Like most everyone these days Dr. Edward Yardeni, Deutsche Morgan Grenfells well-known chief economist, has been trying to get a fix on the year 2000. Specifically, hes looking into what impact the computer problem, a.k.a., the millennium bug, is going to have on the financial services industry. Among the places his research has apparently taken him are movie theaters. Speaking in mid-April to a gathering of top Wall Street analysts in New York, Yardeni couldnt help but offer a few pop-culture driven metaphors. Our roaring economy, he told the group, is a lot like the Titanic. And the iceberg lurking off in the not-so-distant waters may just be the year 2000.
Although not everyone shares Yardenis pessimistic view that worldwide systems problems at the turn of the century are going to usher in a period of financial chaos, it is generally agreed that businesses, and financial services firms in particular, are going to be in for a bit of a rough ride if all the systems linked throughout the industry are not made fully Y2K compliant.
How can that be achieved? Yardeni asked, continuing the celluloid theme. We need the Full Monty on the year 2000, he said. Everyones got to be honest and face up to whats really going on.
Although, hopefully, the industry can do that without literally disrobing to the disco strains of Donna Summer, the task of finding and fixing all the potential Y2K problems in time continues to be a daunting task. The General Accounting Office (GAO) appears to be in agreement with Yardeni that the industry has not been as revealing as it could be on the specifics of the problems and their solutions.
John Dingell (D-Mich.), the ranking minority member of the House Commerce Committee, became alarmed earlier in the year after reading Securities and Exchange Commission (SEC) reports on the industrys preparedness. Those reports, Dingell said, provided an overview of the problem but not enough of the specific detailed information that Congress will require to assess progress. Rep. Dingell then asked the GAO, the investigative arm of Congress, to look into the matter. In mid-March, the GAO filed a report asking the SEC to begin providing it with reports spotlighting systems critical to the continued functioning of the markets, the progress being made toward moving those systems through each phase of year 2000 compliance, the time frames required to complete each phase, the efforts made to address systems that are behind schedule, and contingency plans for systems that may not be ready in time.
Further, the GAO requested the SEC provide Congress with this information on a quarterly basis. The SEC responded with a letter from Richard Lindsey, director of the division of market regulation. Lindsey wrote that the SEC agreed with the need for more frequent reports to Congress, but claimed the SECs reporting could be accomplished through informal, rather than scheduled quarterly briefings. In his comments to the GAO, Lindsey said that due to the size of the securities industry, obtaining specific quantitative information on both mission critical and non-mission critical systems for these entities would be beyond the capacity of the SEC staff resources.
Lindsey added that, nevertheless, the SEC would adopt many of the GAO suggestions and would move toward providing more detailed and quantitative reports to Congress. He also said that contingency planning would become a priority of the SECs efforts throughout 1998.
Privately, some SEC staffers complain that the GAO is engaging in a bit of grandstanding. According to one SEC staffer, If you look back, youll find it was the securities industry that called everyones attention to the millennium problem in the first place.
And yet the SEC hasnt been able to provide much in the way of specifics about industry systems because it doesnt have the information to share.
Although the industry is working very hard on this problem, many firms consider the work they are doing on their own systems to be proprietary, says an SEC staffer. I doubt if Merrill really wants to share the specifics of their technology problems with the world.
Chain Crash? The problem as most everyone should know by now is that computers use two, rather than four digits to record years. As a result, on midnight, Jan. 1, 2000, systems that have not been fixed will assume that 00 represents 1900, and that will likely trigger a wave of computer confusion and crashes.
With industry systems linked throughout the world to exchanges, brokerages, transfer agents and countless vendors, a failure of one system could clog even the most Y2K compliant systems.
Whats at risk? Pretty much any transaction or computation that is date sensitive--just about everything. For example, without proper coding computers wont accept settlement dates for the year 2000, which means some systems have to be ready by 1999 to handle futures and options transactions. All computation models, such as those used for risk analysis, hedging, derivatives pricing and trading could fail. Calculations for interest payments for bonds maturing after 2000 would be equally affected.
Despite the GAOs call for more disclosure and the doomsday scenarios put forth by some economists, the industry is not exactly dogging it on its Y2K labors. The Securities Industry Association (SIA) has been pushing on several fronts to create awareness of the issues among its members and to institute a test plan and schedule for industry systems. The SIA also has been pushing everyone from vendors to the local telephone companies to get going on millennium preparations. The trade group also proposed a trading holiday on Dec. 31, 1999, to serve as a test day. Even after the regulators, including the Federal Reserve, shot down the idea, the SIA put out a press release saying it respected the Feds decision. The release didnt say, however, that the test day wasnt needed.
Although clearly the industry is moving forward on the problem and the SEC has put out for comment tougher reporting standards (it still is not recommending quarterly reporting), the big fear remains that no one knows for sure exactly what is going to happen when the clock strikes midnight.
The problem is that we are facing a situation that has no precedent. No one has ever faced anything quite like this before, says Michael Gazala of Boston-based Forrester Research, a financial services research firm.
The task already is having a significant impact on the industry and its day-to-day operations. The intense need to make systems compliant comes at a time when firms also are coping with the equally time-sensitive introduction of the new euro currency, and the upgrades necessary to be compatible with the National Securities Clearing Corp.s new ACATs account transfer system (see NSCC Gears Up for New ACATs System, June 98 Oddlots, Page 25).
All told, the result is a major drain on firms technology resources. According to a study conducted by the Towers Group for the SIA, Year 2000 projects are overwhelming firms and redirecting such initiatives as the move to T-1, decimalization, and even introductions of new broker workstations and Internet capabilities.
Most other projects are grinding to a halt because of these Year 2000 projects, says Lawrence Tabb, senior research analyst with the Tower Group and the studys author. Never before have so many firms been required to perform such extensive modification to their technologies.
Ultimately, despite a carefully planned test schedule, no one will know until the curtain rises how well theyve really done. Tic ... tic ... tic.
An SIA-sponsored report claims that the deadlines of three large technology projects--Y2K, the euro and the ACATs upgrade--are seriously stressing the industrys budgets and manpower.
When undertaken simultaneously, the difficulty of implementation increases dramatically, says Lawrence Tabb, senior research analyst with the Tower Group and the studys author.
Although the Year 2000 has garnered the most attention, the conversion of systems to handle the euro, the currency of the new European Monetary Union, is in some ways more complicated than performing Year 2000 upgrades. The reason? System planners simply dont know how to plan for the euro.
A basic rule of technology says that you need to know what you are programming for before you can program. When it comes to the euro, little is clear, says Nick Jones, research director at The Gartner Group in London, an international consulting firm that has done several studies on the impact of the European Monetary Union on business and technology.
According to Jones, the millennium bug and the implementation of the euro present a textbook example of how two seemingly similar system-based challenges can be so different.
The Year 2000 presents a clear, defined problem that involves no strategic business decisions--fix computers to switch from 99 to 00, he says. The euro is the exact opposite; programmers need to know what the front office wants before they can program. And the front offices just arent in a position to answer that yet.
On the surface, euro programming wouldnt seem much more complicated than upgrading the functionality of multicurrency trading systems. But systems must be taught to triangulate--move from one currency to the euro to a third, and properly round off figures in the new currency. Computers also must be taught to read the euro symbol, and spreadsheet formats need to be overhauled, along with invoices, contracts and statements. Even archived trading records might need to be converted to the euro for accurate historical comparisons. Then all the new systems must be installed and tested for compatibility.
Adding further complication to the process, Jones says, is the fluid timetable surrounding these changes. Every firm will face different demands from each of their overseas trading partners.
The current framework for the implementation of the European Monetary Union calls for the phasing in of the euro between 1999 and 2002. Most European nations are hoping to have everything in place early on. If so, they may demand the same of U.S.-based partners.
Non-European firms may feel insulated from euro challenges, but In this era of globalization, all firms are going to have to prove their ability to transact euro-based business if they want to retain clients and trading partners, says Malcolm Stirling, a director of KPMG Consulting in London. But I doubt whether even 10% of firms have a dedicated budget for euro IT adaptation.
When a Congressional committee made noises a year ago about forcing the nations stock exchanges to price in decimals, several industry officials testified that such a mandate might, among other problems, interfere with the industrys push to fix its Year 2000 problem.
Nevertheless, under congressional pressure, the NYSE in June 1997 agreed to stop the 200-year-old tradition of pricing in fractions and adopt dollars and cents pricing as early as June 1998. At the same time, NASD chief Frank Zarb ordered a study on the issue and promised to take a position by Labor Day, 1997. Zarb predicted that the NASD could be ready as soon as the end of 1998 to make the conversion.
But by October, those plans were changed. Zarb and SEC Chairman Arthur Levitt Jr. told a House panel they wanted to delay until after the year 2000 any move to decimal pricing. Converting to decimals would distract from the industrys efforts to ready its computer systems for the new millennium, they said.
Lawmakers then asked the Government Accounting Office (GAO) to look into the issue. Last May, the GAO finished its report and concluded that Year 2000 efforts are too important to the continued functioning of the industry to risk failure by attempting to implement decimal trading before the Year 2000 effort is completed.
The GAO relied on interviews with industry officials.
Meanwhile, a working group of exchange representatives has proposed that testing of decimal pricing be done beginning in January 2000, with full implementation in September of that year. The SEC is considering the idea and has said it wants implementation by June 2000.
But the industry, which has long opposed moving to decimals, may still have an out: The working group says if major processing problems are encountered, exchanges should be allowed to revert back to fractions. Message traffic over the markets quote and linkage systems is expected to increase if minimum trading increments are narrowed further.
The latest decimal delay is one in a series dating back several years. In late 1993, in releasing proposed rules on payment-for-order flow disclosure, the SEC noted that one-eighth trading increments might be one reason dealers could pay for orders. In its January 1994 Market 2000 report, the SECs Division of Market Regulation opined that decimal pricing is preferable and may be inevitable. ... Later that year, Levitt told reporters at an industry conference that he had a built-in impatience in seeing decimal pricing and other recommendations in the report implemented.
Nevertheless, the issue went nowhere until then SEC Commissioner Steven Wallman issued a public challenge in the fall of 1996 for the industry to convert to decimals before the year 2000. Wallman was instrumental in bringing the issue to Congress attention.