in recent years with a proliferation of discount models.
So I think the core of—the second panel has spent much time on this—and I would like to
highlight that this is a feature that we should study and we should pay attention to. In many other
industries, you pay for the service. Why can’t we do it here? Having the seller paying for both
creates confusion, limits price competition, and potentially prevents us from getting the benefit
of technology innovation.
BILL ADKINSON: Thanks very much, Panle. Brian, could you close us out?
BRIAN LARSON: Sure. I don’t have any summative comments to make. Thank you to the DOJ
and FTC for inviting me to be here today. As you mentioned, I’m a law professor at Texas A&M.
And I’m of counsel with the law firm of Larson Skinner. My expenses to be here today are paid
by the Council of Multiple Listing Services or CMLS, but I’m not speaking on their behalf. Any
stupid things I say here today are my own views, and there may be one or two of those.
I come here today as a 25-year veteran from the MLS business, having run an MLS for about 10
years and having been an attorney for MLSs for another 15 after that. So although I’m not
speaking on CMLS’s behalf today, I would like to direct people to CMLS’s white paper that it’s
submitted as commentary to FTC and DOJ on Friday. Because I think it makes the key points
that I’d like to emphasize today.
First of all, I think there’s a consensus here that, probably on this panel, but certainly on earlier
ones, that MLSs function to make a more liquid market for real estate. They decrease search
costs for consumers. And that there’s a value to that. And they also foster this cooperation among
competitors that we see in the MLS context.
We also know that they’re critical conduits for listing data content. In the last 10 years, there’s
been an explosion of distributions of listing content by MLSs or by the contractors of MLSs—
people working for with MLSs. And MLSs are working to make that distribution more effective
by bringing it into compliance with standards like the RESO standards.
I want to emphasize here today too that I see listing data content as valuable intellectual property
of brokers and sellers. I know there are some commentators who have talked about how brokers
are in the real estate selling business, not in the database business or information business.
But the way that you present a listing as the listing broker, the way that you present a listing
online in the various media where you have an opportunity to present it, that is your business.
That’s how you’re going to earn your keep is by making that place look attractive, by presenting
it in a way that’s consistent not only with your brokerage’s brand, but with your seller’s interests.
I think that it’s also important to remember therefore that the listing broker and the seller should
have choices about where the listing data goes. The folks who seem to think that it should be
forced on them to distribute listing data places that neither the seller nor the listing broker has
asked for the listing data to go, that’s a mistake. If we’re worried about buyers, I think we already
have evidence that they’ve got access to probably more information than they want or need let
alone not enough.
And then lastly, when it comes to this question about government intervention—I mean, we’re
meeting here at two federal competition agencies’ request. I think it’s important to remember that,
so far in this space, government initiatives have been limited both in scope and probably in
And I think it’s also important then to think about, if we’re going to consider policy changes, we
want to make sure that they’re carefully motivated by good research and that we are confident
that implementing some kind of policy change or some kind of enforcement change is going to
improve the competitiveness of the market. That we don’t try to remedy one potentially
anticompetitive aspect and, in so doing, destroy the significant procompetitive aspects of the
MLS space and the brokerage industry. So those are my opening thoughts.
MATTHEW MANDELBERG: Thank you all. As you can hear, we have quite a diverse array of
views on our panel. And I imagine that most of our panelists will actually want to speak to many
of the questions. So I’ve already informed them that they will put their name tag up like this
when they want to answer one of my questions.
So my first question, I think going back to the first panel, the consent decree that was entered
into against the National Association of Realtors about 10 years ago arose—and I often think that
one of the best ways to think about where we are today is to take a step back and look at where
we’ve been. And so, since it’s already come up, I wanted to ask the panelists to address what it
was that the consent decree, from their perspective, was seeking to address. And what has been
its effect, and whether the decree remains relevant to the industry today. So I’ll hand it over to
Katie, who may have some views.
KATIE JOHNSON: Well, thank you very much. I do have some views. So I just think it’s
helpful to maybe paint the pictures, since we’re going to go back in time and go a little bit farther
back in time. How does the MLS work? Well, the associations of realtors that own or operate, or
are part shareholders of MLS abide by the NAR MLS policies. And the NAR policies are created
by brokers and MLS executives through a committee process. Very standard—you submit a
proposal and the committee debates it, deliberates it, they vote on it. If it passes, it goes to the
board of directors of which we have 900 professionals. And they pass or decline to pass a policy.
So many years ago, there was the creation of internet display listings. It was a policy that
facilitated the ability for brokers to advertise other brokers’ listings on their websites. Every state
regulates the display and marketing of real estate for their state. So this was a contractual way to
allow brokers to get that permission and to facilitate the display really easily. So it was a win.
There was limited information in that display because it was the first time people were having
this mass display of property information on the internet, and there was just a subset of
Then soon after, simultaneously, there were internet brokers who have been termed “virtual
office websites” wanting to do business online. And they wanted more information. They didn’t
have bricks and mortar offices. And they wanted it to display more information than what was
available in that advertising IDL feed. And so the committee debated the policies and came out
with a policy that allowed brokers to opt out of sharing their information with virtual office
And the Department of Justice disagreed with that policy and that led to the consent decree,
which said that brokers cannot opt out of virtual office websites. There is some structure to
providing that information to consumers, namely, having them register and consent to a special
relationship between the broker and the consumer viewing this more robust data. So that takes us
back, I think, to 2008. And now, as you heard, especially on the first panel, IDX, which is now,
the advertising mechanism called IDX, has transformed so much that it has a lot more robust
information, very similar to the VOW feed.
MATTHEW MANDELBERG: So thank you, Katie. We also heard earlier, I think, a diverse
perspective on the consent decree—I think it might have been Glenn Kelman at Redfin who
seemed to be relatively positive on it. And by contrast, I think Art Carter seemed to think that the
decree was more technology- and time-focused. I’d be interested to hear views from panelists on
that perspective as well. So Brian—
BRIAN LARSON: I’ll reset up my name tag. So I think there are a couple of respects in which
the consent decree embodies a policy that is not really relevant now. Now, that’s not to say that
there is not a place for antitrust regulators to observe what’s going on in the industry and make
sure that there aren’t anti-competitive practices. But the consent decree embodies a policy for
virtual office websites or virtual brokerage that would have made sense in 2007, but doesn’t
make much sense in 2018. It refers only to websites, for instance.
And NAR, of its own volition, without any nudging from regulators, extended its IDX policy to
cover mobile applications and also newer technologies like voice-driven technologies—Alexa
and the like. And there was nobody holding a gun to NAR’s head to do that. And so now the IDX
policy has sort of leapfrogged what’s available for virtual brokers in a way that’s meaningful.
So I think that that’s one thing to keep in mind, is that the old policy that’s embodied in the
consent decree is an old policy. And there’s no evidence that NAR intends to make it harder to be
a virtual broker once the consent decree expires. And extending the access consent decree,
however that might actually happen, wouldn’t make much sense because it doesn’t embody a
MATTHEW MANDELBERG: So another question, along with that question, I’ll just add the
sub-question of, to what extent do the principles underlying the consent decree remain relevant?
I think we had Brian mentioning that he still thinks the antitrust agencies should be following
this industry closely. So I’ll, again, extend that question to the extent that people want to weigh
BRIAN LARSON: I don’t know if I said closely. But in the same way that you’re keeping your
eyes on any industry in this country, which I assume is part of your portfolio, certainly real estate
is an industry in this country.
MATTHEW MANDELBERG: Stephen—
STEPHEN BROBECK: Well, as I said in my introductory remarks, it’s critically important that
the DOJ and FTC continue to actively monitor, scrutinize this industry and intervene when
there’s evidence of lack of effective competition. The state agencies really are largely influenced
by the industry and are not that effective—in fact, pay virtually no attention to competition
except, actually, to support a lack of competition to the extent that the states have approved antirebate statutes, minimum service statutes.
A number—in a minority of states, they still exist. We think it’s very important for the two
agencies to try to eliminate those statutes as I indicated. I actually think that when you have those
statutes, it influences the whole culture of the entire brokerage industry. And so let’s target that.
But let’s also continue focusing on the lack of price competition that exists—that has been
documented, not just by Panle and her research colleagues, but also by a number of other
This is complicated. But one important role that they have played is really evidenced by this
workshop today and the second panel, in particular, in which they invited in nontraditional
brokers, who explained why price competition does not really exist and why it doesn’t exist.
MATTHEW MANDELBERG: Panle?
PANLE JIA BARWICK: I want to basically explain a couple of research studies that may share
some insights about the importance of the consent decree. It doesn’t speak to it directly. It’s about
two other industries. But it has important insights. So there are two recent studies. One looks at
the 2011 event where American Airlines had the dispute with Orbitz and decided to withdraw
their ticket pricing information.
So the study looked at this and discovered that in that period, about six months or so, the visits to
Orbitz went down by 11%. And it significantly affected Orbitz—it’s about 11%, maybe around
that number. It significantly affected Orbitz’s business model.
And another study that looked at another aggregator, news aggregator, Google, and followed the
2014 event where Google News withdrew from the Spain market after the legislation has
allowed the news publishers to charge Google for summarizing the news. And what the study
found is—looked at how the news consumption by consumers reduced significantly, and they did
switch to some of the traditional publishers. But their consumption of all news went down.
And so I think those are—we don’t yet have studies that look at the internet brokers or the portals
in this industry. If any of you have access to that data and would be willing to share that, I think
economists would love to have this opportunity to study them. But I just wanted to use those
studies to basically voice the importance of having free information. And if we allow some
parties to withdraw some information, that could have important consequences.
MATTHEW MANDELBERG: So we’ll come back to the flow of data and who has access to it.
But I wanted to pick up on something that Stephen mentioned, which was the anti-rebate and
minimum service laws. And so, as we’re walking back and then looking forward, I also would be
interested if someone could explain just what are anti-rebate and minimum service laws, and are
they an ongoing area of concern?
STEPHEN BROBECK: Well, the anti-rebate laws are so important because most buyers, as was
indicated by an earlier panelist, don’t really think that they’re paying anything for their brokerage
services. And so you need an incentive to get them to focus on price. So you offer them a rebate,
which was really starting to explain to them that, in fact, you are paying.
Basically, the economic research is pretty clear—both the buyers and the sellers over time are
paying roughly equal amounts of the commission. You’re paying the commission, I’m going to
split, I’m going to rebate part of that to you. So if the buyer brokers are not allowed to do the
rebating in today’s marketplace, with the coupling of listing and buyer brokerage, there’s just
really no hope for effective competition on the buyer’s side—price competition.
MATTHEW MANDELBERG: Katie.
KATIE JOHNSON: So with all due respect to Stephen, of course, I took away a different
takeaway from the second panel. I think that they illustrated very well that there is price
competition and that there are varying business models and that consumers are understanding
that there is not a fixed commission and that it is either negotiable or you can choose a flat rate
versus a percentage commission.
So I think that that was a good illustration of price competition. And I agree with Panle that it
should be studied and it should be evaluated, and we should care about this because we, the real
estate brokers and agents, want the best experience for consumers just like everybody else does.
But you should use that information that’s happening today. The studies shouldn’t look back at
2005 or before that didn’t have these varied business models that we’ve learned about. They
should be as current as possible.
And to that point, I think it would challenge you—you’d have to look at the closing disclosure.
Because of the way the system is set up and that the seller’s broker is paying the buyer’s agent, a
lot of that rebate, a lot of the buyer price competition happens at the closing and in the closing
disclosures that we just—I don’t think are being reflected in the studies.
MATTHEW MANDELBERG: Brian?
BRIAN LARSON: So this is where I’m speaking entirely on my own behalf. I think that antirebate laws and laws that set minimum service standards for brokers generally are a bad idea,
that consumers and brokers should be able to negotiate the relationships they want to have. And
unless something is patently problematic—and most of these situations aren’t—I think it makes
sense to let them do that. And I’ll note that, at least as far as I know, most MLSs have worked to
facilitate, to some extent, on the minimum services side, again, with NAR’s leadership, which
nobody forced on NAR—I’ve never been identified as a huge supporter of NAR.
But in this case I have to say that they adopted policies to help MLSs figure out how to
communicate through the MLS whether a listing that was going to be in the MLS was subject to
a minimum services relationship. Because the cooperating brokers need to know that. It changes
the nature of the transaction for them. So I think that that’s been supported. That’s been widely
supported. And I would like to see those laws go away. I agree with Stephen there. I don’t see
that there’s any value to the consumers. Katie might have to be more circumspect in terms of
speaking about those laws since NAR doesn’t lobby the state legislatures.
I just want to put the marker down, Matthew—I would like to get back to Panle’s examples about
American Airlines and Google. And as long as we’re going to come back to that, I’ll wait to talk
about that because that’s not a good question.
MATTHEW MANDELBERG: All right. We will.
STEPHEN BROBECK: Just a brief reinterpretation of the second panel comments. Yes, there
are a number of new competitive options out in the marketplace. There actually were in the ’90s
when I started working in this area. And they were full of hope and optimism. But if you look at
market share and you look at where the prices are, the market share is very low. I am hopeful
that that will expand. And maybe it will. We really can’t accurately foresee the future. But I’m a
little—because of the barriers that I outlined in my opening remarks—just a little skeptical that
that will be, essentially—represent a sufficient set of conditions for price competition to
MATTHEW MANDELBERG: So I’d like to take our conversation to the present now, which we
seem eager to do. And I actually put it to you, Panle. Katie was mentioning, and others have
mentioned today, that there are low barriers to becoming an agent today. There is an enormous
number of agents across the industry. And there has been, I think someone said, an explosion of
consumer access to information. So why isn’t all of that enough to discipline the market? Why
shouldn’t we be happy with this status quo?
PANLE JIA BARWICK: So good question. So I think it appears there are many, many agents
who are competing aggressively for consumers. But if you look at the industry, so we call this
market structure—meaning, how many firms are out there—it’s not equivalent to
competitiveness of the industry. A key measure that economists use is how much is price above
cost, price-cost margin. If we use that criterion, the cost of selling or buying properties is
unlikely to be higher than 20 years ago with all the new technology, all the new innovations.
But if you look at the commission the consumers are paying today relative to 20 years ago,
they’re nearly paying twice as much. So if we use that, from that perspective, it appears that the
status of competitiveness in this industry has probably not improved. Are we going to talk about
the competition pricing? Or should I talk about it now or should I wait?
MATTHEW MANDELBERG: Well, why don’t we let Stephen jump in and then we’ll come
back to that.
STEPHEN BROBECK: You’re right. There’s been an explosion of information. And it’s been
very good for consumers. And it’s about the characteristics of the properties that are available on
the marketplace. Information about more properties is accessible to consumers. That’s a great
thing. That’s the first thing and probably the most important thing that consumers focus on is the
characteristics of the houses and also the prices of the houses, not the brokerage fees. But there’s
been no explosion of information about the actual fees that are being charged.
You have to ask yourself the question, why aren’t any of the traditional brokers talking about
price? Why don’t they advertise price? Why don’t they put it on their website? Why do they even
make it difficult for consumers, their customers, to get access to that information? I think what
existed when we called hundreds of brokers in the ’90s and tried to pry price information out of
them, and the standard response—and we really had to work hard to get any information at all—
was, oh, it’s just all competitive. So there’s been no explosion of reliable information about
compensation to consumers.
MATTHEW MANDELBERG: Brian.
BRIAN LARSON: So I’m not sure, first of all, Panle, could you clarify? When you said that the
commission has doubled in the last 20 years, essentially—
PANLE JIA BARWICK: In many cities.
BRIAN LARSON: You’re talking about in dollars, right—not as a percentage of the transaction?
PANLE JIA BARWICK: No. I’m talking about dollars, which matters most, I think. It’s the
amount you pay that matters. That’s the price you pay for the service. That’s the price you pay for
BRIAN LARSON: But you’re also talking about a service that’s priced as a percentage of
something else, right? The cost of real estate has doubled in that time too, more than doubled.
PANLE JIA BARWICK: We can debate whether the cost has increased or not. I think the
argument seems to be—this is actually, if we have more data on this, it would be great for us to
study. But if you believe that the technology has made it easier to DocuSign, to allow electronic
signatures, you can do things much faster, you can now easily aggregate information of different
housing—that seems to indicate that the cost might not have doubled.
And if you look at the cost of selling a $500,000 property versus the cost of selling a $250,000
property, it’s also hard to argue the cost of the former is double—twice the cost of the latter. And
if you look at—across markets, it’s hard for us to believe that in Boston, where the housing price
is so much higher, and in some other places—Denver or where the housing price isn’t that high—
the cost of selling, it’s just, somehow, costs twice as much. So I think that’s where we often view
this as a commission puzzle.
BRIAN LARSON: So when you say it’s hard to believe, though—that you’re saying that as a
matter, sort of a priori, of theoretical constraints, you wouldn’t expect that to be the case. Do we
know whether the costs are—I mean have you actually examined those costs?
MATTHEW MANDELBERG: I’m sorry to interject here. We’ll leave the follow-up questions
BRIAN LARSON: OK. That’s all right. I didn’t know if you wanted us to have a conversation or
not. So I guess my comment is to ask the question. I mean, I think it’s a fair question. If Panle is
asserting something, it’s fair for us to ask her to clarify it. And I’m not even challenging your
assertion—I just want you to explain what it means and how it fits into the real world.
PANLE JIA BARWICK: Right.
MATTHEW MANDELBERG: I think it’s great to have, maybe, a question of other panelists, but
we’ll leave the follow-up questions to me. But I’ll ask Katie—and sort of following-up on this—
whether we see the commissions, they’ve stayed high and consistent in the sense that I think on
the previous panels, people were speaking on the order of 3% or 2.5%, or if we took Panle’s
suggestion that from a dollars value, they’ve actually increased. If this truly is an industry that
reflects robust competition, why do we see commissions so high, so consistent?
KATIE JOHNSON: So I thank you for this opportunity to dispel a lot of misconceptions that are
just accepted as fact, I think. In the earlier panel, as well as this panel, we’ve heard that
commissions are high in general. That’s not true in all markets. We heard from major metros
brokers. There are rural areas.
There is price competition. There are differences based on geography. We’ve also said they’re
high, but we’ve acknowledged that the comparison is other countries that only have one
representative in the transaction. So one of the two consumers is represented and getting paid 1%
to 3% in those other countries. Here, we have two representatives that are both representing
consumers. So I don’t think that they’re high.
But also, there’s been a misconception that consumers don’t know. I think consumers are smarter
than we’re giving them credit for on these panels in that they do know. And when they work with
a good broker and agent, they get educated. When you take a listing and you have a contract—
contracts are obviously negotiable, they include the price, you have the discussion.
The discussion with the seller’s agent says how much they’re going to pay the cooperating
buyer’s broker in order to bring a ready, willing, and able client and help them get to the success.
Again, commission being a percentage of the sales price is pro-consumer because it’s a success
fee. Consumers don’t have to pay if they’re not happy. If they don’t get what they wanted through
the transaction, the purchase, or the sale of the house, they don’t have to pay anything.
And yet, all of the brokers and agents are going to do the work to try to get them there. So I just
think that—and then finally, the misconception that they don’t know unless it’s published on a
publicly facing website that aggregates the national inventory of listings—is also a misnomer
because the agents, the brokers are having these conversations at the kitchen table all the time.
MATTHEW MANDELBERG: Stephen.
STEPHEN BROBECK: Just some of you may not be aware of a survey that we undertook, a
consumer survey with AARP. It was a very in-depth survey. And it was in 2007. And we
learned, among other things—to be fair, most of the—as I’ve reported in my written comments—
most of the consumers who had experience with agents recently said that they were satisfied with
that experience. On the other hand, we also learned that they don’t know very much about
compensation. Fifty-eight percent of recent home buyers and sellers thought that commissions
were set by the industry or by the agents or by the government.
And my comment on that is, if you ask them if they were really certain about that, my guess is
they would say they weren’t. But the fact is they don’t know very much—they have some
impressions. And the buyers never are really forced to deal with the compensation issue.
MATTHEW MANDELBERG: So I want to actually—there’s a lot to unpack in the two most
recent answers. And one of them was the—Katie, you were highlighting that here in the US, we
tend to have two agents involved in the process. And so to provide some education for our
viewers, maybe you could just explain a little bit about that relationship and how cooperating
KATIE JOHNSON: Sure. As I started to say, the listing agent, after they’ve won the listing and
convinced the seller to use them to sell their house, enters into a listing agreement that sets forth
what the listing agent will get paid. And that’s usually a percentage, but it could be a fixed
amount. And it sets forth an explanation of what they will offer the cooperating broker who
brings the ready, willing, and able buyer.
So in the MLS, the selling broker communicates to all of their competitors who know potential
buyers that they will pay them to help them sell this house if they bring the buyer and they help
them through every step of the complex transaction to the successful closing. And so the
compensation gets paid to the buyer’s broker at the closing table in the closing disclosure.
MATTHEW MANDELBERG: Thank you. And just sort of following up on that particular
question, I’d be interested in know what panelists see as both some of the benefits of this
particular model—and I think Katie alluded to some of those already—as well as some of the
drawbacks of this model.
PANLE JIA BARWICK: Sorry. This is for a previous question.
MATTHEW MANDELBERG: OK. Does anybody want to speak to the benefits or drawbacks of
the cooperating commission model?
STEPHEN BROBECK: Well, I think that a principal reason that the commission levels are so
low in a half a dozen Western European countries and in Australia is because, essentially, the
estate agents are functioning as facilitators—what are called transactional brokers here. And I
think consumers should have the right and have access to fiduciaries. And many, if not most—I
don’t know the percentage—of consumers may want fiduciaries. And so if you want a fiduciary,
probably the other side is going to have to get a fiduciary. And that’s going to tend to drive the
price up. So CFA has never suggested that the price—the commission level be down at 2% if
there are two fiduciaries involved.
But, what doesn’t exist in this country is the option, the real option, of transactional brokerage. I
mean, after we helped to expose subagency, agency was just—it was a mess. And different states
passed different laws. But some states, Florida, for example, defined transactional brokerage as
the default brokerage and did not require disclosure of that.
So I’d like to see a marketplace where there are agents out there competing as fiduciaries. I’d like
to see facilitators out there saying, look, I’m not going to do everything a fiduciary would do for
you—I will do less, but I will charge less. But that option, basically, is not available in today’s
MATTHEW MANDELBERG: Stephen, do you see the iBuyers that were discussed earlier as
maybe a movement in that direction?
STEPHEN BROBECK: Well, yeah, there are several movements in that direction. But they
haven’t really got traction yet in the marketplace.
MATTHEW MANDELBERG: Panle.
PANLE JIA BARWICK: There’s no doubt the realtors are extremely valuable. They’re providing
valuable services to consumers both on the buying and for the selling side. And what I think I
actually want to bring back to my opening remark where I see an issue arise, when the seller is
specifying the compensation to the buying agent and to the listing agent. So, effectively, that
limits consumers choices. What’s the fraction of agents who would be willing to rebate 100% of
the commission to the consumer if the buyer would prefer to do the legwork in order to save on
So I think that there’s actually value in presenting the buyer and value in presenting the seller—it
would be great for consumers to choose. The buyer can choose how much service they want,
how much to use the internet. As Mr. Inman has commented, the internet now is becoming the
buyer’s agent. So if we can have both of them choose, I think that will facilitate more new
business models and be better to consumers.
MATTHEW MANDELBERG: Brian.
BRIAN LARSON: So in a partial response to Stephen’s comments, I think it is important for
consumers to understand the kind of relationship they’re entering with the brokerage, and also to
understand the flow of money. And I’m pretty sure the NAR Code of Ethics requires that, once
you list a home with a seller, that you have to disclose to the seller how much money you’re
going to pay the cooperating broker. I’m pretty sure that most state licensing laws I’ve ever
read—and I haven’t read them all and I don’t know of a 50-state survey—require brokers or both
sides to disclose to their principals, even in a facilitator-state, to disclose to their principals how
much money they’re receiving in the transaction. And a great many states require an agency
disclosure at the point that a broker makes a first substantive contact with a consumer.
So there’s a bit of a patchwork going on there. And I’ll definitely acknowledge that. And I’ll also
suggest that it’s unclear how effective disclosures in general are, right? I mean, how many of you
have been getting the GDPR emails in the last week? And how many of you had even thought to
read privacy policies before that?
It’s because you’re presented with a disclosure and you’re focused on the transaction, not on the
disclosure. So you move to the transaction and skip the disclosure. So I’m sure there are plenty
consumers who just sign on the bottom line when they get an agency disclosure. And there may
be more effective ways to communicate those disclosures so that people actually get some real
value out of them. I just don’t know what that is.
I know not everybody loves the new TRID, the thing that replaced the HUD-1 and the good faith
estimate. But it was at least designed with design principles in mind, with an intention to get
people to understand what it is they were agreeing to. And perhaps similar kinds of moves can be
made in the real estate space, although it seems to me those are probably going to be state-bystate moves because that’s where the regulation of real estate generally happens.
MATTHEW MANDELBERG: And Brian, just for the benefit of the audience, TRD? Do you
know the acronym or just what it is?
BRIAN LARSON: Oh, I don’t even remember what the acronym stands for—Transaction,
something, something, something. I don’t know. You tell me.
KATIE JOHNSON: The new closing disclosure.
BRIAN LARSON: It’s the closing disclosure. It used to be the HUD-1. And then there was the
mortgage side of that too.
KATIE JOHNSON: The Truth in Lending RESPA Integrated Disclosure Document. Yes.
MATTHEW MANDELBERG: Well done. So Katie, would you like to follow up?
KATIE JOHNSON: Sure. Yeah. Just on a couple of comments, the internet being the buyer’s
agent, I just totally disagree with that. I think that the internet has improved search. And buyers
are so much more knowledgeable because they have all of that information through the internet
and that’s great. And then to echo actually what Glenn Kelman said is that the internet—what it’s
done is it’s made it easier to meet consumers. And then we’ve learned that there are some costs to
meeting consumers on the internet, costs that brokers would really like to keep and meet the
Why the commission structure works also is a function of how houses are purchased and
financed. I think it makes most sense to be paid for by the seller. And also why it works is
because, like all the brokers said, actually—we do an annual homebuyer and seller survey and
survey tons of homebuyers and sellers who have been through the transaction. And they
repeatedly say what Stephen said—85% of them would recommend or use again their agent that
they used in their transaction.
The brokers on panel two actually said that too. One said 80%, one said 75%. So it’s totally
consistent with what we find. And that’s pressure. I mean, the real estate agents are going to
deliver services. They’re going to talk about how they get paid because they want to be referred
and recommended again.
MATTHEW MANDELBERG: I think that brings up a good question about how agents want to
be referred and recommended again. And I think, on the previous panel, we heard even some of
these more novel business models rely on something on the order of 70% to 80% of either
referrals or repeat users to build their business. And so one question here though is to what extent
are agents able to compete on quality?
We’ve heard a bit of a back and forth now about what is the quality of information that
consumers actually have. And one of the things that I would be curious about is, how much
consumer information is there about the agents that they are asking to represent them? Stephen.
STEPHEN BROBECK: Well, there’s not very good information about the quality of service
offered by large numbers of agents. As has been noted, a very high percentage of sellers and
buyers rely very heavily on a recommendation. That’s not the worst thing in the world if I talk to
a couple of people that have had a recent experience with the agent. And the fact is, I believe,
that actually encourages competition of service quality because the agents—and remember
there’s a big glut of agents—and so they’re really competing for customers. And that reputation is
So that’s good in disciplining in the marketplace. But the fact, nevertheless, is, if you’re a
consumer, you just can’t go out and find a Consumer Reports or a Washington Checkbook—in
this area—article on residential real estate brokers and the quality of service. You can go to Yelp
and Angie’s List, et cetera, but as I explain in my written comments that’s very limited and
people don’t do it for a variety of reasons. I am a little surprised—and I wouldn’t ask the
government to do this—that some third party hasn’t developed a website that basically reports on
a large number of consumer experiences with agents.
And I will take the opportunity, at this point, to say that some of the information on websites is
not very accurate. And I’ll talk about Zillow because they’re gargantuan at this point—getting
larger. Basically, for any listing, agents can basically buy advertising space. Now, there are
several problems with the way they do that. First of all, they call those agents “Premier Agents.”
That suggests to consumers that they are somehow superior. Well, they’re not. They just spent
money to be listed.
Secondly, they include a whole array of consumer reviews. I’ve looked through the reviews of
100 agents and there’s nary a contrary word, a negative evaluation. And I’m not sure how they do
it. My guess is that Zillow says, look give us some of your comments from your clients and those
are, you know, washed. But the worst thing of all is that there is a rating number. It’s between 0.0
and 5.0. And so you would assume the average was, the mean score was around 3. Not so. An
overwhelming percentage, I’m guessing at least in the couple of hundred I looked at, are 5.0%
regardless of whether they do include the number of transactions that that agent has been
involved in. And actually I do think that’s one measure, not a sufficient measure, that’s helpful to
But you have agents with one sale getting a 5.0. And you have agents with 40 sales getting a 5.0.
And maybe they’re—I occasionally saw a 4.9 or a 4.8. I don’t know what that agent did to merit
a lower rate. But in any event, they’re all very, very high. That is very misleading, almost
bordering on the fraudulent.
MATTHEW MANDELBERG: Stephen, do you mind?
STEPHEN BROBECK: I’m finished.
MATTHEW MANDELBERG: I need to keep it going. Brian, and then Panle after.
BRIAN LARSON: There’s a lot of information out there. But it tends to fall prey to one of two
flaws. One is that it’s quantitative. So I might see how many sales a particular real estate agent
took part in or how many listings they have. Now, you can find that by looking on Zillow or
Realtor.com and I have no reason to believe that those statistics are particularly wrong. I don’t
know that they’re right either. I don’t know if anyone has studied that. But I don’t think that’s a
very good reason to choose an agent either. Because you don’t know where those numbers arise
The second problem is the hamburger problem. I eat a lot of hamburgers. I like hamburgers a lot.
And so when I go on Yelp, let’s say, and I look at reviews for hamburgers, I read reviews by
other people who eat a lot of hamburgers, right? And so I can expect that the people writing the
reviews actually know what they’re talking about when they write a review of a hamburger joint
because they also have eaten a lot of hamburgers.
When someone writes a review of a real estate agent, they’re writing a review of someone who
has provided them services that they may not fully have understood in a transaction they do once
every five to seven years in their lives, right? So they’re not as good a judge of what it is they’re
getting. And so I can’t rely on their evaluations as well. And so when I’m choosing, I have to
choose a lot more carefully. And I did a transaction last fall where I found realtors who did a
bunch of transactions in the town where the listing I had was located. And I interviewed them
and I read the reviews.
But the reviews were mostly useless and the interviews are what helped me choose the agent that
we worked with. So I think the challenge is that there’s not—I just don’t know yet what kind of
information we could use to make it easier for consumers to choose agents and brokers on
quality. Because that information is either not available or not reliable.
MATTHEW MANDELBERG: Panle.
PANLE JIA BARWICK: So I agree with both Stephen and Brian that reputation is very
important and agents compete aggressive on quality. And there are a couple of studies that
show—there’s a recent paper by Sonia Gilberg in her job-market paper in which she used
CoreLogic to study 500,000 agents in the past, more than a decade, using 60 MLS markets. And
what she found, which is similar to the previous studies, is the past performance of the agents are
very useful at predicting their future performance.
As a result, I think this would actually be an extremely valuable service that NAR or the MLS
could provide to consumers. Part of the debates about the ratings on Zillow is, are they used or
not. We can actually be a lot more modern. Use that past data we have. We have so much
information on agents’ transactions, at least in price, how many transactions we have, how long
the property stayed on the market, all the different attributes, metrics that we can use. And
actually, NAR does this, to some extent, to give to the very top best agents. But what if we
provide this information on many agents?
What if we provide this using historical data to speak about the quality of the agents? I think
having this will reinforce the importance of reputation. This will also help consumers to get
better service. I do feel this is one area that we could improve on and benefit everybody. The
high-quality agents will now have an easier time accessing consumers. Consumers will be
matched with higher quality. This is a win-win situation.
MATTHEW MANDELBERG: Katie.
KATIE JOHNSON: So I guess first, no idea is bad. Every idea should be considered and so
worth considering. But what you said was that we have that data. And we, NAR does not—
PANLE JIA BARWICK: MLS, sorry.
KATIE JOHNSON: But MLS has that data, but I’d say no—the brokers have that data. And
that’s the brokers’ and the agents’ productivity. That’s theirs to use in the free market to use
however they want. And they should be empowered to do that. Like most professionals and
service providers, they have to market themselves in the way that they think will be most
effective to reaching the customers that they want.
Some of them might think that their transaction history is impressive and they’ll use that. Others
won’t. Others will use price, others will use quality, others will use their therapy skills. So, like,
they have a lot of different ways to market themselves. And they should just be allowed to do
that. I don’t think that it should be mandated to them on how they do that.
BRIAN LARSON: Matthew, can I just chime in? I think that if there is a model of the kind that
you described, that can be described in a way that is unambiguous so that a broker, for instance,
could say, OK, I’m going to look at my agents’ work and I’m going to look at the work of other
agents in the MLS, and I’m going to calculate their Panle index. Their Panle index—and I’m
going to show that we have a higher Panle index so that consumers should work with us. I think
brokers would be interested in that. Some—well, the brokers who had low Panle indexes would
not be interested in it. But the ones who had high ones would be interested. And it would be
good to have a model that was unambiguous because then, the comparisons would be apples to
Some of the comparisons that brokers have done—and anyone who’s managed an MLS knows
that brokers will slice and dice the data any way. And transactions that close on Tuesdays during
the winter months, our firm outdoes every other firm in the metro area, not counting this one
county. That’s the kind of thing that we’ll do to be number one. So I mean it has to be
unambiguous, I think. But there’s no reason that it isn’t something that could be implemented.
But I agree with Katie that the implementing of it happens at the brokerage level. It doesn’t
happen with the MLS’s blessing a broker with a high Panle index. It happens with the brokers
proving they have it themselves.
STEPHEN BROBECK: So a quick question to you, Brian. How do you make that happen?
BRIAN LARSON: Well, I think you have to give them the model. So I’ve read some of Panle’s
articles. And I have to admit I can’t understand a lot of them because there’s a lot of Greek letters
in them. So I think you have to make that model interpretable, both to consumers and to brokers.
And then once the model is interpretable, you have to see, well, is there a way to map that onto
Because the MLS may not be recording—MLSs record data to achieve a purpose and it’s not that
purpose. So the question is, does the data the MLS records, does it map onto that model? And if
it does, then I think you tell brokers, look, here’s a model. Go for it. Have fun. And I think the
MLS rules and NAR policies would actually require an MLS to allow brokers to do exactly
that—if they had a model.
STEPHEN BROBECK: Consumers get access?
BRIAN LARSON: Well, the brokers who are doing well in the model would definitely want
consumers to have access. Some other brokers might want to be kind of quiet about it.
MATTHEW MANDELBERG: Katie, did you—I saw your—
KATIE JOHNSON: I—Yeah.
MATTHEW MANDELBERG: So I had two directions I wanted to go here. So I think actually,
we’ve been talking a bit about quality competition. I’d like to switch back to price competition.
Earlier in the day, we heard that—I think a number of the portals mentioned that they didn’t have
visibility into the data on the commissions.
And then on panel two, we heard varying types of comments about that they would like to make
available certain data that includes commission numbers, but that they felt that they could not
actually do so. And so I wanted to just understand to what extent there is data that gets to price
transparency in the industry? Katie?
KATIE JOHNSON: So the NAR Code of Ethics requires that that price information be revealed
to consumers through the seller or the buyer. So the easiest ways are for the broker to proactively
give that information, the buyer’s agent giving that information to the buyer. Absolutely, if a
buyer ever asked where that information is, they would get it. There is no bar to that. And then
whether or not it be distributed publicly for the nation to see is just a local discretion of the MLS.
And Joshua Hunt did mention that he’s doing that now, so he must have petitioned his MLS to
change their rule on that.
MATTHEW MANDELBERG: Brian.
BRIAN LARSON: So I think, at least with regards—it’s important to understand, first of all, that
MLS’s don’t collect information about the gross commissions. I think that’s already been said
today, but I think it’s worth repeating, right? So the MLS doesn’t know what the listing broker is
charging the seller. The only commission number that the MLS gets is the number that the listing
broker is offering to cooperating brokers.
And with regard to disclosing that information, I think there’s a tension. And it’s one that I think
Panle, in our conference call, we talked about last week or the week before. I can’t remember any
more now. But the fact is, disclosing the commission that the listing broker is offering to the
cooperating brokers creates an opportunity for that price to end up somewhat coordinated. And it
allows brokers to discipline each other along that coordination. So there’s a sense in which we
don’t want to disclose it. We’d like to have a way not to disclose it.
And further, disclosure doesn’t seem to me that is going to help that problem. If you were to
move to a more flexible model, different brokers might get paid different amounts by the listing
broker. But then you wouldn’t necessarily want that disclosed on a consumer-focused website
because then brokers would go back to being able to discipline each other over cooperating
compensation. So I think this question—and Katie has pointed this out before—brokers in the
industry are very anxious to talk about commission at all because they’re afraid of getting sued
for antitrust violations.
In fact, we’ve had Twitter comments during the day today, why are you guys talking about
commissions? That’s dangerous, dangerous. I don’t think anyone has said anything dangerous
here so far today. But I think that is a concern or a question. So there is an ethical obligation to
disclose. Most states require disclosure at an opportune moment. That seems to me like, as long
as brokers are complying with those requirements, that we’ve satisfied that requirement. I’m not
sure what more disclosure is going to achieve.
MATTHEW MANDELBERG: Stephen?
STEPHEN BROBECK: Yes, the laws do require disclosure of the overall commission. And
maybe buyers learn during the closing a little late what the commission split was.
But the fact is that that information is not widely known. In fact, it’s not well known to very
many consumers at all. The entire industry has access to those commission splits. Consumers and
everyone else ought to have access to those splits as well. And nontraditional brokers should not
be disciplined by MLSs for revealing those splits, particularly to their buyer clients.
MATTHEW MANDELBERG: Panle?
PANLE JIA BARWICK: I hate repeating myself. But I think, if we think about the alternative to
this model, where the buyer pays either the buying commission and the seller pay the selling
commission, then we don’t have to have this current arrangement where the MLS listing, the
agent, knows the commission to all of the properties. The agent knows the buying commission
on every property. And we already argued that the literature shows evidence of—as Brian
mentioned—perhaps this could, as a facilitating device, for agents to discipline each other or
maybe there is evidence of steering.
So here’s where I hazard that, I could see in that ideal world, we wouldn’t have some of the issues
that we are discussing. But right now, in the arrangement, the agent has all the pricing
information. The consumers, if they don’t ask, they’re not presented the information. So making
this information public to anybody, I agree with Stephen, is very valuable.
But I also worry that, given that consumers don’t seem to search aggressively across agents—the
NAR 2017 survey showed they mostly just work with the first agent they interviewed—this
might not work to the best interest of the consumer. I think the ideal world is still to have them
separate so we don’t have this issue of disciplining each other or having this—
MATTHEW MANDELBERG: I just want to ask a follow-up to Brian. Generally, as antitrust
lawyers, we distinguish between sharing price information among competitors versus advertising
prices to your consumers. And so I’m just wondering if, when you say that there is a concern
about sharing price information with consumers, where that antitrust concern would arise?
BRIAN LARSON: Well, because it had to first be shared among competitors before it could be
shared with consumers. The only reason I would have my competitors’ prices is because they’ve
given them to me, which raises exactly the issue that antitrust regulators are frequently
concerned about. And I think Panle’s suggestion and possible solution is well-taken. But I think
what we need to study or what we need to understand is how it could be how it could be done.
So for instance, my understanding is that Fannie and Freddie will allow a buyer to finance a
buyer’s broker’s commission that’s paid at closing as part of the transaction costs for a mortgage.
But my understanding is also that a lot of mortgage underwriters pushed back against that. They
don’t believe that that’s the case. So if we’re going to say, well, we’re just going to make this
change without first making sure that the rest of the structures that are out there—the mortgage
industry structures—are prepared to deal with it, we could introduce a lot more destructive
disruption than we want.
So I think that’s important to think about first. And we also just don’t know whether it works,
right? So I think it would be interesting or useful to see if there’s a way to test that empirically
before we say, well, let’s just throw up our hands and do it a completely different way. Because I
think theoretically, what you say makes sense, but practically, all sorts of things can crop up. I
mean in theory, theory and practice are the same, but in practice they’re not.
STEPHEN BROBECK: And if the disclosure is just to the individual buyers, there are all kinds
of opportunities for deception. What’s really essential is that those commission splits, that
information needs to be available or easily accessible to researchers, to the FTC and the DOJ, to
the Consumer Federation of America, and to intrepid reporters. That would really increase
transparency. That’s the way it works in so many different markets.
You can’t expect all consumers to do a thorough search. They don’t. We have the research on
that. What you do is you have their representatives in government, in public interest groups, who
have access to information and who analyze that information and make it available through the
press and social media. And that’s what disciplines markets to a large extent.
MATTHEW MANDELBERG: So another thing that’s come up here has to do with portals and
the role that they’re playing in this market. Obviously, we’ve talked a lot about the information
that is getting to consumers. And portals are clearly playing an important role in that process.
And so I wanted to just, as an open-ended question for the group, to what extent are they playing
a positive role in terms of competition. And what, if any, are there concerns that could be raised
about the role they’re playing in the industry?
KATIE JOHNSON: I’ll just start by just explaining a relationship that I think a lot of—there’s
maybe some confusion about. And that’s our relationship with Realtor.com. Realtor.com is an
excellent portal that provides advertising of listings on a national scope. But it’s not owned or
operated by the National Association of Realtors.
It’s owned and operated by Move, which is owned by News Corp. We license our trademark to
them for their use, so we’ve had different relationships through the years. So it’s independently
operated outside of our rules or our oversight. And then I’ll just add that—no, that’s it. Thank
MATTHEW MANDELBERG: Anyone else want to weigh in there on the portals?
STEPHEN BROBECK: Yeah. Well, there are positives, there are negatives. A big positive is a
lot more information available to consumers about a larger number of properties and more
information about those properties. The Zestimates, I think, they’ve been criticized from time to
time. And then Zillow works real hard to try to improve the algorithm. And then also, as I’ve
learned recently, at least Zillow, maybe some of the other portals, will list for sale by owners.
I think that’s very important for there to be a truly healthy competitive marketplace—for the
FSBOs to be able to get access to those large public databases. Now, on the other hand, if the
portals are really large they can wield oligopolistic or even power. And interestingly, it’s first
against the agents and the brokers. They charge them a high price for the advertising.
And there’s some evidence—I haven’t seen a definitive study—some evidence that that tends to
keep the commission levels up and that’s not a good thing. And I’ll stop there. I don’t know where
the large portals are going. They could go several different directions. Some would be very proconsumer. Others would not be so pro-consumer.
MATTHEW MANDELBERG: Brian?
BRIAN LARSON: Well, we seem to have a history of a dominant portal. So for a while it was
Realtor.com, and now it appears to be Zillow. And I think that Stephen has really put his finger
on the two sides of that coin. I mean, one is that it’s great to be able to go someplace and get
accurate listing data if you’re a consumer. I’m not sure the extent to which those sites deliver
accurate listing data. That’s a subject of debate in the industry. But they seem to be winning in
terms of getting people to come and check them out.
But then they also sometimes exhibit practices that are intended to hold onto their dominant
position. And that was true of Realtor.com before. I think it’s been shown to be true of Zillow
now. And particularly, in reference to Stephen’s comment about commissions, if you listen to
Zillow’s earnings calls, they say that they get their money from real estate brokers and they hope
to get a whole bunch more money from real estate brokers.
That’s not removing costs from the transaction. That’s not introducing efficiencies in the
transaction. Now, it may be introducing a kind of efficiency in terms of search for the consumer,
right? I’m not saying that they’re not valuable. But it doesn’t press dollar costs out of the
MATTHEW MANDELBERG: We also heard a lot, earlier in the day, about the role of lead
generation. So isn’t that possibly a very favorable role that these portals are playing for the agents
BRIAN LARSON: If they get it for free, sure.
MATTHEW MANDELBERG: I wish all of my services were free as well.
BRIAN LARSON: It could be a very valuable service, sure. But the agents are going to pay for
that service. And that’s a service they might not have been paying for before. And so it’s not
going to remove cost from the transaction. It might, again, result in a kind of finding cost
efficiency, a search cost efficiency. But in terms of the dollar cost, it doesn’t remove any costs of
MATTHEW MANDELBERG: So one last set of questions I wanted to raise for the group is the
role that listing off an MLS is or could be playing in the industry right now. We heard earlier
examples that this might be more prevalent. I think two cities that were mentioned were San
Francisco and Seattle. And so I’m wondering, what are the concerns that might arise, the more
prevalent this practice is? And what might be the arguments in favor of pursuing such a practice?
KATIE JOHNSON: You want me to start? So again, just a little background—there was a nice
discussion, I think, about pocket listings in the first and second panels. But just to reiterate that
there is no generally accepted definition of what a pocket listing is. There are lots of terms.
You’ve heard pocket listings, office exclusives, coming soons, off MLS, off market. But, in
general, they all mean trying to sell the house without using the MLS—so a broker-agent using it
outside the MLS.
And so an office exclusive would be sharing it with their other brokers and agents within their
office before it goes to the MLS. Coming soon might be using the Zillow tool, “Coming Soon,”
to upload the listing before you upload it into the MLS and try to sell it before your cooperative
competitors have a chance to show it. So as they said too, I think it is growing. It is a problem in
that it’s increasing in prevalence in markets and in market share.
I think Art Carter from California might have said it was about 10% to 20% of the off-market
activity, which sounds very high. But it’s not exclusive to California either. It is growing across
different geographies. So a concern is that the information is not in the MLS. The listing may not
be exposed to the widest market, so it may not be getting the best price. That’s really the
consumer concern, I think, in withholding them.
MATTHEW MANDELBERG: Stephen.
STEPHEN BROBECK: Well, I think that all consumers should be able to buy and sell property
without necessarily utilizing the services of an MLS. My concern here is that—and I haven’t seen
data—but my strong suspicion is that quite a large percentage of those pocket listings are
basically in-house sales. In-house sales, there are some inherent conflicts there.
If, in fact, they make a quick sale—an in-house sale allows a very quick sale. Maybe the buyer
and the seller is prepared to sacrifice something financially for that. But the commission is less
likely to be lowered if, in fact, there’s an in-house sale. So it’s a complicated issue. And it will be
interesting to see if there’s research on this, which we haven’t seen, as to whether it’s growing or
declining and, in fact, what form it takes.
MATTHEW MANDELBERG: So I’m also curious. To the extent that we have the internet
today, why not just list directly to the internet through one venue or another, through a portal
directly, when they accept it? So Brian, what do you see as some of these advantages or
drawbacks to listing off of an MLS?
BRIAN LARSON: Well, are you asking a question about a single portal? Or are you talking
about the off-MLS?
MATTHEW MANDELBERG: So I guess I’m sort of following up both, putting the original offMLS question to you. But I also am just sort of adding a question of, if it is the case that you can
just go straight to—post the information about your house on a Facebook, or to the extent that
portals of the world will accept your listing directly without going through an MLS—why is that
a negative from a competition perspective?
BRIAN LARSON: OK. So in terms of brokers not putting things in MLS, I think there is a
variety of reasons they might not. And Stephen has addressed some of them. I think that the
challenge for the MLS—I’m guessing there aren’t many MLSs that like off-MLS listings. I know
a number of them have worked in a variety of ways to accommodate them in one way or another.
But I don’t think they do it out of delight for their existence.
The problem for the MLS is that the alternative really for them is to be strict and say, well, you
have to put it in. And an MLS can’t really do that, right? If the seller says I don’t want my listing
in an MLS, and the broker says I’m in an MLS that requires me to put all my listings in, I’m
going to lose this transaction if I can’t comply with the seller’s wishes. That’s the kind of thing
that gets MLSs sued for antitrust violations, right? It’s an inequitable limitation on membership.
So I think that MLSs have to allow sellers to make these choices. There might be a disclosure
issue. And again, maybe we can make better disclosures to sellers about the consequences of not
having their listings in the MLS. But certainly, this question is sort of like, can consumers put
their listings where they want or can they have their brokers do that? I think that’s true. The
notion of a single portal seems to go in the opposite direction of where we want to go.
I think Stephen and I have both expressed some concerns about the market power of the portals
that exist already. If you replace them with some kind of single, monolithic place where
everything goes, then that single monolithic place becomes one single point of control. And
that’s the kind of—it’s not divide and conquer, it’s unite and conquer, bring something together in
a bigger entity, then you only have to control one entity to control the market. And I don’t think
that’s something that we want in real estate.
MATTHEW MANDELBERG: Katie.
KATIE JOHNSON: So in response to your question about why not just let that happen, I think
it’s because it would lead to the total annihilation of the MLS. And that’s not good for consumers.
MLS is the envy of the world, of regulators and countries across the world. They recognize the
pro-consumer benefits of having an orderly marketplace where all of the information is located
for that market. Real estate is local. There doesn’t have to be a national database because people
who are buying and selling are local.
But they’re in the MLS, it’s orderly, it’s there, it’s accurate, it’s reliable. That’s very pro-consumer.
That allows consumers to have more accurate comps and understand the whole inventory
without having to go to broker to broker or website to website to figure out what’s even available
to them, let alone, how much it should cost. So I think that is a huge risk that we’d face if we
allowed this fragmentation of listings.
MATTHEW MANDELBERG: I’m going to ask just a—we have time for just one question from
the audience. And then I’m going to ask our esteemed panelists to give their closing remarks. So
the one question, as I read it, I think is directed towards you, Panle. And it is, can you please
flesh out what you mean when you propose that buyers pay their own agent?
PANLE JIA BARWICK: Good question. So I’m basically explaining the results from the
literature that document some of the not-so-desirable consequences of having the seller paying
both commissions. And I’ve identified several. And so it seems like the natural question, since
that seems to be creating the issues—competition, potentially, some of the disciplining that Brian
was talking about—it seems that the most direct answer and solution is to separate them.
While I do recognize from this panel discussion and research on my own, that this is not easy. So
I’m not necessarily advocating this saying we do this right now. So I actually think that Brian had
an excellent suggestion. If there are cities or MLS listings that are willing to experiment, maybe
we can see what happens. And that would be great for the society—more data for us to study
MATTHEW MANDELBERG: So I think we’ve heard a very lively and dispersed point of views
with a range of answers and perspectives.
BILL ADKINSON: So we’re really out of time. So I’d asked each panelist to offer a minute of
either closing thoughts or takeaways, starting with Brian.
BRIAN LARSON: Well, I’d like to thank you, again for allowing me to be here today. It’s been a
real pleasure. This has been a great panel. And the other panels I enjoyed very much as well. I
think the takeaway I have here is that there are definitely points of friction in the real estate
business. And I would expect that any business we can think of probably has points of friction. I
think continuing to study what those points of friction are and identifying potential places where
we can improve practices is always a good idea.
And I’ll just reaffirm the comment I made and I think that Panle just echoed, it would be good to
have lots of information or lots of knowledge about why we’re making changes if we’re making
changes. So knee-jerk reactions to anecdotal stories, which is some of what we’ve heard here
today, or under-tested solutions to problems that are identified based on theoretical or empirical
perspectives that are limited in their scope, they come with risks. And so we need to think about
what those risks are and make sure we’re not throwing the baby out with the bathwater. Thank
BILL ADKINSON: Thanks, Panle?
PANLE JIA BARWICK: There’s one thing that I think that the panelists haven’t touch on that I
want to bring up, which is, this is an industry where the consumers don’t participate very often.
In a book called Fishing for Fools by two Nobel laureates, it makes the point very clear. Most
consumers will sell on average a couple of times. As a result, educating consumers is very
And I think we can—in the contracts, perhaps besides providing the commission rate, it would
be very useful to provide guidance of the actual commission that consumers are paying. If they
recognize it’s going to be $30,000, perhaps they’re going to be more mindful of searching for
alternative solutions. And similar for buyers too. So I think that’s an area where we can, perhaps,
bringing more information and educating consumers, and that would be useful.
BILL ADKINSON: Thanks, Katie?
KATIE JOHNSON: Thank you very much for inviting me. It’s occurred to me that we all have
the same interests. We have the same endgame up here. And that’s to deliver a better experience,
the best experience, in the home buying process for consumers. We do it in a variety of ways.
And you guys have good ideas on how we can continue to do that. So that’s encouraging, I think.
Chairman Simons said in his opening remarks that we the FTC want to help consumers feel more
confident when they buy and sell a home. But that’s exactly what we want to do too. And that’s
what I think the current market environment supports. And that’s what, you know, we continue to
strive to educate our agents and brokers and raise the level of professionalism so that consumers
are ultimately very confident in those transactions.
BILL ADKINSON: Great. And last, Stephen.
STEPHEN BROBECK: Well, first, I just want to thank the FTC and the Department of Justice
for decades of work to ensure a competitive residential marketplace. They’ve made progress.
This is a very complicated marketplace and in part deliberately so with huge barriers to effective
And I guess I would just say that one thing that would really help is more transparency in prices.
And if you look at my written comments, there are several ideas that we have. And I think we
need to focus more attention on price transparency. And that will really discipline the market as
effectively as anything. But the marketplace, in general, is working fairly well for consumers.
BILL ADKINSON: Great. Thank you. If you could keep your seats, I think Matthew is going to
introduce our closing speaker.
MATTHEW MANDELBERG: Yes. So I just wanted to introduce Robert Potter, who’s the Chief
of the Competition Policy and Advocacy Section at the Antitrust Division of the Department of
ROBERT POTTER: Thank you, Matthew. I will keep my remarks brief so everybody can go
Why did we spend the day here? All the speakers on, I think, every single panel noted that
residential real estate brokerage market is extremely important to consumers. Buying or selling a
home is often the largest and most important financial transaction individuals and families can
undertake. It is important that the antitrust agencies focus on issues that are crucial to consumers.
The antitrust agencies are focused on competition and the benefits it has proven to provide to
American consumers—lower prices, higher quality, more innovation, and greater choice. These
workshops are an important tool to delve into the dynamics of the market and understand recent
developments from market participants, experts, and other key stakeholders to enable us to
determine the potential for future efforts to protect and promote competition in this important
The question of how far we have truly traveled in the past decade was a theme throughout the
day. We learned how to technology has led to consumers now taking the lead in searching for
their homes. We learned about the roles of MLSs and agents in that process, and the role of new
industry players in getting some, but clearly not all, of that data to consumers.
We learned of innovative fee and service models and how industry developments may be
impacting their adoption. A number of the panelists mentioned that maybe we are near a tipping
point in the industry. And I am interested to know whether, when we have this workshop again
in the next decade, that tipping point will have been reached by then.
Lastly, we heard diverse perspectives on the state of competition in the industry, particularly,
with respect to compensation. All of this reminds us about the importance to the agencies, the
stakeholders, and experts that we continue to study and deepen our understanding of competition
in this vital and evolving industry.
I want to highlight, for the public, that the opportunity to share insights on this industry
continues. Public comments for the workshop will be accepted through the FTC’s workshop until
July 31 of this year.
Now, moving on to my thank yous. First, I want to thank all of our panelists for their knowledge
and viewpoints that they have, in many instances, more than willingly shared with our audience.
One of the panelists earlier today noted that he couldn’t believe that he had to spend four
preliminary calls preparing for this. But that just goes to show that the preparation that these
people put in, and the thankfulness that we have for them for spending their time and efforts to
make this a good workshop.
I want to express a special thanks to the FTC for being a good partner with us in this workshop
and for hosting the workshop at their facilities. If you will bear with me just for a couple of
seconds, often the people who work on this don’t get mentioned by name and just get mentioned
in a thank you for the staff. I would actually like to mention some names with respect to this one,
if you could just indulge me.
For the DOJ, I want to thank Matthew Mandelberg, Matt Siegel, Erica Mintzer, Steve Kramer,
John Martin, and Emily Tifft for all their efforts in this regard. For the FTC, Bill Adkinson,
Keitha Clopper, Jess Drake, Karen Mills, Kelly Signs, Andrea Zach, Betsy Lordan, Nicole Jones,
Nathan Ottlewski, Jennifer Leach, Kristal Peters, Richard White, Philip Zanfagna, Dillon Smith,
Kathleen Curlee, Michael Greenstein, Cecilia Valadez, Ruxandra Giura, Jason Wolf, Rich
Custer, Bruce Jennings, James Murray, and Somethea Mam.
Finally, I would like to thank the audience, both you who have attended in person and those
watching on the web. We hope you found the discussion illuminating. I know we certainly did.
Finally, as a quick housekeeping matter, I think most of you got this on the way in. If you could
remember just to return it on your way out. And with that, our workshop is over. Thank you