Telling Fortunes

A San Francisco firm rolls the dice in the multibillion-dollar game of global finance

A mammoth computer monitor sits atop a pulpit in the conference room of KMV Corp.'s Montgomery Street office. Studying it, one imagines peering for the first time at a foretelling Mayan calendar, or prophetic passages of the Old Testament. In the manner of cuneiform glyphs in ancient clay, the squiggly lines on KMV's computer screen divine the heretofore hidden fates of nations; the collapse of economic belief systems; the straightest path to the financial promised land.

High-flying Internet companies, this electronic crystal ball says, might not be poised for the spectacular collapse Silicon Valley doomsayers predict. But some more traditional technology companies, like Motorola, are fraught with risk. The spectacular recovery in Southeast Asia, meanwhile, may be an apparition. Companies there stand within shouting distance of bankruptcy.

These insights are the work of a miraculous computer model designed by technicians at KMV, a financial engineering firm launched a decade ago by a trio of San Francisco entrepreneurs. With this model, KMV lays odds on which companies and banks -- and, by inference, the countries where they are headquartered -- will flourish, and which will perish. In its 10 years, KMV has emerged as one of the hottest comers in the field of numbers-based risk assessment.

In the age of globalization, there is no theology more powerful than risk assessment. Risk is God.

Pension fund managers, currency speculators, banks, insurance companies, corporate financial officers, bond traders, stockbrokers -- even firemen and nursery-school teachers -- are competing in a fevered contest to move billions of dollars around the world profitably. To do that, they need a way of measuring risk. More than ever, they are turning to computer models like KMV's.

In this modern financial age, a shift in investors' ideas about, say, the relative risk of Third World sovereign debt and Japanese stocks can cause billions of dollars to suddenly sweep across continents, sundering some nations while bringing riches to others. Brazil's unexpected decision to devalue its currency last week, for example, caused traders to move money away from Third World countries and other similarly risky investments, roiling financial markets around the globe.

The short-term winners in this frenetic game are those with nerves of steel, lightning reflexes, and a keen appreciation of mob psychology. But the real sharpies are the ones with the best insight into where on the globe -- at any given moment -- the greatest reward is possible with the least risk.

In this game, computer scientists like those at KMV are the gurus of a new age. To handicap the future, KMV crunches an amalgam of raw numbers: 150,000 corporation-years of industry data; cash flow; stock-price swings; and defaults. Sifted and massaged through KMV's computer instructions, this data produces a comforting, seemingly concrete number called a default-probability rating. This number is supposed to predict the future -- Company A is likely to fail; Company B is destined to sail -- so that investors can make decisions. By lumping together companies in a particular country, a nation's destiny can be similarly divined.

Banks have invested heavily in KMV crystal balls. Sixty-five of them pay up to $500,000 per year to get the company's default-probability ratings. These bankers consider KMV's model a financier's version of the Rosetta stone, deciphering the true meaning behind otherwise mysterious financial instruments. The Bank of Montreal, for example, has $2.5 billion worth of corporate debt funds crafted specifically with KMV's model in mind.

"We only buy companies with publicly traded equity in order to use the KMV model," says Barry Campbell, managing director for leveraged debt management at the Bank of Montreal. Using a computer program, instead of a huge staff of analysts and loan officers, Campbell is able to buy and sell the debt of around 400 companies with the help of only four people. It's simple, he says: "We look at KMV's estimated default-probability ratings, and sort things out based on the rating."

Other traders buy corporate bonds that KMV's model says are low-risk, and place bets against bonds chosen from the list of dicey issuers.

KMV Managing Director Tim Kasta won't reveal the names of customers who use the model for arbitrage -- as the exploitation of such price discrepancies is known. But the Bank of Montreal's Campbell says it's being done by more and more KMV customers.

"We're not doing that, but I know others who are," he says. "It's not completely dissimilar from the kind of arbitrage Long Term Capital was doing."

Long Term Capital, of course, is the Nobel Prize winner-packed investment fund that last fall nearly brought down the world financial system. It is the reigning poster child for the pitfalls of computer modeling.

Long Term Capital's most famous players were economists Robert Merton and Myron Scholes, who won the 1997 Nobel Prize for developing complicated mathematical formulas in 1973 that put a price on risk.

The fund specialized in bond arbitrage, using Merton and Scholes' math formulas to make $100 billion worth of bets on certain types of bonds and stocks. But the calculations were based on historical data. They didn't allow for what would happen when the world's financial markets melted down, as they did on Aug. 17, 1998, when Russia devalued the ruble and defaulted on some of its debt.

Risk trumped intellect. The fund collapsed.
To forestall a worldwide financial collapse, the U.S. government pressured New York banks to pony up $3.5 billion to keep the fund from going bankrupt.

Yet Merton and Scholes' 1973 calculations -- particularly those dealing with the wealth of information inherent in a company's stock price -- beat at the heart of computer models such as KMV's. Like Long Term Capital's calculations, KMV bases its projections on historical data, and on Merton and Scholes' ideas about handicapping risk.

Computer models shouldn't be blamed for putting the world at peril of financial disaster, says Cornell University derivatives professor Robert Jarrow. The math is sound, he says. "A crude analogy is fire. If you don't use it right, it burns things down," he says. "If you use it right, you can cook, heat, and provide comfort."

But it's hard to cook on a leaky gas stove without sparking a conflagration. And it's harder to predict the future when your computer model is threatened by everything from Boris Yeltsin's health and Hurricane Andrew, to the integrity of South Korean bankers and the vicissitudes of the Brazilian real.

The dizzying developments in international finance over the past decade have underscored the age-old idea of financial risk, which holds that the more unlikely your bet, the greater your potential reward. Tom Wolfe may have been facetious when he called his Bonfire of the Vanities characters "Masters of the Universe." But the fact is that a series of events during the past decade-and-a-half have colluded to turn a gaggle of financiers in New York, Paris, Amsterdam, Hong Kong, and San Francisco into the world's all-knowing, all-being masters of finance.

New York financiers lose confidence in the safety of Mexican treasury bills, spelling economic ruin for millions of Latin American families. Other traders get nervous about the Thai currency, and set the world's greatest industrial region back 10 years. And, perhaps, a San Francisco Internet stock fund manager's attack of cold feet will cause financial carnage to tear along the San Andreas fault.

"The way the world is now, a 24-year-old trader wakes up, he has a cup of latte, and he says, 'I want to destroy the Czech currency today,' " says Robert Theleen, head of Chinavest, a San Francisco group of investment funds. "And traders in Czechoslovakia will help him do it."

Traders bet against the currency, compelling the Czech government to raise interest rates. The standard of living for Czech families collapses, and members of the middle class move to the impoverished class. They will stay that way for a decade, until some other investor decides that the Czech Republic is a good place to park money, and pulls a few billion dollars out of, say, the Philippines, to bet on the Czech koruna.

Before making these decisions, the masters frequently turn to the new computer risk models. Banks spent millions of dollars during the 1980s buying high-powered computers, and hiring Ph.D.s in physics -- as opposed to business or accounting -- to crank out mile-long risk-analysis algorithms. These computers mix and match bonds and stocks in different markets, and in different countries. While humans make the decisions to move money, computer calculations form the basis for more and more of the assumptions these brokers trade on. Or, in perhaps easier-to-understand gambler-speak: It is new computer-risk models like KMV's that set the house rules.

International finance has been around for centuries, of course. Speculative bubbles followed by spectacular collapses are nothing new, either. Cycles of risk, reward, and catastrophe repeat every generation, as parents pass along greed, hubris, and an innate misapprehension of economic swings. Bookstores are full of tomes about the disastrous tulip craze in 17th-century Holland, and pretty much every financial column these days includes mention of 1929 or some other financially catastrophic date.

But the game really is different now. The failure of the fixed exchange-rate regime that governed the globe from the end of World War II to the mid-1970s, the collapse of the Berlin Wall in the 1980s, and the concurrent opening of economies in Latin America and Eastern Europe have created a different economic world.

Growing financial markets in Asia, Latin America, and Eastern Europe have given financiers more tables on which to play. The massive growth of mutual funds and 401(k)s has given them fantastically greater sums of money to bet. And the absolute lack of limits on where or when they can move this money has made the global economy more fragile than ever.

All decisions -- whether to buy government bonds denominated in Mexican pesos, sell debt pegged to the Thai baht, bet on the probability of movement in the Russian ruble, or buy shares of -- are based on individual acts of risk assessment. This can mean quantitative calculations, subjective assessments, ordinary hunches, spectacular leaps of faith, or, most commonly, a fanciful blend of all four.

But there's one thing that hasn't changed at all, even in the brave new world of financial-risk assessment. Nobody has proven able to predict financial collapses. The global economic disasters that used to roll around every generation -- and lately seem to happen every couple of years -- have always caught financiers unawares. Indeed, the masters seem most giddy in the moments before financial collapse.

In the words of economist John Kenneth Galbraith, "Recurrent speculative insanity and the associated financial deprivation and large devastation are, I am persuaded, inherent in the system."

In other words, the world survives only by the grace of a tribe of fools. This fact became apparent to me between 1993 and 1997, while I worked as a Mexico City-based staff writer for Dow Jones & Co.'s financial newswires. This period spanned the buildup to, and immediate aftermath of, the Mexican financial and economic collapse, a debacle famously described by International Monetary Fund Director Michel Camdessus as "the first financial crisis of the 21st century."

The speed with which investors fled the peso, and the devastating effects on other countries, had the appearance of an entirely new kind of event. It was the world's first such collapse triggered by vast international capital flows gushing into -- and out of -- an emerging economy ill-prepared to deal with them.

In Mexico, I saw the surreal euphoria Galbraith says has historically preceded financial collapse. I saw a trillion dollars' worth of U.S. pension-fund managers breakfast with Mexico's president, then leave with thoughts of weighting their retirement portfolios slightly southward. Four-hundred-dollar-a-night Mexico City hotels were jammed with Wall Street bankers seeking takers for their money.

When the collapse began with a devaluation of the Mexican peso just before Christmas 1994, the world's financiers were caught off guard -- despite a seemingly sophisticated system for analyzing the financial health of countries. Investors had based their risk-reward calculations -- that peso-denominated treasury bills, for example, were worth the 18 percent interest the bonds paid before the collapse -- on the absurd idea that Mexico had entered the First World. They had listened to the scores of Ivy League-educated Mexican government officials who traipsed through the conference rooms of Manhattan every day, repeating the words "modernization" and "free trade."

Wall Street bankers had sent fresh-out-of-college, $80,000-a-year "analysts" to Mexico City to write country-risk reports. They lunched with Mexico's MIT-trained finance minister, and its Harvard-trained president. They drank with Mexican officials at Latin America investment conferences in Cancun, Mexico City, Miami, and New York.

Then, they returned to their offices and placed huge bets. Later, banks and private investors lost billions when the currency collapsed, financiers adjusted their Mexico risk-return calculations, and the country's entire financial system, economy, and standard of living cratered.

And Mexico was only the first.
The 21st century's second international financial catastrophe came in Southeast Asia, in the summer of 1997. The field of risk analysis had advanced since the peso debacle, but it is difficult to find a financier who foretold Asia's demise.

"Timing a collapse is impossible to predict," says Mark Headley, manager of Matthews International Funds, which lost millions of dollars during the Asia collapse.

While it may be impossible to predict the timing, the basic shape of such disas-ters appears quite predictable when they do happen.

The scene before, during, and after the century's second financial disaster was just as surreal as its first. Euphoria was boundless. Korean companies had mammoth financial and business interests so convoluted as to hide billions of dollars in loan losses. Banks had loan portfolios growing at a rate of up to 60 percent a year. Asian skylines grew like bamboo.

Andrew Foster, a San Francisco investment research analyst, worked as a financial consultant in Thailand during 1996, and at a Malaysian bank when the collapse hit. Before the collapse, he remembers looking out of a Bangkok high-rise onto a forest of construction cranes.

"The crane was the national bird of Thailand, and I remember looking out over Bangkok seeing all these cranes. The country was growing so quickly, and banks were making so many loans that they ran out of loan officers, so they were actually using security guards as loan officers. We used to joke that this gave new meaning to the word loan 'officers.' "

The crisis officially began with the July 1997 collapse of the Thai baht. Within months, the shock waves were felt around the world. And by the end of last year, the International Finance Corp.'s emerging-market index had lost 45 percent in dollar terms during the year-and-a-half since the crisis began.

Nobody knows where, or when, the 21st century's third global financial crisis will begin. But some seers hint that it may occur in California. More precisely, in the San Francisco Financial District that funds much of America's miraculous technological economy.

While the possibility of economic troubles seems remote in the fabulously nouveau riche San Francisco Bay Area, here and there one hears murmurs of doom. The Association of Bay Area Governments plans a conference this month called "Will the Bubble Burst?" Financial columns have been warning of an inflated stock market for several years now. Internet fund executives talk of a coming "downdraft." Even U.S. Federal Reserve Chairman Alan Greenspan has warned of "excessive exuberance" in U.S. financial markets. Greenspan jiggled interest rates, the exuberance continued. This December, the prices of already stratospheric Internet stocks doubled. Some observers have gone from scared to terrified.

But inside the amazing universe of Internet stocks, people are much more sanguine.

"It's important to think of this as a game of musical chairs," explains Keith Benjamin, Internet stock analyst at the investment firm of Robertson Stephens. "You need to do your homework and know which chair you want to be sitting in when the music stops."

For now, the music is playing, and the accompanying dance of millions in some ways resembles Galbraith's pre-crisis euphoria. Like Mexico four years ago and Thailand the year before last, California is now a land of surreal wonders.

Consider companies like, which has never turned a cent of profit, but whose stock nearly tripled in value between November and mid-January. At more than $20 billion, Amazon is now worth about three times the value of the bank J.P. Morgan & Co.

Every Californian with a job and a computer seems to be buying Internet stocks. Technology workers quit their positions to play Frisbee for a month, convinced there will be another, better, job waiting for them when the mood arises. Closet-sized apartments are renting for a couple grand a month.

And the U.S. stock market, which has been a lot of fun for a lot of years now, became truly delightful during the past few months.

"Valuations don't make sense. The only thing I can think of is that people buy on the belief that they won't be the last one in. Yikes -- not exactly an investment strategy. It does however seem to be working for many of my friends ... hmm, maybe I should give it a try!" writes Paul Moody, a member of a tech-stock Internet bulletin board.

Jim McGinn, responding to Moody's bulletin board posting, is more dour: "The meltdown will occur. The gamble is when, and how much can be made in the meantime."

Perhaps. Or perhaps it's best to listen to the true believers. They're most comforting during times like these.

Brian Mutert is the self-assured director of Stratagem, a San Francisco Financial District matchmaker who arranges marriages between small technology start-up companies and bigger ones. After 10 years and 70 such deals, he has a suite of 19th-floor offices on Sansome Street and status as a technology-world sage. Mutert was a player in the very first software IPOs, and he was hip to the Web before it was a blip on the rest of the investment world's computer screens, he says. "We were online before online was cool," he explains over coffee.

Rather than a collapse, Mutert expects to see a shakeout of high-tech firms, with the Yahoo!s, Netscapes, and E-Bays of the world eventually trouncing smaller pretenders. Mutert's in his early 40s, a fact he thinks will come in handy during the coming Internet share-price downdraft.

"I clearly think there's going to be a lot of consolidation," he says. "This means increased opportunity for the young and the nimble. While the going's good, you've got to position yourself in the strongest way possible."

Emerick McDonald, who heads a technology stock fund for Amerindo Investment Advisors Inc., is even more enthusiastic. He worked for years as a computer engineer before getting into the fund managing trade, and he says the Internet is now only barely coming into its own. There is no reason the technology stock boom should abate, given this virtually unlimited potential, he says.

Now, around $100 billion worth of commerce is conducted on the Internet each year. There's little reason to think that shouldn't rise to $1 trillion before long, McDonald explains. If one considers the number of households in the world with enough income to buy a computer, then considers the amount of commerce they might be able to do over those computers, the possibilities boggle the mind. Given this projection, the doubling in value of profitless firms like isn't alarming -- it's tame. After all, wasn't Microsoft overvalued in earnings-multiple terms 10 years ago? And wouldn't a 1989 Microsoft investor have made out all right?

"That investment would have been justified even if the stock was five times as expensive, or even 10 times as expensive," McDonald says. "As soon as I saw what a computer could do, I saw that it could be huge. It's been a long time in coming, but the Internet is finally allowing the computer to live out its potential."

McDonald might be encouraged to know that the seers of the modern age, KMV Corp., agree with his optimistic assessment about the strength of technology stocks.

Like rating agencies such as Standard & Poor's and Moody's -- which have traditionally told financiers how likely companies, countries, and municipalities are to go belly up -- KMV takes a close look at data from the balance sheets before making its ratings.

But according to KMV, its model gets a better sense of a company's future by considering fluctuations in the company's stock price, and using a mathematical formula to predict the vicissitudes of the industry a particular company happens to be in. Car manufacturing, for instance, is more stable than making computers, and that is reflected in KMV's model.

As it watches day-to-day fluctuations in a company's stock price, KMV's computer keeps an eye on whether the business' value dips below its debts. Just like in ordinary households, the greater the cushion between net worth and debt, the less likely a company is to go bankrupt.

But stock price is more than a measure of total worth. KMV considers it a telling indicator of a company's fate.

That's because the price of a company's stock is distilled from every conceivable force that comes to bear on financial markets. The perceptions of thousands of investors, the best guesses of hundreds of analysts, history, current events, the weather, and national and global politics meld into a single, all-encompassing number -- the stock price. That's why the Dow Jones Industrial Average jerks on every tidbit of financial news.

In effect, the price put on a "financial instrument" -- as economists call anything worth money -- is the result of a massive, ongoing opinion poll. And unlike a grocer's opinion expressed as the price on a head of lettuce, stock price is an opinion about a company's future. It expresses thousands of people's feelings about a company's potential risk, and potential return.

The KMV model presumes that you can take all available information about a company, beam it through an algorithmic prism, and shine a focused ray of light into the future.

According to KMV's model, for example, high-flying is in much better financial health than traditional ratings agencies such as Standard & Poor's believe. According to KMV, Amazon stands only a quarter of 1 percent possibility of default. Standard & Poor's, the most-watched rating agency, puts the likelihood of default at around twice that. Moody's, the other big rating agency, says Amazon's chances are even worse.

Even KMV Managing Director Tim Kasta is candidly astounded at what the market is saying about Internet companies.

"Amazon's a wild one," says Kasta. "We believe in the market, so I guess this is right, but this is amazing."

Still, he says, one banker he knows bought stock awhile ago based on KMV's optimistic assessment -- and made a killing.

KMV customers owe their new crystal ball to entrepreneurs John McQuown and Oldrich Vasicek, a pair of former Wells Fargo bankers who got the idea in the early 1980s that they could apply the Merton and Scholes financial theories to bond markets.

McQuown and Oldrich created KMV in its current form 10 years ago, just when computer technology was reaching a point where off-the-shelf machines could handle the massive amounts of data that make the KMV model work. The model now monitors around 20,000 companies in North America, Asia, and Europe by running publicly available financial data through top-secret algorithms developed by Vasicek.

"To develop these algorithms, Vasicek solved mathematical equations that had never been solved before," says Kasta, the KMV managing director.

These algorithms allow KMV to use past changes in a company's stock price to come up with a number predicting future changes. A company whose stock has swung vigorously in the past has the potential to dip perilously close to bankruptcy in the future.

Banks and other customers pay between $100,000 and $500,000 per year for the right to view computer screens displaying the sum results of KMV's data crunching. With a few mouse clicks, these bankers can pull up numbers describing the likelihood that General Motors will default on a particular day. A few more clicks produce the combined default probability of numerous Hong Kong companies, illuminating the health of the country as a whole.

The amount of data required to launch the KMV project 10 years ago was mind-boggling, McQuown says.

"We started in February 1989, but it wasn't until April 1991 that we had it up and running," McQuown says. "We had to process a lot of information in order to compete in this world. And in the background was the development of the personal computer -- that was just when the 80386 chip was introduced, and it was fast enough, and with enough storage capacity, that around $5,800 would get you a reasonable amount of computing power. You could load up a database, and all of a sudden, you could distribute information in a consistent, effective way."

Like any new product, selling the model to banks was difficult at first. Sales grew as computer analysis pervaded more and more areas of banking. During the last three years, a number of computer-risk-analysis competitors have entered KMV's field. In 1997, JP Morgan launched a similar risk-analysis program, as did Credit Suisse Financial Products. The British consultancy McKinsey & Co. is marketing a package called CreditPortfolioView, and another consultancy, KPMG, has a risk-measurement package of its own. According to Risk magazine, at least two more such products should be on the market by the end of this year. And many banks have in-house models that help them analyze their investment portfolios.

But KMV remains atop this heap by virtue of the massive amount of information its computer models crunch, the proprietary mathematics contained in its estimates, and the impressive track record of some of its predictions.

KMV's model foresaw Apple computer's improving financial health at the beginning of 1997, months before Standard & Poor's decided to make new, more optimistic projections about the quality of Apple debt. And the KMV model saw Zenith Corp. heading for its May 1998 default a year-and-a-half before Standard & Poor's sensed trouble.

Some investors have found the model similarly useful for divining the health of entire countries. While investors were unanimously caught off guard by the 1997 Asian meltdown, KMV's model had been chronicling a steady deterioration in credit value in countries like South Korea since 1994. Standard & Poor's, meanwhile, didn't downgrade Korea's sovereign debt until after November 1997, when Korea went to the International Monetary Fund begging for a rescue package.

As more and more financiers turn to models like KMV's, the world just might become a safer place, argues KMV Chairman John McQuown.

The risky 21st century described by Michel Camdessus could be over nearly as soon as it begins. Rather than living in a brave new world fraught with financial risk, we would inhabit a planet where investors really knew what they were getting into when they bought, say, the debt of a Korean shipbuilder. Financial bubbles based on economic folly would be less likely, and investors might be forewarned of financial collapses.

"There used to be no commonly shared frame for estimating default risk," McQuown says. "With the enormous proliferation of companies around the world, people are ready to invest money anywhere. But they need to better understand risk."

They need to have been able to predict, for example, that on Thursday, Jan. 14, Francisco "Chico" Lopes would become Brazil's finance minister following the abrupt resignation of his predecessor, Gustavo Franco, throwing the world into financial turmoil.

They need to peer into the future and predict on what date investors will begin changing their minds about the Silicon Valley miracle. They need to know when living-room investors will start moving their mouses to the "sell" icon. They need to know ahead of time when, or if, our San Francisco dance of millions will end.

But they can't, really. All risk assessment, including computer models, can never be anything more than reading the entrails of history.

Predicting events in the 21st-century financial world by studying events of the 20th isn't possible. Capital moves about the globe like a schizophrenic jet stream. We can measure with absolute accuracy every degree of temperature and every drop of rain that falls for a thousand years, and still not say with 100 percent certainty what the weather will be tomorrow. Or when the sky will fall.

But history's all we've got, and it can be telling. Take KMV's history, for instance.

Eight years before they got KMV up and running in 1991, McQuown and Vasicek made an early stab at convincing bankers to buy into their ideas about quantifying default risk. They pooled $10 million from venture capitalists, and went door to door trying to convince banks to sell them especially risky corporate loans at a cut rate. The plan was to resell the loans to other banks with more stomach for risk.

But McQuown and Vasicek were ahead of their time. Banks weren't interested in selling and buying risky loans. There was trouble with the model, too. Affordable computers didn't exist that could process the amounts of data Vasicek and McQuown needed to evaluate the corporate loans. Instead the pair worked with boxes and boxes of data printouts, crunching numbers on companies individually, without a quick way to cross-reference or compile company data the way KMV's model does now.

As it turned out, their business model involved trying to buy loans nobody wanted to sell, analyzing the data with a model that didn't work as well as it needed to, then trying to sell loans nobody wanted to buy.

"Everything that could go wrong, did go wrong," McQuown recalls.
Within a couple of years, they had burned through their $10 million of venture capital.

They went bankrupt.
They defaulted.
They suffered a financial collapse.

But in 1989 they picked up the pieces, refined the model, and found there was more money to be made assaying risks for others than taking risks themselves.

Now billions of dollars are moving across global borders based, in small part, on the prognostications of a group of failed investors with the fanciest rainmaking machine on the market.

Who could have predicted that?

Show Pages
My Voice Nation Help
Sort: Newest | Oldest
©2014 SF Weekly, LP, All rights reserved.