This is part 3/3 of a series of blogs on the history of the Market Data Wars. Parts 1 and 2 are here and here. Feedback and suggestions more than welcome at benjamin.connault@iextrading.com
Part 1 of this blog series took us from December 1999 to January 2011 and described the beginnings of the Market Data Wars. Part 2 covered January 2011 to March 2015, focusing on the “ArcaBook II” series of market data court cases.Part 1 of this blog series took us from December 1999 to January 2011 and described the beginnings of the Market Data Wars. Part 2 covered January 2011 to March 2015, focusing on the “ArcaBook II” series of market data court cases.
Today’s Part 3 covers March 2015 to January 2020 and focuses on the rise of “platform theory” arguments in the Market Data Wars. Each Part contains multiple sections; each section revolves around one expert report. Details and link to each expert report are provided at the beginning of each section.Today’s Part 3 covers March 2015 to January 2020 and focuses on the rise of “platform theory” arguments in the Market Data Wars. Each Part contains multiple sections; each section revolves around one expert report. Details and link to each expert report are provided at the beginning of each section.
8) 2018–2019: Platform theory enters the Market Data Wars
Ordover and Bamberger (with Compass Lexecon) for Nasdaq, February 2019. Available here.Ordover and Bamberger (with Compass Lexecon) for Nasdaq, February 2019. Available here.
In October 2018, the SEC issued a final opinion in the “ArcaBook II” case. In it, the SEC stood once more by the 2010 findings of the U.S. Court of Appeals for the D.C. Circuit (“D.C. Circuit”). It used the June 2008 market-based framework, but, within that framework, found that the exchanges had not submitted enough evidence to show that the fees were constrained by competitive forces. NYSE and Nasdaq immediately appealed to the D.C. Circuit. (This appeal was recently granted on procedural grounds; see Parts 1 and 2).In October 2018, the SEC issued a final opinion in the “ArcaBook II” case. In it, the SEC stood once more by the 2010 findings of the U.S. Court of Appeals for the D.C. Circuit (“D.C. Circuit”). It used the June 2008 market-based framework, but, within that framework, found that the exchanges had not submitted enough evidence to show that the fees were constrained by competitive forces. NYSE and Nasdaq immediately appealed to the D.C. Circuit. (This appeal was recently granted on procedural grounds; see Parts 1 and 2).
This opinion reignited the Market Data Wars. The same month, the SEC held a roundtable on Market Data and Market Access. In May 2019, the staff of the SEC’s Division of Trading and Markets published a “Staff Guidance on SRO Rule Filings Relating to Fees.” The Staff Guidance provided a high level of detail regarding the standards that market data fee filings should meet, within the market-based framework used since 2008. Ten years after their “joint products with joint costs” report, Nasdaq hired Ordover and Bamberger to submit yet another expert report on Nasdaq’s behalf. The report was published as a comment to the roundtable on Market Data and Market Access.This opinion reignited the Market Data Wars. The same month, the SEC held a roundtable on Market Data and Market Access. In May 2019, the staff of the SEC’s Division of Trading and Markets published a “Staff Guidance on SRO Rule Filings Relating to Fees.” The Staff Guidance provided a high level of detail regarding the standards that market data fee filings should meet, within the market-based framework used since 2008. Ten years after their “joint products with joint costs” report, Nasdaq hired Ordover and Bamberger to submit yet another expert report on Nasdaq’s behalf. The report was published as a comment to the roundtable on Market Data and Market Access.
The 2019 Ordover-Bamberger report mostly reiterates Nasdaq’s old argument that “Nasdaq incurs ‘joint costs’ to produce ‘joint products’” so that “it is not possible to appropriately evaluate the pricing of connectivity services in isolation from the pricing of trading and other ‘joint’ services offered by Nasdaq.” The report does not really attempt to provide evidence which could fit into the market-based framework suggested by the SEC and supported by the D.C. Circuit. The report contains no direct argument that competition for order flow or availability of alternative market data products constrain the prices of Nasdaq’s market data feeds.The 2019 Ordover-Bamberger report mostly reiterates Nasdaq’s old argument that “Nasdaq incurs ‘joint costs’ to produce ‘joint products’” so that “it is not possible to appropriately evaluate the pricing of connectivity services in isolation from the pricing of trading and other ‘joint’ services offered by Nasdaq.” The report does not really attempt to provide evidence which could fit into the market-based framework suggested by the SEC and supported by the D.C. Circuit. The report contains no direct argument that competition for order flow or availability of alternative market data products constrain the prices of Nasdaq’s market data feeds.
One notable feature of the report is the renewed emphasis on “platforms.” This is likely because the Supreme Court’s Ohio v. Amex opinion, which was decided in June 2018, legitimized (multi-sided) platform theory as a valid theoretical framework in the context of antitrust and legal economics.[1] The theory is widely seen to be deeply laissez-faire in nature. Unsurprisingly, NYSE and Nasdaq, which had been trying to resist regulatory oversight on the pricing of their products since the beginning of the Market Data Wars, saw Ohio v. Amex as a huge opportunity.One notable feature of the report is the renewed emphasis on “platforms.” This is likely because the Supreme Court’s Ohio v. Amex opinion, which was decided in June 2018, legitimized (multi-sided) platform theory as a valid theoretical framework in the context of antitrust and legal economics.[1] The theory is widely seen to be deeply laissez-faire in nature. Unsurprisingly, NYSE and Nasdaq, which had been trying to resist regulatory oversight on the pricing of their products since the beginning of the Market Data Wars, saw Ohio v. Amex as a huge opportunity.
Early attempts by the exchanges at leveraging platform theory to their advantage were not very convincing: “Exchanges, like credit card companies, operate two-sided platforms that must attract two sets of customers purchasing two related products the pricing of which is inextricably linked” (August 2018). The long-standing confusion between multi-product and multi-sided “platforms” did not help. The Ordover-Bamberger report sticks to Nasdaq’s old multi-product platform theory, but adds a classical discussion of exchanges as multi-sided platforms when it comes to trading (with the two sides being liquidity providers and liquidity takers). How that discussion is supposed to impact market data prices is very unclear: “Competition is enhanced and consumers benefit when competitors with joint products and/or multisided platforms can employ varying competitive strategies and adjust their strategies in response to changes in competition and consumer demand.”Early attempts by the exchanges at leveraging platform theory to their advantage were not very convincing: “Exchanges, like credit card companies, operate two-sided platforms that must attract two sets of customers purchasing two related products the pricing of which is inextricably linked” (August 2018). The long-standing confusion between multi-product and multi-sided “platforms” did not help. The Ordover-Bamberger report sticks to Nasdaq’s old multi-product platform theory, but adds a classical discussion of exchanges as multi-sided platforms when it comes to trading (with the two sides being liquidity providers and liquidity takers). How that discussion is supposed to impact market data prices is very unclear: “Competition is enhanced and consumers benefit when competitors with joint products and/or multisided platforms can employ varying competitive strategies and adjust their strategies in response to changes in competition and consumer demand.”
9) December 2019: NYSE argues new flavor of exchanges as platforms
Rysman (with Cornerstone) for NYSE, December 2019. Available here.Rysman (with Cornerstone) for NYSE, December 2019. Available here.
Rule filings are the frontlines of the Market Data Wars. Given the new SEC’s stance on market data fees, as evidenced by the October 2018 ArcaBook opinion and the May 2019 Staff Guidance, exchanges had to adapt their market data fee filing strategy. As we saw in the previous section, Ohio v. Amex provided them with a very timely theory on which to lean, even though early attempts to leverage “platform theory” were not very convincing.Rule filings are the frontlines of the Market Data Wars. Given the new SEC’s stance on market data fees, as evidenced by the October 2018 ArcaBook opinion and the May 2019 Staff Guidance, exchanges had to adapt their market data fee filing strategy. As we saw in the previous section, Ohio v. Amex provided them with a very timely theory on which to lean, even though early attempts to leverage “platform theory” were not very convincing.
In December 2019, 18 months after Ohio v. Amex, NYSE finally produced a more substantial argument based on platform theory. To do so, it submitted an expert report in support of SR-NYSENAT-2019–31. The report proposes a new flavor of platform theory for exchanges: “exchanges as platforms between consumers of market data and consumers of trading services.”In December 2019, 18 months after Ohio v. Amex, NYSE finally produced a more substantial argument based on platform theory. To do so, it submitted an expert report in support of SR-NYSENAT-2019–31. The report proposes a new flavor of platform theory for exchanges: “exchanges as platforms between consumers of market data and consumers of trading services.”
Since the same exchange customer typically both buys data from and trades on a given exchange, this description is surprising at first sight — how can those be the two sides of a multi-sided platform? Furthermore, any multi-product firm appears to fall under such a multi-sided description, which would make the argument vacuous and inoperative. The report somewhat addresses these two points. According to it, “[t]he ‘sides’ of the market served by platforms need not be distinct sets of agents.” And “almost every firm has some elements of a platform to it […] it is not generally useful to try to distinguish whether firms are platforms or not, as we can most often find platform elements in a firm.”Since the same exchange customer typically both buys data from and trades on a given exchange, this description is surprising at first sight — how can those be the two sides of a multi-sided platform? Furthermore, any multi-product firm appears to fall under such a multi-sided description, which would make the argument vacuous and inoperative. The report somewhat addresses these two points. According to it, “[t]he ‘sides’ of the market served by platforms need not be distinct sets of agents.” And “almost every firm has some elements of a platform to it […] it is not generally useful to try to distinguish whether firms are platforms or not, as we can most often find platform elements in a firm.”
Ultimately, the report recognizes that the issue is “how important platform issues are in understanding a particular firm’s activities” and that “an assessment of the degree of competition between U.S. stock exchanges (and other trading venues for U.S. equities) is beyond the scope of this paper.” In summary, the report argues that exchanges may theoretically be analyzed as multi-sided platforms, but leaves it to NYSE to do so. This is precisely what NYSE has been attempting to do in various rule filings. Since December 2019, SR-NYSENAT-2019–31 was suspended by the SEC, withdrawn by NYSE, refiled under SR-NYSENAT-2020–05, suspended again, and subject to a rare SEC Request for Information: the frontlines of the Market Data Wars.Ultimately, the report recognizes that the issue is “how important platform issues are in understanding a particular firm’s activities” and that “an assessment of the degree of competition between U.S. stock exchanges (and other trading venues for U.S. equities) is beyond the scope of this paper.” In summary, the report argues that exchanges may theoretically be analyzed as multi-sided platforms, but leaves it to NYSE to do so. This is precisely what NYSE has been attempting to do in various rule filings. Since December 2019, SR-NYSENAT-2019–31 was suspended by the SEC, withdrawn by NYSE, refiled under SR-NYSENAT-2020–05, suspended again, and subject to a rare SEC Request for Information: the frontlines of the Market Data Wars.
10) January 2020: SIFMA offers non-platform theory of exchange competition in which unregulated market data products are priced at supra-monopoly levels
Glosten for SIFMA, January 2020. Available here.Glosten for SIFMA, January 2020. Available here.
SIFMA, the industry organization opposing NYSE and Nasdaq in the ArcaBook cases, did not stand still after NYSE filed SR-NYSENAT-2019–31. It submitted an opposing comment letter and attached an opposing expert report in January 2020.SIFMA, the industry organization opposing NYSE and Nasdaq in the ArcaBook cases, did not stand still after NYSE filed SR-NYSENAT-2019–31. It submitted an opposing comment letter and attached an opposing expert report in January 2020.
The Glosten report offers an alternative economic model of the market for market data to that of the Rysman report. In this model, “a broker looks at the prices of all the data and decides either it is worth buying the data or it is not.” This leads to “supra-monopoly prices,” meaning the prices set independently by each exchange group are higher than those that would be set by a monopoly group which owned all exchanges. In the report’s model, prices are not competitive and regulatory intervention is called for.The Glosten report offers an alternative economic model of the market for market data to that of the Rysman report. In this model, “a broker looks at the prices of all the data and decides either it is worth buying the data or it is not.” This leads to “supra-monopoly prices,” meaning the prices set independently by each exchange group are higher than those that would be set by a monopoly group which owned all exchanges. In the report’s model, prices are not competitive and regulatory intervention is called for.
The Glosten report does not directly address the platform arguments of the Rysman report. However, it does provide a few arguments that indirectly go against those of the Rysman report. For example, that “reduction in data price is unlikely to cause an increase in market share,” which would negate the linkages between the two sides of the Rysman report’s platforms.The Glosten report does not directly address the platform arguments of the Rysman report. However, it does provide a few arguments that indirectly go against those of the Rysman report. For example, that “reduction in data price is unlikely to cause an increase in market share,” which would negate the linkages between the two sides of the Rysman report’s platforms.
July 2020: Where we stand and what comes next
In July 2020, the Market Data Wars are raging as intensely as ever. The June 5, 2020, D.C. Circuit decision was a clear victory for NYSE and Nasdaq but did not advance the debate on merits. How free should exchanges be in designing and pricing their market data product offering? What type of standards does the Securities Exchange Act mandate for the market for market data? Clearly, some exchanges and the SEC have widely different opinions on these questions, with Federal Courts acting as slow and unpredictable arbiters between the two camps.In July 2020, the Market Data Wars are raging as intensely as ever. The June 5, 2020, D.C. Circuit decision was a clear victory for NYSE and Nasdaq but did not advance the debate on merits. How free should exchanges be in designing and pricing their market data product offering? What type of standards does the Securities Exchange Act mandate for the market for market data? Clearly, some exchanges and the SEC have widely different opinions on these questions, with Federal Courts acting as slow and unpredictable arbiters between the two camps.
Active battlefronts in the Market Data Wars include various exchange market data fee filings, such as SR-NYSENAT-2020–05, but also the SEC’s major rulemaking agenda around the public tape in U.S. equities.Active battlefronts in the Market Data Wars include various exchange market data fee filings, such as SR-NYSENAT-2020–05, but also the SEC’s major rulemaking agenda around the public tape in U.S. equities.
Although a comparatively more subdued battlefront over the 2000–2020 period, the public tape is a crucial piece of the Market Data Wars. The SEC’s ambitious rulemaking agenda in this area was reaffirmed in a very recent SEC policy speech and involves the end of effective-upon-filing fees for the public tape, a restructuring of its governance structure, and proposed major changes to the data content of the public tape and the economic model through which it is provided.Although a comparatively more subdued battlefront over the 2000–2020 period, the public tape is a crucial piece of the Market Data Wars. The SEC’s ambitious rulemaking agenda in this area was reaffirmed in a very recent SEC policy speech and involves the end of effective-upon-filing fees for the public tape, a restructuring of its governance structure, and proposed major changes to the data content of the public tape and the economic model through which it is provided.
In the short term, expect an all-out war on both the private market data products and the public tape battlefronts. In the long term, hope for a resolution that enables market data to serve investor protection, competition and price formation to its fullest potential. [1] In Ohio v. Amex, the Court found that American Express did not violate antitrust laws by contractually prohibiting merchants from incentivizing customers to use forms of payment other than American Express (so-called “anti-steering” provisions). The Court relied on a multi-sided analysis of the credit card market, with the two sides being merchants and consumers. In analyzing the impact of anti-steering provisions on both sides, it accepted the argument that the provisions ultimately did not harm consumer welfare. American Express had argued that the fees it charged merchants allowed it to offer a higher quality of service to consumers, the other side of the platform. The 5–4 opinion was immediately recognized to have wide-ranging implications for antitrust practice, by endorsing platform theory as a valid theoretical framework for antitrust purposes.In the short term, expect an all-out war on both the private market data products and the public tape battlefronts. In the long term, hope for a resolution that enables market data to serve investor protection, competition and price formation to its fullest potential. [1] In Ohio v. Amex, the Court found that American Express did not violate antitrust laws by contractually prohibiting merchants from incentivizing customers to use forms of payment other than American Express (so-called “anti-steering” provisions). The Court relied on a multi-sided analysis of the credit card market, with the two sides being merchants and consumers. In analyzing the impact of anti-steering provisions on both sides, it accepted the argument that the provisions ultimately did not harm consumer welfare. American Express had argued that the fees it charged merchants allowed it to offer a higher quality of service to consumers, the other side of the platform. The 5–4 opinion was immediately recognized to have wide-ranging implications for antitrust practice, by endorsing platform theory as a valid theoretical framework for antitrust purposes.