LSEG response to market data report 2025

Market Structure Partners (MSP) published a report in February 2025 entitled “There’s No Market in Market Data”. The report contains assertions about the equity market data businesses of five European exchange groups selected for inclusion by MSP, including LSEG’s London Stock Exchange and Turquoise.

Turquoise is the only pan-European MTF included in this analysis. The other pan-European MTFs, Cboe Europe and Aquis [Note1], were not included. This creates inconsistencies in the report which, in several instances, mixes reference to Turquoise, LSE and LSEG, to artificially include LSEG in analysis that in fact relates to Turquoise, in some places, and LSE in others.

In addition to this unbalanced foundation, the report contains a number of factual errors, misleading comparisons and unfounded conclusions. This article covers some of the most obvious ones we have detected so far. Given the volume of errors, we have not highlighted them all. The absence of statements on any other aspects of the report does not constitute our agreement or endorsement.

Had LSEG been approached in the process leading to the publication of this report, we believe these factual errors could have been avoided, as well as the erroneous analysis, which supports unfounded conclusions (and would do so even with accurate data).

Market data is available for either i) both venues combined or ii) TGHE (EU entity) only. For the combined package, TGHL (UK entity) is the contracting entity and accordingly, the revenue is recognised on the TGHL RCB disclosures. The report fails to understand the contracting arrangements and split of trading volume across the two Turquoise venues, and accordingly has arrived at wholly incorrect conclusions.

On several occasions, the report mixes arguments relating to LSEG, LSE and Turquoise. It goes on to speculate that the alleged reduction in Turquoise turnover could partly be attributed to LSEG’s sale of Borsa Italiana in 2021. There is no link: Borsa Italiana is an independent business and has never been part of Turquoise (or vice versa). The divestment of a completely separate trading venue had, by definition, no impact on Turquoise volumes or revenues.

Retail

The report states that 'LSEG' increased its data charge for private investors by 151.2% between 2017-2024 [Note 3]. Firstly, and as previously stated, it is mixing up different trading venues and inconsistencies to reach an ‘LSEG’ number.

Secondly, the report fails to indicate that:

• Turquoise waived private investor data charges throughout this period.
• In 2023, LSE simplified its price list for private investor fees by removing unused bandings. This had no impact on the data charges actually charged to customers.
  

Assuming the report is referring to London Stock Exchange, looking at the number itself, it is grossly inaccurate. For instance, the 151.2% increase stated in figure 2 was derived by MSP’s use of a theoretical “avatar” firm being subject to £123k monthly fees for London Stock Exchange data. In fact, the avatar firm would have been charged £25k monthly for LSE’s Unlimited Private Investor licence from 2018, a 79.7% reduction in monthly fees from the quoted 2017 fees. This fee is publicly available in the LSE Market Data price list.

Finally, the report does not acknowledge the benefits of London Stock Exchange’s announcement that we are waiving data charges for private investors from January 2025, supporting industry objectives to democratise access to capital markets: (Retail investors: democratising UK market access).

Display and non-display

Figure 14 purports to show the magnitude of change in cost for a single level 2 user, by comparing London Stock Exchange fees for a single display user in 2017 to an enterprise-style licence which conducts non-display principal trading, agency trading and operation of trading platforms in 2024. We have no use cases of a single display data user who serves all three of these functions. This is an unrealistic and misleading comparison.

Indeed, running similar analysis (based on public price lists) of market data fees levied by another pan-European MTF that is a competitor to Turquoise shows the same result, despite the report commending that same pan-European MTF for its approach to market data fees (as mentioned elsewhere in this blog, this MTF is excluded from the core analysis of the report). If we apply the figure 14 analysis to this pan-European MTF, for example, the cost of display data is £624 (2017 fees for L2 display user, annualised). The cost for the same data as non-display in 2024 would be £43,200, representing a change of 6823%. Nonetheless, we believe this comparison demonstrates that this measure is unrealistic and unhelpful.

The report also talks in percentage terms without clearly demonstrating the actual fee levels being discussed. For example, a Turquoise display user in 2017 would have paid a maximum of £10.50 per month for consumption of that data. The same user will be paying a maximum of £15.91 per month in 2025 (average annual change of £0.68 per user per month). Turquoise remains extremely good value for market data, which the report overlooks.

The report states London Stock Exchange’s price list has reduced to 6 pages due to removal of Borsa Italiana and formatting. However, it omits that we have consolidated the Turquoise agreement into the London Stock Exchange agreement during this period, simplifying the process for customers to sign up to data from both venues. A simple zoom-in on figure 12 shows this very clearly: the summary of fees shows that several of the additional lines are linked to Turquoise.

The report also claims that of LSEG’s venues, only Turquoise Europe makes reasonable commercial basis disclosures. Such disclosures are required to be published under both UK and EU regulation. All of LSEG’s equity trading venues publish these on the London Stock Exchange website [Note 4], in line with FCA and ESMA requirements. This claim is therefore inaccurate.

The report includes Turquoise in its core analysis alongside primary exchanges, instead of grouping it with the other pan-European MTFs.

The report claims that incumbent exchanges simply 'contribute' to price formation, and MTFs play a ‘similarly proportional role in price formation to the incumbent exchanges.’ Yet it also refers to MTFs and systematic internalisers as ‘price takers’, and primary exchanges as ‘price makers’, which would rightly indicate that the roles are in fact different.

In reality it is the primary exchanges that provide the price formation used throughout the rest of the value chain, including by other trading venues that reference the price formation from the primary exchanges, without incurring any of the substantial costs of running a price forming venue. Other trading venues pay market data licences to reference the product of the price formation, but the costs of these pale in comparison to the costs incurred to run a primary venue.

We wish to correct the assertion throughout the report that the cost of producing market data is directly correlated to transaction volumes. Transaction volumes correlate closely to the amount of post-trade data published. However, as set out in the LSE RCB disclosure documents (which are used as a source for the report), the market data ratio between pre-trade and post-trade data is 98.36% to 1.64% respectively. By mapping trends in transaction volumes relating to post-trade data, the report provides fragmented analysis, only tracking trends of 1.64% of the market data published but presenting it as representative of 100%.

Finally, the report questions our pricing model [Note 5], which again is publicly available information, and we refute the inference that we might not apply the same pricing to intragroup companies as we do to third party customers. London Stock Exchange and Turquoise apply the same fees to all customers, whether they are intragroup or external.

In line with prevailing regulations, exchanges are required to publish their market data policies and fee schedules on their websites, and to disclose their costs to regulators upon request. Relevant information about LSEG’s equities trading venues is therefore either available to the public, or to those regulators with oversight of the venues.

The report claims that there is a “lack of transparency in exchanges’ public accounts with respect to the revenues and costs associated with equity markets and equity market data”, Despite this, the report relies on such public disclosures to present firm conclusions about such costs and revenues, which we find contradictory.

The same available data sets were used in a similar report from Oxera in 2024 (which came to very different conclusions), however Oxera also utilised confidential information provided by the FESE member organisations [Note 6].

The report claims in figure 43 that there has been a “notable reduction in costs for the exchange business” without having any visibility on this. The assertion is wholly incorrect (actual costs have increased over this period) and calculated by misallocating costs attributed to other group businesses within group annual reports. For instance, figure 5 omits LSEG’s acquisition of Refinitiv as the main driver for the change of LSEG revenue and the proportion of Equity revenues at a group level.

Operating resilient venues results in significant costs, which the report fails to reflect. Operation of primary trading venues also incurs significant costs which are not required to operate MTFs or systematic internalisers. As LSE plc operates a primary venue, and Turquoise operates MTFs, we can attest to the magnitude of costs being very different between the two, which is ultimately reflected in MTFs charging less for data generally.

The report states that there are no “specific costs” for creating market data, without providing supporting evidence. However, market data is a defined Important Business Service under PS21/3 requirements, meaning providers are required to maintain levels of operational resilience for all users, not just the low latency users referred to in the report. Additionally, our core market data feed provides subscribers with not only the incremental order book feed that is used by our lowest latency trading customers, but also aggregated “market by price”, top of book and last trade only feeds. Our investment in building and operating high-quality feeds with choice of data formats benefits all our customers, not just customers that are trading participants.

The report makes numerous unsubstantiated assertions about investment in equity trading platforms, including a false statement that LSEG has not changed data centre throughout the period. LSEG undertook a high-profile and well-publicised migration to a new data centre in February 2023 [Note 7], which reduced the LSE order book equities average round-trip latency by 50%, providing better outcomes for investors.

The MSP report fails to accurately measure market volumes and market data costs. Given these inaccuracies and misleading conclusions, amongst many others, we question the merit of the report in its current form. We wish to correct the gross inaccuracies and flawed methodologies used in support of its conclusions.

As demonstrated in this blog in relation to LSEG:

• The narrative that market data prices have increased to compensate for lower volumes of trading is factually inaccurate.
• The assertions that LSE does not publish several types of information, including RCB are inaccurate.
• Many of the assertions published are based on flawed understandings and/or representations about information we have published.

Readers should be mindful of making conclusions based on reporting that is incorrect and misleading.

LSEG continues to invest in its market data offerings and engage with customers to bring increased transparency to investors in UK markets, including waiving LSE market data user fees for private investors from January 2025.

Further to this, we agree there is a need for greater transparency on the total view of volume across different market mechanisms to demonstrate the increased fragmentation of volumes (as referred to in page 90 of the report), which is why, in the UK, LSE launched a Total Market Volume report in 2024 without any data charges (UK Trading Volumes).