Here Are The Pieces, Now Let's Put Them Together: MTR Interviews ADP's John Gaven

Automatic Data Processing, Inc. -- a long-time fixture in back-office processing in the securities industry -- has in recent years been positioning itself to be a competitor in the front office as well. Through its acquisitions first of GTE Telenet's and then of Bunker Ramo's quote businesses, ADP has bought about a 45 per cent share of the market.

However, market share alone does not a successful business make. ADP faces the daunting task of pulling all of its pieces under one sales and systems umbrella, while at the same time moving everything forward into the next generation of office automation and broker productivity enhancements.

We spent part of a recent morning discussing the task with John Gaven, president of ADP's Financial Information Services division:

Gaven: I guess in the case of ADP, we think the acquisition of Bunker did strengthen us to the extent of getting real technical talent with knowledge of this industry. So I think we've got a stronger force in terms of both the systems programming development side as well as the field service organization. So now we have, basically, a 450-man field service organization. So we've doubled the size of the field service group, doubled our terminal presence in the marketplace, and increased our capabilities from the technical side. Also we are in the process of integrating the system such that we will have a common ticker processing plant in both Trumbull and Mt. Laurel, so now we will have a true backup facility now where we will be processing the same information and feeding all terminal products with the same market data information. So we think it's really strengthened us overall, especially in those areas. Also, Bunker had about six regional data centers, which had all UPS and raised-floor computer facilities where we're now co-locating our equipment. This will take place over the next twelve months. So we think that it really enhanced our capabilities.

Gaven: From the customer's side, it will take replacement of their existing terminals with our FS Plus or FS Partner, in that we will have the same exchange information updating the terminals, that will be common, but what won't be common is the network. We will have two separate networks so that the local loop between the data center and the customer's premises on Bunker's side will be using their protocol, and on the ADP-FIS or the old GTIS terminal system, the FS One, we'll have a unique network. We'll be running two separate networks until such time as the customers are converted from the FS One or the MDS 7/90 to our new terminal products, the FS Plus or the FS Partner. MTR: Regarding these recent acquisitions, if you look at them in terms of the cost per market share point, there seems to be a big disparity between what ADP paid for GTE and Bunker, and then what Citicorp paid for Quotron. The numbers that we've come up with seem to be under $2 million per share point that you paid, and more than $10 million per share point that Citicorp paid. Did somebody get a bargain? Did somebody pay too much?

Gaven: Well I guess you have to ask yourself what the rationale was for the acquisition. You know, was it financial? What were the strategic implications of the acquisition? My personal feeling is that Citicorp wanted Quotron for more than its market value, for more than its current business. I think Citicorp had more strategic implications as a result of distributing their data over the Quotron network to Citicorp customers. But I see it as more of a strategic issue with Citicorp, whereas ADP was acquiring just to grow that business. I think you have to look at the acquisition much differently from the ADP side versus the Citicorp side.

Gaven: Yes. I would agree that it's not developing as rapidly. I think part of that is the customer's perception of what his requirements are, what he'd be willing to pay for it, and what are his advantages. I believe right now what we see in the industry is directionally, the brokerage community, especially the large firms, are saying, "I should now start doing something in the front office," and then they want to understand what the costs are, and then they ask themselves, "OK, how does this help me? How does this benefit me? What are my advantages?" And I think it's been a difficult time in quantifying those advantages. The other side of that coin is, most of the major firms, based on a personal observation, are now in a period of evaluating, testing, and setting the direction. So I think it's a matter of, "Are we a year away or two years away?" -- not "Are we five years away?"

Gaven: I think the large firms are more proactive because they probably have the staff and wherewithal to do it working with the vendors on the next solution. The smaller firms, I think, are probably waiting to see, directionally, if it feels right, what the large firms, the wire houses are doing, and then probably look to respond quickly on that. And let me say that not all regional firms, because we are working with certain regional firms that are very, very active, and maybe think that there might be an advantage of being right with the majors. I think it really depends on the characteristics of the firm. MTR: Is it your feeling, or ADP's philosophy, that every registered rep needs a set of these high-tech tools, productivity enhancements, and so forth, like the Intelligent work station?

Gaven: Well, said differently, I think every RR can use the capabilities that that might give him. How you package it, and what you put on it, that is a function of what you really need to do with it. So I think you can architect a system where you put intelligence on every RR's desk, and as long as it's functionally doing the things he wants, that's fantastic. Conversely, if you put dumb terminals, and you put all your power in the controller or the minicomputer in the office, that's another approach. Our feeling is, you let function drive what technical requirements are. And we believe that functionally, the rep wants tools to allow him to be more productive and the best way of probably delivering those tools would be putting in some intelligence. And when you say intelligence now, I guess that's kind of nebulous now too, because you could put half a meg of memory on a desk at almost the same price as it costs you to build a dumb terminal. So the cost curve on technology is being driven down so far that I think that's going to become a moot point over the next year or two. As an example, to build a dumb terminal today probably costs you $600-700. You look at some of these PC clones, you can get them for $800-900, so it almost becomes a moot point.

Gaven: I see that the big firms are looking at alternative ways to supply quotation information. Outside of Hutton and Merrill Lynch through Imnet, I don't see anyone actually putting it in place, actually going out and spending the dollars. My feeling is, that the firm wants the ability to somewhat control their own destiny, and we've attacked the problem a little differently, what we're doing is giving the function of the integration of market data, client information, and other financial information between the client's host, where the customer's data is resident, and our terminal systems which are located, obviously, at the branch side. And we're building tools that will run both on a customer's host as well as our terminal systems out in the field. It will allow him to control all of that process. We think that as long as he has control, then that's really all that he cares about. He doesn't necessarily care that he has a database of prices. We think we can do a good job there, and that it doesn't make sense for the brokerage community to do it.

Conversely, if we get the tools to be able to integrate market data with his own customer information, that's what he really wants to do, to be able to control that environment, and that's what we've built. I think we've taken the tack that this will give you the control, we think we can do the market data side, and almost in a way that we become part of the process and he's somewhat controlling that process. That approach gives us the best of both worlds. He doesn't have to build the infrastructure, and all of the cost implications that are required to build a ticker plant and a network, and terminal products, and keep them current, so we think that's the ultimate answer. So far with the major firms, the direction sounds right to them as well. I think we're getting some good acceptance on it, some of the preliminary discussions we've had with them.

Gaven: Yes. We basically will support both product lines. We think that the Convergent Technologies line, if you look at it from a price-performance standpoint, is a better alternative than the PC. Conversely, the PC has the name IBM on it, and that has certain strategic implications especially as it relates to the national firms. Many of the firms have given a strong, strong indication that they would like an IBM solution, and I think we're just hitting the market because that's what our customers are telling us. But certainly we will support both product lines. We're calling our product the ADP Producer on both the CT as well as the IBM, and really the difference is the hardware, in terms of the implementation of these applications.

Gaven: You're going to need a certain of those applications for the trading rooms, whether they be the equity traders or the option traders or the government traders. So you're going to need to provide long-term solutions. The other thing you're going to need to do is have the pricing information of the governments and the other debt securities. So I think you're going to have to get this information, and then you're going to have to do the analytics for the traders. So I think, long-term, that absolutely has to be done. What you're seeing today is niche players -- you're seeing Sharks -- that are really geared to the option traders and the basket traders, you're seeing Telerate as a marketplace for the government traders, and on the institutional side as well. Telerate has been very, very successful. Shark -- Wang -- is just starting in the trading environment and there's probably three or four others that are smaller -- Knight-Ridder has trading system -- and just designed for the trading room and more specifically for certain areas of the trading room.

Gaven: We're probably talking a 12-month time frame for our trading system.

Gaven: Probably not a lot. You're more into the trading room environment where the decisions are being made, and most of the activity is taking place. But on that score, we think it's important, it's another area we must be in because of the expansion of the brokerage firms outside of the U.S. -- and also the foreign investors coming into the U.S. To that extent, we formed a joint venture that was announced in June, with Mercantile House, and we are now bringing a product to market, by October first, in the U.K. principally in London, that basically will provide quotation information to the brokerage firms in the U.K. as well as the foreign branches of the U.S. firm. We think that that's important, we think it's a real growth area in this market place, and as you say, the twenty-four hour market trading is starting, and it will only get stronger over the next few years, so we think we have to be there. We think the U.K. is really important strategically, as the Far East will be strategic. At present we're busy doing more to promote joint ventures -- strategic positioning ventures -- as opposed to trying to go out as a total ADP entity. Because of the culture and regulations you need someone who knows that marketplace. So directionally that's where we're headed.

Gaven: I think certain firms feel that because of the competitive nature of Quotron being acquired by Citicorp, there's some real concern with some of the brokerage firms. And I think in terms of the level of importance, in certain firms it's extremely important, to others it's probably not significant. So I think that they're concerned about it just as they were concerned with the formation of Imnet because of the Merrill Lynch competitive nature of that transaction. So I think that that probably will give us an opportunity with some of the firms. I see that as a positive for ADP. As far as some of the other things we're seeing in the marketplace, we're seeing some of the niche players going out of business. I guess SPMI just closed down on August 1st. I think some of the smaller niche players are, because they don't have critical mass, I think we'll see those areas contracting, but I also see that there could be some other big players waiting on the sidelines.

Gaven: Well, I think Reuters, who basically have announced a product that they're going to bring to the marketplace. I think they want to come in in a big way in the United States. There's definitely a real competitive threat here. I guess that's the primary. You know, you've got a thing with Monchik-Weber being owned by McGraw-Hill. They're a potential. Those types of firms, I think, could see this as a real good marketplace for them to be in.

Gaven: I think what we're going to do is we're going to link the broker, we're going to make it possible for the brokerage firm to be able to integrate market data, customer data, and other financial databases, and then we're going to give them the tools to be able to control it, so that he's really delivering to his user community, which is the RR in our case -- or the trader for that matter -- the information that they really require and allow them to do it. I think we're the first and only firm that really will integrate the customer host, where we actually will write an application running on a customer host and our terminal products, we'll integrate the whole process. I guess in the last few SIA shows, most of the problems that the brokerage community thought that they were going to have in the long run was how to downline load data, how to really distribute the process of customer information, and how to integrate it with market data. I think we basically have that solution, I think we have that solved, and I think that'll be the one methodology that'll be able to allow them to really do distributed data processing. And I don't think anybody's really addressed that, and part of the reason that we're sensitive to it is that we've been in the traditional back office brokerage business for a long time, so we understand that process, and we understand how difficult it is, every night, to be able to distribute information over a large number of branches. I think part of the thing that we're differentiating is that we're just going to integrate everything, and I don't think anybody's looked at that.

Let me give you an example. Today Shearson probably has about 1,000 users using Reveal as a portfolio package as a broker's book. Of those, only a very small number actually integrate with their host by downloading information so that the RR does not have to do local updates. The reason that they do that, and they can only provide that service to a small number of users, is that it's very difficult to make sure it happens every night, and make sure that the files get updated, and make sure that the programs are run with all of the information, and I think that you'll find that most of the big firms that have tried it, that's where they fall short. So what happens is that they leave it up to the RR to keep their database current. Then you have the conflicts of the database that the RR has doesn't equal what the firm's books and records show. There are all kinds of compliance problems. We think we've solved that problem, and we think that that'll give the firm the ability and the comfort level to do something in distributed data processing.