New working group to create open framework for managing rising market data costs

It’s not just about the fees; it’s about the cost.

Each year—or each contract renewal—market data users and procurement professionals can count on the dollar figures printed on their invoices to be bigger than they were last time. The difference is easily calculated, the upward trend easily parsed and graphed.

At the inverse, savings are equally as straightforward to seek, find, and—if the fates will have it—attain.

But Mike Carrodus is less interested in these two things on their own or even together. Instead, through his market data consultancy Substantive Research, he is focused on something much more challenging to put a number on: cost avoidance.

This is a seriously driven, focused, knowledgeable, and hardworking set of people in banks and asset managers fighting with one arm tied behind their backs because they’re facing off against a provider base that is exceptionally powerful

Mike Carrodus, Substantive Research

A metric that is tricky to define and trickier to track and analyze, cost avoidance compares the amount a firm is paying today for data with what they might have paid for it but didn’t—in other words, contextualizing negotiation efforts by said firm and its peers and determining what it all adds up to.

To that end, Substantive Research is assembling a working group of buy-side and sell-side data-consuming firms, with the goal of creating an open-source framework that quantitatively measures the monetary impact of an organization’s market data control and vendor management processes, including benchmarking, relationship management, demand management and re-evaluation, inventory management, and more.

When Substantive Research began working with market data buyers and vendor managers in 2015, Carrodus was struck by the fact that they all seemed to report into different business lines within their organizations: sometimes finance, sometimes technology, and other times operations. But interestingly—and frustratingly—these designations weren’t static. These roles, he observed, are often passed around an organization and can spend a number of years under each domicile before eventually ending up back where they started. Rinse and repeat.

He wanted to know why this group, a massive cost center and a critical function, was essentially homeless.

“This is a seriously driven, focused, knowledgeable, and hardworking set of people in banks and asset managers fighting with one arm tied behind their backs because they’re facing off against a provider base that is exceptionally powerful,” Carrodus says.

He’s referring namely to providers of indexes, benchmarks, credit ratings, ESG data, and trading terminals—industry subsets that Substantive Research has been focusing on for years. In January, the consultancy released the results of its latest study, which found that some market data users are paying hundreds of percentage points more than their peers for the same data and services, for the same use cases, from the same providers.

James Davenport, global head of data management at Columbia Threadneedle Investments, told WatersTechnology last year that the pains feel particularly sharp in the index data market, where there is an oligopoly of three: S&P Global, FTSE Russell (owned by the London Stock Exchange Group), and MSCI. Because these providers are under no regulatory obligation to publish price lists, they can quietly raise prices, offer made-up discounts, and take discounts away each year.

“There’s no reference point, so it basically allows those vendors to go out and create fictional price points to increase their revenue, which is beyond really the price of production and dissemination,” Davenport said in January.

Substantive Research’s previous effort to benchmark data vendors took 2.5 years to complete and measured the inconsistencies in the pricing models and discounts offered by major data providers to 40 buy-side firms, representing $17 trillion of assets under management, and 20 sell-side firms, representing $20 trillion of assets. 

It built upon a smaller study from 2023 by Substantive Research, which found price disparities between asset managers buying comparable products and services from the same providers can reach as high as 2,632% in the ratings, ESG, and index data markets.

There’s the list price that that vendor says. There’s what’s getting proposed. And there’s what people are agreeing upon

Czarina Reinante, Substantive Research

A source who recently worked for a European asset manager with more than $300 billion of assets says the lack of transparency is incredible, especially compared to other industries.

“If you want to buy most products, you can go online and find a list price. I mean, nobody would book a hotel room if they didn’t know what the cost was,” they say. “What if they asked, ‘Well, how many towels did you use? How long did you watch the television? How long were you in the room?’ That’s how market data is charged. And it’s ridiculous.”

Fair or laissez faire?

Regulators don’t seem very interested in intervening. In February, with muted fanfare, the UK Financial Conduct Authority (FCA) declined to directly regulate the prices of wholesale data, ending a year-long investigation aimed at determining whether wholesale data markets are hindered by anti-competitive behavior. 

The FCA also declined to make a market investigation reference to the Competition and Markets Authority, which considers whether features of a given market adversely affect competition and what should be done about them, if anything.

The decisions came despite the FCA’s findings that across all three segments of the study’s scope—providers of market data, credit ratings, and benchmarks—there was a market concentration of no more than three key providers, that those key providers are highly profitable, data from key providers is essential to users, and that key providers face limited competition from challenger firms.

Prior to the conclusion of the Wholesale Market Data Study, the FCA had stated its goals were to measure the amount of harm—i.e., cost—that would be passed on to end investors, as well as determine if competition was functioning effectively. On the first point, the FCA found that there may be a “small proportionate impact on the prices charged to end investors” but this figure “is not easily quantifiable.”

On the second point, the FCA wrote that “users may be paying higher prices for the data they buy than if competition was working more effectively.”

This is where Substantive Research, its previous work, and its upcoming framework, comes into play.

Nobody would book a hotel room if they didn’t know what the cost was. What if they asked, ‘Well, how many towels did you use? How long did you watch the television? How long were you in the room?’ That’s how market data is charged. And it’s ridiculous

Source who recently worked for a European asset manager

“We have a history of being able to take this slightly amorphous stuff and normalize it and quantify it. We want to get the ideas right and make it valuable from a theoretical perspective, so if you wanted to have an internal meeting at your bank and you had this thing in front of everybody, then it would be easy to get the conversation going, and you’d get to some tangible stuff,” Carrodus says. “But how much better would it be if you could then start putting things into Excel and then be able to make some decisions based off quantitative factors?”

He asks this like it’s easy, but he knows it’s not. And he knows it won’t be possible to do everywhere.

For the research company to achieve its quantitative goals, Czarina Reinante, head of market data and ESG spend analytics at Substantive Research, says that much effort must be spent on the basic minutiae.

What can be considered cost avoidance, and what is just good negotiation? What constitutes a list price? Is a list price what the vendor says it is, or is the list price what’s proposed by the vendor to the most clients? Can that be considered a list price if everyone gets some sort of discount anyway?

“So there are three different things,” Reinante says. “There’s the list price that that vendor says. There’s what’s getting proposed. And there’s what people are agreeing upon.”

Second to definitions is attribution—in other words, the attribution of avoidance to certain factors. To put that into investment terminology, Carrodus asks: What’s the alpha and the beta of cost avoidance?

If the beta is that everyone gets 20% off in the first 10 minutes of conversation with a particular provider, then in the cost avoidance framework, no one wins any medals for that first 20% discount. It’s the next 20% where a vendor management lead or market data procurer can show they’re doing something, and the framework will aim to provide insight into the practices that achieved, or failed to achieve, those additional savings.

The source who recently worked for a European asset manager recalls a time, six or seven years ago, when a large data provider made changes to its data license, prompting complaints from users. In response, the vendor had said it hadn’t raised prices in about 10 years, which few could disprove due to the lack of price transparency.

The source says that they hear a lot of “bitching and moaning” from other market data users, but his response is always the same: What are you going to do about it?

They believe that a particularly wasteful area of the industry is the amount of duplication across data providers, and that this is where the aggrieved should expend their effort.

“Why do we need three credit rating agencies? Why do we need multiple identifiers?” they say.

Earlier this month, nine US regulators recommended exclusively using the FIGI identifier in financial reporting to government agencies as part of the proposed rules of the Financial Data Transparency Act. Notably, the regulators—including the Securities and Exchange Commission, the Commodity Futures Trading Commission, and the Federal Reserve—chose not to recommend the premier identifier for North American securities, the Cusip, which has a record 60 years long, nor its international counterpart, the ISIN.

Cusip, for its part, is the subject of ire among industry participants, who claim that Cusip Global Services’ licensing policies are too aggressive and, perhaps even illegal, as they are also the subject of an ongoing antitrust suit in federal court.

While the source supports Substantive Research’s aims to fight opacity and give leverage to powerless consumers, they believe there are far too many systemic factors currently at play to bring significant change to the market.

Overly complicated contracts, the undertaking of migration, the “nonsense” of data compliance, and the implications of artificial intelligence come together to create a “spaghetti” of roadblocks that prevent a serious overhaul of data licensing policies and practices—which is what is sorely needed here, they say.

The source, a self-described pessimist by nature, is doubtful the leverage and impact sought by Substantive and its framework-makers can be realized.

“I think it’s very good what [Carrodus] is doing, but sometimes it leads people up the garden path, believing they can change things, and only if they make different decisions will they change things. But if it’s the same input in and the same output out—nothing will change,” they say.

Substantive Research is calling for participants from the buy side, the sell side, other consultancies, and vendors to join its working group. The first draft of the framework will be published in January 2025 after a series of meetings and webinars that will run between September and December 2024.