[ICE, NDAQ, CME, CBOE] Exchange Operators: Fragmentation and New Points of Differentiation

Intercontinental Exchange (ICE)

Intercontinental’s multifaceted data and exchange platform began as an electricity marketplace, founded in 2000 by current CEO Jeffrey Sprecher with the backing of two investment banks and seeded by volume from energy majors Shell, Total, and British Petroleum (the collapse of Enron, the dominant electricity broker at the time, fortuitously diverted a flood of trading volume to ICE as well). Through its acquisition of International Petroleum Exchange the following year, ICE picked up the Brent contract (against which 2/3 of crude futures volumes are benchmarked) and expanded into oil futures trading in direct competition with the New York Mercantile Exchange (which would be acquired by CME in 2008).  

Since going public in November 2005, the company has acquired its way into various non-energy domains, including soft commodities (NY Board of Trade, $1.1bn; 2007)(1); credit default swaps (Creditex, $625mn; 2008)(2); exchange-traded emissions (Climate Exchange, $606mn; 2010); and, most notably, equities. Through its industry-quaking acquisition of NYSE Euronext in November 2013(3), an $11bn cross-border deal ($2.7bn in cash and 42.4mn shares representing ~37% of the company; 13x/8x pre/post-synergy EBIT), ICE more than doubled its operating profits, shifted its revenue mix towards data and listing fees, and brought cash equities trading into its stable of exchange operations. Here is a before (2012) and after (2014) decomposition of revenue:

The

ICE/NYSE merger was part of a frenzy of deal activity in the exchange

space during the mid/late-2000s(4), most of them motivated by the same

industrial logic:

1/ Eliminating redundant technology, trading platforms, and corporate overhead. The high fixed costs of exchange operations make them highly scalable. In a typical merger, 25%-40% of the target’s cost structure might get hacked within a few years. The cost cutting opportunity is especially rich for a technology-focused operator like ICE, who did not mature in the opulent, stained oak halls corseting the pens of the flesh and blood exchanges. ICE has, without compunction and to the great consternation of floor traders, replaced open-outcry with electronic market making systems.

2/ Bolstering

scale, especially in light of intensifying fragmentation. Exchanges

benefit from network effects…more buyers and sellers –> deeper liquidity

–> lower transaction costs and better price discoverability —> more

buyers and sellers. 

Just under half of ICE’s total revenue, net of transaction expenses, comes from exchange transactions and clearing, broken down across asset types like this: 

While

only 17% of net transaction revenue, “cash equities and equity options” punches

above its weight, driving listing revenue – NYSE remains a marquee listing

venue for companies going public and NYSE Arca is the flagship listing venue

for ETF listings(5) – and data fees. 

Equity

trading venues have multiplied over the last 20 years. In the late ’1990s/early 2000s, advances in technology

coupled with accommodative regulation gave rise to a host of non-exchange Alternative

Trading Systems (ATS) and dark pools who didn’t provide the listing functions

of a regular exchange and so weren’t required to publicly disclose their

trading operations. They competed fiercely for order flow, offering fees

and rebates to liquidity providers as part of a “maker-taker” payment model that

was later adopted by exchanges and became ubiquitous by the mid-2000s. Non-exchange share of trading in NYSE listed

stocks has gone from 17% in 1990 to 31% in 2015, per the OECD(6). Major US exchange operators have been losing

share since the mid-2000s, as this SEC memo elaborates: 

“…lit venues [as opposed to “dark pools”] have been losing market share

for a variety of reasons to non-exchange dark venues that do not display quotes

or orders. From February 2005 to February 2014, the share of volume

executed by displayed venues declined from 70.6% to 61.4% for NASDAQ stocks and

from 87.0% to 65.4% for NYSE stocks. During the same period, the collective

share of dark venue trading in NASDAQ stocks increased from 29% to 39%, and the

collective share of dark venue trading in NYSE stocks increased from 13% to 35%”.

Source: Bloomberg

Source: Wall Street Journal

In response to the proliferating number of trading venues, stock exchanges consolidated. There are 13 approved stock exchanges in the US today. All but one, IEX, are owned by one of three publicly traded companies: Intercontinental, Nasdaq, and CBOE. But there are many other trading venues outside these exchanges, testament to lower entry barriers brought on by electronic trading. Of course, an unintended consequences of pro-competition regulation, which was supposed to reduce exchange fees, is that exchange operators began charging connectivity and market data fees – fees that are the subject of so much controversy today – to traders seeking price discovery in an increasingly fragmented landscape. But the point remains that equity exchanges, on their own, are a less good business than they were a decade ago.

Relative to equities, futures trading in asset classes like energy, commodities, and rates have proven more immune from competition, grounded as it is in extensively referenced benchmarks – including well-known ones like Brent, US Dollar Index (USDX), ICE swap rate, and LIBOR – to which ICE owns the underlying IP and against which many $trillions of assets are managed and hedged(7). 

Energy trading is dominated by CME and ICE. A few competitors have tried to challenge the duopoly. In early 2015, NASDAQ OMX predicted that by dramatically undercutting transaction prices on a small sliver of CME’s and ICE’s energy complex, its futures and options exchange (“NFX”) would grab ~10% of market within a few years. But after coming out the gate to great fanfare in July 2015, interest in NFX seems to have petered out. An oil price benchmark launched by Shanghai International Energy Exchange (INE) has seen more promising uptake, grabbing an impressive ~6% spot market share within its first year of launch. Still, unlike Brent and WTI, which are traded by market participants with real commercial interests, INE’s futures are traded for speculative purposes by Chinese retail traders and suffer from jerky trading volumes. It doesn’t appear that this new benchmark functions as an efficient hedging tool for global oil producers and shippers, at least not yet.

Banks, brokers, and energy merchants regularly consult ICE’s commodity, rates, and FX benchmarks and trade the futures underlying them, in turn further reinforcing their benchmark status and attracting still more transaction activity on ICE’s exchanges. A benchmark can be monetized multiple times. Moody’s makes money not only by charging issuers for ratings, but also by charging investors for the research backing those ratings(8). Likewise, Intercontinental makes money not just from transactions on its exchanges, but also on the data spilling off them.  

At

first, ICE’s Data Services segment (~45% of revenue and EBITA) mostly

disseminated the data generated on ICE’s exchanges to third party data vendors

and directly to end users who, rather than wait for delayed aggregated data

from wholesalers, paid for direct data feeds that powered their custom

applications and risk models. But

through acquisition, ICE has expanded into Pricing and Analytics [bond and

futures valuation, risk analytics, regulatory compliance solutions, index

design – i.e. the benchmarks I discussed earlier], Desktops [ICE’s

version of Bloomberg, basically, where “10s of thousands” of users trade, chat,

and manage risk with analytic tools], and Connectivity [secure and

low latency connectivity to exchanges and clearinghouses].

2015 was a pivotal year in this journey. Around that time, European regulators were crafting new regulations under MiFID II, aimed at expanding the reach and tightening the constraints imposed by MiFID I with nebulous goals related to pricing transparency and consumer protection. This ambitious, unwieldy piece of legislation went into effect early last year and has far reaching implications across the financial landscape, but there are two that seem relevant here.

First, the

regulation imposes onerous record keeping and reporting requirements on their

trades, including price and volume information, that demonstrate best

execution. Second, it expands the number

of regulated platforms authorized to host commodity derivatives trading,

potentially fragmenting the markets and obscuring price discovery(9). ICE

seized on the prospect of heightened demand for data and analytics catalyzed by

this regulation and in December 2015, bought two companies that more than

doubled its data services revenue, transforming a data business that was

captive to its exchange operations to one that stands firmly on its own,

carving out unique avenues of value creation.

ICE’s major foray into this space was its $5.2bn (5.5x revenue / 10x post-synergy EBITDA) purchase of Interactive Data Corp (IDC), which it “won” by outbidding rival suitors Nasdaq and Markit. Whether by through broad distribution or better execution, ICE has double IDC’s standalone ~3ish% growth rate. At the time of acquisition, around 70% of IDC’s revenue came from valuing thinly traded fixed-income securities and selling this data to portfolio managers and risk managers at banks and asset managers (with 98% annual retention)(10). Fixed income securities are far more complicated and varied than equities. There is only one AIG common stock, but a thicket of AIG fixed income SKUs, some rather arcane, with different maturities, covenants, coupons, and liquidity profiles. 

During the mid/late 2000s, I’d get bond pricing data from broker “runs” delivered to my Bloomberg inbox, and if I wanted to know the bid/ask on a particular security, I’d have to ping various dealers. TRACE quotes were too sparse to be meaningful, especially for off-the-run securities. Through a manual, ad hoc, and somewhat dubious process, you could triangulate historical bond yields from third party data sources and dealer runs or impute them by averaging different points along the curve. It sucked and even back then there was bubbling acknowledgment that reliable, systematic pricing of debt securities was long overdue. Little progress has been made in the intervening dozen years; only 20% of the bond market trades electronically, mostly through RFQ (“request for quotes”, where an electronic expression of interest for a certain security is sent to market participants). IDC, with continuous pricing on 2.7mn fixed income securities, sourced either directly from trading data or indirectly from market data-fed models, is a step in that direction(1). 

In its ongoing effort to automate bond trading – to go from emailing dealers for bond quotes to having a central order book with resting bids and offers that wealth managers, family offices, and others dealing in small volumes can click to buy/sell, as they can with stocks – ICE acquired several other fixed income properties post-IDC, including ones related to fixed income securities and derivatives pricing (S&P’s Securities Evaluations and Credit Market Analysis; Oct. 3, 2016, $431mn cash); trading technology (BondPoint; Jan 2018, $400mn); and indexing (Bank of America Merrill Lynch’s Global Research Division’s Index business, with $1tn of AUM benchmarked against its indices). [Capitalizing on the growing popularity of passive investing in fixed income, Bloomberg acquired Barclays’ fixed income indices in 2016. Both ICE and Bloomberg now compete with Markit, who owns the IP for widely referenced indices like iBoxx and iTraxx and CDX] 

ICE is gripping various parts of the fixed income trading lifecycle, starting with data (IDC) and more recently backing into execution (BondPoint, TMC). This is a notable change to how value was realized in the past. Five or so years ago, I might have summarized the process of value creation at ICE as follows: market participants trade a panoply of products linked to soft commodities, energy, equities, fixed income across ICE’s 11 exchanges, cleared on one of ICE’s 6 globally distributed clearinghouses(11). Trading generates proprietary data that is then sold to both 3rd parties and to the buyers and sellers whose activity spawns the data in the first place. In the rates and commodities, that data evolves into reference points off which market participants benchmark a wide range of exposures relevant to their real-world commercial activities. And as those benchmarks congeal, more trading is drawn to derivatives tied to them. 

This

description is less applicable to ICE’s strategy today. Only ~1/3 of

ICE’s data services revenue today comes explicitly from proprietary exchange

data. ICE is unbundling of exchange,

clearing(12), analytics, and data services, treating each of these services as either

distinct on-ramps to other ICE services or as standaloneofferings with broad horizontal

reach, rather than as a vertically integrated set of services. The center

of gravity for all this is less so exchanges and more so a base of data and

connectivity, a rather dramatic re-orientation away from the vertically

integrated exchange/clearinghouse/data model. 

Now, one might say that data was really at the heart of the exchange

business too, as in:

“And at the end of the day, what an exchange and clearing house is, it’s really an information asset. It’s an information asset and a market structure asset that enables customers in as easy a fashion as possible to gather the information that they need to determine what’s the next trading strategy I want to have?….We haven’t changed our strategy because what are those businesses at the heart? At the heart of those businesses, they’re informational assets”. (Deutsche Bank Global Financial Services Conference; Benjamin Jackson, ICE President; May 30, 2018)

But the difference is that in the past, this “information asset” was spillover from trading activity on the exchanges, whereas today, differentiation starts with content. ICE is acquiring data assets directly, leveraging distribution, and, at least in the case of fixed income, spinning up an electronic trading network. So now you’ve got this unbundled collection of analytics, valuation, and pricing data resting atop a secure private network called the Secure Financial Transaction Infrastructure (SFTI, aka ICE Global Network) that asset managers can use to design indices, price bonds, trade bonds, attribute performance, and otherwise streamline their workflow. 

[SFTI was built by NYSE to help comply with Reg NMS, which required every exchange to route to all other exchanges in order to find the best price for customers. When ICE acquired NYSE, it enhanced this network with cyber security, expanded the infrastructure into Asia, Eastern Europe, and Canada (through acquisition), fenced it off from the public web, and transformed it into a secure, low latency private cloud (it does not communicate with the broader web) that now connects 1,200+ firms to 120+ exchanges and 600+ data and news providers.] 

ICE

splits Data Services into three parts, with hsd orgnic growth in Pricing & Analytics

(fixed income data, 49% of segment revenue) and msd growth in Exchange Data

(32%), diluted by lsd growth in Desktops & Connectivity (19%). The latter subsegment can be further

disaggregated into strength in connectivity services offset by weakness in

Desktops, with recent divestitures in the latter further testify to something

amiss here. From what I can tell, Desktops

is a customer facing service that mimics Bloomberg, Thomson Reuters, and FactSet

as a general workstation terminal solution offering a wide range of financial

content, tools, and analytics to a wide range of market participants. I don’t know how ICE competes effectively in

this domain. Bloomberg is Bloomberg. Thomson Reuters is a basket case that, in the

course of acquiring 300+ companies over the last decade, forgot how to run a

business, but its prospects may brighten now that its Financial & Risk

segment (the part of Thomson Reuters that competes with Bloomberg, now rebranded

Refinitiv) is majority owned by Blackstone. 

Both Bloomberg and Refinitiv draw and retain users with widely used

messaging platforms, the latter’s now integrating with Symphony, an open source

chat function launched by certain banks and asset managers that has amassed

over 235k users (vs. Refinitiv Eikon’s 300k and Bloomberg’s 325k terminal

subscribers). Factset has built on its

historical strength in equities to now cater to buy-side clients across

multiple asset classes, and like many others, is gathering alternative datasets

(weather, sentiment, credit cards, etc) to flesh out its portfolio.

In

other words, these workstation providers are trying to be everything to their

institutional customers. Competing with them

on their own terms is a mistake. Where ICE

differentiates is in its niche credit focus, the bedrock of which is its

comprehensive real time credit and CDS pricing from IDC, which customers can

consume through direct feeds and customize to their liking. For the most part, IDC competes at a layer

below Bloomberg and Factset, who in fact are customers that consume IDC

data. Factset also has analytics for

fixed income and has moved into trade automation through its recent acquisition

of Portware, but it does so as part of a multi-asset class lifecycle solution

and doesn’t have IDC’s depth of real time pricing and reference data. Bloomberg facilitates electronic credit

trading, but mostly relies on RFQ’s, whereas ICE is trying to jump start small,

dollar click-based trading. IHS Markit

is a formidable competitor to IDC but doesn’t offer trade execution.     

But it starts with data. Data is valuable for its own sake but supplementing stale TRACE quotes with continuous algorithmic pricing also improves transparency, giving traders a clearer sense of where they might buy/sell certain credits (IDC’s models take the liquidity profile of securities into account), and might facilitate trading, reinforcing ICE’s execution platforms (BondPoint and Creditex) and clearing capabilities. Moreover, besides licensing the thousands of fixed income indices acquired via Bank of America Merrill, ICE can provide asset managers like BlackRock the pricing data, reference data (security identifier, terms and conditions), and analytics to design and build their own branded indices. So, an asset manager like BlackRock or State Street that wants to create an ETF off a proprietary index (“self-index”) – either to reduce their costs of differentiate their product – can rely on Intercontinental to supply the core ingredients (reference and pricing data) and calculate the index NAV. For instance, in partnership with BlackRock, ICE is leveraging the trading technology acquired over the last few years – BondPoint, TMC, ICE Credit Trade – to launch an “ETF hub”, a platform for traders to buy and sell the fixed income securities underlying the creation and redemption of ETFs. 

In both

cases, trade facilitation and index construction, ICE is leveraging data to

monetize other adjacencies in credit, rather than simply selling exchange

exhaust.       

Which

is just as well since the market maker community has expressed increasing concern

over the costs of proprietary data feeds, which they claim, according to this Business

Insider article, have escalated by ~20%/year for the last 5 years. 

The exchanges maintain that this high-speed data is optional;

broker-dealers retort that it is table stakes in providing

competitive trading solutions for clients.

Per the

above linked article, Brad Katsuyama of Flash Boys fame and CEO of IEX Group

asserts: 

“Exchanges

don’t create any unique content — market data is generated by their members and

other market participants including real investors — so it’s very hard to

believe that exchanges can perpetually charge their members more every year to

look at the members’ own data”. 

This strikes me as a strange argument since aggregating data from disparate market participants, like sharing poll results with surveyed participants, is its own kind of value. But I get it. Three companies, all of whom generate greater than 50% EBITDA margins, own the 13 exchanges that host more than 60% of US equity trading. Because brokers are required by regulatory mandate to send customer orders to the exchange displaying the best price, they have no choice but to pay connection and data fees to all those exchanges. The power imbalance is subject to abuse and the SEC, which regulates the exchanges, has more or less auto-approved fee hike proposals. But, perhaps under court pressure, last spring the SEC started suspending or rejecting fee increases, and in October 2018, per the Wall Street Journal:

“…the

Securities and Exchange Commission unanimously ruled against the New York Stock

Exchange and Nasdaq Inc. in a 12-year legal dispute over market-data fees. The five-member commission shot down a pair

of requests to raise fees for certain NYSE and Nasdaq data, saying the

exchanges hadn’t justified the price increases. The decision is the first time the SEC

has rejected fee increases for the exchanges’ most lucrative class of

stock-market data feeds….

The SEC also

put into limbo over 400 other market-data fee increases that were challenged by

Sifma. The regulator didn’t reject those price increases but told the exchanges

to review brokers’ complaints that they undermine competition, giving the

exchanges a year to complete the effort. The move puts pressure on the stock

exchanges to restrain fees or disclose more about why they are necessary, which

exchanges have historically resisted”.

I don’t want to sound dismissive, but whatever pockets of ICE’s data segment that have been subjected to extortionary fees – like real-time stock market data (which accounts for 2% of revenue, though that doesn’t include connection fees) where it pricing has allegedly been hiked by 20%/year – represents a subset of exchange data revenue, which has been only been growing by 5%-6% per year (~2% of that from pricing), and moreover, exchange data taken as a whole only represents 15% of ICE’s total revenue. Also, as far as I know, while the SEC’s recent action may stymie ICE’s ability to raise prices, they don’t mandate that the exchanges reverse previous price hikes. 

And then there’s the years-long LIBOR scandal. As a reminder, it is the banks who submitted dubious rates who are in the hot seat and exposed to injurious fines, not ICE, who merely administered the benchmark. In fact, ICE assumed LIBOR benchmark oversight from the British Bankers’ Association in 2014, years after the LIBOR scandal was brought to broad public awareness. That said, ICE could be impacted to the extent that LIBOR’s credibility concerns give rise to an alternative benchmark administered by another entity, and in that vein in July 2017, the Financial Conduct Authority in Britain announced that it wanted to phase out LIBOR by 2021. But with something like $300tn of global assets pegged to LIBOR, much of that long duration, good luck with that. And just to keep this in perspective, ICE recognizes around $40mn-$50mn in revenue for administering the LIBOR benchmark and some unspecified subset of $354mn in interest rate and financial futures transaction revenue can be tied back to the benchmark, so we’re talking at most a high-single digit exposure. In the meantime, ICE has seen some success standing up a few alternative rate products (SONIA and SOFR) in the US and Europe, though this is a jump ball as CME is also competing vigorously for SOFR volume.

Nasdaq (NDAQ)

Like ICE, which spent $20bn on 20 acquisitions over its lifetime, NASDAQ has pursued an aggressive acquisition agenda, though it has taken ICE’s strategic posture towards information services further afield, bundling a whole set of buzzy technologies as part of a managed technology service offering. 

In 2008, Nasdaq went a little nuts. Following its transformative $4.3bn merger

with OMX, a Nordic stock and derivatives exchange, the company acquired or took

partial ownership of 6 exchanges, clearinghouses, and trading networks, in an ambitious

drive towards becoming a vertically integrated “premier global exchange

company”, profiting from various exchange-related services like market data,

listings, clearing, issuer support. These

acquisitions were scared into existence by competition from dark pools and

alternative trading systems, as well as by aggressive consolidation activity

from rival, NYSE, which acquired Euronext in 2007 and Archipelago in 2006. 

By the looks of it, putting these

exchanges under one roof and on a common technology platform has done little to

enhance the company’s competitive position.

US equity options volume has grown at a low-single digit rate since 2010 and Nasdaq’s share of those volumes have been roughly flat:

Nasdaq has lost share in US-listed equity securities, whose average daily trading volumes have declined:

It has also lost share in Nordic and Baltic securities (albeit, off a high base), though strong volume growth at the industry level has fueled 10% annual growth in equity trades executed on Nasdaq’s exchanges.

Perhaps disquieted by flagging exchange share, Nasdaq went on another buying spree, diverting its attention away from exchanges. Since 2010, Nasdaq has generated $5bn in free cash flow; nearly 80% of that has been plowed into a wide range of disparate assets, including Thomson Reuter’s IR, PR, multimedia solutions business (2013), an options exchange operator (ISE, 2016), an index provider (Mergent, 2012), a smart beta index provider (DWA, 2015), a dealer-to-dealer Treasury trading platform (eSpeed, 2013), and other analytics, risk management, and content providers. Today, Nasdaq’s revenue mix looks like this:

[“Market services” includes trading and

clearing revenue]

[55% of Information Services revenue comes

from market data fees]

See that sliver of pie labeled “Market Technology”? After growing by a modest high-single digit annual rate since Nasdaq picked it up in its merger with OMX, MT amounts to only 11% of revenue, but the way management talks about this segment, you’d think it was the centerpiece of Nasdaq’s whole operation. MT is where management smuggles all sorts of frontier initiatives like blockchain and quantum computing, but, at its core, what this segment does (and what it’s been doing since 2008), is implement technology solutions for capital market participants – other exchanges, clearinghouses, central securities depositories, buy/sell-side firms – leveraging the decades of experience it has amassed building and operating its own exchange operations. For instance, Nasdaq provides the trading engine for the Indonesian Stock Exchange and has helped at least one broker build out its dark pool. Brokers, exchanges, and buy-side firms(13) use Nasdaq’s trade surveillance solution to comply with market regulations, detect anomalous trading patterns, and monitor market risk. And now, in a somewhat speculative but not insensible extension, the company is taking its technology to two-sided markets outside the financial sphere, like ad tech, loyalty points exchanges, and horse racing(14).

Like many other technology companies,

3 or 4 years ago Nasdaq decomposed its applications into microservices

(autonomous, commonly used functions that can be mixed and matched to quickly

launch new solutions), accessible both to its own proprietary applications and to

those of its clients through a set of well-defined APIs. This effort was part of a grander mission of

getting clients to host their technology on what the company has dubbed the

“Nasdaq Financial Framework”. NFF is

similar to ICE’s SFTI network or LSE’s Technology Services in that it supports

and distributes the various solutions that Nasdaq has built and acquired over

the years, but it also functions as a PaaS, hosting cloud-based managed services

for exchange operators, making the technology it uses in-house available to

others. Compared to having each exchange

individually build out its own matching engine, train its own machine learning

algorithms, and integrate all its applications, there are system-wide benefits

to centralizing those functions with a single party. 

But of course, this is also a highly competitive space, with different aspects of NFF addressed by public cloud vendors, exchange operators (notably, ICE and LSE), and independent software vendors. Given its decades of experience operating exchanges, NFF may have an advantage selling into market infrastructure operators, but this point of differentiation blurs as Nasdaq carries its platform outside of exchanges and clearinghouses and into the buy/sell-side firms and non-financial marketplaces (i.e. ad exchanges) who comprise a disproportionate amount of the TAM opportunity. Moreover, my sense is that a significant part of the value that an exchange client realizes with Nasdaq relative to a specialized technology vendor is Nasdaq’s unique expertise operating an exchange – the company has referred to itself as the outsourced R&D arm of exchange clients and often has its own personnel staffed at client exchanges – which carries with it a consultative element that does not yield the scale benefits of a self-serve platform. 

Nasdaq has been investing aggressively in NFF over the past few years, taking segment EBITDA margins down from 36% in 2016 (when NFF was first introduced) to 20% in 2018 and plowing 2/3s of MT EBITDA into capital investments, while offering assurances that costs will scale over time (starting next year) as subscription revenue expands relative to labor intensive advisory services. So far, this “capacity to suffer” hasn’t yet translated into a discontinuous kink in the revenue plot, one that can be unambiguously tied back to investments in this new platform initiative. MT revenue has grown by just 6%/year since 2016, slower than the high-single/low-double digit growth realized in the 2 years prior to NFF, and the contract value of orders signed in 2018 are no higher than they were two years ago. Maybe I’m being myopic, but I am having a hard time seeing MT grow into a significant business. This is partly an issue of competitive differentiation (diminishes as Nasdaq moves further afield from exchanges and clearinghouses), partly an issue of scale (does not command the network effects of the exchange business or the profitability of licensing indices and selling market data), and partly an issue of execution (management has a spotty track record executing new growth initiatives).

Nasdaq has various other things going on. Like many other enterprises, it has venture capital investments in fintech startups. It also got involved in blockchain 5-6 years ago and now has a few early permissioned blockchain pilots/proof-of-concepts tied to proxy voting and backend/recordkeeping functions like settlements and asset transfers. Distributed ledger technology has significant potential in these domains (see this prior post for more on blockchain as it relates to integrity of asset ownership and disintermediation potential), though, of course, intermediaries who would have no place in an apparatus designed from first principles today are, in actual fact, deeply embedded and protected by regulations and institutional inertia. And then one questions the wisdom of examining speculative frontier technology like quantum computing “very, very closely”, as management claims to be doing, considering that the company has yet to bear fruit in several other commercial areas. 

In pursuing diversification and chasing fast growing trends (smart beta) to compensate for a trundling exchange business, it seems to me that Nasdaq has not been thoughtful about how its acquired assets interlock as part of a broader strategic framework. Performance has been unsurprisingly mixed. Corporate Solutions – a set of solutions that helps companies manage their investor and public relations, coordinate communications among Board members, and produce webcasts – has struggled to generate any revenue growth outside of acquisitions and is now scaling back on the services it offers, recently divesting its PR and digital media assets. Revenue from market data and listing services have compounded at what looks like a ~low/mid-single digit rate after stripping out acquisitions. eSpeed, one of Nasdaq’s largest transactions, has been a disaster, donating 30-40 points market share to rival BrokerTec since damn near the day it was acquired. NLX, Nasdaq’s London-based interest rate derivatives exchange, shut down 4 years after launch(15). Last year, Nasdaq announced a new US interest rate futures product on the Nasdaq Futures Exchange (NFX). With CME’s dominant position in US rate futures further bolstered by complementary dominance in cash treasuries via its recent acquisition of NEX, I suspect Nasdaq will find it difficult to gain critical mass. Its attempt at energy derivatives has likewise fizzled out. Relative to those disappointments, The Nasdaq Private Market, a marketplace for private growth companies, has been a relative success, though this initiative, too, has performed below management’s ambitious expectations. And then, of course, whether Nasdaq wields a competitive advantage in acting as a purveyor of managed technology outside the arena of market infrastructure operators, is an unresolved point.

The uncharitable interpretation of all this is to say Nasdaq is too clever for its own good and sorely in need of focus, execution, and capital allocation discipline (management uses repurchases to offset share count dilution…ugh). On the other hand, one can hardly fault management for exploring other avenues of growth given the challenges of ongoing exchange fragmentation and share losses, not to mention the possible pressure on the data fees that comprise 15% of its revenue. 

In further defense of Nasdaq, the company’s missteps have, for the most part, been largely obscured by decent performance from its two largest profit contributors, Market Services (the exchange business) and Information Services (market data and index licensing). In the 7 years since 2011, the first year Nasdaq began breaking out segment level profits, Market Services’ EBITDA margins have expanded from 52% to 67%, with profits growing by 2x revenue (8% vs. 4%), testament to the scale economies inherent in exchange operations as well as the cost synergies realized in its purchase of ISE. Squinting my way past several meaningful acquisitions, index licensing and services revenue has grown organically (albeit, inconsistently), with $AUM linked to Nasdaq indices rising from $99bn in 2014 to $180bn in 2018, no big surprise given the growing popularity of indexing. Segment-level pre-tax profits have grown by $180mn against incremental assets of $1bn, with return on segment assets drifting higher, suggesting, though by no means proving(16), that management has more than earned its cost of capital on incremental investments.

CME Group (CME)

CME doesn’t label itself a tech company with the same alacrity that Nasdaq and ICE do. On the contrary, it touts the advantages of its vertically integrated exchange/clearing model. As a futures and options exchange, it also doesn’t enjoy the same bounty of market data fees that cash equities exchanges and by the same token, is not facing the same scrutiny over allegedly extortionary price hikes(17). Instead, ~85% of CME’s revenues are rooted to exchange transactions and clearing, where it commands dominant transaction share across a range of derivatives products. 

In US Treasury futures, whose daily volumes how exceed those of the cash market, the company effectively has a monopoly. But CME is extending its dominance to the cash market too through its recent $5.5bn acquisition of NEX Group (the $91mn of cash markets revenue in the above table represents ~2 months ownership of NEX transaction revenue; there’s another $12mn in market data fees and $31mn in post-trade services revenue grouped elsewhere). There are 23 Primary dealers who buy and sell Treasuries directly from the US Government (you can think of interdealer as a “wholesale” market through which dealers buy and sell Treasuries between themselves in the service of providing liquidity to institutional investors in the “retail” market). NEX, through its ownership of BrokerTec, owns the electronic platform that intermediates 80% of interdealer trades (on the retail side, Bloomberg and Tradeweb control 90% of electronic trading)(18). By concentrating cash and futures trading and clearing on a single venue, the combination creates efficiencies for traders and expands CME’s data. And then, of course, there are the usual synergies that come from cross-selling and migrating volumes onto a common trading platform.

Source:

Greenwich Associates via TradingPlaces

Prior to BrokerTec, there was another interdealer platform called eSpeed – launched in 1999 by broker Cantor Fizgerald, who at the time had 90% share of what was mostly a voice-based interdealer market – that opened up the interdealer market to non-primary dealers. Primary dealers, fearing disintermediation as buyside firms eventually sourced direct from eSpeed, launched a competing platform, BrokerTec, but anti-trust scrutiny forced them to divest their ownership in 2003. eSpeed and BrokerTec roughly shared the market for the next decade until the former was sold to Nasdaq, who intimated that it would open the platform up to the buyside firms and launch dark pools for Treasuries on the eSpeed platform, causing dealers to balk and move their trading over to BrokerTec. Here we are, 5 years after the sale, and eSpeed’s share has been reduced to just 11%. Perhaps recognizing how fragile BrokerTec’s dominance could be, management has adamantly maintained that it would do nothing to change the structure of the interdealer market. So far, it doesn’t appear that BrokerTec has lost any share(19).

In

energy – CME’s second largest market, comprising another 20% of revenue – the

company, through NYMEX, which it acquired in 2008, operates the primary

exchange for trading the futures of WTI, one of the two primary crude oil

pricing benchmarks. It also has 80%

share of futures trading in Henry Hub, the primary benchmark for US natural gas,

one that is seeing increased global adoption. 

Via its ownership of COMEX (obtained through its acquisition of NYMEX),

CME has 93% share of precious metals futures. 

It has 70% share of FX futures trading, which account for a small (~2%)

but growing share total FX volumes. 

In

listening to CME’s earnings calls alongside Nasdaq’s, it can seem as though

these companies are not even operating in the same industry. You will not hear CME’s management talk about

blockchain or quantum computing or other innovative technology plays. Nor are they scooping up data assets (since

CME’s 2007 merger with CBOT in 2007, the combined company has made just two

acquisitions – NYMEX and NEX Group – the first an exchange, the second a

trading platform). That’s not to say

that CME is starved of innovation, just that its innovation manifests in

product – electronification of Eurodollar options; Total Return Futures pegged

to the Dow Jones and Russell 1000; crude futures for WTI crude at Houston, and

that sort of thing – rather than in business model. The company creates and maintains liquid

markets across a range of derivatives and it does that job very well.

Across several foundational derivatives markets, it hosts the deepest pools of liquidity. Cross-side network effects, combined with mounting trade volumes on a largely fixed cost base, have given rise CME’s stratospheric margins. And yet, in isolating the period from 2009 to 2019, a 10-year stretch that doesn’t include any meaningful acquisitions, EBITDA margins have contracted (though, at 66%, they are still extraordinary), with growth in revenue (~6%/year) outpacing EBITDA (~5%). At 24x earnings, CME is appropriately valued for what it is and will likely be: a low growth, well-moated dividend payer with a solid balance sheet. [With few reinvestment opportunities, the company dividends more than all its free cash flow (the company has an explicit policy of dividending any excess cash above $700mn that builds on its balance sheet).] 

CBOE Global Markets (CBOE)

For most of its 46-year history, CBOE

was an exchange operator that derived all its revenue from options and futures. In the decade leading up to 2016, the decline

in equities options volumes was more than offset by double-digit annual growth

in index options and an explosion in futures volumes, the latter carrying

substantially higher fees per contract than equities options ($0.70, $1.78, and

$0.06 for index options, futures, and equity options, respectively). Moreover, pricing on multiply listed equity

options had declined by around 70% during this time, so that by 2016, index

options and futures represented nearly 90% of CBOE’s transaction revenue, up

from 50% in 2010. EBITDA margins

expanded from 53% to 64% over that period. 

Then CBOE paid $4bn (14x pre-synergy

EBITDA; 69% stock) to acquire BATS in Feb 2017, three years after BATS merged

with Direct Edge, to create one of the largest stock exchange operators in the

world, its ~20% US market share ~equivalent to those of NYSE Euronext and

Nasdaq, its 22% share in Europe unrivaled.  

CBOE’s revenue mix shifted like this by asset class:

…and like this by type:

The BATS acquisition ushered CBOE into the world of US and European equities exchanges and delivered a heap of the market data fees. It gave rise to a larger, more diversified, more levered company…

…and presented enormous cost synergies

that summed to 40% of BATS standalone operating costs(20). Moreover, BATS brought with it superior

technology platform on top of which all the combined entity’s markets –

including 6 cash equities markets and 4 options markets – are now hosted(21).

CBOE is different from the other

exchanges in that a significant part of its competitive advantage stems from a

scarce resource can be traced upstream, to the companies that own the IP for various

S&P 500, Russell 2000, Dow Jones, and MSCI indices against which CBOE has

struck exclusive long-term licensing agreements(22) to offer options and

futures trading. The value of these

exclusive contracts can be proxied by the revenue per contract that CBOE

realizes for index options exclusively traded on its exchanges vs. equity options

traded on multiple exchanges, $0.74 for the former (up from $0.55 in 2006) vs.

just $0.07 for the latter (down from $0.21 in 2006). Options and futures on VIX and SPX,

complementary products that institutional investors often pair, are the CBOE’s

most frequently traded products. Nearly

2/3 of CBOE’s net transaction fees comes from futures and index options, the

“overwhelming majority” of which is tied to proprietary products, namely

S&P Index options and VIX options and futures.

You may recall the hailstorm of

negative press that buffeted volatility products in early 2018: a few leveraged

exchange traded products (ETPs) that paid 2x the inverse daily return of the

VIX, blew up in spectacular fashion when volatility spiked. Casual retail investors, for whom these

products were dubiously suited,

found themselves unsuspecting victims. 

But obscured in the reporting on the carnage is the reality that most

volatility trading is done by institutional traders – 60% of VIX futures

volumes comes from prop traders – who use vol instruments to hedge other

exposures. They aren’t naked shorting

the VIX with leverage, which is why there weren’t widespread collapses among

funds and trading shops who were short vol. 

Even while the AUM for ETPs linked to the VIX have been cut in half from

their $5bn peak in early 2018, the volatility complex, in total, is doing just

fine. More than fine. VXX, the most popular ETP tied to the VIX,

has seen record trading volumes over the last year(23), as have VIX futures and

options (month-to-month fluctuations notwithstanding). 

The S&P 500, Russell 2000, and the rest are globally recognized benchmark indices that can monetized in at least four ways up and down the value chain. The owners of the S&P, MSCI, DJIA, etc. receive licensing fees from CBOE (and other exchanges who license their indices on an exclusive basis). CBOE collects fees from traders trafficking in futures and options tied to those indices, and fees for offering access to data tied to that trading activity. It also sub-licenses its proprietary VIX Index, which it builds (and owns) on IP licensed from S&P Global, to Barclays, who uses it to create a popular product like VXX (an exchange-traded note pegged to VIX).

Prior to BATS, CBOE, as an operator of

derivatives exchanges, did not have a

listing venue for VXX(1), and so while it benefited from VXX trading activity,

it could not enjoy the fees from listing the underlying. Now that it has BATS, CBOE can pinch the end,

vertically integrating listings(24) for exchange traded products created off

the proprietary indexes it sublicenses, and also complementing the market data

from derivatives with market data from the underlying security(25). Notably, the VXXB ETN – which, like its

predecessor, tracks the VIX and was launched by Barclays in 2018 in preparation

for VXX’s pending expiration(26) – is listed on CBOE, not the NYSE.

While ICE and Nasdaq have positioned themselves as data and technology companies, CBOE’s main emphasis, like CME’s, continues to be exchanges. Now that cash equities account for ~25% of ICE’s net revenue and data has become a meaningful earnings contributor, one might wonder if a post-BATS CBOE hones its focus on data, but with all the push back that these fees are under, I somehow doubt it. The company seems leery about overcharging for data and obstructing the trading volumes that it is really optimizing for. To the extent data revenue does grow, that growth will come from new products and new market participants and new markets (i.e. cross-selling US derivatives data to European traders on BATS exchanges)…in other words, from volume rather than price.   

But part of CBOE’s exchange business may nonetheless find itself collateral damage to the actual or perceived abuse that exchanges have heaped on brokers. This past January, a group of broker-dealers, banks, and market makers announced that they would be launching a rival equity exchange, the Members Exchange (MEMX), ostensibly to “to increase competition, improve operational transparency, further reduce fixed costs, and simplify the execution of equity trading in the U.S” but practically to reclaim the margin that they have for so long donated to third party exchanges. Members Exchange hasn’t yet filed for regulatory approval and this move may just a bargaining chip to extract concessions on data and access fees. Whether or not lower data fees offset the system-wide costs introduced by more fragmentation is an open question.

On the one hand, back in 2005, BATS brought to bear superior technology that lowered transactions fees to a fraction of what Nasdaq and NYSE were charging, at a time when Reg NMS imposed best execution requirements on brokers. It is unlikely that MEMX can boast a similar cost advantage relative to today’s exchanges and ATSs. After transaction rebates and routing/clearing expenses, Nasdaq extracts just ~$0.05 for every 100 shares traded. CBOE captures around $0.03. For multiply listed options contracts, CBOE realizes just $0.07 per contract (down from $0.21 in 2006). You need huge volumes to profitably scale the fixed operating costs of running an exchange on those gross profits. Also, around 1/3 of equity trading takes place off-exchange, a lot of it in the dark pools operated by MEMX founders, and CBOE, NYSE, and Nasdaq could compete for volumes that get pulled into the exchange domain.

On the other hand, two-way network effects can be fragile when participants on one side of the network are sufficiently consolidated and can coordinate activities for their collective interest in a way that fragmented, atomized suppliers cannot. Exchanges are nice businesses, but they don’t dominate the value chain, squeezed as they are between powerful interests up- and downstream. Most of CBOE’s revenue relies, one way or another, on IP owned by 3 companies: MSCI, S&P Global, and the London Stock Exchange. The brokers and market makers who trade on its US exchanges are an increasingly consolidated bunch: Virtu and Citadel, two of the nine trading firms backing MEMX, account for 40% of US equity trading volume.

There is precedent for market makers acting as king makers: ICE, dominant in energy futures, was stood up by energy majors; BATS was 90% owned by broker-dealers and market makers who accounted for nearly half its transaction fees; Direct Edge went from less than 1.5% of US trading volume to nearly 13% just two years after parts of it were sold to Goldman Sachs and Citadel, who subsequently directed volume to it (today, Bats + Direct Edge, who together comprise CBOE’s equities business, command nearly 20% of US equity trading volume); eSpeed saw its market share plummet from ~50% to 11% in just five years after its sale to Nasdaq prompted disintermediation-fearing brokers to shift their dealer-to-dealer treasury volumes to BrokerTec. But all things considered, given its disproportionate reliance on proprietary benchmark indices, I find CBOE relatively more insulated from competitive threats than Nasdaq.

Footnotes

(1) The

NY BoT also brought with it clearing technology that ICE has since leveraged as

a strategic asset, to clear rates and equity derivative trades. Around the same time, ICE tried buying the

Chicago Board of Trade for $9.9bn but lost to CME Group.  

(2)

Following the financial crisis, there was a big industry push to standardize

and centrally clear single-name CDS trades, which had historically been

executed OTC. ICE’s Clear Credit is the

largest CDS clearinghouse today and has been deemed a systemically

important financial market utility by the Financial Stability Oversight

Council.

(3) The

European exchange business, the “Euronext” part of “NYSE Euronext”, was

divested from ICE through an IPO in 2014. NYSE Euronext was the product of the

April 2007 merger between NYSE and Euronext and Euronext itself was the fusion

of Paris, Amsterdam, Brussels, and Lisbon stock exchanges and a London

derivatives exchange called Liife. EU regulators blocked a proposed $17bn

merger between Deutsche Borse and NYSE Euronext in 2012, arguing that the

merger would create a “quasi-monopoly” with 93% share of European

exchange-traded derivatives. At the time, exchange-traded derivatives

made up just 15% of all derivatives trading, substantially all of which

occurred OTC. But the Commission argued that, for various reasons

including size and cost per trade, OTC and exchange venues did not really

compete with one another and so were treated as separate markets for purposes

of calculating market share. This was not the first time DB’s

consolidation efforts were thwarted by regulators. In March 2017, its

merger with the London Stock Exchange was blocked on grounds that it would

create a monopoly in the clearing of fixed income securities…this, following 2

prior attempts at a union, one in 2000 and one in 2005.  

(4) Here are some of the major ones:

This

table doesn’t include several botched deals including CME/DB (2007), Boerse

Dubai/OMX (bidding war with Nasdaq, 2007), Singapore Exchange/Australia’s ASX

(blocked by regulators, 2010), nor the complicated web of minority interests weaving

through the above listed exchanges. Regulatory

resistance and more mundane factors like the “home bias” of retail investors

and sell-side brokers, and differences in regulatory, legal, and accounting

regimes, may have played a bigger role in confounding the once great hope of

active cross-border exchange trading. 

(5) with

over 90% of all listed ETFs, it has twice as many as the next largest venue and

will likely benefit from continued flows into passive investment vehicles

(6) As dark pools and other off-exchange venues account for more trading volume than any single exchange, they have come under intensifying regulatory scrutiny and oversight following alleged market abuses that compelled Deutsche Bank, Barclays, and Credit Suisse to pay nearly $200mn in fines. In Europe, under MiFID II regulation, dark pool trading volumes may be dramatically limited. Maybe this helps turn the tide a bit, we’ll see.

(7) 54%

and 75% of Brent and Henry Hub futures open interest is claimed for commercial

purposes. The ICE Brent Crude futures

contract refers to light, sweet crude oil extracted from 4 different fields in

the North Sea and represents the price benchmark for ~2/3 of the world’s

internationally-traded physical oil.  

(8)

This revenue is classified “Research, Data, and Analytics” segment of Moody’s

Analytics

(9)

MiFID II places limits on dark pool trades, which goes against the point about

fragmented markets, but a) less than 10% of the shares

constituting major European indices trade in these pools and b) MiFID does

not place such limits on “Systematic Internalizers”, essentially, investment

banks who frequently and systemically buy and sell securities from their own

inventory (as opposed to simply matching buyers and sellers) and are required

to publish post-trade and reference data (among other things) for the sake of

transparency.

(10)

According to management, around 88% and 75% of trading for municipal and

corporate bonds, respectively, are still done through voice orders. 

(11) By

stepping into the middle of every trade, a central clearinghouse

alleviates counterparty risk for buyers and sellers and lubricates transactions

(12)

MiFID

II’s Open Access policy unbundles exchange services from clearing, requiring

captive clearinghouses to clear trades executed on third party exchanges,

allowing netting and cross-margining to occur across different trading venues. Notably, in early 2018, after its agreement

to acquire LCH from LSE fell apart following the collapse of LSE’s merger with

Deutsche Borse, Euronext agreed to have its financial and commodities

derivatives trades cleared by ICE’s clearinghouse in the Netherlands for the

next decade. 

(13) “Market Infrastructure Operators”

(exchanges, clearinghouses, CSDs, regulators) make up ~2/3 of MT’s customers; buy/sell-side

firms make up nearly all the remainder. 

(14) With 5 clients bringing in ~$4mn

of revenue, this is a teeny part of Nasdaq. 

(15) Per this January 31, 2017 article from Reuters:

“The most

active NLX contracts, referencing three-month Euribor, had open interest of

around 42,000 contracts as of the end of December. By contrast,

InterContinental Exchange’s suite of futures referencing three-month Euribor

had open interest in excess of 3.3m contracts”.

(16) For instance, it may just be that

index licensing profits, which require no incremental capital to grow, are

disguising lackluster returns on this segment’s acquisitions.

(17)

The market data revenue is does realize comes mostly from real time data feeds

sold to Bloomberg and Reuters and, to a lesser extent, from licensing deals

with creators of ETFs and structured products. 

(18)

according to Flextrade: “In the

wholesale D2D market, 6% of Treasury trading is voice and 84% is electronic. In

the dealer-to-client market, 47% is voice and 53% is electronic, according to

Greenwich data”.

(19)

The background to the eSpeed/BrokerTec origins can be found in this Trading

Places article.

(20) 2020 run-rate, or $85mn, an

upward revision from the $65mn year 5 and $50mn year 3 synergies projected at

the time of acquisition

(21) prior to BATS, CBOE had tried,

unsuccessfully, to repurpose its options trading technology for futures trading

(22) through Dec 2033 for the DJIA and

Dec 2032 for the S&P 500/100/Select Sector indices

(23) VXX, because it rolls into ever

costlier VIX futures, is designed to lose value over time. It is meant to be used as a tactical,

short-term trading tool, not as a way to express a long-term view on VIX levels.

(24) Now that all trading is

electronic and securities are easily traded at comparable cost across different

exchanges, I don’t think it really matters so much to a company where its stock

is initially registered. These days,

exchange operators compete for listings by providing services like corporate

governance, IR, and publicity.

(25) In 2011, the SEC approved Bats as

a primary listing venue

(26) as an exchanged-traded note, VXX has a hard maturity date

Disclaimer: As of the time this report was posted, Forage Capital did not have a position in the securities of Intercontinental Exchange, Nasdaq, CME Group, or CBOE Global Markets. This may have changed at any time since.