This document is available in two formats: this web page (for browsing content) and PDF (comparable to original document formatting). To view the PDF you will need Acrobat Reader, which may be downloaded from the Adobe site. For an official signed copy, please contact the Antitrust Documents Group.

Comment No.: REW-0408
Received: 10/31/2005
Organization: AEI-Brookings Joint Center for Regulatory Studies
Commenter: Hahn, Robert
State: DC
Attachments: None


Joint Center Logo


Bringing More Competition To Real Estate Brokerage

Robert W. Hahn, Robert E. Litan and Jesse Gurman*

November 2005

* Robert W. Hahn and Robert E. Litan are the co-founders and directors of the AEI-Brookings Joint Center for Regulatory Studies. Jesse Gurman is a research assistant at the AEI-Brookings Joint Center. The authors would like to thank Scott Wallsten, John Weicher, Susan Woodward and Norm Miller for helpful comments. The views in this paper are those of the authors and not necessarily those of the institutions with which they are affiliated.

Joint Center Logo

AEI-Brookings Joint Center for Regulatory Studies

In order to promote public understanding of the impact of regulations on consumers, business, and government, the American Enterprise Institute and the Brookings Institution established the AEI-Brookings Joint Center for Regulatory Studies. The Joint Center's primary purpose is to hold lawmakers and regulators more accountable by providing thoughtful, objective analysis of relevant laws and regulations. Over the past three decades, AEI and Brookings have generated an impressive body of research on regulation. The Joint Center builds on this solid foundation, evaluating the economic impact of laws and regulations and offering constructive suggestions for reforms to enhance productivity and welfare. The views expressed in Joint Center publications are those of the authors and do not necessarily reflect the views of the Joint Center.

Executive Director


Stanford University
University of Maryland
Common Good

Massachusetts Institute
of Technology

Stanford University
Stanford University

University of Michigan

Milken Institute
Massachusetts Institute
of Technology
Harvard University
University of Chicago
Harvard University

All AEI-Brookings Joint Center publications can be found at
© 2005 by the authors. All rights reserved.

Executive Summary

This paper provides an economic analysis of the residential real estate brokerage industry. We find that the traditional model for residential real estate brokerage services may be dated, and could be improved substantially with some public policy interventions that spur innovation.

We believe that there are numerous barriers to entry that are slowing the emergence of new models for serving consumers. Some of these barriers are likely to be anti-competitive. Examples include discrimination against new brokerage models and online brokers who wish to join multiple listing services; state legislation that would require minimum service requirements, effectively preventing "a la carte" offerings; and prohibitions by real estate commissions on providing rebates to customers. In our opinion, none of these practices should be allowed.

We offer three broad policy recommendations: First, antitrust authorities should carefully scrutinize efforts to limit competition in the residential real estate brokerage market. Second, state governments should refrain from adopting laws or rules that inhibit competition in real estate brokerage. Third, Congress should allow the Federal Reserve Board and the Treasury Department to permit banks, which have long been natural potential entrants into this business, to offer residential real estate brokerage services through separately capitalized affiliates.

We do not know which business models are likely to succeed in the marketplace for residential real estate services in the future. We do believe, however, that judicious public policy interventions could have a marked impact on improving services and lowering costs for home buyers and sellers.

Bringing More Competition To Real Estate Brokerage

Robert W. Hahn, Robert E. Litan and Jesse Gurman

1. Introduction

For most Americans, the largest single purchase they will make is a house.1 Most of them will do it by engaging the services of a licensed real estate professional.2 For brokers, selling houses is a big business. In 2004, the estimated revenue of the brokerage industry was $60 billion.3

In recent months, the residential real estate brokerage industry has been featured in the news. A number of publications have reported the rise of newer discount brokerage services, operating on and off the Internet, that have begun to offer buyers and sellers cheaper alternatives to the commissions of between 5 and 6 percent that have traditionally been charged by real estate brokers.4 In March, the U.S. Department of Justice (DOJ) filed an antitrust suit against the Kentucky Real Estate Commission for prohibiting agents from giving consumers a rebate on some of the commissions they pay.5 Legislatures in other states have recently introduced or enacted bills to prohibit real estate agents from offering more limited service, which they can perform at a lower fee. Several states tabled the bills only upon direct warning from the Department of Justice and the Federal Trade Commission that such laws were anti-competitive.6

In perhaps the most important action of all, in September, 2005 the DOJ filed an antitrust suit against the industry's national trade association-the National Association of Realtors (NAR)-over their website rules. The lawsuit came after a 20-month investigation of the trade association's rules, issued in response to the advent of online brokerage.7 These rules would allow NAR member agents and brokers, if they so choose, to withhold their listings from being displayed on other brokers' websites, including the websites of online brokers.8 Representatives of the NAR have denied that their rules thwart competition and instead argue that brokers are entitled to control how their customers' listings are distributed and displayed.9

So what should consumers really think? Do the new brokerage models promise real cost savings for buyers and sellers? Is the traditional industry and its trade association truly making it more difficult for new entrants to compete, to the detriment of home buyers and sellers? If so, is the DOJ on the right track, and what else can or should be done to promote competition in this industry? These are among the questions we address in this paper.

In brief, we reach several conclusions:

  • It is difficult to tell whether the industry is acting anti-competitively by simply examining commission rates. From 1991 to 2004, average commission rates have declined nationally from about 6.1% to 5.1%.10 According to these numbers, claims that commission rates are fixed do not appear to have empirical support. Nonetheless, we cannot rule out the distinct possibility that the structure of the industry, as well as certain practices of some of its members, has led to rates that are above what they would have been in a more competitive environment. Furthermore, U.S. rates appear to be above those of other advanced countries. Average commission rates in other developed countries appear to range from 1 to 10% with a mean of 3.5%. Thus, U.S. commission rates are about 1.5 to 2.5 percentage points above the average.11 While foreign commission rates may not be appropriate for the U.S. because of differences in the housing brokerage markets and services rendered, at the very least they argue for more scrutiny and study of the U.S. industry.
  • The Internet has the potential to significantly reduce the costs of buying or selling a house as well as increase the value of real estate services. First, it can dramatically reduce search costs for buyers and sellers, and also improve the matching of buyers and sellers. In a competitive market, cost savings to brokers should be reflected in lower commission rates, but this does not appear to be happening yet on a wide scale. Second, the Internet should facilitate the disaggregation of real estate services, such as listing and searching for houses, negotiating a contract, and getting title insurance. This disaggregation should improve efficiency by allowing consumers to choose only those services they need and pay accordingly. For example, sellers can use the multiple listing service in many areas to show their house for a relatively modest fee.12 Recent developments indicate more Internet-based entry into the listing business, from both existing and new competitors. If this entry is not inhibited, it could encourage more competition in real estate brokerage, and in turn, lower fees.
  • There are numerous barriers to entry that are slowing the emergence of new models for serving consumers. Some of these barriers are likely to be anti-competitive. Examples include discrimination against new brokerage models and online brokers who wish to join multiple listing services; state legislation that would require minimum service requirements, effectively preventing "a la carte" offerings; and prohibitions by real estate commissions on providing rebates to customers. In our opinion, state governments and regulators should reject policies that restrict competition in the real estate market. Furthermore, while private entities such as real estate associations are free to pursue their own economic interests, it is perfectly appropriate for both federal and state authorities to scrutinize their practices and act to ensure that they do not violate antitrust laws.
  • Finally, federal financial policy makers can and should enhance competition in real estate brokerage by authorizing banks to engage in this activity through separately capitalized affiliates. The Federal Reserve Board and the Treasury Department have the authority under the Gramm-Leach-Bliley Act to give this permission.13 Resistance to the participation of banks in real estate brokerage is yet another instance of traditional brokers being protected by a barrier to legitimate competition. Congress should allow this competition to proceed because it is likely to benefit consumers. One possible beneficial effect of recent federal antitrust investigations is that it may encourage elected officials to promote more competition in real estate brokerage, or, at least not to interfere with regulators who are trying to accomplish this important objective.14

2. The Traditional Real Estate Brokerage Model

For decades, buying and selling a home in the United States has been a routine process. Residential real estate agents have traditionally provided a bundle of services to both buyers and sellers. For example, the local multiple listing service (MLS), which is a directory of listings typically maintained and paid for by local real estate firms, enables sellers to list their properties and buyers and agents to view these properties. Agents also often help with marketing a house through advertising and open houses, negotiating a price, and addressing contracting and closing issues. Examples include helping with inspections, mortgage insurance, and financing, to mention a few.

Home sellers are typically required to enter into an exclusive agreement with the agent trying to sell their home, known as the "listing agent." This agent represents the seller in dealing with all potential buyers and is typically paid a predetermined fee, in the neighborhood of 6% of the sale price, which is frequently split among agents and brokers who facilitate the deal. To help reach interested buyers, listing agents usually offer part of this fee to any other agent who brings in a buyer upon completion of the sale. If the listing agent is able to find a buyer without the help of another agent, he can act as a "dual agent" and keep the whole fee. Under this system, home buyers do not have to contract with the agent helping them, or pay them directly for their services. However, the home buyer does share some of the cost of her agent's services, since the seller prices his house to account for the fee.15

Upon completion of the deal, the seller's agent and the buyer's agent (if there is one) often split the commission with their brokers. A broker typically helps and supervises agents, and requires an additional license.16 The split between broker and agent may be 50-50 for newer agents, though agents often get to keep a higher percentage once they become more skilled and experienced. Under some models, a broker will simply charge a flat fee for resources, such as office space, phones and copying, and let the agent keep the whole commission.17

A brokerage, which can include one or more brokers, agents and support staff, can either operate independently or belong to a local, regional or national real estate network. Brokerages that choose to franchise can benefit from the advertising and name recognition of these networks. A local franchisee pays fees to the franchisor for the brand name, office space and other support by splitting revenues in some proportion or paying a fixed amount per month, quarter or year and keeping the commissions themselves.

This traditional structure is notable for two reasons. First, we believe that the involvement of multiple parties and the unique splitting arrangements make it difficult for buyers and sellers to pay for services according to their needs. Second, the commonality of the structure across firms and its persistence over time suggest the possibility that alternative models have not had a fair chance to compete. Both of these issues are examined further below.

3. Is the Real Estate Brokerage Market Efficient? Academics Aren't Sure

There is a substantial literature on whether the residential real estate market for brokerage services is efficient. By efficient, we mean maximizing the benefits to consumers as well as profits to brokers. The literature on this subject is inconclusive, in part because there is an absence of available data on how these markets actually operate. A number of academics have examined the relationship among various factors, including commission structure, housing prices and the quality of service.18

Many scholars and casual observers suggest that commission rates have been relatively stable, around 6%, over the years, suggesting there is not much price competition among agents. Given the relatively free entry into agent market, some conclude that a form of price fixing, or at least informal collusion, keeps fees at an above-competitive level. This, in turn, could result in losses to consumers.19

One study developed a theoretical model and calculated the socially optimal commission rate to be 3%.20 Another argued that commission rigidity results in a socially wasteful oversupply of underproductive agents in high-priced cities, compared to lower-priced ones.21 An empirical study of commission rates across 30 countries using data from 1999 found that fees in the United States are significantly higher for comparable real estate services offered in other industrialized nations.22

Other academics have maintained that the market does exhibit competitive characteristics, and that a theory of widespread collusion is unconvincing. Some studies suggest that paying an agent something above his minimum acceptable commission rate could yield benefits for the seller because of the additional effort the agent might spend on selling the house.23 Other studies argue competitive rates are unlikely to be achieved due to some of the fixed costs involved in providing real estate services.24 Furthermore, one could argue that the use of the MLS platform to bring buyers and sellers together could give rise to prices that deviate from costs.25

Many of the arguments on both sides are based, to varying degrees, on the assumption that fees are stable. But some literature suggests otherwise. Using a large sample of housing data for a single area, one study suggests that commission rates are sensitive to local market factors, including house prices.26 This finding is consistent with other studies that have found a connection between higher housing prices and lower commission rates.27

Furthermore, there is some limited evidence that commission rates have generally fallen over time. As mentioned above, the average national commission rate charged on a house transaction is estimated to have fallen from 6.1% in 1991 to 5.1% in 2004-about 17%. During that period, housing price increases have outpaced inflation by 55%, particularly in the last five years.28 Commission revenues from the sale of an average priced house have thus risen, but at a rate that is only slightly higher than inflation.29 But these are national estimates. It is possible that in areas of the country where home prices have been rising most rapidly-such as on both coasts-that commission revenues there have considerably outpaced inflation. We cannot know definitively without having more detailed commission data by state and locality. Clearly, there is a need for more data to be collected, especially by an impartial source.

While there is not much evidence to support the view that agents have competed vigorously on commission rates for the average home, this situation may be changing.30 Still, if commission rates are artificially high due to barriers to entry in the industry, then more vigorous competition could lead to more dramatic benefits for consumers. The commission on the exchange of a house frequently represents a significant portion of the transaction cost. If this transaction cost could be reduced, then it is highly likely that buyers and sellers of homes would be better off.31 Under the current system, it is very difficult for a buyer to reduce the cost of the transaction by searching for a home and representing himself.32

One of the ways that has been suggested for increasing competition has been to change rules governing access to the MLS. One review of the literature suggests that full access should be provided to all buyers and sellers in order to promote economic efficiency.33

4. Making Sense of the Academic Mishmash

While the academics fight amongst themselves about whether this industry is reasonably efficient, we think that there are reasons to be skeptical. In particular, the rules governing access to the multiple listing services have not received the attention they deserve in the academic literature.34

The MLS is a very powerful and useful tool for bringing together buyers and sellers. One reason the MLS in most locales has become a staple of the residential real estate industry is that the MLS delivers substantial efficiency gains to all who use it. Collecting all of the listings for a given region in one place significantly reduces the amount of time buyers and sellers-and their brokers-have to spend gathering information that is crucial to potential transactions.

But the cooperative networking relationship among agents in a regional MLS has the potential to give rise to uniformity in services provided and fees charged. In fact, up through 1950, at least some MLSs explicitly set fees that their members could charge, until such actions were deemed in violation of federal antitrust law.35 Since then, it has been illegal to set standardized rates among members, but there are other, more tacit ways of reinforcing such cooperation.

One possible means of tacit collusion is through the multiple listing service admission process. Terms of MLS membership are generally decided by the MLS board, which usually overlaps with the dominant local brokers and the state or local Realtor associations. Since MLSs are private entities, they set their own admission criteria, but for most MLSs, the terms are coordinated though NAR's national policy, and they can design it to keep membership limited to firms who will conduct their business in a particular manner.36

Another potential avenue for collusion is through boycotts, either as a group or by the dominant player. An MLS listing typically shows which agent is representing a home seller. Other agents can avoid bringing buyers to houses listed by agents who offer lower commissions or operate under an alternative business model.37

Skeptics may argue that incentives to compete on price are high, and that relatively free entry should inevitably result in such competition. But there are relatively few large brokerages in many areas, and they may be able to keep rates artificially high. Even some large cities have only a few successful brokerage companies, and since brokers are allowed to control fee policies among their agents, sustaining collusive behavior is easier than if each agent sets his own fee. Also, in many cases, membership in the MLS is predicated on membership in the National Association of Realtors, which requires membership in the local and state association of Realtors, an arrangement which further encourages a cooperative spirit and similarities in service offerings.

Some brokers view the listings as their property, as opposed to the seller's, and believe they should have some control over who gets to utilize them. They also may believe they have a duty to protect their customers from unscrupulous agents who may bring down the quality of the system.

While these concerns are legitimate, they must be viewed in light of the informational bottleneck the MLS system creates, and therefore must be balanced against any anti-competitive inefficiencies the MLS facilitates.38 Our own views on how to strike an appropriate balance will be discussed shortly. But we first turn to what is rapidly becoming a critical factor in the evolution of this industry-the Internet.

5. The Role of Internet-Based Services

The Internet has revolutionized the way business is done in many sectors of the economy, and surely will continue to produce more change in the years ahead. The Net has had an especially significant impact on all activities that involve searching by consumers for the best deal. By dramatically lowering search costs-giving consumers access to a vast array of suppliers without having to physically go to them-the Net has led to new business models (think "bricks and clicks," or having both a Net and physical presence in retail, as one example). The Net has also led to lower costs and prices in a range of services, including such traditional "brokerage" activities as selling securities and travel services.

A few recent studies have revealed some of the economic benefits available to real estate service providers and consumers who utilize the Internet. Two empirical studies have used survey data to demonstrate gains to brokers, and found that firms who utilize computer technology and the Internet tend to have superior productivity and profit margins.39 Another study compared home buyers who used the Internet for searching with buyers who did not and found that those using the Internet located more houses meeting their needs in the same amount of search time.40 Others have speculated that residential real estate transaction costs could be reduced by as much as 50% if technology were better utilized.41

The Internet could, and probably should, have a radical impact on real estate brokerage services, where searching is much of what real estate agents have traditionally done. Displaying listings online is an obvious step for the industry, to facilitate quick and effective matching between potential buyers and sellers. Identifying both the inevitability and usefulness of this technological step, the NAR began a national MLS,, consisting of listings from local MLSs throughout the country. Most major brokers and national franchisors now maintain their own websites (often including MLS listings) in order to advertise their services to increasingly web-savvy customers. According to one survey, about 70% of home buyers do use the Internet when searching for a home.42

In addition to these technological cost savings and marketing advantages that have been similarly recognized in other industries, more widespread use of the Internet is challenging the conventional process for selling a home. Some newer startups have emerged to try to harness technological innovation to better meet consumers' real estate needs. National firms such as eRealty and ZipRealty operate largely on the Web and pass on some of their savings from low overhead and high productivity to their customers.43

Other recent entrants have attempted to disaggregate the process of buying or selling a home, specializing in particular parts of the process and trying to perform them more efficiently. LendingTree, which is also web-based, assesses buyers' and sellers' needs and then matches them to a network of local real estate firms, where contracted agents can help with the transaction. The network agents split a portion of their commissions from completed sales with LendingTree for helping recruit the customer, and LendingTree in turn passes a rebate on to the consumer as a discounting incentive to use their services. HomeGain operates on a similar web-based, networking model.44

Other Internet-based services, such as and, offer to place home sellers' listing information in local MLS databases for a flat fee, often from $500-$700. This presents potentially large savings for sellers willing to do some of the work themselves, such as showing the property to potential buyers and negotiating the sales price. The seller would still likely have to offer half of a commission, typically 3% of the sales price-in order to attract buying agents to bring their customers-but this still represents a 50% savings.45

Some "bricks-and-mortar" brokers are experimenting with alternative models as well. Foxtons, originally an English company, gained some traction in New York, New Jersey and Connecticut by offering stripped-down services at a two percent fee. It now provides full-service brokerage for three percent, saving costs and boosting productivity by breaking the transaction up among its specialized staff.46 Other "fee-for-service" or "menu pricing" firms have emerged that charge flat or hourly fees for specific services, such as listing in the MLS or closing contracts, and thus allow customers more flexibility to pay for the services they want. Franchisors such as Help-U-Sell, for example, are helping to popularize flat-fee alternatives at a national level.47 Even traditional companies such as Col dwell Banker and Century 21 have experimented with discount and menu price services.48 This suggests that demand for these options is becoming mainstream, and that new entrants present a competitive threat to older models and fees.

Splitting up real estate services into smaller parts could result in substantial gains for consumers. We may be moving toward a new era in which firms develop a variety of different business models. Some will specialize in specific parts of the process, allowing consumers to choose exactly what fits their needs; and some may offer bundles of services that match or exceed those of traditional agents.

While the Internet could have a large positive impact for home buyers and sellers, there is no guarantee that this change will occur rapidly. As discussed earlier, access to the MLS appears to be a potential bottleneck. This access is tightly controlled by board members selected by existing members, and often requires local Realtor association membership.49 This control is likely to have benefited traditional incumbent brokers, but does not necessarily benefit consumers or reward innovative producers. If home buyers, and firms with new business models, were able to access the MLS listings more easily and could take a proactive role in the search process, the dynamics of the industry likely would change. As noted above, this is already beginning to occur, with both Internet companies and some bricks-and-mortar companies offering more specialized services on attractive terms.

It may be the case that a significant number of consumers prefer the traditional system, where they pay a single agent a percentage of the sales price to help with every step of the process. They may find that paying a commission is an important effort motivator that leads to better results than flat fee payments. But the best way to learn what consumers really want is to allow them to choose from the rich array of services the market is capable of delivering.

In the end, no one knows exactly what the real estate market of the future will look like. But the experience of other industries, such as the deregulation of rates in the buying and selling of stocks, suggests that the impact could be substantial.50 The key, in our view, is to create an environment that encourages rather than suppresses competition. As we suggest in the next section, there are reasons to believe that special interests may succeed in stifling competition that could lead to dramatic innovations that would benefit consumers.

6. Resistance to Innovative Competition by Traditional Realtors

The National Association of Realtors and state and local real estate boards have a direct economic stake in keeping out potential competitors to the bricks-and-mortar model. They do so using a variety of mechanisms, many of which are likely to hurt consumers.

Though some traditional brokers have embraced the marketing capabilities and cost savings afforded by the Internet, there has been active resistance in the industry toward the use of innovative business models that threaten to make the standard one obsolete. Some of the bigger, established companies may fear that the importance of branding and franchising, on which they rely heavily for profits, will be significantly diminished if independent brokers can recruit customers through web-based or network models more cost-effectively.

One of NAR's stated objectives is to "advocate for expanding the REALTOR® role in managing the entire transaction." Another is to "support development of real estate-related business models that enhance the first point of contact and member profitability."51 While it is expected that a trade group will do everything it can to protect the success of its members, we believe some of the actions taken by NAR and state realty organizations are likely to violate antitrust law and their own ethics codes. Furthermore, these actions are likely to hurt consumers and overall economic welfare. We focus on the following five barriers to competition being erected by real estate associations, and by some legislators and regulators who are influenced by them: 1) NAR's rules for online listings; 2) discrimination in MLS membership; 3) state minimum service licensing requirements; 4) state anti-rebate laws; and 5) Congressional limits on banks participating in real estate brokerage.

1) NAR has been adjusting its rules for displaying MLS listings on the Internet since 2000, when an MLS in Texas filed a copyright infringement suit against eRealty for showing listings to customers through its website. eRealty claimed that, as a member of the MLS it had just as much a right to distribute listings to potential customers through its "virtual office website" as any other member did in its physical office. It filed a counterclaim alleging other MLS members were stifling competition by trying to restrict MLS access. To settle what threatened to become a widespread problem in local MLSs, NAR drafted a voluntary sharing policy, which allows for the creation of MLS-sponsored websites that would display listings of member brokers in a given local MLS who chose to participate.52

eRealty was dissatisfied with NAR's solution and continued to pursue its right to share all MLS listings with customers via the web. It claimed that it shared the listings in a controlled environment, requiring a login and acceptance of terms from customers and thus allowing for a similar broker-customer relationship as a physical office.53 In 2003, the Department of Justice launched an investigation of the matter to determine whether NAR was trying to restrict the ability of Internet competitors like eRealty to compete.54

The provision of NAR's virtual office website rules that has received the most attention is known as "selective opt-out." The policy says that MLS members are free to distribute listings to customers over the Internet, but that any member who does not want its listings shared can choose the firms it wants to exclude from sharing its listings on the Internet, without any requirement to notify the home seller.55

The problem with selective opt-out is that it makes it even easier to tacitly collude against entrants with new business models.56 Indeed, NAR has admitted that including selective opt-out was a concession to the two largest national companies in the industry.57 Without an equal ability to utilize the MLS, legitimate competitors with new business models may not be able to compete against established companies.

In our view, there is little justification in being able to discriminate selectively in sharing Internet listings. Brokers cannot prevent each other from showing the pooled listings to potential customers who walk in their offices.58 Making such a distinction for the Internet is not only arbitrary but discourages innovative models that may improve the efficiency of service delivery.

The NAR has signaled a willingness to abandon the selective opt-put provision of their virtual office website rules, in favor of a new provision known as "blanket" or "general" opt-out. General opt-out would allow an MLS member an all-or-nothing choice to let other members display its listings online. This option leaves less room to target specific competitors.59 Nonetheless, apparently after some attempts by the NAR to persuade the Justice Department that a general opt-out did not violate the antitrust laws, DOJ proceeded to file suit against the NAR in September, 2005, alleging that the new rules still impeded competition in real estate brokerage. The NAR presumably will defend against the suit largely on the grounds that brokers own their listings and are not required to share them with anyone, even if such sharing leads to wider exposure of the listings to potential buyers.

There is no question that firms should have some rights over listings they acquire in order to give them appropriate incentives to get those listings. The question is where to draw the line in terms of limiting a firm's ability to fend off unwanted competition that may be socially desirable.60 DOJ is implicitly drawing the line where competitors can agree among themselves on actions that limit the sharing of those listings. If this case ever goes to trial, a court could provide definitive guidance as to who is right: the DOJ or the NAR. As of this writing, we believe that the Justice Department's position is not unreasonable. If the rules are not struck down on their face, we hope that relevant authorities will ensure that they are not used anti-competitively in individual markets.

There are also ethical and fiduciary concerns under state law that should be prominent in a broker's decision to permit other brokers to show the listings to potential buyers. Selling brokers have a duty to promote the home seller's interests, which in most cases surely means giving the seller's listing maximum exposure, to get the highest price possible or the quickest sale.61 Even if firms are not explicitly trying to limit competition, banning some competitors from displaying listings to potential buyers may be inconsistent with such goals. Brokers should be obligated to serve their customers' best interests, which includes informing sellers if they are limiting the pool of available buyers for a particular house.62

2) MLS rules at the local level are also impeding competition. For example, many local MLSs will not allow listings by "network brokers"-those that refer potential home sellers and other agents for a fee, as described above. The ostensible rationale is that the online referral sites are not sufficiently "engaged in real estate."63 We believe this rationale is a device for inhibiting legitimate competition.64 Online companies that perform marketing services and refer customers to agents who bid for the customer's business are simply offering real estate brokerage through a different business model. These requirements do not appear to be legitimate bases for denying access to the MLS, and the justifications that the rules protect consumers are unproven and even somewhat ironic.65

3) Another way in which competition may be impeded is through state legislation, often crafted by state real estate commissions who oversee real estate licensing policies. These commissions are supposed to protect consumer interests, but their members are often themselves practicing brokers, nominated and supported by prominent brokerages, whose interests they typically share. One type of regulatory restriction that has been pushed by some state commissions would prohibit agents from offering a la carte services by establishing "minimum service requirements" for obtaining and keeping a broker's license. As mentioned above, several states are considering these bills, and some have postponed consideration only in the face of warnings of their anti-competitive effects from the Department of Justice and the Federal Trade Commission. The experience of Illinois and Texas (which have passed minimum service statutes), Missouri, Oklahoma, and other states suggests this trend was actively supported by the state Realtor associations and is likely still underway. In a memo to state Realtor associations, the NAR general counsel stated that federal competition authorities can do little since state legislative actions are insulated from the federal antitrust laws under the "state action" doctrine.66

The argument for maintaining service requirements as a condition for having a license is not persuasive. Licenses exist to ensure a standard of service quality in a given profession, but there is no reason to believe that agents who offer more narrowly tailored services and charge accordingly will do any worse of a job or harm consumers in any way. To the contrary, they have the potential to better serve consumer demand and make the market for services more efficient.67 For example, some home sellers may desire the exposure the MLS affords, but feel comfortable handling all other aspects of the sale themselves. Others may feel confident showing their home to buyers and negotiating their own sales price, but want to hire a contract specialist to check over the final terms of the agreement. In general, consumers can be expected to be willing to pay for different types of services and levels of service quality. Some may want an entire bundle that is of very high quality while others may want the bare minimum.

Thus, minimum service requirements are vehicles for inhibiting competition from brokers or other specialists who are willing to provide services more selectively. These potential competitors, who might charge flat fees or hourly fees for unbundled services, rather than percentage-based commissions for the entire package, could be expected to put downward pressure on commission rates. It is true that the full service brokerage bundle has prevailed in the market for many years, but it appears to us that more individualized services have become increasingly popular in the wake of soaring housing prices and the emergence of the Internet. Whether or not a la carte-style real estate will catch on more widely remains to be seen, but there is no rationale for preemptively impeding such competition other than to protect the interests of those who feel threatened by it.

4) Another way in which legitimate competition is being thwarted at the state level is through anti-rebate laws. We think that rebating to the consumer, either directly or through a third party, will have a positive impact on consumer welfare. The DOJ is essentially making this point in suing the Kentucky real estate commission for blocking rebates.68 We see no compelling economic rationale for not allowing rebates as they represent a form of price competition that should improve efficiency by putting pressure on brokerages to provide better services at lower prices. Again, it may turn out that higher commissions are necessary to ensure the quality of service most customers want; this result, however, should be determined by consumers and producers via the market, rather than real estate commissioners via rules that may be anticompetitive.

5) Lastly, NAR has strongly opposed the entrance of banks into the real estate brokerage market, further shielding its members from legitimate competition via political lobbying. In 2001, the Federal Reserve Board proposed that real estate brokerage be added as a permissible activity for financial holding companies to engage in under the Gramm-Leach-Bliley Act of 1999.69 But Congressional action, mounted at the behest of the brokerage industry, has repeatedly impeded the Federal Reserve and the Treasury from implementing any regulatory change.70

The brokerage industry's main argument in opposition is that brokerage is a commercial, rather than a financial, activity and thus should be off limits to financial holding companies.71 But this is misplaced. Brokering the sale of a house is fundamentally no different than brokering insurance or securities. Both of the latter are permissible activities for financial holding companies. Indeed, hundreds of financial organizations have brokerage affiliates. If banks can sell insurance on homes, why can't they broker their sale?

In summary, the various efforts of the NAR and local real estate boards are best understood in terms of rent-seeking. We don't blame the traditional bricks-and-mortar industry for trying to protect its turf. After all, its members are acting rationally to maximize their profits. But policy makers should focus on setting ground rules that will work best for consumers.

7. What Should Be Done?

We offer three broad policy recommendations-one related to effective oversight of possible anti-competitive activities, a second related to state-level actions that are likely to cause inefficiencies, and a third related to introducing more competition by allowing banks to enter into this industry through the use of affiliates.

Recommendation 1: Federal and state antitrust authorities should carefully scrutinize efforts to limit competition in the residential real estate brokerage market.

Federal authorities have an important role to play and are increasingly playing this role. The claims that real estate sales are purely intrastate activities and should be regulated only by the states cannot withstand scrutiny. Millions of people move across state lines each year, and many of them buy homes.72 It is perfectly appropriate for federal competition authorities to investigate and prosecute anti-competitive actions taken not only by the industry's national trade association, but also activities at the state level, and to urge state lawmakers and regulators to act in a way that promotes competition in real estate brokerage. Investigations by bodies such as the Federal Trade Commission and the Government Accountability Office can also be helpful in educating legislators and the general public on these issues.73

The Department of Justice is right to show concern about NAR's rules governing online listings and is also right in attempting to persuade states that prohibitions on rebates are inappropriate. The Department also seems on solid ground in urging state legislatures not to erect new hurdles to competition-for example, by banning or limiting a la carte services-even though the federal government may lack the authority to overrule state legislation.74

If the evidence of anti-competitive rules or practices by local MLSs is true, state attorneys general could also pursue legal action. They could use antitrust law, as well as state and common law regarding the fiduciary duty of agents to their principals.

A key feature of any policy related to the MLS is that it should not discriminate against legitimate market players. Currently, the most important concern is that rules governing entry and operation do not discriminate against online brokers, or brokers that deviate from the traditional bricks-and-mortar model. We believe that selective opt-out rules and constraints on network models constitute such discrimination. While discrimination based on the cost of providing the service may be reasonable, discrimination that is unrelated to operation costs and that may negatively impact competition should be scrutinized carefully and must be justified in order to comply with antitrust law.75

We recognize that under the current system, brokers are entitled to some control over their listings. One possible outcome of allowing more brokers into the MLS, or allowing listings to be more widely displayed, is that some of the bigger brokers will no longer participate in the MLS system. However, companies that withdraw will risk losing employees and customers who value the benefits of sharing listings. Thus, we think the dissolution of the current MLS system is unlikely unless a more attractive alternative presents itself. Over time, we hope that this platform, or any possible replacement, will evolve to better meet the needs of Internet-savvy consumers.76

Recommendation 2: State legislatures and real estate commissions should refrain from adopting laws or rules that inhibit competition in real estate brokerage.

Some policies that have been pursued by state governments are likely to limit competition in this market. In particular, rebate bans and minimum service requirements inhibit the emergence of innovative models with alternative price and service offerings.

We realize that local real estate associations can wield significant political power at the state level. But now that the Justice Department has put a spotlight on the industry at the national level, we are hopeful that elected officials and the regulators they appoint will weigh more seriously the economic consequences of their actions, and resist enacting rules that protect incumbents at the expense of consumers.

Recommendation 3: Congress should not prevent the Federal Reserve Board and the Treasury Department from allowing banks to offer residential real estate brokerage services through separately capitalized affiliates.

Federal authorities can and should enhance competition in real estate brokerage by authorizing banks to engage in this activity through separately capitalized affiliates. The Department of the Treasury and the Federal Reserve Board, which have joint power under the Gramm-Leach-Bliley Financial Modernization Act of 1999 to authorize financial holding companies to enter businesses that are financial in nature.

Banking and broader financial organizations are natural entrants into real estate brokerage. They have a natural clientele-their home borrowers. Any concerns that banks might subsidize their real estate activities have already been addressed by the holding company structure mandated by the 1999 Act, through which banks are allowed to affiliate with permissible non-banking organizations only through separately capitalized affiliates.

We cannot predict how much commissions would come down if financial holding companies entered the brokerage industry, but the additional competition is likely to put downward pressure on rates and stimulate innovation. The financial authorities should renew their efforts to drop this anti-competitive barrier that inhibits competition.77 The benefits of entry would be additional to those from intensified competition from Internet-based and discount brokers.

As for Congress, it should let the regulators do what they were authorized to do. Up to now, the real estate industry has successfully persuaded Congress to prevent regulators from allowing financial holding companies to enter the housing brokerage market. Given the ongoing changes in the industry-and specifically the Justice Department's antitrust investigation-Congressional representatives who have traditionally sided with the real estate brokerage lobby have good reason to keep their distance now. There is no defense for a Congress ostensibly interested in limiting the tax burden on Americans to defend what amounts to an artificial "tax" on home buyers resulting from an industry that has yet to take full advantage of the digital revolution.


The preceding analysis suggests that the traditional model for residential real estate brokerage services may be dated, and could be improved substantially with some public policy interventions that spur innovation. Unfortunately, there are numerous barriers to entry that are slowing the emergence of new models for serving consumers.

Moreover, some of these barriers are likely to be anti-competitive and violate antitrust laws.

We do not know which models are likely to succeed in the market place for residential real estate services in the future. We do believe, however, that judicious public policy interventions could have a marked impact on improving services and lowering costs for home buyers and sellers.


Anderson, Randy I., Ken H. Johnson, and Leonard V. Zumpano. 2003. "Internet Use and Real Estate Brokerage Market Intermediation." Journal of Housing Economics, 12, pp. 134-50.

Anglin, Paul and Richard Arnott. 1999. "Are Brokers' Commission Rates on Home Sales Too High? A Conceptual Analysis." Real Estate Economics, 27:4, pp. 719-49.

Bahney, Anna. 2005. "Are Commission Rates Really Negotiable? Sort of." New York Times, May 15, pp. 11.1.

Benjamin, John D., Peter Chinloy, G. Donald Jud, and Daniel T. Winkler. 2004. "Technology and Real Estate Brokerage Firm Financial Performance." 20th Annual American Real Estate Society Conference: Captiva Island, FL.

Benjamin, John D., G. Donald Jud, and G. Stacey Sirmans. 2000. "What Do We Know About Real Estate Brokerage?" Journal of Real Estate Research, 20:1/2, pp. 5-30.

Birger, Jon. 2005. "How come we still pay 6%?" Money Magazine. Published May 20. Accessed June 17, 2005 at

Birger, Jon and Joan Caplin. 2004. "The 4 1/2% Solution." Money, October, pp. 105-09.

Bruce, Donald and Rudy Santore. 2004. "A Note on Optimal Real Estate Commissions." University of Tennessee.

Bureau of Economic Analysis. 2005. "National Income and Product Accounts Table: Table 5.4.5BU. Private Fixed Investment in Structures by Type." Published May 27. Accessed June 23, 2005 at underlying/TableView.asp#Mid

Capper, Russell. 2002. "Testimony Summary." Federal Trade Commission, Office of Policy Planning, Public Workshop on E-Commerce ed.: Washington, DC.

Crockett, John H. 1982. "Competition and Efficiency in Transacting: The Case of Real Estate Brokerage." AREUEA Journal, 10, pp. 209-27.

Delcoure, Natalya and Norm G. Miller. 2002. "International Residential Real Estate Brokerage Fees and Implications for the US Brokerage Industry." International Real Estate Review, 5:1, pp. 12-39.

Deparment of Housing and Urban Development. 2005. "HUD Testimony: Statement of The Honorable Alphonso Jackson, Secretary, United States Department of Housing and Urban Development, before the United States House Committee on

Financial Services." Published March 2. Accessed June 10, 2005 at

Department of Justice, Antitrust Division. 2005a. "Complaint, United States v. Kentucky Real Estate Commission, Civ. Act. No. 3:05CV188-H." Published March 31. Accessed April 10, 2005 at

__. 2005b. "Letter from R. Hewitt Pate, Assistant Attorney General to Oklahoma State Representative Todd Heitt." Published April 8. Accessed April 10, 2005 at

Department of Justice, Antitrust Division and Federal Trade Commission. 2005c. "Letter to Texas Real Estate Commission Re: Proposed Amendments to 22 Tex. Admin. Code 535.2." Published February 22. Accessed April 10, 2005 at

Evans, Blanche. 2004. "CAR Report Shows Increased Use of Internet By Homebuyers." Realty Times. Published June 30. Accessed June 15, 2005 at carreport.htm

__. 2005. "Steel Magnollia -vs- The Men In Black." Realty Times. Published April 26. Accessed April 27, 2005 at magnoliavsmib.htm

Evans, David. 2003. "The Antitrust Economics of Multi-Side Platform Markets." Yale Journal on Regulation, 20:2, pp. 325-82.

Federal Reserve System and Department of the Treasury. 2001. "Proposed Rules." Federal Register, 66:2, pp. 307-14.

Federal Trade Commission. 2002. "Public Workshop: Possible Anticompetitive Efforts to Restrict Competition on the Internet." Published September 30. Accessed May 10, 2005 at

Francese, Peter. 2004. "America's Area Dilemma." American Demographics. Published July 1. Accessed June 10, 2005 at http :// 1/is_6_26/ai_n6121421

Frew, James R. and G. Donald Jud. 1987. "Who Pays the Real Estate Broker's Commission?" Research in Law and Economics, 10, pp. 177-87.

Goolsby, William C. and Barbara J. Childs. 1988. "Brokerage Firm Competition in Real Estate Commission Rates." Journal of Real Estate Research, 3:2, pp. 11-20.

Hagerty, James R. 2003. "Real-Estate Sites Raise Commissions." Wall Street Journal, December 24, pp. D1.

__. 2005a. "It Pays to Negotiate Your Agent's Commission." Real Estate Journal. Published June 22. Accessed June 22, 2005 at sell/agentsandbrokers/20040622-hagerty.html

__. 2005b, "Antitrust Suit Could Reshape Residential Real Estate Sector." Wall Street Journal, September 12, pp. B2.

Hagerty, James R. and Ruth Simon. 2005. "Realtor Commissions Face New Pressure." Wall Street Journal, May 10, pp. D.1.

Hahn, Robert and John Hird. 1991. "The Costs and Benefits of Regulation: Review and Synthesis." Yale Journal on Regulation, 8:1, pp. 233-78.

Ham, Shane and Robert D. Atkinson. 2003. "Modernizing Home Buying: How IT Can Empower Individuals, Slash Costs, and Transform the Real Estate Industry." Progressive Policy Institute: Washington, DC.

Hsieh, Chang-Tai and Enrico Moretti. 2002. "Can Free Entry Be Inefficient? Fixed Commissions and Social Waste in the Real Estate Industry." NBER Working Paper: Cambridge, MA.

Jarrell, Gregg A. 1984. "Change at the Exchange: The Causes and Effects of Deregulation." Journal of Law and Economics, 27:2, pp. 273-312.

Jud, G. Donald, G. Stacey Sirmans, and Daniel T. Winkler. 2002. "The Impact of Information Technology on Real Estate Licensee Income." Journal of Real Estate Practice and Education, 5:1, pp. 1-16.

Levitt, Steven D. and Chad Syverson. 2005. "Market Distortions when Agents are Better Informed: The Value of Information in Real Estate Transactions." National Bureau of Economic Research, Working Paper: 11053.

Miceli, Thomas J. 1991. "The Multiple Listing Service, Commission Splits, and Broker Effort." AREUEA Journal, 19:4, pp. 548-66.

Mullaney, Timothy J. 2005. "Real Estate's Turf War Heats Up." Business Week, April 18, pp. 38.

National Association of Realtors. 2001. "Comment Letter to the Board of Governors of The Federal Reserve System." Published May 1. Accessed May 10, 2005 at

Office of Federal Housing Enterprise Oversight. 2005. "House Price Index: 4Q 2004." Published March 1. Accessed April 18, 2005 at

Realtors, National Association of. 2003 a. "National Association of Realtors Member Profile."

__. 2003b. "Virtual Office Website ("VOW") Policy." Published May 17. Accessed May 10, 2005 at

__. 2004a. "National Association of Realtors Profile of Home Buyers and Sellers."

__. 2004b. "Newest NAR Members are Committed to Real Estate." Published April 19. Accessed June 7, 2005 at

__. 2005a. "2004 Annual Report: NAR's Core Purpose and Principles." Published February 1. Accessed May 15, 2005 at

__. 2005b. "2005 Midyear Meetings: NAR's Virtual Office Web Site Policy News." Published June 8. Accessed June 8, 2005 at

__. 2005c. "Code of Ethics and Standards of Practice of the National Association of Realtors." Published January 1. Accessed April 10, 2005 at

__. 2005d. "Field Guide to Fee for Service." Published April. Accessed May 1, 2005 at 110

Regional Multiple Listing Service, Inc. (Florida). 2003. "Rules and Regulations." Published December. Accessed May 10, 2005 at

Risen, Clay. 2005. "Realtors vs. the Internet." New Republic, May 2 & 9, pp. 14-15.

Roberts Jr., Glenn. 2005a. "Discount brokerage bills spread through nation." Inman News. Published May 4. Accessed June 6, 2005 at scountbrokeragebills5_4_2005.htm

__. 2005b. "States Limit Limited Service." Inman News. Published May 10. Accessed June 6, 2005 at

Rochet, Jean-Charles and Jean Tirole. 2003. "Platform Competition in Two-Sided Markets." Journal of the European Economic Association, 1:4, pp. 990-1029.

Roth, Stephen. 2005. "Proposal would permanently bar banks from real estate brokerage." The Business Journal. Published April 8. Accessed June 21, 2005 at ttp://

Sirmans, G. Stacey and Geoffrey K. Turnbull. 1997. "Brokerage Pricing Under Competition." Journal of Urban Economics, 41, pp. 102-17.

Sloane, Julie. 2003. "The 2% Solution." Fortune Small Business, pp. 78.

Streitfeld, David. 2005. "A Glut in the Market for Homes." Los Angeles Times, May 20, pp. A.1.

Thompson v. Metropolitan Multi-List, Inc. 1991. No. 90-8724, 934 F.2d 1566.

Tolle, Bill. 2005. "Real Estate Buyers Agents: Fiduciary Duties Of A Real Estate Agent." Published January 1. Accessed June 15, 2005 at

Turnbull, Geoffrey K. 1996. "Real Estate Brokers, Nonprice Competition and the Housing Market." Real Estate Economics, 24:3, pp. 293-316.

United States v. National Association of Real Estate Boards et. al. 1950. No. 428, 339 U.S. 485.

United States v. Realty Multi-List, Inc. 1980. No. 78-2481, 629 F.2d 1351.

U.S. House of Representatives, Committee on Financial Services. 2005. "Protecting Consumers and Promoting Competition in Real Estate Services." Published June 15. Accessed June 16, 2005 at

U.S. Senate. 1999. "Information Regarding the Gramm-Leach-Bliley Act of 1999." Published November 1. Accessed June 1, 2005 at

Wilke, John R. and James R. Hagerty. 2005. "U.S. Plans Antitrust Suit Over Real-Estate Listings." Wall Street Journal, May 9, pp. A1.

Winston, Clifford. 1993. "Economic Deregulation: Days of Reckoning for Microeconomics." Journal of Economic Literature, 31:3, pp. 1263-89.

Yavas, Abdullah. 1994. "The Economics of Brokerage: An Overview." Journal of Real Estate Literature, 2:1, pp. 169-95.

__. 1996. "Matching of Buyers and Sellers by Brokers: A Comparison of Alternative Commission Structures." Real Estate Economics, 24:1, pp. 97-112.

__. 2001. "Impossibility of a Competitive Equilibrium in the Real Estate Brokerage Industry." Journal of Real Estate Research, 21:3, pp. 187-200.


1 See HUD (2005).

2 According to NAR (2004a), 77% of homebuyers used a real estate agent or broker in their purchase.

3 Data from Real Trends yields an estimate of $61 billion in $2004. This figure is commonly cited in popular press. See, for example, Wilke and Hagerty (2005). The Bureau of Economic Analysis estimates that brokers' commissions on residential sales totaled $96.6 billion in 2004, but this calculation assumes commission rates are fixed at specified levels, an issue we address below. See BEA (2005). Unless otherwise noted, all dollar figures are in 2004 dollars.

4 See Mullaney (2005) and Risen (2005).

5 Eleven states currently ban rebates, in addition to others that restrict the form they can take. See Birger (2005). See Department of Justice (2005a) for information on the Kentucky lawsuit.

6 Illinois and Missouri passed such laws before any DOJ intervention. Several other states have passed or are considering similar bills in spite of direct discouragement from DOJ and Federal Trade Commission to the Oklahoma and Texas legislatures. See Roberts (2005b) for a summary of legislative activity. See Department of Justice (2005b) and Department of Justice (2005c) for the responses to Oklahoma and Texas.

7 See Wilke and Hagerty (2005) for a summary of DOJ's action. NAR attempted unsuccessfully to negotiate with the Department and revise their rules in order to avoid a suit.

8 NAR's policies only technically affect "Realtors" and not all real estate agents. The term Realtor is a registered trademark of the National Association of Realtors and refers only to real estate agents who belong to NAR and thus have agreed to abide by their ethics policies and other guidelines. Since a substantial proportion of active agents are Realtors (see NAR (2004b)), it is fair to argue that their policies have a significant impact on how business is conducted. Any licensing requirements or other regulations passed by state legislatures or commissions apply to all licensed agents and brokers, regardless of NAR membership.

9 See Hagerty (2005b)

10 These commission estimates are based on Real Trends data. See, for example, Risen (2005).

11 Average calculated by us from 18 developed countries (out of 30 total) from Delcoure and Miller (2002). 10% is atypical and usually applies to lower-priced houses. According to the study, most rates in developed countries are 2%-4%, typically split between buyer and seller.

12 Numerous websites such as and connect sellers with a local agent who will list their home in the MLS for a flat fee, typically around $500-$700. Many bricks-and-mortar agents are offering limited service such as flat fee MLS listing as well. For other examples, see Hagerty and Simon (2005) and NAR (2005d).

13 For summary and text of the Gramm-Leach-Bliley Act, see U.S. Senate (1999).

14 See the discussion below on the Gramm-Leach-Bliley Act and banks as potential competitors.

15 See Frew and Jud (1987).

16 This is a common arrangement. The exact rights and roles of agents and brokers vary on a state-by-state basis. See for more information.

17 According to NAR (2003a), in 2002, 73% of Realtors split some part of their commission with their broker, 20% kept their entire commissions, and the rest had another compensation arrangement. RE/MAX is well known for its "100% plan" which allows agents to keep their entire commissions.

18 See Benjamin, Jud and Sirmans (2000) for a comprehensive review of residential real estate research.

19 Whether the losses to consumers are merely transfers to producers is another matter. It depends on how the agents and brokers respond. For example, Turnbull (1996) suggests that agents may expend more effort on behalf of clients if rates are higher, thus providing some additional benefits to consumers.

20 See Anglin and Arnott (1999).

21 A seemingly competitive environment in which more agents are competing for the same number of listings still may be inefficient if they are not competing on price. See Hsieh and Moretti (2002). This phenomenon of agent oversupply has been documented recently in articles such as Streitfeld (2005).

22 See Delcoure and Miller (2002).

23 Bruce and Santore (2004) suggest that the equilibrium commission rate may be higher than the least expensive agents are willing to work for, since an increase in the fee is a way to motivate the agent to get a better price when effort is unobservable. Other studies have highlighted this difficulty in aligning the goals of house sellers and their brokers, since brokers are likely to profit more from selling a higher volume of houses in a given time period than by waiting to get the highest possible price on each one. Levitt and Syverson (2005) document this phenomenon by showing that agents leave their own homes on the market longer, and get a higher price on average, than they do for their clients. See also Miceli (1991) and Yavas (1996) for a discussion of possible conflicts of interest among sellers, buyers and their brokers.

24 See Yavas (2001).

25 See Rochet and Tirole (2003) and Evans (2003).

26 For example, Sirmans and Turnbull (1997) analyzed data from a local MLS in Baton Rouge over an eight year period, which included a boom and bust in the housing cycle.

27 Goolsby and Child (1988) showed this using somewhat more limited data. See Benjamin, Jud and Sirmans (2000) for more examples.

28 This price increase was calculated using the Office of Federal Housing Enterprise Oversight's Housing Price Index. See OFHEO (2005). According to this data, the increase from 1991-2004 was 94%.

29 Using the HPI prices, the increase in revenue per house was 51% from 1991-2004 (commission and agents per sale data from Real Trends), slightly higher than the 39% inflation from the Consumer Price Index (see Using Real Trends' price data, the revenue increase was 31%, slightly lower than inflation.

30 Some reports have indicated recent pressure on commission rates, particularly in areas where prices are increasing rapidly. See, for example Bahney (2005).

31 The only caveat could be that a reduction in the commission rate could reduce the effort of agents representing the buyer or the seller. We think the market could handle such changes with suitable adjustments in contracting terms.

32 Exclusive agreements between a seller and his agent often include a preset fixed fee that impedes a buyer's ability to negotiate the half of the commission he is effectively paying (in the form of a higher house price) if he does not need certain agent services.

33 See Yavas (1994). This particular review does not specify the terms on which access should be provided.

34 Yavas (1994) provides a good summary of studies assessing the impact of the MLS system on the brokerage market. He and others (for example, Crockett (1982)) have pointed to the MLS as a major source of inefficiency and argued for wider access to listings, and ultimately unbundling of services.

35 See U.S. vs. National Association of Real Estate Boards (1950).

36 There are around 800 MLSs (see Risen (2005)) throughout the country, some NAR-affiliated, some not. Their structure and operating procedures vary somewhat. We find that the commonalities among MLSs are generally more meaningful than the differences, but realize that our analysis may not apply to some MLSs that differ from the norm.

37 See Birger and Caplin (2004) for examples of discrimination and boycotting of discounters.

38 We do not dispute that brokers have some legal ownership over sellers' listings. However, this does not allow them to pursue strategies that restrict legitimate competition or that violate fiduciary obligations to their customers. See discussion below.

39 See Benjamin, Chinloy, Jud and Winkler (2004) and Jud, Donald, Sirmans and Winkler (2002).

40 See Anderson, Johnson and Zumpano (2003). The California Association of Realtors has also conducted studies showing that buyers using the Internet spend more time searching on their own before contacting a Realtor and less time working with the Realtor, including fewer accompanied house visits. See Evans (2004). In cases where consumers are undertaking more work themselves, the resulting cost savings to agents should be reflected in the form of lower fees.

41 See Ham and Atkinson (2003). The complete empirical basis for this estimate is not provided, but the authors do offer several suggestions for ways in which technology could improve the efficiency of real estate services.

42 See NAR (2004a).

43 These companies used to openly advertise their fees at 4.5%, but now negotiate on a case-by-case basis. They still maintain that their model affords their customers savings compared to traditional brokers. See Hagerty (2003). eRealty is now owned by Prudential Realty. See and

44 See

45 Exactly how much of this 50% savings is kept by either side depends on how much each party values the services of the agent bringing the buyer. The seller typically pays the fee but can raise the house price to shift some or all of the commission to the buyer. We welcome the emergence of different forms of contractual arrangements, such as buyers paying their own agents directly, which may better capture the value the buyer and seller place on the services each receives. One encouraging development in this area is the emergence of "buyer's brokers," who can be hired separately by the buyer and are dedicated to serving his interests, as distinct from the seller's. See, for example, Tolle (2005).

46 Foxtons has teams of marketers, agents and attorneys who handle different stages of a home sale. See The company used to forgo listing its houses in the MLS and hosting open houses (see Sloane (2003)) but now includes these and other traditional services as part of the standard 3% plan.

47 See Other examples of popular fee-for-service companies include Assist-U-Sell and National Association of Real Estate Consultants. See Hagerty (2005a).

48 Blue Edge Realty is a discount offshoot of Coldwell Banker, available in Illinois and Pennsylvania. See Century 21 offers flat fee services, including MLS listings, in select states. See www. c21 clickit. com.

49 For a typical example of MLS bylaws, see Regional Multiple Listing Service Inc. (2003). MLS bylaws are often very similar to each other, particularly when the service is NAR-affiliated, since they are based on national guidelines and must be approved by NAR.

50 See, for example, Jarrell (1984). Today, buying and selling of an unlimited number of shares costs less than $10. In the seventies, it could cost on the order of $1 per share to buy or sell. The drop has been remarkable and is due to deregulation and the resulting increase in competition, and changes in technology-most notably the Internet. See also, Winston (1993) and Hahn and Hird (1991).

51 See NAR (2005a).

52 This came to be known as NAR's "Internet Display Policy." An MLS-sponsored listings site, where members can post listings and also advertise their broker services, is known as an "Internet Data Exchange." For a full description of the eRealty case, see Capper (2002).

53 In particular, it did not freely distribute listings, which was a concern raised by traditional MLS members.

54 NAR has, over the last two years, been re-drafting its policy for virtual office websites, which was set to take effect in July 2005. The rollout of this new policy has been delayed further, until January 2006, due to continued investigation by DOJ. See Wilke and Hagerty (2005) and NAR (2005b).

55 See NAR (2003b). It is possible that some form of disclosure will ultimately be required in the final rules.

56 It is easy to see how firms with large market shares could cooperate, explicitly or implicitly, by withholding display rights from any competing firms whose model or pricing structure threatened their profits. While firms have a right to pursue profits, the Sherman Act of 1890 prohibits them from coordinating business strategy or unfairly restricting entry to the market by legitimate competitors.

57 In Risen, (2005), NAR spokesman Steve Cook explains, "'Cendant and RE/MAX said they would not cooperate without broker opt-out. In a sense, we had the concern that the whole MLS was at risk, that they would create their own database.'" According to the article, these are the two largest national franchises, accounting for roughly 40% of the market. Cendant is the parent company of Coldwell Banker, Century 21 and ERA.

58 To the contrary, they have supported doing so for many years as an effective way to match buyers and sellers.

59 It does uphold the distinction between paper and online listings, treating those who choose to operate primarily online differently from those who do not.

60 Blanket opt-out may still allow brokerages to act anti-competitively-for example, if one or two dominant firms in a region choose to opt-out for the express purpose of stifling competitors. We acknowledge that without proof of coordination between firms or intentional abuse of monopoly power, practices are not actionable under antitrust law, but we would at least encourage consideration of the competitive impact of all opt-out provisions by the appropriate authorities, such as the Department of Justice.

61 See NAR (2005c) for NAR's Code of Ethics. See Tolle (2005) for a basic description of fiduciary duty as it pertains to real estate. Specific laws vary by state.

62 There may be some cases where a customer, for privacy or other reasons, wishes not to have his listing freely displayed on the Internet by other brokerages, but absent such a concern we find it difficult to see why it would not be in a seller's interest to distribute as widely as possible. In the traditional MLS, in which the only opt-out mechanism is not to join, the listing is freely displayed on paper.

63 See section 17.2 of Regional Multiple Listing Service (2003) bylaws for an example of this common, vague "engaged in real estate" requirement. We do not see how a so-called "referral" company who obtains a state broker license and matches its contracted agents to buyers and sellers is not "engaged in real estate." Requirements that those who have access to the listings must be the ones to exchange compensation (see section 1.5(a)) serve only to keep the process bundled together. These results clearly benefit traditional brokers, but have no obvious benefits for consumers or welfare. The NAR online rules sustain discrimination against network models by prohibiting firms from receiving compensation for referring customers to other brokers. See provision "4g" of NAR (2003b). As discussed, the national policy is still being revised, but we feel that provisions that impede alternative models are likely to be unjustified economically.

64 Some MLSs allow network brokers to partner with other member brokers, but this still relegates them to second-class status and reduces the benefits to consumers that would likely result if they were able to operate independently.

65 Cases such as U.S. vs. Realty Multi-List (1980) have found that MLS admission requirements that may negatively affect competition must be balanced by a legitimate pro-competitive purpose and must be well tailored to meet that purpose.

66 See Evans (2005) for excerpts from NAR general counsel Laurie Janik's memo. Janik asserts that "'antitrust laws were not intended to prevent a state government, acting in its sovereign capacity, from putting in place those policies that the state chooses-even if those policies might be anticompetitive.'" She also defends the right of Realtors to "'seek action from state legislatures and state regulatory agencies even if that action would have anticompetitive effects.'" NAR does not have an official position on minimum service requirements, and Janik maintains that these statements do not "'suggest that laws and regulations imposing minimum duties on licensees are anticompetitive,'" but they certainly suggest NAR's willingness to support state-level regulations and lobbying efforts that do not promote competition or benefit consumers.

67 Some regulators have cited complaints from buyer agents who have had to work harder or offer extra service to sellers who opted for discount service, but needed more help. See, for example, Roberts (2005a). While we agree that it is not the buyer agent's job to help an unprepared seller, we believe the market can resolve this issue through the emergence of more specialized services, and that state-mandated services for all agents is likely to cause inefficiencies.

68 See Department of Justice (2005a).

69 See Federal Reserve (2001).

70 Congress has approved a one-year ban on financial holding companies' entering the real estate market each year since 2001. This year, both the Senate and the House are considering bills to permanently ban banks from participating in real estate brokerage. See Roth (2005).

71 See NAR (2001) for a full description of NAR's opposition to the suggested rule change allowing banks to participate in real estate brokerage. It has voiced similar resistance in each subsequent year as the rule has been reconsidered. See U.S. House of Representatives (2005).

72 According to the Census Bureau, about 7.6 million people moved to a different state in 2003, typical of the national average over the past 20 years. See Francese (2004).

73 The Federal Trade Commission conducted hearings on barriers to Internet commerce in 2003, the outcome of which helped in part to bring about the recent Supreme Court decision facilitating interstate wine sales. See FTC (2002). The Government Accountability Office is expected to conduct a study on real estate competition later this year. See Risen (2005).

74 While we feel it is appropriate and economically beneficial for federal authorities to try to prevent anti-competitive actions by states, we recognize that there are limits to what federal authorities may be able to accomplish (since the antitrust laws permit even anti-competitive state statutes under the "state action" doctrine).

75 This framework is supported by past cases that have considered this issue. See U.S. v. Realty Multi-List (1980) and Thompson v. Metropolitan Multi-List (1991).

76 Currently, many online MLSs require potential buyers to go through a seller's broker in order to inquire about a listing, effectively forcing additional help that sellers may or may not want. By contrast, other Internet platforms such as allow consumers to post housing or other offerings and contact each other directly to make transactions. Future services like these could serve real estate needs on a broader basis, disaggregating control of the listings from the agents providing negotiating, contracting or other services.

77 Recent Congressional hearings held by the House Committee on Financial Services are an encouraging step toward promoting valuable competition from banks in real estate. See U.S. House of Representatives (2005). We hope that further action will be taken to overcome the consistent Congressional support of this barrier to competition.