Documentation

Informations sur la décision

The Toronto Real Estate Board - Rapport d'expert de Gregory S. Vistnes

Collection	Documentation
Date de dépôt	2015-08-04
Numéro de dossier	CT-2011-003
Type de document	Autre
Type d'affaire	Abus de position dominante
Article(s) de la Loi sur la concurrence	art. 79
Numéro d'instance	334

Contenu de la décision

Attention : ce document est disponible en anglais seulement.

PUBLIC 1

THE COMPETITION TRIBUNAL IN THE MATTER OF the Competition Act, R.S.C. 1985, c. C-34, as amended; IN THE MATTER OF an application by the Commissioner of Competition pursuant to section 79 of the Competition Act;

AND IN THE MATTER OF certain rules, policies and agreements relating to the multiple listing service of the Toronto Real Estate Board.

BETWEEN: THE COMMISSIONER OF COMPETITION Applicant - and -

THE TORONTO REAL ESTATE BOARD Respondent - and -

THE CANADIAN REAL ESTATE ASSOCIATION

Expert Report of Gregory S. Vistnes, Ph.D.

August 4, 2015

Intervenor

PUBLIC 2

PUBLIC 3

I. SUMMARY I have been asked by the Competition Bureau to respond to, and comment on, the recently submitted reports of Dr. Church and Dr. Flyer, submitted on behalf of the Toronto Real Estate Board (TREB) and the Canadian Real Estate Association (CREA), respectively. 1 I have also been asked to discuss any new information and evidence relative to my own previous reports.

As discussed in my previous reports, certain brokers in the Greater Toronto Area (GTA) want to compete by using a virtual office website (VOW) to provide consumers with information about the excluded data fields. 2 Although brokers routinely provide information about these excluded data fields to their clients through traditional means (e.g., the telephone, a fax, or an email), 3 TREB’s rules prohibit brokers from using their VOW to provide that same information to their clients. Many brokers, however, believe that clients want to access that information over a VOW, and thus want to offer a VOW so that they can better compete for clients. It follows that TREB’s rules that prohibit brokers from offering a full-information VOW constitutes a direct restriction on how brokers can compete. 4 TREB’s restriction on how brokers compete results in a substantial lessening of competition in the relevant markets. By prohibiting brokers from offering innovative new VOW-based services that many consumers value, TREB’s restriction causes direct consumer harm. TREB’s restriction further reduces competition by increasing brokers’ costs and by distorting competition associated with principal-agent problems. 5 1 See Church 2015 Reply Report at ¶ 2 and Flyer 2015 Reply Report at ¶ 3. The focus of the economic experts’ 2015 Reply reports concern the ongoing harm to competition attributable to TREB’s continued restrictions in how firms can compete. Neither the Church 2015 Reply Report nor the Flyer 2015 Report respond to the evidence that TREB’s more extreme pre-2011 restrictions on brokers’ use of VOWs resulted in a substantial lessening of competition. (See the Vistnes June 2012 Report at ¶¶ 184-200 and the Vistnes August 2012 Reply Report at ¶¶ 39-40 for a discussion of that pre-2011 harm.)

2 As discussed in my June 2012 report at ¶ 256, the excluded data fields consist of sold listings, the offered buy-side commission, and withdrawn, expired, suspended and terminated (WEST) listings. Significantly, information about the sold and WEST listings is not limited to just the price of those listings, but all information about those listings (including days on market and home characteristics such as number of bedrooms, age of the home, location, etc.).

3 There appears to be no dispute over the fact that brokers routinely provide information about the excluded data fields to their clients through more traditional means. As Dr. Flyer and others have consistently acknowledged, brokers commonly provide this information to their clients. See, for example, the Flyer 2015 Reply Report at ¶ 7 (“TREB’s VOW policy does not prohibit consumers from acquiring the information in the disputed fields, only that the information be obtained through means other than a VOW”) and ¶ 14 (“such information [the excluded data fields] is currently widely available and disseminated”).

4 I refer to VOWs that can offer the full range of MLS information, including the excluded data fields (but excluding a limited set of information such as home alarm codes), as “full information VOWs.” See Vistnes 2015 Report at footnote 16.

5 Throughout this report, I use the terms “reduced competition” and “a substantial lessening or prevention of competition” synonymously.

PUBLIC 4

Despite brokers’ belief that consumers will value new VOW-based services, and those brokers’ desire to offer those services in an effort to better compete for customers, Drs. Church and Flyer conclude that these new services are of no real value to consumers. Thus, Drs. Church and Flyer implicitly conclude that GTA brokers are misguided in believing that using VOWs to provide information about the excluded data fields will help them better compete for clients.

In their most recent reports, Drs. Church and Flyer offer a variety of arguments for why they believe consumers, and competition, do not suffer as a result of TREB’s restrictions on how brokers can compete. They also present arguments why these restrictions are more likely to benefit competition than harm competition. This report addresses those arguments and demonstrates why those arguments fail to show that TREB’s conduct is unlikely to cause a substantial lessening of competition. 6 After evaluating the new arguments put forth by Drs. Church and Flyer, it remains my opinion that consumers in the GTA value the ability to obtain information about the excluded data fields over a VOW and that full-information VOWs will provide substantial competitive benefits. It also remains my opinion that TREB’s conduct increases brokers’ costs and distorts competition associated with principal-agent problems. Accordingly, in my opinion, TREB’s rules result in a substantial lessening of competition in the relevant markets.

II. THE MANNER IN WHICH INFORMATION IS PROVIDED MATTERS Drs. Church and Flyer conclude that TREB’s rules do not constitute an important restriction on how GTA brokers can compete. In reaching this conclusion, Drs. Church and Flyer rely heavily on the argument that the information that brokers want to provide over their VOW is already available through other means. 7 Drs. Church and Flyer give little or no weight, however, to the fact that the manner in which consumers obtain information can be just as important as the content of that information. Yet the manner in which information is delivered is clearly important. Few would argue that travel sites have not affected competition and consumers even if those sites offer the same information that could otherwise have been obtained by visiting the local travel agent. Similarly, few would argue

6 This report addresses what I believe to be the most important arguments that Drs. Church and Flyer put forth in their latest reports. I do not mean to suggest that I agree with other arguments that they make but which I do not address.

7 See, for example, Flyer 2015 Reply Report at ¶¶ 11 and 14 and Church 2015 Reply Report at ¶¶ 27-28, including footnote 51.

PUBLIC 5

that Google or other internet search engines have not had an impact on consumers because all of that information was available through other means (e.g., a trip to the local library). 8 The manner in which information is provided is important not only in terms of how easily consumers can access the information, but because it affects how consumers can use that information. The benefits of information available in machine readable formats are widely recognized – for instance, economists and accountants benefit from receiving data electronically instead of on paper, legal professionals benefit from online law databases that are easy to access and search. In the real estate industry, Gary Simonsen, CREA’s CEO, speaks to the fact that it is not just what information consumers get, but how they get and use that information, that is important. In particular, Simonsen speaks to how consumers embraced improvements in Realtor.ca that allow consumers to search listings with internet-based apps (including one for the new Apple Watch). More generally, CREA’s Simonsen testifies to how consumers are changing the manner in which they choose to access real estate data. 9 VOWs represent an important innovation in the manner in which consumers can access real estate information. Rather than having to rely on their broker to provide information about the real estate market (including information about the excluded data fields), consumers can now use their broker’s VOW to obtain that same information at any hour of the day, seven days a week. This can significantly increase consumers’ ease of access to that information. But just as important, VOWs provide consumers a new means by which they can use that information. 10 Brokers’ VOWs can offer tools that allow consumers to sort, summarize and analyze this information by helping them perform tasks such as reviewing sold listings in particular neighborhoods, analyzing how commission rates affect sold homes’ days on market, or determining comparables. Thus, VOWs help transform how consumers view and use information. 11 Put simply, the value of information does not depend solely on the content of that information, but also on the accessibility of that information. To conclude that consumers should be

8 Similarly, consumers are hardly indifferent to receiving emails rather than letters, watching movies at home (or on their tablet) rather than going to a movie theater, or getting traffic alerts on their phone rather than the radio.

9 Updated Witness Statement of Gary Simonsen (June 1, 2015) at ¶¶ 32 and 38. 10 A VOW data feed that includes the excluded data fields can also directly benefit brokers by allowing the broker to more readily conduct their own data analyses that can, in turn, be provided to consumers. Brokers can also use a full-information VOW data feed to conduct analyses that allow the broker to provide better service to their client. See Second Witness Statement of John Pasalis (February 2, 2015) at ¶¶ 5-9 and 12.

11 See Vistnes June 2012 Report at ¶¶ 51 and 58-71, Vistnes August 2012 Reply Report at ¶¶ 10, 13 and 110 and Vistnes February 2015 Report at pp. 5-6. In the remainder of this report, my discussion of how brokers can use VOWs to change the manner in which information is delivered should be understood to include how consumers can use that information.

PUBLIC 6

indifferent to how they get information, as long as the content of that information is the same, contradicts that basic fact.

III. COMPETITION CAN BE REDUCED BY NON-PARTICIPANTS EVEN IN “COMPETITIVE” MARKETS

In concluding that TREB’s conduct has not reduced competition, Dr. Church erroneously relies on the belief that, “[s]ince the downstream markets are competitive, TREB’s conduct cannot be an anti-competitive practice or result in a substantial lessening or prevention of competition.” 12 As I have previously discussed, the observation that a market is “competitive” does not prevent TREB’s conduct from reducing competition. Competition is not a zero/one determination: markets can be characterized by different degrees of competition. Thus, even though the GTA real estate market can be fairly characterized as “competitive” despite TREB’s restrictions, the relevant economic inquiry is whether TREB’s conduct reduces competition relative to a but-for world without those rules.

Economists and policymakers recognize that certain conduct can restrict competition even in seemingly “competitive” markets. 13 In effect, TREB’s restriction on how brokers can compete changes the rules by which competition occurs. Thus, Dr. Church’s observation that there are many competing brokers is of little economic significance if TREB’s rules reduce competition between those brokers. By restricting how brokers can use VOWs to compete, TREB reduces competition relative to a but-for world without those constraints. Thus, even in a market with thousands of competing brokers, TREB’s conduct results in a market that is less competitive than it would have been otherwise. And that less competitive market is one in which consumers are worse off.

Dr. Church also argues that, because TREB does not itself compete in the market for real estate brokerage services, TREB cannot have market power and that competitive concerns can only arise if TREB’s conduct creates market power on the brokers’ behalf competing in the relevant

12 See also Church 2015 Reply Report at ¶ 39: “since the downstream markets are competitive, then TREB’s conduct cannot [reduce/prevent] competition,” and Church 2015 Reply Report at ¶ 36: “[if] the conduct of TREB does not create, enhance, or maintain market power ... [then] therefore it [TREB] cannot have substantially lessened or prevented competition.”

13 This recognition goes back to at least 1995. See, for example, Hovenkamp, H. (1995). Exclusive Joint Ventures and Antitrust Policy. Columbia Business Law Review, Vol. 1995 (No. 1), pp. 1-125 (hereafter referred to as Hovenkamp). For the most pertinent discussion, see pp. 65-70, 83 and 86 of Hovenkamp. Further, Lande and Marvel observe how changing the rules under which firms compete “fixes the fight” so that firms need not compete as fiercely (see Lande, R. H. & Marvel, H. P. (2000). The Three Types of Collusion: Fixing Prices, Rivals, and Rules. 2000. Wisconsin Law Review, Vol. 2000 (No. 5), pp. 953-84). See also Lande, R. H., & Marvel, H. P. (2002). Collusion over Rules. Antitrust Magazine, Vol. 16 (No. 3), pp. 36-40.

PUBLIC 7

market. 14 It is, however, largely irrelevant that TREB is not a market participant in assessing the economic effect of TREB’s conduct. 15 TREB is an organization run by, and for the benefit of, its member brokers. 16 Economists and policymakers recognize that such trade associations and other organizations can often raise serious anti-competitive concerns and reduce competition even though the trade association or organization is not a competitor itself. 17 This role that TREB can have in reducing competition is critical in understanding why Dr. Church’s oft-stated emphasis on the significant competition that exists among brokers is misplaced and misleading. 18 Dr. Church relies on his belief that as long as there is significant competition among brokers and agents, TREB’s conduct can raise no serious competitive concerns. 19 Yet a trade organization such as TREB that can control how its members compete is a well-recognized vehicle by which that competition among members can be reduced. 20 In other situations, trade organizations operated by or for the benefit of members have changed the rules by which members can compete, e.g., by making it more difficult for others to enter the market and thus increase competition; by imposing rules that prohibit advertising among competitors or

14 Church 2015 Reply Report at ¶ 14. 15 TREB is not a market participant because TRB does not compete to provide buy- or sell-side residential real estate brokerage services. See Vistnes June 2012 Report at ¶¶ 22, 116-129 and 137-149, Vistnes August 2012 Reply Report at ¶¶ 22-33 (particularly ¶¶ 26-28) and Vistnes February 2015 Report at pp. 9 and 11-12.

16 Vistnes June 2012 Report at ¶¶ 121-125. 17 Vistnes June 2012 Report at ¶ 120. See also, for example, a speech by former U.S. FTC Commissioner Leibowitz, where he stated, “... self-regulation sometimes has a dark side because industry members can develop rules and so-called ethical provisions that restrict price competition or that keep out new competitors that threaten the established firms” (Jon Leibowitz. “The Good, the Bad and the Ugly: Trade Associations and Antitrust,” March 30, 2005. https://www.ftc.gov/sites/default/files/documents/public_statements/good-bad-and-ugly-trade-associations-and-antitrust/050510goodbadugly.pdf (last accessed 7/31/2015)). See also “Self-Regulation and Antitrust,” speech by FTC Chairman Pitofsky, February 18, 1998. A summary is available at https://www.ftc.gov/public-statements/1998/02/self-regulation-and-antitrust (last accessed 7/31/2015). The former FTC Chairman Robert Pitofsky states therein: “[the] self-regulation process may enable producers to exclude from the market the products of entrants that threaten to take market share from the incumbents. In response to the competitive threat of product innovation, incumbent producers may respond by engaging in self-regulation such as promulgating standards that exclude the innovators’ products from the market, rather than by improving their own products. Attempts to impede competition on the merits, and without strong justification, is a kind of ‘self-regulation’ that cannot be tolerated.” Pitofsky further noted, “Competitors may use the self-regulatory process to disadvantage new rivals or new forms of competition. When that happens, enforcement must be forceful and firm.”

18 See, for example, Church 2012 Report at ¶¶ 249 and 281-284 and Church 2015 Reply Report at ¶¶ 9, 15, 41, 42, Schedule C ¶ 6 and footnotes 11 and 26.

19 See, for example, Church 2015 Reply Report at ¶ 15 where Dr. Church argues that downstream competition among brokers is “completely inconsistent” with antitrust market power (emphasis in original). See also footnote 18 of this report.

20 See, for example, Hovenkamp at pp. 5, 23 and 65. These competitive concerns exist despite pledges such as CREA’s that members will act “with integrity.” (See CREA’s Pledge of Competition, available at http://www.crea.ca/content/pledge-of-competition (last accessed 7/31/2015). TREB also has a similar pledge and is available in Exhibit A in the Updated Witness statement of Donald Richardson.)

PUBLIC 8

reduce price competition among members; 21 by restricting the types of services that competitors can offer; or by allocating markets between competitors or otherwise restricting direct competition. 22 IV. TREB’S RULES REDUCE COMPETITION RELATIVE TO THE BUT-FOR WORLD

Drs. Church and Flyer argue that TREB’s restrictions on how brokers are allowed to use VOWs to provide information about the excluded data fields will not reduce competition. 23 Evaluating the competitive effect of this restriction on competition involves several key questions that economists commonly recognize to be the appropriate economic framework for analyzing conduct similar to TREB’s: 24 • Does TREB have substantial market power in a relevant market? • Can brokers readily substitute away from TREB’s MLS and obtain the excluded data fields from alternative sources?

21 Opinion of the Sixth Circuit Court of Appeals in re Detroit Auto Dealers Association Inc., et al., Bartlett Pontiac-Datsun Inc., et. al., v. F.T.C. (Decided January 31, 1992; rehearing and rehearing en banc denied May 22, 1992).

22 Porter, R. H. (2005). Detecting Collusion. Review of Industrial Organization, Vol. 26 (No. 2), pp. 147-167. At p. 154: “The Joint Executive Committee was a railroad cartel in the 1880s that transported grain and other provisions from Chicago to the eastern seaboard of the U.S. The trade association allocated market shares to its members, and kept records of shipments and prices charged by the various railroads.” Porter discusses collusion and enforcement via the garbage hauling associations in New York City at p. 158. 23 As I discussed in the Vistnes August 2012 Reply Report at ¶¶ 8-10, 71, 156 and 160, Dr. Church emphasizes that his analyses and opinions depend upon his application of a “essential facilities” analysis. Although there are common elements between an essential facilities analysis and a foreclosure analysis, they are not the same and cannot be used interchangeably. Unlike the current case, I understand that an essential facilities analysis is generally viewed as applying to a situation in which one firm is being required to make its assets accessible to its competitors. (See, for example, Pitofsky, R., Patterson, D., and Hooks, J., The Essential Facilities Doctrine Under U.S. Antitrust Law. Antitrust Journal, Vol. 70 (No. 2), 2002, pp. 443- 462. Pitofsky, et al. state that the essential facility doctrine concerns a limitation on the general rule that “a firm has no obligation to deal with its competitors” and concerns the situation in which “one firm, which controls an essential facility, denies a second firm reasonable access to a product or service that the second firm must obtain to compete with the first” (emphasis added).) This does not characterize the facts or conduct in this case: the question is not whether TREB should have to provide its MLS data to a competing MLS provider that might offer its services in competition to TREB. Rather, the question is how TREB’s prohibition on how brokers can use the MLS data that TREB already provides to those brokers is likely to affect how brokers compete with each other. In relying on an incorrect economic framework (i.e., an essential facilities analysis), Dr. Church’s analysis and opinions lose their relevance to this case.

24 See Vistnes February 2015 Report at pp. 2 and 8-10. As I noted within that report, this framework is often referred to as a “foreclosure” or “partial foreclosure” (see Vistnes February 2015 at p. 8 and footnotes 24 and 26). Although Dr. Church believes this foreclosure framework represents a change in my analytic framework, see Church 2015 Reply Report at ¶ 21, he is mistaken; my approach has been consistent throughout.

PUBLIC 9

• If brokers cannot readily substitute to an alternative source of information, does this harm VOW-based brokers by reducing their services’ attractiveness to customers, and does that harm to competing brokers substantially lessen competition?

• Is competition in real estate markets distorted by principal-agent problems, and is TREB’s conduct preventing VOWs from reducing those problems, thus raising prices and harming consumers by distorting competition?

• Are there pro-competitive benefits from TREB's discriminatory conduct in which TREB allows brokers to provide clients access to the excluded data fields through traditional means, but not through the brokers' VOWs?

As previously discussed, the answers to these questions demonstrate why TREB’s conduct likely reduces competition: TREB’s control over the MLS gives it the power to impose rules that brokers cannot ignore, and since brokers cannot practically turn to other equivalent sources of information regarding the excluded data fields, brokers are effectively prevented from providing that information on their VOWs. Inasmuch as the evidence shows that brokers would like to compete by offering those innovative VOW services that many consumers value 25 , this results in a substantial lessening of competition for which there is no offsetting efficiency justification. 26 In this report, I discuss the extent to which Drs. Church and Flyer introduce any new evidence bearing on the above questions and why I believe their new arguments fail to show that TREB’s conduct is unlikely to result in a substantial lessening of competition.

A. TREB has substantial market power 1. Summary of my previously stated opinion TREB has substantial market power in, and control over, the relevant markets. 27 This market power stems from TREB’s control over the MLS and brokers’ reliance on those MLS data in

25 See Vistnes June 2012 Report at ¶¶ 252-255, Vistnes August 2012 Reply Report at ¶¶ 10-13 and Vistnes February 2015 Report at pp. 18-19. See also the second witness statement of Tarik Gidamy (¶¶ 7, 14), John Pasalis (¶¶ 3, 9, 11), and Scott Nagel (¶¶ 4, 17) and the updated witness statement of Donald Richardson (¶¶ 176,189), Evan Sage (¶ 11.5) and Pamela Prescott (¶ 32). Note that Mark Enchin has invested over $60,000 in an incomplete VOW (¶ 5 of his third witness statement).

26 The full scope of likely anticompetitive effects from TREB’s conduct cannot be quantified with precision or reliability. These effects include both the short-run benefits consumers would receive from what they perceive to be more attractive VOW-based broker services and the long-run benefits associated with lower broker costs, reduced principal-agent problems that otherwise harm consumers and distort competition, and increased innovation involving services that allow consumers to better use and understand the information associated with the excluded data fields.

27 Vistnes June 2012 Report at ¶¶ 19, 74-79 and 155.

PUBLIC 10

order to compete. 28 TREB’s ability to deny brokers’ access to the MLS provides TREB with the power to exclude or disadvantage competitors and thus gives TREB substantial market power. 29 TREB’s market power associated with its ability to control access to the MLS data, and thus control over whether and how brokers can compete in the market, is important. Absent market power, brokers would not be compelled to follow TREB’s prohibition on brokers providing information about the excluded data fields over their VOW. TREB’s ability to threaten brokers with the loss of MLS access, however, gives TREB the power to enforce its prohibitions on how brokers can compete. 30 Thus, TREB’s market power associated with its control over brokers’ ability to compete is directly relevant to the question of whether TREB’s conduct (i.e., its prohibitions on how brokers can use their VOWs to compete) is likely to reduce competition. 31 2. Discussion of Drs. Church’s and Flyer’s positions in latest reports Neither Dr. Church nor Dr. Flyer present any new evidence or arguments showing that TREB does not have the ability to exclude competitors from the market or that TREB does not have the market power necessary to enforce its prohibitions on how brokers compete.

a) TREB’s market power is not defined with respect to the excluded data fields

Dr. Church suggests that TREB’s market power depends on consumers’ willingness and ability to substitute with respect to the excluded data fields. the analysis of TREB’s conduct, it is not the correct question to ask when assessing TREB’s

28 Vistnes June 2012 Report at ¶¶ 144-164 and Vistnes August 2012 Reply Report at ¶¶ 17-18 and 22-24. 29 The power to exclude or disadvantage competitors is evidence of market power. See Vistnes June 2012 Report at ¶¶ 142-144, 155, 160-164, 249 and 250 and Vistnes August 2012 Reply Report at ¶ 24. See also Hovenkamp at pp. 5, 88-9 and 93 (stating, “the market power inquiry ... [must consider] whether the exclusion rule permits members committed to an established technology or way of doing business to prevent the growth of a rival and perhaps superior technology or mode of doing business.”). See also Vistnes June 2012 Report at ¶¶ 145-151 and 157 and footnotes 154-161 and Vistnes August 2012 Reply Report at ¶¶ 161-163.

30 Vistnes June 2012 Report at ¶ 156-157. See also White, L. J. (2006). The Residential Real Estate Brokerage Industry: What Would More Vigorous Competition Look Like? Real Estate Law Journal, Vol. 35(Issue 1). White writes at page 16, “[T]he ability of the collective members of a MLS to exclude rivals, especially if those rivals are ‘mavericks’…, can be a powerful way of enforcing a high-fee structure and … market power.” He goes on to note: “if [access to the MLS was denied to brokerage], it would be a substantial disadvantage vis-à-vis its MLS-member rivals in attracting both sell-side and buy-side customers.” 31 Vistnes June 2012 Report at ¶ 164 and Vistnes August 2012 Reply Report at ¶ 28. See, for example, Hovenkamp who discusses how an organization’s enforcement power is an important aspect to consider when assessing that organization’s market power and ability to reduce competition among its members (Hovenkamp at pp. 88-90).

32 Church 2015 Reply Report at ¶¶ 27-29.

32 Yet, while that question is important to

PUBLIC 11

market power. 33 Instead, TREB’s market power depends on its ability to exclude competitors by denying brokers access to MLS data, thus ensuring that TREB has control over how brokers compete. Dr. Church fails to provide any evidence that brokers can compete absent access to TREB’s MLS data, and thus safely ignore TREB’s prohibitions on how they compete. 34 b) Market power does not depend solely on the ability to set supra-competitive prices

Dr. Church also argues that “market power is not defined as the ability to exclude or disadvantage competitors” but rather that it is “typically defined as the ability of a firm to profitably raise price above competitive levels.” 35 Yet, while Dr. Church is correct that market power is typically defined in the context of an ability to raise price above competitive levels, he is mistaken in suggesting that this is the only way to assess market power. As previously discussed, market power properly encompasses the ability to exclude or disadvantage competitors. 36 Under this metric, TREB clearly has market power. c) The power to exclude is not solely relevant to entry questions Dr. Church argues that “the mere ability to exclude competitors is not the exercise of market power that Dr. Vistnes states.” Dr. Church suggests that the power to exclude competitors is not evidence of market power, but is instead only relevant to the market power issue because the power to exclude can help a firm prevent entry and thus enjoy durable market power (i.e., market power that a firm can enjoy over “a considerable period of time”). 37 In making this argument, Dr. Church confuses two important economic issues: how the power to exclude entry can affect whether a firm can enjoy market power over “a considerable period of time,” and whether the power to exclude is evidence of market power.

Consider first the issue of how the power to exclude entry can affect whether a firm can enjoy market power over time. Dr. Church correctly observes that firms are unlikely to enjoy significant market power over time if entry is easy. Thus, if TREB were likely to face entry from

33 In essence, Dr. Church confuses the question of market power with the question of competitive effects. Although both questions are important, the evidence bearing upon each question differs and needs to be distinguished.

34 Dr. Church is also wrong as a matter of economics in claiming that “whether TREB does or does not have market power in the provision of MLS services is irrelevant” (Church 2015 Reply Report at footnote 24). To the contrary, TREB’s market power in the provision of MLS services is central to the question of whether TREB can enforce its prohibition on brokers’ use of VOWs to display the excluded data fields. Absent that market power, TREB could not enforce its prohibition.

35 Church 2015 Reply Report at ¶ 12. 36 See Vistnes June 2012 Report at ¶¶ 141-143, as well as this report’s Section IV.A discussing how price is not the only dimension by which the exercise of market power can be assessed. See also Hovenkamp, especially pp. 79- 89.

37 Church 2015 Reply Report at ¶ 13.

PUBLIC 12

other competing MLS providers, Dr. Church would be correct to conclude that TREB would not have significant market power. But Dr. Church has not claimed that TREB is likely to face competition from a new, competing MLS. 38 Thus, entry is unlikely to affect TREB’s ability to exclude brokers from the market if they violate TREB’s rules, and thus does not bear on the issue of TREB’s market power.

Yet, while Dr. Church is correct that the power to exclude can help a firm enjoy enduring market power, he errs in suggesting that this is the only relevance of TREB’s power to exclude competitors from the market: TREB’s power to exclude competitors in the brokerage services market also provides direct evidence of TREB’s existing market power. Dr. Church would be correct in concluding, however, that given that TREB is unlikely to face entry by a competing MLS, TREB’s market power is likely to last “a considerable period of time.”

d) TREB need not compete in the market in order to have market power Dr. Church argues that, even if TREB does have market power, entities such as TREB that are run on their members’ behalf have no incentive to exercise market power. 39 This claim, however, is at odds with the general economic literature which recognizes that trade associations and other organizations run by, and on behalf of, their members can raise serious competitive concerns. 40 Hence, it is wrong to assume that they will not exercise their market power. Economists and policymakers also recognize that trade organizations that are set up to promote and protect their members’ self-interest can sometimes impose rules that limit new forms of competition in ways that benefit some members to the detriment of other members. 41 38 Further, in my opinion, entry by a competing MLS in the GTA is extremely unlikely. First, I am aware of no evidence that any entity is contemplating such entry. Second, such entry would likely be prohibitively expensive and require many years. Third, I am aware of virtually no instances in the U.S. or Canada where there have been competing MLS entities. Rather, even in the very few instances in the U.S. where I am aware of two MLS entities operating in geographically proximate regions, those MLS entities largely operate in different areas rather than compete.

39 Church 2015 Reply Report at ¶ 17 and footnote 29. 40 Porter, R. H. (2005). Detecting Collusion. Review of Industrial Organization, Vol. 26 (No. 2), pp. 147-167. At p. 151, Porter writes “[c]ollusion will be abetted by any practice, such as information gathering and dissemination by a trade association, which speeds the detection of, and hence response to, defections from an agreement” (emphasis added). On the same page, it is noted, “a legal cartel could require that a defector sacrifice a bond with, or pay a fine to, a trade association” (emphasis added). See also Levenstein, M. C. & Suslow, V. Y. (2006). What Determines Cartel Success? Journal of Economic Literature, Vol. 44 (No. 1), pp. 43-95. Specifically, see Levenstein and Suslow at p. 44 where it states that “[t]here are in fact many successful cartels in quite unconcentrated industries, but they almost always rely on industry associations.” See also footnote 17 above regarding the competitive threat that industry organizations can pose.

41 See Vistnes June 2012 Report at ¶¶ 116-139 on how TREB may have incentives to implement rules that benefit some competitors (e.g., incumbents) at the expense of others (e.g., entrants). See also Hovenkamp at p. 65.

PUBLIC 13

B. Brokers cannot readily substitute away from TREB’s MLS data 1. Summary of my previously stated opinion In my previous reports, I provided evidence that brokers cannot readily obtain the information associated with the excluded data fields from other sources: much of the information associated with the excluded data fields is simply not available elsewhere, and the information that is available is more difficult to obtain, more costly to obtain, less accurate, and less timely than what would otherwise be available through TREB’s MLS system. Thus, to the extent that substitution is possible, it would be to an inferior, more costly, alternative. 42 2. Discussion of Drs. Church’s and Flyer’s positions in latest reports Dr. Church provides no new evidence regarding brokers’ ability to substitute to alternative information sources in the GTA.

Instead, he argues that TREB’s prohibition on brokers using their VOWs to provide information about the excluded data fields will not affect brokers because, if brokers really want that information, their demand will possibly stimulate “entry” by new firms that will offer that same information. 43 Dr. Church’s reliance on the possibility of entry is, however, misplaced. • Dr. Church’s suggestion that entry will ensure that brokers will have ready access to the excluded data fields from alternative sources 44 is misguided. Although Dr. Church cites to several firms that have real estate information, Dr. Church provides no evidence that those firms have access to the same timely, comprehensive, data that is otherwise available at no incremental cost to brokers through the MLS system.

• Dr. Church seemingly acknowledges that firms with certain public record real estate data (e.g., CoreLogic and RealtyTrak) are not currently offering data to brokers in the GTA, but he suggests that those data firms could enter that market and begin competing with the local MLS. 45 Simply asserting the mere possibility of entry, without any evidence about the timeliness, likelihood or sufficiency of such entry (including a more careful analysis of exactly what data those firms can, and cannot provide and the likely price at which they would make those data available), renders Dr. Church’s reliance on such

42 See Vistnes June 2012 Report at footnotes 234, 247, and 248 and ¶¶ 222-238, Vistnes August 2012 Reply Report at footnotes 128, 133 and 143 and ¶¶ 23, 85-90 and 91-105 and Vistnes February 2015 Report at pp. 12-14.

43 Church 2015 Reply Report at ¶¶ 30-31. 44 Church 2015 Reply Report at ¶¶ 29-31. 45 See Church 2015 Reply Report at ¶ 31 where he speculates about the possibility of a “supply-side response” if problems arose in the GTA.

PUBLIC 14

entry largely speculative. 46 As such, it does not provide a sound basis for concluding that brokers in the GTA can readily substitute from TREB’s MLS to other data providers.

• Dr. Church emphasizes that firms such as CoreLogic and RealtyTrak rely on “public record information” regarding properties. Even if such firms wanted to begin offering a product in the GTA, those firms would first require a license from Teranet. As discussed in my previous report, not only is Teranet’s data a poor substitute for the excluded data fields contained in TREB’s MLS data, . 47 This creates further uncertainty as to whether brokers could reasonably rely on entry by such firms as an alternative to the MLS data.

C. TREB’s conduct harms VOW-Based brokers and substantially lessens competition

1. Summary of my previously stated opinion Economists generally believe that markets are most efficient, and consumers best served, when competing firms decide how to compete and whether to try to better compete by offering a new product or service. Economists also typically accept that the market and consumers themselves are the best judge of whether new products or services are, in fact, valued by consumers, and thus typically try to avoid substituting the market’s judgment with their own. 48 Here, certain GTA brokers believe they can provide innovative services, and thus better compete, if they can use their VOW to provide clients with information relating to the excluded data fields rather than being forced to provide that information through other means. By using a VOW, those brokers believe that consumers can not only access information in a more convenient way, but those consumers can also analyze that information in ways that cannot be readily performed if that information were instead provided, e.g., over the telephone. 49 Thus, despite Dr. Church’s 46 Dr. Church identifies Zillow as a success story that somehow shows that other real estate data firms can be expected to enter the market. Zillow’s business model, however, is dramatically different from the presumed business model that Dr. Church seemingly believes that real estate data firms would follow were they to enter the market to sell data to brokers: while data firms would presumably earn revenue from selling data, firms such as Zillow earn money by attracting website visitors and then selling advertising “space” on its website. Further, Zillow does not offer comprehensive, timely information on the excluded data fields, making Dr. Church’s analogy to Zillow even less informative.

47 See also Vistnes August 2012 Reply Report at ¶¶ 90- 92.

48 See Vistnes February 2015 Report at p. 6. 49 Vistnes February 2015 Report at p.15. One way in which consumers can analyze that information is by using tools that the broker makes available over the VOW, e.g., a tool that allows the consumer to analyze sold prices for homes with particular characteristics in particular regions. Brokers’ use of VOWs to provide information about the excluded data fields can also lower brokers’ costs and allow them to better compete.

PUBLIC 15

opinion that the means by which those brokers seek to compete is of no real value, the brokers that live and work in the GTA disagree: they do expect this form of innovation to help them better compete and offer services that consumers value. 50 In short, those brokers believe that TREB is prohibiting a competitively significant service.

In my previous reports, I provided evidence that supports brokers’ belief that using VOWs to provide information about the excluded data fields is likely to be a competitively significant service. 51 As a result, TREB’s prohibition on brokers’ offering that service substantially lessens competition and harms consumers in the GTA.

2. Discussion of Drs. Church’s and Flyer’s positions in latest reports a) VOWs provide incremental value to consumers Dr. Church argues that TREB’s conduct is unlikely to reduce competition because the service that TREB prohibits brokers from competing to offer has no incremental value to consumers. To support that argument, Dr. Church suggests that, because consumers also value other types of information, the excluded data fields are not “uniquely useful.” 52 By this, Dr. Church presumably means that the services that certain GTA brokers want to offer are unlikely to provide any incremental value to consumers relative to what is already available.

In support of his belief that brokers’ VOWs offer no “uniquely” useful information to consumers, Dr. Church notes the popularity of real estate websites such as Zillow, Trulia, and Realtor.com. 53 From that, he concludes that prohibiting brokers from using their VOWs to provide this information should be of no competitive significance. Yet, while clearly popular, websites such as Zillow, Trulia, and Realtor.com are not substitutes for the services that GTA brokers provide. 54 • Those websites do not offer the same timely, comprehensive information regarding the excluded data fields that brokers could otherwise provide on their VOWs. 55 Thus, 50 See footnote 25 of this report. 51 See Vistnes June 2012 Report at ¶¶ 47-53, Vistnes August 2012 Reply Report at ¶¶ 54-58 and Exhibits A, C, E.1-3 and Vistnes February 2015 Report at pp. 11-12 and 18-19 and Updated Exhibits A, C, E.1-3.

52 See Church 2015 Reply Report at ¶ 32. 53 Church 2015 Reply Report at ¶¶ 32-33. 54 See Vistnes August 2012 Reply Report at ¶¶ 45-53, 57-58 and 62 and Vistnes February 2015 Report at pp. 16-18 and footnote 53. Similarly, those websites would not be good substitutes for brokers seeking to increase their own productivity by using VOW data feeds to conduct their own data analyses that the broker can then provide to their client or use to better serve their client. See footnote 10 of this report.

55 Those websites’ reliance on non-MLS data means their information relating to the excluded data fields is typically incomplete (e.g., not all sold listings are shown, there is little or no information regarding WEST listings or commissions) and is not up-to-date. Realtor.com, for example, provides no information about sold listings. Sites

PUBLIC 16

consumers in search of information about the excluded data fields simply cannot turn to those portals as substitutes to brokers’ VOWs. 56 • Those websites are not operated by, or on behalf of, brokers. Thus, brokers cannot turn to these websites as an alternative means by which they can compete to provide innovative new services.

Dr. Church’s definition of a “uniquely useful” product or service appears to be one that is not consumed together with other products or services. Thus, even if consumers value the information in the excluded data fields, Dr. Church appears to believe that the information is not “uniquely useful” because consumers also value other MLS information (e.g., information about current listings) even though the other MLS information is not a substitute. 57 In depending on such logic in reaching his conclusions, Dr. Church reaches the erroneous conclusion that consumers do not value the services that TREB prohibits brokers from using their VOWs to provide. 58 b) Consumers’ desire to work with brokers does not diminish VOWs’ value

Dr. Church suggests that consumers’ continued desire to work with a broker is evidence that VOWs are not useful. 59 That suggestion, however, is incorrect. There is little question that consumers value the services they receive from brokers, and that brokers are likely to remain an integral, and important, part of the real estate industry. This does not mean, however, as Dr. Church seems to imply, that VOWs have little or no value. VOWs are

such as Zillow offer some information about sold listings, but that information is neither comprehensive nor timely. See also Vistnes August 2012 Reply Report at ¶¶ 23 and 85-105.

56 Dr. Church fails to recognize that some of the data providers that those websites rely upon with respect to showing U.S. listings are not available in the GTA. For example, Trulia and Zillow rely upon public data sources and data supplied by agreements with MLSs, so similar websites in Canada would still not have access to the excluded data fields through these sources. And Realtor.ca, the GTA’s closest counterpart to Realtor.com, is not providing information about the excluded data fields.

57 Dr. Church does not claim that consumers treat information about sold listings as equivalent to information about current listings. Thus, he makes no effort to argue that the information provided on sites such as Zillow is a good substitute for the excluded data fields.

58 By this logic, one would conclude that because consumers both watch television and listen to the radio, neither is “uniquely useful” and thus consumers would not be harmed by the loss of either of those products. Similarly, Dr. Church’s logic would seemingly suggest that, because consumers enjoy both fiction and non-fiction books, or enjoy reading about both local news and international news, or sometime watch television but other times watch movies on Netflix, that neither is “uniquely useful” and thus offers no incremental value to consumers. Such a conclusion would be clearly erroneous: in each case, while there may be some similarities between the products or services, they each offer incremental value to consumers.

59 Church 2015 Reply Report at ¶ 54.

PUBLIC 17

used as a complement to the other services that brokers provide – not as a substitute for all services. 60 In fact, even pure-model VOWs such as RedFin and ZipRealty have emphasized how their brokers continued playing an important part in the real estate buying and selling process. 61 VOW-based brokers such as ZipRealty simply emphasized that VOWs could increase the effectiveness of brokers and increase consumers’ overall satisfaction with how they interacted with their broker.

It follows that consumers’ continued desire to work with brokers says nothing about the value of VOWs. Similarly, VOWs’ failure to drive brokers from the industry can hardly be taken as evidence that VOWs, and the additional services that could be offered were there no TREBimposed restrictions on how brokers can use those VOWs, are not valued by consumers.

It is also instructive to observe that, although ZipRealty and eRealty no longer exist as independently owned and operated firms, neither of those firms were forced from the market because consumers found their technologies to be unattractive. To the contrary, both of those firms were purchased by existing brokerages seeking to incorporate those firms’ VOW technology as a means of better competing for clients. The firm eRealty, and its technology, was purchased by Prudential in 2004 so that Prudential could incorporate that VOW technology into its own operations in an effort to better compete for more business. 62 Similarly, ZipRealty was purchased by Realogy 63 in 2014 with Realogy not only continuing to offer ZipRealty as a distinct product, but also incorporating ZipRealty’s technology to improve the services that its other brokers can provide to consumers. 64 60 More generally, VOWs provide consumers with a different way of interacting with brokers. VOWs allow consumers to obtain certain services directly from a broker’s VOW rather than having to contact the broker to obtain those services. As discussed in previous reports, many consumers prefer this way of interacting with their broker and brokers can also reduce their own costs, thereby compete more effectively, by using VOWs in this fashion.

61 A “pure-model” VOW is one in which brokers are not operating out of traditional brick-and-mortar offices. See Vistnes June 2012 Report at ¶ 66, Vistnes August 2012 Reply Report at ¶¶ 34-36 and Vistnes February 2015 Report at pp. 2-3.

62 See, for example, Evans, B. (2004). Prudential Explores Ways to Capitalize on eRealty Acquisition. RealtyTimes. http://realtytimes.com/todaysheadlines1/item/12878-20040213_prudential (last accessed 7/31/2015).

63 Realogy owns NRT, Coldwell Banker, Century 21, ERA and Sotheby’s International. 64 Realogy emphasized the importance of ZipRealty’s technology for how brokers can provide service to clients: “‘We like their software platform, which this industry lacks,’ Richard Smith, Realogy's chief executive, said in an interview …. ‘It can fulfill a void that we have recognized for some time.’” (Li, R. (2014). Realogy Holdings Corp. to Buy ZipRealty for More Lucrative Leads. The Wall Street Journal. http://www.wsj.com/articles/realogy-holdings-corp-to-buy-ziprealty-for-more-lucrative-leads-1405906684 (last accessed 7/31/2015)). In a statement at the time of the acquisition, Realogy also stated, “This transaction has two uniquely attractive facets for Realogy. First, we are acquiring an established, highly productive, technology-based national residential brokerage operation. Second, we will capitalize on the innovative technology platform that ZipRealty has honed over more than a decade of development. ZipRealty's technology provides a seamless digital experience for consumers, brokers and sales associates across the entire real estate transaction life cycle. We intend to fully leverage ZipRealty's comprehensive suite of world-class technology tools across our business, enabling both our franchise brands and our company-

PUBLIC 18

c) The manner by which information is accessed and analyzed can be as important as the information itself

Dr. Flyer argues that TREB’s prohibition on brokers’ ability to compete to provide this information on their VOWs, rather than provide it through more traditional means is unimportant: “consumers currently are not deprived of the information in the [excluded] data fields, as those who want the information in the [excluded] data fields have various other means to obtain it.” 65 Dr. Flyer also states, “Therefore, why is it so important to be able to get this information from your broker via the broker’s VOW versus through the other means of communication that are likely taking place?” 66 Thus, Dr. Flyer suggests that, as long as a consumer has access to the information associated with the excluded data fields, it should not matter whether that information is provided over a broker’s VOW versus over the telephone, via a fax, or through an email.

The answer to Dr. Flyer’s question, as I have previously discussed, is that the manner by which this information is provided to consumers (i.e., a VOW) can be just as important as the information itself, and that VOWs constitute a method of information delivery that can provide important additional benefits to consumers – both by allowing consumers more direct access to the data, but also by allowing the consumer to use tools to automate the analysis of those data. 67 Further, by suggesting that the manner in which information is provided does not matter, Dr. Flyer is seemingly at odds with his apparent recognition that digital technology (such as VOWs) is an important innovation relative to traditional technologies and that the mechanism by which consumers receive information is important. In particular, Dr. Flyer appears to acknowledge that digital technology is an important business innovation that has resulted in consumer benefits in numerous areas of commerce. 68 owned operations to be more productive, efficient and better serve their customers.” (Realogy Announces Agreement to Acquire ZipRealty for Approximately $166 Million. PRNewswire. http://www.prnewswire.com/news-releases/realogy-announces-agreement-to-acquire-ziprealty-for-approximately-166-million-267223371.html (last accessed 7/31/2015)).

65 Flyer 2015 Reply Report at ¶ 14. 66 Flyer 2015 Reply Report at ¶ 11. Thus, Dr. Flyer clearly is not suggesting that the excluded data fields contain confidential information that should not be provided to clients. (See also Flyer 2015 Reply Report at ¶ 14 where he notes that “consumers currently are not deprived of the information in the disputed fields, as those who want the information in the disputed fields have various other means to obtain it.”) Rather, his argument rests on the belief that the broker will provide that information – just not over the broker’s VOW.

67 See Section IV.C.1and Section II of this report. Also see Vistnes August 2012 Reply Report at ¶¶ 10-11 and 108-110 and Vistnes February 2015 Report at pp. 3-4, 21 and 22.

68 Flyer 2015 Reply Report at ¶¶ 13-14.

PUBLIC 19

D. TREB’s conduct distorts competition because of principal-agent problems

1. Summary of my previously stated opinion In my previous reports, I discussed the evidence showing that competition in real estate markets in the GTA is distorted by principal-agent problems. These principal-agent problems include buy- and sell-side steering to homes that give brokers higher commissions and result in inefficient matches. These principal-agent problems would likely be mitigated, and thus competition would likely increase, if brokers could use their VOWs to provide information to consumers about the excluded data fields. 69 In these reports, I discussed evidence showing: • The prevalence of Dual Agency transactions is evidence of principal-agent problems in which brokers steer their buy-side clients to purchase a home the broker is listing, and steer their sell-side clients to sell to a client the broker is representing.

• There is additional evidence of Buy-Side steering in which brokers steer their clients away from homes offering low commissions to the broker.

• There is evidence of Sell-Side steering in which, by offering low commissions to the buy-side broker, a sell-side broker can make a home less attractive to other brokers and their clients, and thus increase the likelihood of making the sale to one of their own clients.

These principal-agent problems arise in substantial part because of information asymmetries between brokers and their clients. Allowing brokers to use their VOWs to show information about the excluded data fields would leave consumers better informed about the real estate market, thus reducing the information asymmetries that give rise to these principal-agent problems that distort and reduce competition.

2. Discussion of Drs. Church’s and Flyer’s positions in latest reports a) Competition among brokers will not prevent principal-agent problems Dr. Church argues that TREB’s prohibition on brokers’ ability to compete by showing the excluded data fields on their VOWs is unlikely to increase principal-agent problems and distort competition because:

69 See, for example, Vistnes June 2012 Report at ¶¶ 39, 47-53, 133-140, 190-197, 201-215, 216-221 and 299, Vistnes August 2012 Reply Report at ¶¶ 22-33 and 40 and footnote 58 and Vistnes February 2015 Report at pp. 2, 15 and 22-27.

PUBLIC 20

“competition and the ready availability of substitutes with relatively similar information seems likely to preclude the possibility that TREB’s restrictions [relating to the excluded data fields] would preserve the ability of agents to exploit this information at the expense of home buyers and sellers.” 70 In making this argument, Dr. Church seemingly concludes that principal-agent problems are unlikely because brokers will compete to provide the information that could otherwise be used to exploit consumers. Yet TREB’s prohibition on brokers’ ability to provide that information with their VOWs thwarts the very competition that Dr. Church believes would solve that principal-agent problem.

b) Dr. Church acknowledges that principal-agent problems exist in GTA real estate markets

In discussing the National Association of Realtors (NAR) Profile, Dr. Church appears to acknowledge how increased information of the type associated with the excluded data fields can reduce principal-agent problems in GTA real estate markets. Dr. Church notes that “consumers often take the lead in deciding which homes they want to see because they have access to listing data on the Internet” and seemingly acknowledges that this information helps prevent customers from being “steered by the agent, who no longer leads the process.” 71 Thus, Dr. Church does not dispute that greater consumer information helps prevent steering problems, but instead appears to limit his dispute to whether consumers already have enough information: in Dr. Church’s opinion, consumers do not need the information associated with the excluded data fields, and if they do, then they can just as well get it over the telephone, the fax or other traditional means. 72 c) Analyses based on data subsets are routinely relied upon by economists

In previous reports, I discussed how, by offering a low commission rate, a sell-side broker can discourage other buy-side brokers from showing their client a listing, thus increasing the likelihood that the sell-side broker can sell the home to their own client and thus earn the entire

70 Church 2015 Reply Report at ¶ 61. 71 Church 2015 Reply Report at ¶ 67. 72 Dr. Church also appears to again conclude that consumers simply do not need the information that brokers want to provide over their VOWs (and thus clearly believe that consumers do want) when he argues in his 2015 Reply Report at ¶ 69 that much of the information embodied in the excluded data fields would be too difficult for most consumers to interpret. Yet Dr. Church provides no evidence suggesting that the information that brokers believe their clients want is information that those clients are unable to interpret or use.

PUBLIC 21

commission themselves. I then showed how dual agency outcomes are much more common when the sell-side broker sets a low commission rate. 73 Dr. Church discounts the significance of this finding by observing that there are only a limited number of instances in the GTA in which those low commission rates are offered. As a result, he suggests that this means steering is not a significant problem in the GTA. 74 Dr. Church’s interpretation of the data, however, is too simple. 75 The very nature of steering – providing inaccurate, misleading, or incomplete information to clients in an effort to change the home they purchase – makes it very difficult to empirically measure. The analysis I conducted, and that Dr. Church comments upon, is simply one way in which one can test for a fairly extreme form of steering – and here, despite the relatively limited number of instances relevant to this particular test, the evidence confirms the hypothesis that brokers use commission rates to steer.

This test for steering should not be interpreted, as Dr. Church suggests, as an indication for the prevalence of that steering problem. Rather, the test is simply meant to show the problem’s existence without speaking to its prevalence. Rather than rely on this test, one can look to the general industry recognition that steering is a pervasive problem, and that assuming it away would be “naive.” On this point, CREA’s Corporate Legal Counsel was clear, stating:

“As it stands now, hordes of real estate agents are constantly in breach of their agency duties because they are not showing buyers all the properties available that meet the buyer’s requirements, because many of those properties only pay $1.00 …. The licensee has an agency obligation to s[h]ow everything whether or not a commission is being paid. But we’re being naive if we really think that’s happening” (emphasis added). 76 73 See Vistnes February 2015 Report at p. 25 and Updated Exhibit 14, Vistnes August 2012 Reply Report at ¶¶ 132-137 and 145 and Vistnes June 2012 Report at ¶¶ 289-294 and 302-304 and Exhibits 13a, 13b, 14a and 14b.

74 Church 2015 Reply Report at ¶ 62. Dr. Church also suggests that dual agency situations may be efficient. In support of this, he cites to an article stating, “[B]uyer-initiated dual agency can be a desirable outcome when it creates search efficiencies … although in the negotiation phase potential incentive misalignments do exist ….” (emphasis added). Concluding that an outcome is efficient when and if it creates efficiencies, however, is hardly evidence that such efficiencies are likely (Church 2015 Reply Report at footnote 148).

75 I also note that the other empirical analyses I discuss that provide evidence of steering are not subject to this criticism: the analyses summarized in Vistnes February 2015 Report’s Updated Exhibit 12 are not limited to observations in which buy-side commissions are less than 1 percent, and thus cover many more listings.

76 Previously cited at Vistnes June 2012 Report at ¶ 277. See also the government reports, industry articles, newspaper articles, broker quotes and testimony cited in my previous reports that all suggest that steering and principal-agent problems are a serious problem in the GTA and, more generally, the real estate industry (Vistnes June 2012 Report at ¶¶ 276-304 and Vistnes August 2012 Reply Report at ¶¶ 131-139).

PUBLIC 22

E. There are no offsetting efficiency justifications for TREB’s conduct 1. Summary of my previously stated opinion In my past reports, I discussed the evidence showing that TREB’s conduct provides no pro-competitive benefits that offset, or justify, the restrictions that TREB imposes on how brokers compete. In particular, I discussed how there was no evidence that allowing brokers to use VOWs, rather than more traditional means such as telephones or faxes, to provide their clients with information about the excluded data fields would likely threaten the MLS system’s future viability, or that a fear of “free-riding” would cause brokers to provide fewer or lower quality services to consumers. 77 Rather, the evidence is more consistent with TREB seeking to protect brokers’ self-interest and profits, and protect brokers from the need to respond to new, innovative competition that appealed to many consumers. 78 2. Discussion of Drs. Church’s and Flyer’s positions in latest reports a) Efficiencies cannot be inferred because individual brokers have no market power

Dr. Church suggests that, as long as brokers have no market power, TREB’s conduct must be “efficiency-related.” 79 As discussed above, however, individual broker’s lack of market power does not prevent TREB from reducing competition by changing the rules by which brokers compete. 80 In fact, as commonly recognized, organizations such as TREB can be particularly worrisome because they can impose anti-competitive rules that members, none of which may individually have market power, would be unable to impose. 81 Thus, individual brokers’ lack of market power is not evidence that TREB’s conduct is efficiency-related.

77 Dr. Church writes that my previous analysis of why TREB’s conduct is unlikely to yield any pro-competitive benefits “no longer applies” because he claims (without explaining why) that my use of a foreclosure framework to analyze TREB’s analysis is “effectively an admission that the confidential price data is an essential facility” (Church 2015 Reply Report at ¶79). Dr. Church is mistaken. Not only do I continue to believe that an essential facilities framework is not a useful construct to analyze TREB’s conduct, but my (continued) reliance on a foreclosure framework to analyze TREB’s conduct is largely irrelevant to assessing whether TREB’s prohibition on how brokers use VOWs to compete is likely to be pro-competitive.

78 See Vistnes June 2012 Report at ¶¶ 79-80 and Vistnes August 2012 Reply Report at ¶¶ 152-174. 79 Church 2015 Reply Report at ¶¶ 9 and 78. Other places where Dr. Church justifies TREB’s conduct as pro-competitive include the Church 2015 Reply Report at ¶¶ 39, 41 and 61 and footnotes 11 and 26.

80 See this report’s Section III and footnotes 13, 15, 16, 17, 20, 21 and 22. 81 In other contexts, an organization such as TREB might be said to serve as a “cartel ringmaster.” See, for example, Krattenmaker, T. G. & Salop, S. C. (1986). Anticompetitive Exclusion: Raising Rival's Costs to Achieve Power over Price. The Yale Law Journal, Vol. 96(No. 2), pp. 209-293.

PUBLIC 23

b) Reducing competition to increase broker profitability is not pro-competitive

Dr. Church suggests that TREB’s prohibition on brokers’ use of VOWs to provide information about the excluded data fields preserves incentives for brokers to invest and to restrict free-riding. 82 In essence, Dr. Church appears to argue that, but for TREB’s VOW policy, the GTA’s brokerage markets would become more competitive. That increased competition would lower broker profits, thereby “expropriat[ing] investment made by TREB members.” 83 Thus, TREB’s policies are necessary to reduce competition and protect member profits so that they have a continued incentive to “invest.”

Dr. Church’s argument that the increased profits associated with less competitive markets is seemingly an efficiency defense that could be used to justify otherwise anti-competitive conduct in any industry. Yet, Dr. Church offers no evidence of the specific types of investment or innovation that would suffer if TREB members faced greater competition. 84 Dr. Church’s suggestion that reducing competition is necessary to encourage such investment and innovation is also at odds with the belief that competition is an important motivation and stimulant of innovation. 85 c) VOWs’ failure to harm MLS systems is not a pro-competitive justification

Dr. Church seemingly acknowledges, or at a minimum does not dispute, that VOWs have not had any significant negative impact on MLS systems in other areas. 86 Rather than acknowledge that this provides evidence that rules restricting how brokers can use their VOWs are unnecessary to protect MLS systems, Dr. Church argues that this evidence should be interpreted as evidence that VOWs are competitively insignificant.

82 Church 2015 Reply Report at ¶ 9. 83 Church 2015 Reply Report at ¶ 78. 84 In fact, the evidence seems to be precisely to the contrary: TREB’s conduct is very clearly preventing members from using their VOWs to offer an innovative service.

85 See Vistnes June 2012 Report at ¶¶ 47-53 and 58-66 where I discuss how brokers offered VOWs in response to the competition they faced and as a means of differentiating the services they offered so as to better compete. Similarly, economists and policymakers have long recognized that competition can benefit consumers at the expense of competitors. “We should not block efficient transactions or conduct just because they will make life harder for rivals. Making life harder for rivals will generally force them to work even harder to keep up, multiplying the benefits to consumers.” Kolasky, W. J., “The Role of Competition in Promoting Dynamic Markets and Economic Growth.” November 12, 2002. http://www.justice.gov/atr/speech/role-competition-promoting-dynamic-markets-and-economic-growth (last accessed 7/31/2015).

86 Church 2015 Reply Report at ¶ 80.

PUBLIC 24

In making this argument, Dr. Church does not advance a pro-competitive justification for TREB’s rules. Rather, he effectively argues that TREB’s conduct is inconsequential because the form of competition it seeks to prohibit is inconsequential. Thus, his argument reduces to the previously examined issue of whether TREB’s conduct likely reduces competition.

d) Pocket and whisper listings do not provide a pro-competitive justification

Dr. Church argues that brokers in the U.S. “are withdrawing listings from MLS systems and instead marketing them on pocket listings or whisper listings,” and that the number of listings shown on non-MLS portals is increasing. 87 From this, he appears to conclude that TREB’s rules are necessary in order to protect brokers’ willingness to continue investing in, and participating in, the MLS system. Dr. Flyer makes a similar argument that off-MLS sales and other services that bypass the usage of an MLS system represent threats to the MLS system’s future viability. 88 I do not understand the basis for Dr. Church’s apparent conclusion. • Neither Dr. Church nor Dr. Flyer provide any evidence that brokers’ use of VOWs to show information regarding the excluded data fields will reduce the brokers’ desire to use the MLS system.

• Neither Dr. Church nor Dr. Flyer provide any evidence regarding the likely magnitude of the relationship they hypothesize: by just how much would TREB’s conduct affect MLS bypass, and just how much impact would that bypass have on the MLS systems’ future viability?

• Dr. Church provides no evidence that brokers’ use of pocket and whisper listings has been changing over time or is in any way related to VOWs.

• Dr. Church is correct that brokers are increasingly providing listing information to portals such as Trulia and Zillow. Dr. Church does not point out, however, that those brokers typically continue to list those homes on the local MLS. In other words, brokers are not abandoning their MLS. Instead, brokers are recognizing that displaying those listings on additional websites gives greater exposure to their listing and thus increases the likelihood that the listing will sell quickly.

87 Church 2015 Reply Report at ¶ 81. Dr. Church cites a variety of industry trade press articles (Inman News) in support of this claim.

88 Flyer 2015 Reply Report at p. 7.

PUBLIC 25

e) There is no plausible free-riding argument Dr. Flyer argues that harm to the MLS might arise because many VOW users do not end up purchasing a home. As a result, brokers have incentives to use their VOW websites as an advertising vehicle, and the benefits of that advertising go to the VOW broker rather than the listing broker. Thus, Dr. Flyer introduces a type of “free riding” argument as a pro-competitive justification for TREB’s restriction on how brokers can compete.

As discussed in my previous report(s), there is an important difference between “free riding” and a “free riding problem.” A free riding problem only results from free riding when free riding leads to a less desirable outcome such as higher prices, reduced service or lower quality. Yet Dr. Flyer introduces no evidence, or even argument, for why market outcomes will be worse if some brokers use their VOWs in order to advertise. This lack of evidence is telling given that brokers have been offering VOWs in both the U.S. and parts of Canada for years: 89 if Dr. Flyer’s feared outcome were realistic, one would expect to see widespread evidence of harm. 90 It is also instructive to note that both Drs. Church and Flyer argue that brokers are increasingly providing their listings to websites such as Zillow and Trulia despite the fact that those websites use those listings to earn advertising revenues. 91 Those advertising revenues are not shared with the brokers. Thus, brokers are willingly working with those websites despite what Dr. Flyer calls website “free-riding” off those listings. This highlights that there is no real free-riding going on; rather, the relationship between brokers and those websites is more symbiotic, with both the website and the brokers benefitting from the relationship.

V. SUMMARY Having reviewed the Reply reports from Drs. Church and Flyer, as well as the additional evidence cited in this report, it remains my opinion that by restricting how brokers can use their VOWs to show information about the excluded data fields, TREB’s conduct is likely to substantially lessen competition in the relevant markets.

89 These VOWs generally do not show the excluded data fields, but Dr. Flyer’s argument is just as applicable to those VOWs: many consumers using those VOWs do not purchase homes, and by Dr. Flyer’s argument those brokers will have incentives to use their VOWs as advertising vehicles that will, in turn, create ruinous incentives for listing brokers.

90 Of particular note, the National Association of Realtors (NAR) in the U.S. operates its own website showing listings (Realtor.com). Like VOWs, individuals often view this website, though they ultimately do not purchase a home. In addition, NAR’s website earns money by showing advertisements for a variety of products (e.g., Dyson Vacuum cleaners, Nissan automobiles, Staples Office Equipment, and Panera Restaurants). It would be very surprising that a member-oriented organization such as NAR would operate such a website if Dr. Flyer’s concerns were legitimate that such advertising poses a serious threat to the viability of MLSs.

91 As noted above, brokers typically place their listings on those websites in addition to, not as an alternative to, placing them on their local MLS.

PUBLIC 26

PUBLIC 27

Appendix A Materials Relied Upon

Expert Reports Expert Report of Gregory Vistnes, June 22, 2012 Reply Expert Report of Gregory Vistnes, August 23, 2012 Expert Report of Jeffrey Church, July 27, 2012 Expert Report of Fredrick Flyer, August 13, 2012 Expert Report of Gregory Vistnes, February 6, 2015 Expert Report of Jeffrey Church, May 15, 2015 Updated Expert Report of Fredrick Flyer, June 2, 2015

Witness Statements Second Witness Statement of Tarik Gidamy, January 20, 2015 Second Witness Statement of William McMullin, February 4, 2015 and Exhibits Second Witness Statement of John Pasalis, February 2, 2015 Second Witness Statement of Sam Prochazka, February 3, 2015 Witness Statement of William McMullin, June 18, 2012 Trial Testimony: September 10 - October 18, 2012 Closing Submissions of the Commissioner of Competition, October 15, 2012 Second Witness Statement of Scott Nagel, February 5, 2015 Third Witness Statement of Mark Enchin, February 2, 2015 Third Witness Statement of William McMullin, July 31, 2015 Updated Witness Statement of Donald Richardson, May 13, 2015 and Exhibits Updated Witness Statement of Evan Sage, May 15, 2015 Updated Witness Statement of Pamela Prescott, May 15, 2015 and Exhibits Updated Witness Statement of Timoleon Syrianos, May 13, 2015 and Exhibits Updated Witness Statement of Tung-Chee Chan, May 15, 2015 and Exhibits Updated Witness Statement of Gary Simonsen, June 1, 2015 and Exhibits

Third Party Sources "The Good, the Bad and the Ugly: Trade Associations and Antitrust", speech by FTC Commissioner Jon Leibowitz, March 30, 2005 https://www.ftc.gov/sites/default/files/documents/public_statements/good-bad-and-ugly-trade-associations-and-antitrust/050510goodbadugly.pdf Opinion of the Sixth Circuit Court of Appeals in re Detroit Auto Dealers Association Inc., et al., Bartlett Pontiac-Datsun Inc., et. al., v. F.T.C. (Decided January 31, 1992; rehearing and rehearing en banc denied May 22, 1992) Kolasky, W. J. (2002, November 12). The Role of Competition in Promoting Dynamic Markets and Economic Growth. Tokyo, Japan: Speech presented TokyoAmerica Center. http://www.justice.gov/atr/speech/role-competition-promoting-dynamic-markets-and-economic-growth

“Self-Regulation and Antitrust,” speech by FTC Chairman Pitofsky, February 18, 1998. Structural Engineers Association of Northern California, 112 F.T.C. 530 (F.T.C. 1989). West Penn Multi-List, Inc., 81 F.T.C. 167 (F.T.C. 2009). Williamsburg Area Association of Realtors, Inc., 61 F.T.C. 268 (F.T.C. 2006).

Websites http://realtytimes.com/todaysheadlines1/item/12878-20040213_prudential http://www.wsj.com/articles/realogy-holdings-corp-to-buy-ziprealty-for-more-lucrative-leads-1405906684 http://www.prnewswire.com/news-releases/realogy-announces-agreement-to-acquire-ziprealty-for-approximately-166-million-267223371.html http://www.crea.ca/content/pledge-of-competition https://www.ftc.gov/public-statements/1998/02/self-regulation-and-antitrust http://www.crea.ca/content/pledge-of-competition

Scholarly Articles Baker, J. B. (2013). Exclusion as a Core Competition Concern. Antitrust Law Journal , 78(3), 527-589. Hovenkamp, H. (1995). Exclusive Joint Ventures and Antitrust Policy. Columbia Business Law Review , 1-125. Krattenmaker, T. G., & Salop, S. C. (1986). Anticompetitive Exclusion: Raising Rival's Costs to Achieve Power over Price. The Yale Law Journal , 209-293. Land, R. H. & Marvel, H.P., “The Three Types of Collusion: Fixing Prices, Rivals and Rules,” 2000 Wisconsin Law Review, No. 941. Land, R. H. & Marvel, H.P., “Collusion over Rules,” Antitrust Magazine , Vol. 16, No. 3, 2002, pp. 36 - 40 Levenstein, M. C., & Suslow, V. Y. (2006). What Determines Cartel Success? Journal of Economic Literature , 43-95. Pitofsky, R., Patterson, D., and Hooks, J., “The Essential Facilities Doctrine Under U.S. Antitrust Law, Antitrust Journal , Vol. 70, 2002, pp. 443 – 462 Porter, R. H. (2005). Detecting Collusion. Review of Industrial Organization , 147-167. White, L. J. (2006, Summer). The Residential Real Estate Brokerage Industry: What Would More Vigorous Competition Look Like? Real Estate Law Journal , Vol. 35(Issue 1).

Vous allez être redirigé vers la version la plus récente de la loi, qui peut ne pas être la version considérée au moment où le jugement a été rendu.

Ne plus afficher ce message