Paul Caplin, founder of Caplin Systems, discusses in which direction he sees the trading world moving, with special regard to data networks. Interview by Parveen Bansal.
As activity picks up in the financial services sector, there is renewed demand to boost order flow while keeping costs down. One area in which the waves of change are lapping is market data.
Paul Caplin, founder and CEO of Caplin Systems, provider of real-time web technology, says: “There is a rapid change in the online connectivity between the buy side and sell side. In the late 1990s many financial institutions were experimenting with web connectivity, especially the exchanges and brokers who got caught up in the euphoria of the web.”
He suggests that many of them invested in the web in the hope of improving their share prices and did not think about the business case. “Many projects that were started were mainly speculative rather than being based on business logic.
Going to extremes
“After the dotcom bubble, they then went from one extreme of a lot of investment to the other, a sudden dramatic withdrawal of investment from anything with an ‘e’ on it.”
This slow-down has lasted two and a half years and only recently has picked up again since the general recovery of the wholesale and stock markets began.
“So now they [the brokers] are racing to catch up because they realised there are real benefits to be had by modernising communication. The oldfashioned ways are slow and expensive, especially verbal communication as it is not susceptible to automation.” Mr Caplin sees a continuous move to more electronic forms of communication because they are cheaper and better.
The traditional methods
The main issue is how the sell side gets its information in real-time to the desktops of its clients. Until now there were three chief ways of doing this electronically: the first is via a private network of the big vendors such as Reuters and Bloomberg, which have screens on all desktops.
However, with each screen costing $1000 per month or more, per user, this is not the most cost-efficient option. More significantly, says Mr Caplin, the sell side has little control over how the data is displayed. “The data display is often rigid and without branding, and there is no mechanism for feedback, so you can’t tell if anyone has looked at the price or not, you can’t send out different prices depending on relationships, so institutions can only send out an indicative price and not an executable price.”
Another fault is that the data is mixed with competitors’ prices. Mr Caplin notes: “Ideally, the buy and sell sides want a dialogue, not a broadcast.”
The second way to communicate pricing data is to put proprietary applications on customers’ desktops via a private network, but this too is expensive to implement and maintain, and organisations don’t always want different dozens of different applications on their desktops.
The third way, traditionally, is to provide a transactional connection to an order management system (OMS). This is commonly via a FIX (financial information exchange) network. “But,” says Mr Caplin, “this is also limited in its functionality, as it was mainly intended for trade order routing, and it does not work well with market information. It does not have a sophisticated ability to manage orders and is mainly used in the equities trading world with slow take-up as yet in fixed income and others.”
A better approach
Mr Caplin suggests an alternative method that is more efficient, secure and reliable: web delivery. “It is much cheaper and more flexible and becoming widely accepted as the best way.
“I believe the internet is the only WAN [wide area network] the world will need in the future,” says Mr Caplin, “and that owning and managing alternative communication channels outside the LAN [local area network] is pointless because this is not only expensive, but also not so reliable.”
Mr Caplin promptly refers to the fact that the internet was the only communication network that was still working in Lower Manhattan immediately after the events of 9/11.
According to Mr Caplin, laying private networks is out of the question. “If you want to talk to someone in another location, no matter how urgent, you use the public telephone network. You never lay a new cable. If you want to make it a private conversation, you may use a scrambler over the public telephone network or you may lease a line if the communication is frequent. But you won’t build your own telephone network.”
Mr Caplin believes the managed IP (internet protocol) networks offered by firms such as Radianz and Savvis are simply intermediate measures that are becoming increasingly unnecessary as the internet develops. “The internet is now mature enough to be used for critical communications without the need for private networks,” he explains.
Fear of the unknown
There is a clear trend of migration toward the internet. The question is how fast this transition can happen as institutions are often held back “by structural and emotional inertia, as well as a fear of the new”.
Mr Caplin names Instinet, JP Morgan Chase and UBS as some of the Caplin clients that are already using the internet routinely for outside communication of data and mission-critical transactions.
Mr Caplin sees internet use going deeper into the organisational structure from broad, high-level acceptance by the large web of users. When questioned on the feasibility of this, given security concerns about the internet, he says: “Actually, this is more of a superstition, as the internet is as secure and reliable as you want it to be.”
Where do we go from here?
The argument for migration to the internet is clearly one of cost and convenience, just as it has been in the retail banking sector.
“The internet has been slow at coming to the wholesale banking arena for four reasons: they are risk-averse; they are wedded to their big investments in complex legacy technology [in terms of client communication]; the marketplace is relatively small [around one million users globally] so that the advantage of the web is not so spectacular; and the final critical reason is the absence of application technologies to deliver real-time information over the web – because the internet in its generic form does not support real-time information,” he says.
Mr Caplin continues: “In the short term, there is an increasing realisation that anyone providing trading services direct to end users needs to offer a web-based individual connection to those services. This is because the cost savings, portability and usability benefits associated with web-based access are being taken increasingly for granted by the buy side.”
This migration is noticeably led by the foreign exchange market, in which there are no antiquated exchanges to hinder the progress and which is one of the most global and liquid markets. The fixed-income market is also moving towards this model, although it is relevant only where there is no need for aggregation of multiple dealers – in other words, for the less liquid securities.
Idea is catching on
Mr Caplin says: “The equities market is the slowest to adopt a direct connectivity model because most clients have an OMS connected via FIX. But more and more firms are offering a web-based trading interface for equities because this can provide a level of visibility and control to end users that is simply not available through their own OMSs.”
The greatest challenge is unfamiliarity with the internet at this deeper level. “A major hurdle is that organisations often try to build their own real-time streaming technology, not realising how complicated it is to build an efficient, secure, scalable, web-messaging system that connects and delivers reliably,” he says.
Mr Caplin believes that in the long term, the web will become the primary aggregation medium for market data because it is efficient and cheaper than private networks. The only thing stopping the web is organisational inertia resulting from massive private investment in legacy technology. But the web’s clear advantages make it likely to become a preferred channel – it is, ultimately, inevitable.
‘Democratisation of the markets’
The ultimate effect of the broker-dealer world adopting the internet as its communication network will be the “democratisation of the capital markets”, says Mr Caplin. “Once the web infiltrates the capital markets to a certain extent, it will have a huge democratisation effect, so that, capital availability aside, there will be no need to work for a specific organisation to participate.”
Mr Caplin also foresees a convergence of retail and wholesale access to market data. “Retail market data systems will be become more sophisticated and professional systems more web-based, and the two will converge,” he predicts.
It is clear that complacency is foolish. One only needs to remember how the fax killed the telex machine to see that private networks are short-lived and cheaper public networks are the way forward.
