By Jim Greco
This column originally appeared on The Medium
It had long been an open secret that KCG was attempting to offload their corporate bond trading platform, BondPoint. Before the Virtu acquisition, KCG went through a multi-year process of selling off all the “non-core” assets that it had accumulated over the years including a reverse mortgage business, a futures clearing merchant, an options market making group, a large stake in BATS, two NYSE floor trading businesses, and Hotspot, a currency trading venue (whew!). BondPoint was next on the auction block, but discussions were halted when the KCG-Virtu merger was announced in April 2017.
Recently, Virtu, having closed on its acquisition of KCG, re-started the bidding process for BondPoint. Virtu took on a lot of debt to complete the acquisition, which Virtu is looking to pay down as fast as possible (especially with shrinking revenue due to low volatility). Virtu hopes to raise $400 million in the sale.
In this week’s newsletter I walk through the history of BondPoint, the economics of the business, and how BondPoint might be accretive to different acquirers. I spoke with five sources with intimate knowledge of the sale, all of whom asked to remain off the record to speak more freely.
The History of BondPoint
BondPoint (then ValuBond) was started all the way back in 1999 at the peak of the internet bubble. Stock trading had gone online five years earlier with firms like E*TRADE commanding multi-billion dollar valuations. Bonds were next, or so the founders believed. The firm, based as far away from Wall Street as you can get in Atlanta, Georgia, was backed by a number of prominent FinTech venture capitalists. It raised over $25 million before being acquired by Knightin 2006 for $18 million.
Like many of its peers, Knight never integrated any of the many acquisitions it made over the years and operated nearly all of them, including BondPoint, as separate businesses under the larger holding company. This strategy allowed Tom Joyce, Knight’s CEO, to grow revenues quickly, but it resulted in a bloated company with many duplicative front-office and back-office systems.
For many years, BondPoint operated as an afterthought inside of Knight. The firm never generated much revenue, but it took few resources to run. When the financial crisis happened, there was renewed hope that corporate bond trading venues would become more popular as banks reduced trading staff and the industry electronified.
KCG was formed when GETCO bailed out Knight (and effectively its BondPoint subsidiary) in 2013 after Knight lost $440 million in 45 minutes. KCG’s management team promised investors that the acquisition would be successful because they would streamline the company’s bloated infrastructure and finally do the hard work of moving everything onto a single technology platform. Part of the original plan included moving BondPoint off its 15+ year old code base to take advantage of the newer technology that underlies KCG’s dark pool.
Virtu has made similar promises to its investors about the KCG acquisition. This time though, in keeping with Virtu’s lean corporate culture, CEO Doug Cifu does not appear to be wasting time. Entire management layers have already been eliminated, and everything is being migrated onto Virtu’s technology stack. Legacy KCG employees are exhausted with the pace of change but respect the progress that is being made. Here is more from Doug in Virtu’s recent earnings call:
At December 31, 2016 the headcount of legacy KCG was 952 people. As I speak, the headcount is 610 people. …
In addition, we have taken the following actions. Offices in Singapore and India have been closed. In London, we have closed unprofitable trading operations and focused the business on the significant opportunity in Europe with regard to Mifid 2017. …
… Actions taken to date, barely three weeks after the closing, along with further actions will result in at least $125 million of annual run rate savings by the end of the third quarter of 2017, which is 50% of our stated growth synergy target of $250 million.
One team inside of Virtu that has been completely untouched is BondPoint. Bill Vulpis, who runs the group, and the rest of the management team have ordered business as usual. This has been relatively easy because BondPoint has its own management team, software development team, sales team, and back-office system. Any sale can be completed quickly and without much disruption to the rest of the firm.
I know what you’re thinking: $400 million! How is that price even remotely justified?
It turns out that while KCG’s overall revenue was crumbling, BondPoint grew into a pretty good business. In 2016, BondPoint averaged $200 million in trading volume per day. Although corporate bond fee schedules are complex, with specific fees based on the customer’s negotiated deal, credit rating, and maturity of the bond, it is safe to estimate that BondPoint makes about a sixteenth of a percentage point on each trade. That brings in an average daily revenue of $125,000 or $31 million per year. Not too shabby for a venue that is operated by roughly thirty people.
Furthermore, the market is growing, and a rising tide lifts all boats. BondPoint has posted average daily volumes of $300+ million in recent months. Meanwhile, MarketAxess and Tradeweb are hitting new records. It is not to hard to imagine revenue at BondPoint doubling over the next few years as long as it continues to execute.
Why not just keep the platform? The problem is one of valuation. Trading firms are priced based on book value. Exchanges and trading venues are priced with multiples similar to software firms (e.g., MarketAxess and NASDAQ have a P/E of 50 and 60, respectively). BondPoint is valued much higher detached from Virtu.
It is always risky to stick your neck out there and do an analysis of the potential acquirers of a corporate bond trading platform in a weekly newsletter. When I’m wrong (which I inevitably will be), it will be memorialized for all, but despite that risk, let’s go for it. Below are some potential acquires and my thoughts on what could motivate them to drop $400 million on BondPoint.
ICE. Intercontinental Exchange made a big bet on the corporate bond market in 2015 when it purchased IDC for $5.2 billion. The newly-christened ICE Data Services is the industry’s end-of-day reference price for millions of illiquid securities. Anyone involved in the corporate bond market has some kind of ICE data contract.
ICE also owns NYSE, which has the NYSE Bonds platform that has gone nowhere through multiple relaunches. The synergy of a leading corporate bond platform with IDC data services might be too much for ICE to pass up. ICE does not typically do small acquisitions like this, so ICE has to believe BondPoint is going to grow into a multi-billion dollar business. I think it is, but ICE is also busy with integrating a bunch of a number of other firms at the moment.
Tradeweb. Tradeweb has been frequently mentioned as a potential acquirer, but I think they are unlikely to end up buying the company. Tradeweb and BondPoint operate very similar retail businesses. Most of their customers are connected to both firms. As a liquidity provider, you will often see the same bonds for sale on Tradeweb and KCG simultaneously.
Tradeweb could potential be a takeout bid to consolidate the retail marketplace, but that is more of an eight figure purchase price than a $300 to $400 million price, to say nothing of the integration costs, disruption, and headaches.
MarketAxess. MarketAxess is also frequently mentioned as an acquirer of BondPoint, but I think the difficult integration of the two firms is going to give MarketAxess pause.
With a market cap of over $7 billion, MarketAxess would not have any problems coming up with the capital to fund the purchase. MarketAxess dominates the corporate bond trading space with 16% of the high-grade TRACE prints. And it is clearly thinking a lot about maintaining that market share as competition in the space heats up, throwing shade Tradeweb’s way:
“To me they are confusing the reader rather than clarifying the result and that is not up to industry standard,” [Rick McVey, MarketAxess’ CEO,] told the Financial Times.
MarketAxess and BondPoint do not have a tremendous amount of overlap among customers with MarketAxess historically focusing on the institutional space. However, MarketAxess has a mature integrated technology platform with trading protocols that are much more flexible than BondPoint’s. MarketAxess would likely have very little use for much of BondPoint’s intellectual property and face similar costs, disruption, and headaches as Tradeweb in bringing BondPoint under MarketAxess’ tech roof. MarketAxess would just be buying the customer connectivity, which would be unlikely to justify the $400 million price tag. The only real option would be for MarketAxess to operate BondPoint as a separate business and slowly integrate the platform over a long period of time.
LSE. The London Stock Exchange owns MTS Bonds, which runs several dominant European corporate and sovereign bond trading platforms. MTS has attempted to crack the U.S. market for a long time without much success. MTS built its European network through a number of acquisitions that are not well integrated with each other, so it is used to dropping entire platforms into the larger MTS umbrella. How much more MTS is willing to put down on that already-failing trade remains an open question.
CBOE, CME, NEX. These market operators have an outside chance of making a bid for the one area where none of them has any presence, an electronic corporate bond market. CBOE acquired Hotspot, so it may want to add another market from the old KCG portfolio. NEX already operates a number of electronic platforms largely separately, as well as having the leading US government bonds platform, which might be complementary to expanding BondPoint in the institutional space. At the end of the day these outliers may be the only firms still willing to shell out nine figures for a fixed income market based on legacy tech.
Others. MarkIt, NASDAQ, Deutsche Borse, BGC. There is no shortage of due diligence be done across the street!
Spinout. The final option is for Virtu to spin out BondPoint as a completely separate entity. Virtu has a longstanding relationship SilverLake, who helped setup a special purpose vehicle for Virtu to purpose KCG. If BondPoint continues to grow volumes at the same rate then it is not unfathomable to think about a unicorn valuation down the line. However, Virtu is looking for cash to pay down the debt they took on so this seems unlikely.
In summary. My top three acquirers are ICE, LSE, and MarketAxess. All have the cash on hand to do an acquisition and would see a lot of synergies with other aspects of their business.
Jim Greco is a trader, developer, market structure enthusiast, and founder of Direct Match and Trading Places.
The vendors will put pre-trade, trade, and post-trade data on one screen.
Despite more transparency, will regulation stifle liquidity.
Report says trades over $5m continue to prove difficult to execute.
It's getting easier to execute small corporate bond trades.
Sometimes past performance does indicate future liquidity.