Reinventing the Business Plan
Excerpts from a roundtable discussion sponsored by PricewaterhouseCoopers, held on June 19 in New York City.
Jim Vinci, co-lead partner in financial services, PricewaterhouseCoopers|
Joe Kolman, editor, Derivatives Strategy
Mark Galant, CEO and founder, GAIN Capital|
Peter Fraser, vice president, JP Morgan
Christine Berthet, consultant
Mike DeAddio, chief technology officer, Cygnifi
Ayaz Khan, director of corporate development, eCredit
John McEvoy, president, Creditex
William Margrabe, consultant
Neil Chriss, president and COO, ICor Brokerage
Michele McCarthy, managing director, Deutsche Banc Alex Brown
Jim Vinci: I'd like to focus on a few e-commerce companies and talk about their particular value propositions. What part of the business model are they addressing? Is the focus on cost reduction? Increasing market share? Maintenance of market share? Is it an alternative distribution strategy? Is the company going after someone else's revenues? How many people are in that particular space for e-commerce?
And at the end of the day, we should ask the following question: Would you recommend that your company invest in this business? Or, to put it another way, When the IPO comes out, would you buy it?
The range of bank strategies we see today is quite dramatic. A lot of banks have said, We aren't really sure what's going to be successful, so we will make investments in many companies as they come along, as a hedge in case our basic businesses deteriorates. Some banks have taken their internal operations and put them into a dot-com to generate revenues. Some companies are seeking to gain cost efficiencies internally, while others are looking at outsourcing. Some are trying to capture bigger market share through structuring alternative distribution systems and adding revenues via a web portal-type company.
|I'd like to focus on a few e-commerce companies and talk about their particular value propositions. What part of the business model are they addressing?
— Jim Vinci
I thought we'd start with CFOWeb, one of the more interesting entities. CFOWeb says it intends to provide services and real-time data to the CFO, treasurer and fund manager, as well as links to interfaced financial institutions. The business model involves providing value-added analytics, but it also involves a bit of disintermediation, which is intended to capture the relationship with the customer. Does anybody know what their revenue model is?
Joe Kolman: They charge a fee to banks that want to provide bids to end users. There is also a dealer fee per transaction. I think it's a challenging business model because you need to create critical mass on the dealer side. That scares dealers because it creates margin transparency. I think it's coming. Depending on whom you talk to, it's either around the corner or two years off. But it takes some of the dealer margin away, especially in low-margin areas such as swaps. That's where they make their revenues.
Vinci: To be fair, CFOWeb and similar systems are doing more. They're providing trusted analytics to customers for valuation and pricing, so a customer isn't totally dependent on a product provider's analytics. On the other hand, since the CFOWeb business model is tied to product providers, I'm not sure how much analytic transparency it can really provide.
Mark Galant: The CFOWeb premise is to disintermediate the banks, but the catch-22 is that it will need the banks to provide credit, liquidity and execution. The banks have asked, Why should we give these revenues to CFOWeb? They've decided that they can provide analytics that are just as good, provide the liquidity, and do it themselves. I think you'll find that CFOWeb and other intermediary-type providers will have a more difficult time in the future. A year ago, when Harpal Sandhu first had the idea for CFOWeb, it was difficult to imagine that the largest foreign exchange banks in the world would get together and form an alliance, given that they would have to cut prices to remain competitive. So it will be interesting to see if CFOWeb's first-mover advantage will be enough to hold off the challenge from FXall once they go live.
Mike DeAddio: SwapsWire is another consortium that operates on a similar business model. The banks are banding together and saying, We don't want to be disintermediated. We want to do it ourselves. But CFOWeb can still win if the banks can't agree to work together and if there is too much infighting.
Kolman: How do you see the strengths of the consortium model relative to the competition from these Internet ventures?
Peter Fraser: I don't see SwapsWire as a direct competitor to CFOWeb. SwapsWire's mission is to act as a catalyst for the electronic trading of swaps on-line, and to get that electronic trading done on a standard protocol—one we can all agree on. Our goal is to make the industry more efficient. It's driven by the anticipated savings from middle- and back-office processes rather than by any threat of disintermediation per se. That's not to say we don't recognize that we need to make a specific offering for the client side. But our initial focus is on the interdealer, day-to-day vanilla hedge trading. That's where the bulk of the transactions are, and hence the bulk of the savings that we think can be delivered.
Christine Berthet: The key is who has the skin in the game. Everybody's analytics will be more or less the same. The real issue is staying on the line to execute and then buy back.
Vinci: Let's go on to the next company: Cygnifi. Mike, what's Cygnifi's value proposition?
DeAddio: I don't think we're competing with CFOWeb. We're not about execution, we're not about getting in the middle between the client and the dealer. What we've done is take a large portion of JP Morgan's derivatives and risk management technology built up over the last seven years and spin it out.
We're going to provide these portfolio and risk management services to a broad base. We're targeting the buy side with an application service provider model that allows us to deliver simplified tools in derivatives to a market segment that's not willing to pay the kind of prices associated with traditional models today.
|The key is who has the skin in the game. Everybody's analytics will be more or less the same. The real issue is staying on the line to execute and then buy back.
— Christine Berthet
We want to do that to the bigger buy-side firms, as well as through a portal strategy with providers. Existing dealer portals will be able to provide these services to their clients through an ASP model.
Vinci: Can you talk about the revenue model for Cygnifi?
DeAddio: It is unclear if the clients will pay for it or if the providers will be sponsoring their clients. It will probably be a combination of both.
Berthet: Is there a preference to transact through JP Morgan?
DeAddio: Not at all. It's a purely independent company. JP Morgan isn't a majority shareholder. Our partners are JP Morgan, Sybase, Bridge and NumeriX. But JP Morgan isn't preferred in any sense.
Michele McCarthy: It's difficult to do proper risk management if you can't get your whole portfolio into the system. It sounds like you're adding some sophisticated credit and collateral features, but what happens if people have complex portfolios with lots of providers and instruments? How can you handle that?
DeAddio: That's a good point. There's no doubt that when you get into certain markets, you need to provide breadth in your product line. It's all going to take time to accomplish. But we're beginning by shooting for the vanilla market that makes up 80 percent or 90 percent of the portfolios of most of the clients.
Vinci: Is there a range of costs for the ASP?
DeAddio: We're trying to shoot for a broad range of costs. To be successful, we need to be able to make money on clients who'll pay us $10,000 or $20,000 a year for Internet-based portfolio evaluation tools. We also need to be able to service the higher-tier government agencies that haven't yet bought a big risk management package. Those start to get into the more traditional numbers. As a chief technology officer, my job is to have an infrastructure that goes both ways, which is by no means trivial.
|It's difficult to do proper risk management if you can't get your whole portfolio into the system. What happens if people have complex portfolios with lots of providers and instruments?
— Michele Mccarthy
Vinci: Would the lower-priced service include integration costs?
DeAddio: No. That kind of pricing is a commoditized, standardized kind of product offering that doesn't involve integration. I've seen a report from the Tower Group on ASPs, which is quite illuminating. It's very difficult to offer standardization, which creates low-cost offerings, as well as the customization you need for the higher-tier clients. If you get it right, you'll be successful. If you get it wrong, you fail with one group or the other.
Vinci: How will clients get their portfolios back and forth? Once a client has assembled an aggregated portfolio from its five or six or 57 legacy systems, how does it send the portfolio to you and how do you map it into your analytics?
DeAddio: The answer is XML. We're placing our bets on FpML, which is near and dear to a lot of hearts in this room. From a vendor-community perspective, we simply want a standard that will open up the market for niche players, for efficiency gains and so on. To represent derivatives transactions in a standard format is the nirvana we're looking for.
Vinci: Let's look at another company. eCredit.com addresses one of the most black-box analytical devices that exist in a bank today—the credit analyst. A big pile of financial information goes into the system and at the end of the day, the analyst says, Well, I really just don't like this name.
Ayaz Khan: The two main constituencies we serve today are merchants, which are the people who sell stuff, and banks, which want to finance the stuff that people sell. We also have three functions operating within our business model.
The first is helping a merchant sell on the basis of better-defined and more standardized credit standards—if Cisco, for example, wants to sell to a consumer or small business and wants to apply uniform credit standards. We have solutions that help the Ciscos and the Gateways underwrite their balance sheet better.
|It's very difficult to offer standardization, which creates low-cost offerings, as well as the customization you need for the higher-tier clients. If you get it right, you'll be successful. If you get it wrong, you fail with one group or the other.
— Mike Deaddio
The second thing we do is that we help the credit analysts in banks represent their underwriting processes and criteria in a more standardized form. We offer an outsourced credit underwriting platform, similar to what we offer merchants. They define and describe their credit underwriting criteria and processes to us, and we have a flexible technology that allows that process to be represented electronically and in a more standardized fashion. Not every decision can be automated, but we will go as far as the credit underwriting people at banks or at merchants want to go. We pull data from standard sources, whether external or internal, to help them represent their decision processes.
The third thing we do is that we provide an automated deal-processing engine for the two models I just described. We also perform industry clearing. We pick up a transaction that's generated at a merchant and flow it to a preselected financial institution seeking to finance that transaction in real time. If the financial institution, for whatever reason, declines the application, the application can be routed to an alternate financial institution. So this is not an auction model. It is the newest piece of our business, but it is growing.
We don't call it disintermediation because we are just adding a new intermediary to an existing process. We call it reintermediation. We are matching a deal to a financier. We're not looking to get into the business of matching buyers and sellers and making markets. We want to take our credit underwriting tools and apply them to more sophisticated markets. We also want to add portfolio tools and other risk management tools to our core offering, so people who buy and sell risk can do so more efficiently.
Vinci: Those comments are geared toward the retail and trade merchant capability. But the technology is applicable to other areas in financial services. What do people think about the reengineering of the credit approval process as a part of the trading process on-line? Do we need better credit processes surrounding the approval of transactions?
John McEvoy: We've taken an interim step. I agree that, ultimately, you want to have all that information on-line, but it is very ambitious to have real-time counterparty limits on day one. Eventually, creditex and the market will get there. In the meantime, we have taken steps toward that by developing a system with multiple credit-screening mechanisms. Among other things, in our system we effectively create something like a buddy's list, where all counterparties on the system give us their approved counterparties. When you look at a bid or offer, the system tells you if the position has been posted by someone who is on your list and vice versa. It is not the end game, but it gives people a lot of counterparty information before they pursue a transaction.
Kolman: The advantage to me is that it doesn't ask the end-user for a lot of information. You don't want to build a system where you say, Please upload 15 megabytes of information every day, so we can do a really good job of handling your credit. People just won't do it.
McEvoy: Even intraday, it would be virtually impossible, because credit limits change intraday.
Vinci: Does anyone see a problem with this?
William Margrabe: I see a bit of a problem. On the one hand, you want to maintain anonymity, and on the other hand you want to give credit information. It seems to me that if you provide enough details, there will be a one-to-one correspondence between the credit information and the counterparty.
Neil Chriss: You're absolutely right. With some of our competitors, people have figured out how to game the system in such a way that they can identify counterparties simply by setting up credit limits in a certain way to help them pinpoint where a particular price is coming from.
We've worked very hard at ICor to get around that problem, but it's not so easy to defeat. Ultimately, you give up some of the fine-mesh credit information that you'd would want to give. But in exchange for that, you get a system that lets people actually put up prices anonymously.
DeAddio: The big banks are now moving to a situation where credit is handled on the trading desk itself and consolidated there. JP Morgan has done that. For any new trade, you get a true marginal credit exposure analysis as to whether that trade will enhance or detract from the credit exposure of that particular counterparty. Traders actually have that information at their fingertips and can manage it on an active basis. That's how the big institutions are managing things because in the summer of 1998, credit was melting down and they didn't know what their overall corporate exposure was.
Vinci: Mike, I have to tell you, though, that there are probably four banks that can really do that.
To sum things up, I think we'd all agree the Internet is having dramatic effects on the financial services business. More people are realizing that the real money is going to be made in re-thinking the traditional value propositions. People are looking anew at all the parts of the value chain and attempting either to disintermediate the customer, outsource, generate new revenues and/or distribute products in new ways. And it's all happening very quickly. In the week that it takes to put together a presentation, you might be out of date. The companies that succeed will clearly be the ones that can think most creatively.