$102 billion. The number tends to get attention, even in Washington where, the late Senator Everett Dirkson famously commented, “A billion here, and a billion there, and pretty soon you’re talking about real money.”

$102.6 billion (you wouldn’t want to entirely forget another $600 million if you were holding the paper) is the size of the debt Argentina decided to restructure through a series of debt swaps. As The Economist reported, the new bonds were worth about 35 cents on the dollar, although it is hard to get too precise since the swap covers 152 varieties of debt in six currencies governed by eight jurisdictions. And, to prevent life for investors from becoming too simple, all the old bonds can be swapped for three new issues.

An example of a mortgage paydown model that has been converted into an interactive Web page using XL2Web.

Investors trying to figure out what to do through the restructuring may need some tools. One large investment bank, a leader in emerging markets, offers its clients very sophisticated Excel models on its Web site by using a Web publishing tool called XL2Web. It draws on the bank’s expertise, its huge quantities of emerging market historical data and indices, plus the research the investment bank’s analysts have conducted over the years and stored in an internal analytics library.

One executive at the bank says the challenge is to distribute information internally to analysts and also make models available to clients without jeopardizing the bank’s intellectual capital that went into developing them. Using XL2Web, clients can plug their data into the bank’s models while the bank protects its intellectual property within its own firewalls.

“With XL2Web, the logic of the spreadsheet itself, as well as the analytics it calls, are all locked up behind the firewall on the server,” explained Jonathan Rochelle, the company’s founder. “An investment bank can integrate XL2Web with several analytic function libraries which contain several thousand functions. In the past, that would prohibit sharing the spreadsheet-based models with clients, because they were proprietary libraries which could not be distributed to clients,” Rochelle explained.

Jed Evans, head of emerging markets analytics at Deutsche, said the bank has a vast analytics library and historical database add-ons that it uses internally. But before XL2Web, development was performed solely in Java. However, the development expense of Java was reserved for large projects that would be accessed by hundreds of clients.

“With XL2Web, we can inexpensively deliver a highly specialized model that may only be accessed by a handful of clients,” he said. “We can link in all the appropriate libraries, convert our models through the XL2Web conversion engine, and make them available on the Web. If we need to modify a model, we can do it in one place and know all our clients have the latest versions, which wasn’t true when we were mailing them out or using downloads.”

XL2Web doesn’t have all the charts of Excel, he added. “But we find if we need one, we call them up and in a week or so we get it.”

Because analysts can do all the work, turnaround is fast.

“We can develop new Web applications at a very low cost and decrease our development cycle so we can give models to our clients in a fraction of the time as having our IT department do it,” he said.

At the other investment bank, analysts in research build models using an internal analytics library and data and models are distributed to clients through Web sites and market data vendors, the bank’s executive explained.

“We have some very complex calculators, and traditional development just takes too long, especially for a fast-moving business challenge like the Argentinean debt restructuring,” said the bank executive.

The bank did a two-month pilot that included “some models that border on the bizarre, and models with near real-time screen calculators and trade analyzers,” said the bank executive. “XL2Web let us turn them into Web tools.”

The bank has added source code management and security management to XL2Web so administration of the tool meets internal compliance requirements.

“I am now able to regulate who has access to the model, who can update it. I can make sure the spreadsheet is now captured in a source code control system, which is difficult to do in a front-office spreadsheet. Now, before any model goes out to the Web, we have it archived and locked down,” the executive said.

Sell-side research has been a big user of XL2Web, said Rochelle. Increasingly, buyers want to see some quantified research, not just opinions, in the reports they receive.

“The buy-side is looking for interactive quantitative research – give me a model and let me plug in some numbers,” explained Rochelle. In the past, the sell-side had a reputation for not making its best models available, because firms couldn’t protect their intellectual property.

“Now they can use the tool to create very quickly an Excel model that the buy-side can see, but the investment banks can control what portions their clients see, and the propriety infrastructure remains within the bank. The banks can also track the usage of the model,” Rochelle said. One recent client took XL2Web and in about three hours integrated 1,500 functions across four different libraries.

With XL2Web, firms can make skills such as credit derivative pricing available as a Web service that they can charge for.

“This opens the door to quick and very highly functional development of Web-based tools. Deutsche Bank has seen up to a 90 percent improvement in time to market – something that would have taken 10 months takes one month,” Rochelle said.

Using XL2Web means firms can provide the functionality through Excel, without relying on the IT shop – something that might make some in-house IT teams worry.

“In some firms, the IT team is defensive because this competes with Java and .NET programmers, but other IT teams see that this creates strong architectural capability that allows the desk to get things done efficiently,” Rochelle said

www.db.com

www.xl2web.com