E-commerce - Bonding on the internet

Phillip Buhannic

Sounds familiar? It should - it's exactly what debt-market bankers have been trumpeting in recent months. Several firms opted to go through e-bond underwriting exercises in January as proof of their commitment to, and leadership in, e-commerce.

Yet these same institutions provoked Fondren's wrath in the past. So much so that last November InterVest filed a lawsuit against several banks and brokers, claiming they had conspired to boycott the firm and keep the bond-market price process opaque. Cantor Fitzgerald, SG Cowen, Deutsche Bank, Liberty Brokerage, Merrill Lynch, JP Morgan, Bear Stearns and Salomon Smith Barney have all been named. InterVest asserts that at one meeting in 1993 executives at Merrill told it that dealers "made an awful lot of money on the inefficiencies of the bond market ... letting customers see where the market really was would crush spreads". Merrill has denied this.

There are other examples besides InterVest: at the start of the 1990s a few bankers at JP Morgan tried to set up an electronic trading platform. It lasted two months.

Mark Ferron

The fracas caused this January by various attempts to underwrite and syndicate bond issues over the internet must be seen against this background. It will alter the way business gets done, but it is just the first step down a road the debt markets have studiously managed to avoid in recent years.

With some justification debt bankers talk about the complexities of their product. "Trading debt products is not like trading equities," says Victor Simone, managing director in fixed-income e-commerce for Goldman Sachs in New York. "What you buy on the New York Stock Exchange is exactly the same type of product as you might buy on Archipelago or Brut [two of the electronic communication networks for trading US equities] ie stocks are routable. There is a lot more variety to fixed-income products."

Sheer numbers is one thing. According to a study by the Bond Market Association last November, on e-commerce and the bond markets, there are about 4.4 million outstanding debt issues in the US alone, compared with some 4,000 equity issues. And equities have exchanges as a central marketplace; the 4.4 million bond issues rely on partisan market-makers. And debt products are far from homogenous: maturities, yields, structures and credit histories all vary.

Benjamin Cohen

Their absence has allowed incumbents to maintain more control over bond market developments than was possible in equities, and by watching the effect of e-commerce in that market, and in forex to an extent, they have learnt not to hang back and hope that nothing changes for them. "One of our main worries is that we might miss out on something that we traditionally haven't imagined to be either a part of or a threat to our business," says the head of e-commerce strategy for investment banking at one of the bulge bracket firms.

Realizing that control can be wrested from them is a major step forward for these banks, but other factors have had an impact. Market forces, for example, are cajoling banks to adopt e-commerce more readily. "The biggest problem all of us face is that our costs, whether technology, people or otherwise, are constantly increasing at the same time as spreads are decreasing and hitting our earnings," says Ben Cohen, managing director and head of CSFB's e-commerce initiative for fixed income and derivatives.

Cost-cutting approaches

Getting appropriate infrastructure in place to make the transition is another matter, as is getting investors to take part after telling them for years that the best way to do business is by phone. That's why most electronic systems up and running so far are there because the banks and brokers want to cut costs.

There are the inter-dealer broking developments: Cantor Fitzgerald, Garban-Intercapital, Tullett & Tokyo and others are all offering or setting up various systems. So far only e-speed, Cantor's system, has had much success. And Brokertec, the consortium set up by Goldman Sachs with six other large institutions last summer is trying to create an electronic inter-dealer broker for cash and futures.

TradeWeb, an electronic brokerage system for US treasuries, was set up by four US banks (CSFB, Goldman Sachs, Lehman Brothers and Salomon Brothers) in 1997, and has been a success. From handling about 1% of volume at the start of 1999, it now controls about 12%, and may soon branch out into other debt products. It started trading US treasuries in January 1998 over the internet. Treasuries are relatively homogenous, have an excellent credit history, and are core benchmarks for the overall debt industry. They are the ultimate safe play, are heavily traded, and their price well known: no-one makes money trading them. So why not do it electronically?

Simone Victor

To an extent, the drive to offer new bond issues on line follows the same logic. Distributing research, prospectuses, filings and other new issue information over the web cuts down on paper, phones and faxes. And it stops salespeople getting bogged down with admin.

But so far that's it. Banks might make a big noise about the amount of orders that have come in over the internet for deals done in January, but the story is not that simple. The first deal was that for Freddie Mac, lead-managed by Warburg Dillon Read and Merrill Lynch. It soon degenerated from an attempt to get a few favourable headlines for selling bonds over the net into a cat-fight between the two leads. Merrill was accusing Warburg of acting in bad faith by trying to take credit for the internet side of the deal. It had been agreed in advance that the announcement of the e-offering would be up to the issuer, said people taking Merrill's side, yet Warburg weighed in early to try to steal the limelight. "Not so" responded the Warburg supporters. Merrill was just sore that its new-issue internet vehicle was not ready in time and that it could only use the Freddie Mac deal for beta testing.

Somehow, the bankers managed to get the deal done despite the insults they had thrown at each other. Two weeks later Merrill was still smarting: new-issue head Jimmy Quigley made pointed remarks about fantastic e-distribution claimed for earlier offerings as he and his colleagues unveiled "i-deal", Merrill's new-issues internet platform for debt and equity products. "The deals which have happened so far have been badly misrepresented," he said.

Judging by the market reaction, he was right. Warburg claims to have sold a significant volume over the internet, but won't release figures. A senior banker at a competitor reckons Warburg sold around $10 million bonds, elsewhere a figure was mentioned of $42 million, and that only after Warburg asked a client if it would mind placing a $25 million order to get the numbers up.

A World Bank deal had its own mix of farce and success. It was announced in the first week of January but not launched for another two weeks, prompting the comment: "With a pre-marketing period that long they could have used carrier pigeons."

The literature said this was to be the first true internet deal: all syndicate members had to be e-underwriting-compliant, even if orders came in by phone. Goldman Sachs and Lehman Brothers were said to be joint leads, because they could each bring a strong retail broker into the equation - part of the rationale was to improve the access of middle-market institutions and retail investors to bond issues. Goldman brought in Charles Schwab and Fidelity tagged along with Lehman - these two formed a binding joint venture last summer, whereas the Goldman-Schwab link is less formal. Unfortunately, Fidelity couldn't yet distribute debt on-line. So the World Bank dropped it from the syndicate and relegated Lehman to co-lead.

This is no disaster, since it is just the beginning of the process, but it is unlikely that Lehman or Fidelity will be consoled by the words of Gumersindo Oliveros, director of treasury products at the World Bank, who said that Lehman and Goldman "participated equally" in the transaction, nor by the thought that all new processes tend to suffer hiccups.

But what is the World Bank trying to do? Sure, let's aim for a broader investor base, and let's use the internet, that great leveller, to achieve it. But to insist that all syndicate members have e-distribution capability is potentially counter-productive. It could shut out pockets of investors that the large firms with their internet toys might not be able to reach. That is, after all, why obscure local banks tend to get a role in the syndicate. Maybe just to limit it to the lead-managers and senior syndicate members would be more productive: after all, they all claim to know all the major investors, so if one isn't up to speed, another can easily step in.

There are some concerns in the market that volumes done over the internet are being exaggerated, or are being talked up by phone. "It would be interesting to find out how many phone calls were behind these internet trades," says an e-commerce specialist at an investment bank. Another banker mentions how he had talked to salespeople at some of the underwriters involved who told him that they had to ring around their accounts and ask them to submit internet orders, and that some refused.

Other banks are speeding up development of their own platforms. Cohen and Buhannic were close to announcing CSFB's on-line new-issue platform, but speeded it up by a few weeks following news of the Freddie Mac and World Bank bond issues. They had thought of everything except the name until, in mid-interview with Euromoney, inspiration came: "Philippe," shouts Cohen, "how about calling it PrimeDebt? Since it's going to be linked to PrimeTrade soon, that'd make sense, surely."

Mark Ferron at Deutsche Bank dismisses a lot of the e-underwriting so far this year as "hype and vapourware". But he predicts that Deutsche will have to roll out its own product sooner than it planned.

Some bankers suspect the internet issues will encourage a new kind of deal support: underwriters, regardless of demand through the old channels, will be so eager to prove their internet presence that they could be tempted to buy back bonds through the internet just to make the numbers look good, before selling them again in the secondary market.

A rather different test might come when WR Hambrecht launches its debt product. The company is better known for its attempts to make IPO bookbuilding more scientific by creating a Dutch auction process called Open IPO. The idea is to try to avoid massive gains for investors on launch at the expense of the company's owners. Now it is trying to do similar things for debt issues, having hired Tim Opler from Deutsche and Mike Evelyn and others from Merrill Lynch. Exactly what they're planning is unclear, and although some expect auctions in the debt markets to fail for all but the very biggest most liquid issuers, because extending credit facilities is too crucial and little start-ups can't do that, others are more hopeful.

Will the internet change the inner workings of the bond market, or are these new developments superficial? Suppose an institutional investor wants to rebalance his portfolio, for example, by dumping poorly performing corporate bonds, and replacing them with a mixture of paper from the secondary market and a few chunks of new issues that look attractive. He'll need to hedge with some futures and do forex trades for some of the bonds. It's a fairly big transaction, so he'll need the use of a market-maker's capital.

Today that trade would take time to organize, and might involve more than one institution. Nor would it come cheap. Surely e-commerce should facilitate it. The hype given to new issues has diverted attention from developments that could effect a more profound business shift - bundling the services of more than one institution through the same portal.

BondHub is the brainchild of Leo Schlinkert and his new firm Communicator. Formerly at Salomon Smith Barney, Schlinkert spent nearly two years laying the groundwork for a simple but radical idea: to create a web portal that allows access to the research of three investment banks, Salomon, Goldman and Morgan Stanley Dean Witter - or to as many of them as the user is a client of.

Then last month three other banks, JP Morgan, Chase and Bear Stearns, announced that they were creating another internet company, MarketAxess, again offering a portal. This one will allow clients access to the whole debt-markets universe from research to new issues to secondary trading and clearing.

Both systems allow for a great deal more self selection by clients without destroying the banks' brands. Clients use Bondhub or MarketAxess to aggregate the information, and then go to the individual bank's site to carry out the business. "Banks are realizing that often trying to do things by yourself is less compelling than a joint effort," says Hamid Biglari, co-head of US investment banking at McKinsey and Co in New York. "The barrier for differentiation has gone up, so in some parts of the fixed-income business it's better to opt for coopetition - competing or cooperating with rivals as circumstances dictate."

An e-commerce philosophy

Speaking just before the MarketAxess announcement, Thorkild Juncker, managing director and head of financial markets e-commerce efforts at JP Morgan, explained what drove his bank's philosophy on e-commerce. "What we're mindful of is whether ventures are utilities or for-value. So many are getting caught up in utility-like efforts where the driving force is the big players seeing margins decreasing. They may hold no value, and there might not be an exit plan. It's more a defensive move. Outside the fully commoditized markets we're looking to create value propositions for the future, which means including technological companies, venture-capital firms as well as the clients in the discussions. It's as if we're acting as a start-up firm."

The major partner with the three banks in MarketAxess is Moneyline, a provider of real-time financial content and trade execution services. So rather than using their own people, the banks are using a software firm that specializes in financial applications to develop and host the site.

That doesn't mean that JP Morgan will only ever be involved in out-of-the-box developments. In January it joined five European institutions in an on-line trading venue for European government bonds, bondclick.com.

But JP Morgan has been the driving force behind MarketAxess, which allows the banks to mount a combined effort to extend their brand: one aim is to increase information flow and, with any luck, business to the small-to-mid-size investor base: those with $500 million to $3 billion under management. But it is the first time that the needs of the client have been put first. And it is a comprehensive offering of products: as well as agency and municipal deals, MarketAxess will offer corporate, high yield and emerging markets bonds.

These three banks collectively underwrote $219 billion of the $1.57 trillion raised last year in the fixed-income arenas targeted by MarketAxess. That they saw their way to cooperate in a hugely competitive space is little short of revolutionary.

On market share, the three combined are still in second place to bond market titan Merrill Lynch, but even this leader of the bond markets for a decade or more is aware that to go it alone might well not be the route of the future. "The e-world is going to unbundle this business we're in more and more," says Kelly Martin, global head of fixed income at Merrill Lynch, formerly chief technology officer for the whole investment bank. "And the industry is going to have to learn to cooperate more, both within the industry as well as with less traditional partners." But for the present Merrill is focussing on its new-issue platform; with its retail broker network it already has access to retail and mid-market accounts.

"We think we've got an excellent package here," says MarketAxess CEO Rick McVey, who left his position at JP Morgan to run the new entity. "We spoke at length with a variety of clients about what they wanted, including about 30 mid-size firms, and the response was consistent across the board: their need to see the bond markets change is high." Greater, and easier, access to research, prices, new issues, and the ability to compare were paramount. "This is more than just making an existing model more economical by shifting it on-line. It's about providing a greater and deeper access to a large and varied market whose participants are crying out for change. It's a great mix."

For one of the other partners, MarketAxess provided a solution to two separate issues. First, says Simon Lack, head of investment banking e-commerce development for Chase, "until now it hasn't been profitable for us to serve clients with assets of less than $3 billion. Now we can, and second, clients increasingly want to be able to use a single access point for conducting their business. If we don't give them the convenience they want they'll look elsewhere."

Does this mean that our theoretical investor can now conduct his complex trade much more simply? No. MarketAxess, for now at least, is designed to allow easier access to services, but not to create an effective merger of offerings. The three banks still want to keep their brands. But this could be the first step. Who's to say that in such a deal a couple of years from now the investor picks out Bear Stearns to do the bonds, JP Morgan to arrange the forex and hedging, and Chase to provide the capital? Or even that the investor simply picks the banks he wants to be considered to take part, and leaves it up to smart software to decide the allocation of service.

Now that is some way off, if at all possible. E-commerce might allow the clients more power to unbundle services, but still they will want to be able to track performance. So brand, capabilities and relationships will be crucial. But they will be much more closely linked to technology. "Once banks start going down the path we're going down with MarketAxess the technology becomes a big part of an institution's brand," says Lack. "There is a bigger premium on having the right technology and it has an ever more direct impact on revenue. So we have to get it right."

The market leader in this respect at present is CSFB and its PrimeTrade system. There are few detractors - criticisms centre mainly on generational development. "The problem first-generation systems such as PrimeTrade and Autobahn will have to face is that they are single-bank offerings and clients increasingly want co-mingled systems," says a competitor.

For now, PrimeTrade's creator Buhannic responds: "We got an e-mail a couple of weeks ago from one of the big hedge funds which uses the system. They thought they ought to examine some other systems and get one or more in as they felt uncomfortable using just our system. They invited our competitors in and discovered that either they weren't doing anything similar, were still in development, or simply weren't good enough. So despite their usual policy of having a variety of systems, they're just using ours."

PrimeTrade was developed first to allow clients to trade futures contracts electronically, whether over-the-counter or, where agreements are in place, with futures exchanges around the world. Foreign exchange has also been added, as has a system called PrimeClear, which allows for straight-through processing of deals and totally transparent data trails. And he is in the process of adding trading capabilities for Eurobonds and Eurocommercial paper, US treasuries and some commodities. At some point the on-line new-issue vehicle, PrimeDebt, will be linked up as well.

"We designed the system to be as flexible as possible," says Buhannic. "So it is multi-product, multi-service and multi-geographic." He has cut costs while competitors talk about it. "The only way to cut costs is effectively to get the customers to the back office themselves." By this he means that the PrimeTrade and PrimeClear system is installed on the client's desktop, so any trades put through it are fully automated from execution to clearing. "I've been able to cut my costs by up to 15% a year by going fully automated," says Buhannic.

His boss, Ben Cohen, who runs CSFB's fixed-income and derivatives e-commerce efforts, explains that PrimeTrade is one of several complementary offerings the bank has invested in. "Our philosophy is to attack the market from several different points at once, because different customer bases might want different offerings, and each client might want to have a variety of options open at any one time." So aside from PrimeTrade, CSFB is a founder member of TradWeb and Brokertec, and between Christmas and New Year announced that it was to invest in TradingEdge, which offers a service called BondLink, an on-line trading network in high-yield and distressed debt

This year Cohen expects to see an increase in co-mingled offerings from banks, and Buhannic has already been thinking about it. "I do occasionally get asked whether, say, Goldman Sachs prices are available on PrimeTrade. I say no, it's a proprietary system, but I do wonder whether that should be the way to go." In fact, he thought even beyond that. "It is entirely possible that we become conglomerates of e-commerce solutions. The business takes place outside these walls in separate legal entities, and we sit here and manage them."

Others are beginning to ask how much e-commerce could change the way banks and brokers operate. "Within five years I'd expect Merrill to have a series of partners, joint ventures, co-companies and other relationships doing our business with us," says Martin. "And they by no means have to be from Wall Street. They could be delivery companies, information providers, tech companies. Little of what we do is specific to finance - look at auctions. What happens if e-Bay sets up a finance auction site, regulations allowing of course?" So what does that mean for institutions such as Merrill Lynch? "Well, that's the difficult part. We have to be willing to reinvent this business, not just develop it."

One catalyst recently that has been largely ignored as a result of the new-issue furore is inter-dealer broker Cantor Fitzgerald's flotation of its electronic broking arm, e-speed. Cantor started offering it to its clients in March last year. Then in December e-speed went public, on Nasdaq, and its president, Frederick Varacchi, now describes Cantor not as an inter-dealer broker, but being "in the business of marketplace technology" or "a business-to-business electronic marketplace".

If a traditional broker used to skimming the ever-thinner cream off debt and equities trading can redefine itself so profoundly, and position itself as a financial and internet company, what do others in similar positions do? It would seem Buhannic's idea of external, separate legal entities is already happening. Another issue is what this does for valuations. "E-speed's IPO has started a change in the way our companies will be examined," says one banker. "We won't be examined as banks so much any more, but as internet or tech companies. Once that catches on there'll be a race to get the limited public money from IPOs which will be available for such companies. And that will begin to drive consolidation in our industry."

Debt-markets bankers, after seeing how the institutional equity market was caught so unawares by retail, are at least on the look-out. "This is no longer in anyone's control," says Goldman's Simone. "We're on the cusp of a sea change which will be nothing short of dramatic."

How do you start-up a start-up?

It's the start of December, and Ronald DeKoven is in his office on the 37th floor of the Citibank tower on Park Avenue in mid-town Manhattan. He has just launched the product he has been working on for a year, ereorg, an on-line site that he hopes will become the top global marketplace for trading bank loans and distressed debt. The first press coverage has appeared, and has been very favourable.

As he is about to start an interview with Euromoney, he's called into another room to take a call. He returns apologizing but saying that he couldn't really refuse a call from a senior executive at a large US commercial bank. "He asked if his institution would be able to invest, and I had to say no," explains DeKoven. It wasn't the first call he'd had on the subject, and probably not the last, either. He is determined that his new business should develop without appearing to have any bias towards particular market participants by making them equity investors and board members.

Instead, he has tapped the private-equity arm of asset management firm Warburg Pinkus, which invested $7.5 million in mid-December, and is set to announce another round of funding soon.

But actively to exclude market players would seem to fly in the face of virtually every other institutionally focused start-up thus far, where industry backing has been seen as crucial. No inter-dealer broker could ever be successful without the support and participation of brokers, and the same would seem to apply to start-up dealer-to-investor trading venues whether for equity or debt products.

Why? First, incumbent banks have usually seen little reason to allow upstart new ventures to take their business away; and, second, if there is going to be any change, the banks would rather try to exercise some control, and make some money from it as well. It drove investments in equity markets last year, as banks scurried to take stakes in various electronic commission networks in the US, or in electronic exchange Tradepoint in the UK.

In debt markets, whether an inter-dealer start-up like Brokertec, which has 12 participating banks, dealer-to-investor shops like TradeWeb, which has eight, or the new on-line trading platform for credit derivatives, Creditex, all either were set up by banks, or have significant investments from them, as well as bankers on the board.

So is this bold independent stance the right move for DeKoven? Or is he in danger of being downtrodden by banks irate at his attempts to change their business model for them?

In this instance, it could well be the best thing to do. Traders at the banks seem keen on it. "We've been getting so many résumés from traders, telling us how they think this is the way for the market to go and asking if we can offer them jobs," says DeKoven.

A more substantive point is that ereorg is seeking to tap markets that are not all that large. A few blocks down Park Avenue, a fellow debt start-up CEO makes the point. "There's a big difference between ereorg and what we're doing," explains Ed McGuinn, the former head of debt operations at Lehman in the early 1990s who now runs LIMITrader, which has just gone fully live trading corporate and high-yield bonds, and distributing new MTN issues "We have a thousand or more eyes focused on us, whereas trading bank debt is a smaller market, with maybe 100 or so eyes trained on him."

For now, at least, DeKoven sees much more value in having venture capital involved. "At some point on an internet start-up's path you need to get a major venture capital firm involved," he says. "On the one hand it lends some credibility, but also they understand the potential and actual weaknesses of a company."

He chose Warburg Pinkus in part because it's New York-based, but also because it focuses investments on companies exploiting the intersection of financial institutions and the internet. It also has a $5 billion fund. "So when they told me that I shouldn't feel constrained by our capital base, I knew it wasn't an empty remark."

Other start-ups have gone a similar route, until they get close to full launch, such as LIMITrader "We started off just with venture-capital money, but are now looking at getting more money from the street," says McGuinn. But he too lauds the role of the venture-capital investor. "When we got our investment from Softbank, it was like receiving a blessing. It was very helpful in increasing our profile and lending us some credibility." Aside from getting even more credibility by allowing brokers to invest, it can also give more confidence that the start-up might actually see some business flow.

By not tapping into street cash for credibility, DeKoven has concentrated more on establishing credibility through the people he works with. His co-founder, for example, is a former head of international equities at Merrill Lynch, Thomas Burnett. And he has hired Phil Andryc, who joined in December. He had taken a few months off after 20 years working at Morgan Stanley, and was looking to get back into finance just as DeKoven was searching for a high-quality president and COO - the former COO of the information technology group at a major Wall Street firm is no bad choice. He has also hired Andy Komarov to run product development. Komarov worked for McKinsey before moving to Ripplewood, where he was one of the key people involved in buying Long-Term Credit Bank in Japan; Kevin Lavin, who was a partner at PricewaterhouseCoopers covering multi-national company restructurings; and Cindy Cunningham, formerly a lawyer specializing in bank debt with Millbank Tweed.

But the street investor's reason, says McGuinn, can be altogether different, namely fear of losing control. "They are afraid of losing out on electronic trading to the likes of Reuters and Bloomberg, so they come to companies like ours."

Or at least to some of them. McGuinn has a particular way of explaining the street's investment strategies. "Let's assume all these start-ups are located around Lake Michigan. At the Chicago end is the crowded marketplace, with your e-speeds [Cantor Fitzgerald's on-line broking tool], TradeWebs and Instinets [which is developing a fixed-income electronic inter-dealer broking system]. Whereas I'm up in Milwaukee where there's less clutter and more space."

So far LIMITrader has taken just one investment from a broking firm - Jefferies and Co, which is usually one of the top 15 banks in US high-yield bonds. The only other broker to have invested in this space is CSFB. It took a stake in a rival to LIMITrader in December, called BondLink, which was developed by a broker-dealer called Trading Edge.

One of the reason the banks are more interested in the clutter of Chicago is because that is where their costs are highest and spreads on trades lowest; up in Milwaukee there is more of a need for discussion and negotiation - and, with less transparency in pricing, more opportunity to take advantage of wider spreads and make some money. What LIMITrader is doing is to take that negotiation on-line, which might worry some brokers. The system first pairs up potential matching orders, and then allows for direct negotiation on exact price on-line, where needed.

As with other electronic and on-line systems, there is a visible order trail, and so much more information about the prices previous deals have been done at. More information and transparency, so the argument goes, usually leads to more business. "The crucial element to a system trading high-yield bonds is the ability for price discovery," explains McGuinn. "That then frees up bankers to concentrate on the more esoteric and structured products."

Creditex has followed a somewhat different course in setting up its business. Its founders, Sunil Hirani and John McEvoy, used to be credit derivatives traders at Deutsche Bank in New York, and left to set up what they hope will become the primary electronic market place for trading these instruments. That posed a problem of being regarded as a pseudo-Deutsche operation, so they decided to approach the whole market with the idea. "We have probably seen the top 70 players in this market in both London and New York, which account for maybe 80% of volume," says McEvoy. "Many of them knew us from our Deutsche days and the response has been overwhelmingly positive and enthusiastic."

As a result Creditex now has a slew of investors, whether dealers such as Bank of Montreal, Morgan Stanley and WestLB, or end users such as Pacific Life Insurance, Capital Reinsurance Company and Financial Security Assurance. And yes, Deutsche is a lead investor. But they purposely left out their former employer - and JP Morgan, the other lead investor - until the others were involved.

And to keep matters simple, only some of the investors will be on the board, with the majority just holding equity stakes. "We made that clear to them all from the start," says Hirani. "We wanted to keep the board structure simple and therefore more nimble to respond to the needs of a fast growing company like Creditex."

That Deutsche and JP Morgan are supporting Creditex without having control might seem odd. They dominate in credit derivatives and might prefer to maintain that lead by keeping the market opaque. Not so, says Thorkild Juncker, managing director and head of JP Morgan's e-commerce. "JP Morgan enjoys a substantial market share in credit derivatives. Any impact of potentailly declining margins in vanilla products and smaller market share will be far outweighed by the benefits of greater transparency and depth in our credit risk management capabilities."

McEvoy and Hirani don't intend to disappoint, asserting that industry estimates put market growth at 110% in notional volume a year. "Credit derivatives are of strategic importance to all financial institutions who have or want to assume credit risk," says Hirani. "Yet the market lacks liquidity, transparency and until recently standardization." That latter point changed in July last year when ISDA introduced standard documentation for credit derivatives, which Creditex has adopted as the default offer document to be used on its site. "The market is adopting it, and so we did. Once it's more widely known, we expect the market to explode," says McEvoy.

Some aren't convinced yet. One banker's reaction was: "It's a case of internet hype meets a market growing at a fantastic rate. It's a good theory, but I'm not sure of the practicalities just yet."

That's rather harsh. The beauty of Creditex's system - and ereorg's - is that it is simple to use, cuts back on time and expense, and is using the distribution vehicle everyone wants to be in on. Having the backing of many of the market participants doesn't guarantee success, of course. Many are investing in start-ups partly as a way of hedging their bets, and might not actually put much or any flow through them. And as the ereorg model shows, sometimes it might be better not to have potential users as investors.

All three either just have gone live, or are about to, as are those at the more commoditized end, such as Instinet and Brokertec. "We'll be in a period of experimentation for the next 18 months," says Hamid Biglari, co-head of investment banking at McKinsey and Co in New York. "Some will make it, but many won't survive."