Goldman Sachs and the $580 Million Black Hole
THE business deal from hell began to crumble even before the Champagne corks were popped.
The deal, the $580 million sale of a highflying technology company, Dragon Systems, had just been approved by its board and congratulations were being exchanged. But even then, at that moment of celebration, there was a sense that something was amiss.
The chief executive of Dragon had received a congratulatory bottle from the investment bankers representing the acquiring company, a Belgian competitor called Lernout & Hauspie. But he hadn’t heard from Dragon’s own bankers at Goldman Sachs .
“I still have not received anything from Goldman,” the executive wrote in an e-mail to the other bank. “Do they know something I should know?”
More than a decade later, that question is still reverberating in a brutal legal battle between Goldman and the founders of Dragon Systems — along with a host of other questions that go to the heart of how financial giants like Goldman operate and what exactly they owe their clients.
James and Janet Baker spent nearly two decades building Dragon, a voice technology company, into a successful, multimillion-dollar enterprise. It was, they say, their “third child.” So in late 1999, when offers to buy Dragon began rolling in, the couple made what seemed a smart decision: they turned to Goldman Sachs for advice. And why not? Goldman, after all, was the leading dealmaker on Wall Street. The Bakers wanted the best.
This, of course, was before the scandals of the subprime mortgage era. It was before the bailouts, before Occupy Wall Street, before ordinary Americans began complaining about “banksters” and “muppets” and “the vampire squid.” In short, before Goldman Sachs became, for many, synonymous with Wall Street greed.
And yet, even today what happened next to the Bakers seems remarkable. With Goldman Sachs on the job, the corporate takeover of Dragon Systems in an all-stock deal went terribly wrong. Goldman collected millions of dollars in fees — and the Bakers lost everything when Lernout & Hauspie was revealed to be a spectacular fraud. L.& H. had been founded by Jo Lernout and Pol Hauspie, who had once been hailed as stars of the 1990s tech boom. Only later did the Bakers learn that Goldman Sachs itself had at one point considered investing in L.& H. but had walked away after some digging into the company.
This being Wall Street, a lot of money is now at stake. In federal court in Boston, the Bakers are demanding damages, including interest and legal fees, that could top $1 billion. That figure is nearly twice what Goldman paid to settle claims that it misled investors about subprime mortgage investments before the financial crisis of 2008.
This account is based on a trove of legal filings — e-mails, motions and roughly 30 depositions, more than 8,000 pages of sworn testimony in all — that open a rare window on Goldman Sachs and the mystique that surrounds it.
JAMES AND JANET BAKER, now in their 60s, are computer speech revolutionaries. Both Ph.D.’s, they became interested in voice-recognition technology in the 1970s, back when a personal assistant like Apple’s Siri would have seemed more science fiction than scientific fact.
They are widely credited with advancing speech technology far faster than anyone thought possible, primarily because of an epiphany Mr. Baker had while doing his doctorate research. He figured out that speech recognition could, in essence, be reduced to math. You didn’t have to teach a computer to recognize accents or dialects, Mr. Baker realized — you just had to calculate the mathematical probability of one sound following another. His algorithms proved remarkably accurate and eventually became the industry standard. (Want to know more? Ask Siri.)
The Bakers founded Dragon Systems in 1982 in an old Victorian house in West Newton, Mass. At that time, despite having two school-age children and a big mortgage, they were determined to take no venture capital and to finance the company’s growth with its own revenue — once they had a product. They figured they could last 18 months, maybe 24.
Their first product was a software program for a British-made PC called the Apricot that let users open files and run programs by voice command. Then came DragonDictate, a groundbreaking speech-to-text system for dictation that still required the speaker to pause. Between. Every. Word.
For years, the Bakers pressed on, convinced that they were on track to create a program that would recognize continuous speech.
To do that, however, they eventually decided that they needed more capital. While Mr. Baker worked on the technology, Ms. Baker brokered a deal with Seagate Technology , the disk drive manufacturer. Seagate bought 25 percent of Dragon for $20 million. Then, in 1997, Dragon introduced Dragon NaturallySpeaking, a program that recognized more words than could be found in a standard collegiate dictionary. It was available in six languages and could handle normal speech, even sentences with words that sound alike, such as, “Please write a letter right now to Mrs. Wright. Tell her that two is too many to buy.”
By this time, I.B.M. and others had piled into the voice technology market, too. As the Nasdaq market raced toward record highs, the Bakers considered taking Dragon public. But in 1999, several companies — including Sony and Intel — expressed interest in buying into Dragon. Finally, unsolicited buyout offers began to arrive. One came from Visteon, a subsidiary of Ford Motor. Another arrived from Lernout & Hauspie.
TO the uninitiated, the mystique of Goldman Sachs may be hard to fathom. Known for what might politely be called ruthless professionalism, Goldman, the thinking goes, is smarter and more plugged in than just about any other investment bank. In the late 1990s, under Henry M. Paulson Jr., who later became the Treasury secretary and orchestrated the Wall Street bailouts, Goldman was the alpha dog in the lucrative game of mergers and acquisitions.
So it was that in December 1999, the Bakers, in over their heads when it came to M.& A., signed a five-page engagement letter drafted by Goldman. In it, Goldman pledged to provide “financial advice and assistance in connection with this potential transaction, which may include performing valuation analyses, searching for a purchaser acceptable to you, coordinating visits of potential purchasers, and assisting you in negotiating the financial aspects of the transaction.”
The 'Goldman Four'
To the Dragon deal, Goldman assigned four bankers, two in their 20s and one in his early 30s. That wasn’t unusual. Although Dragon Systems was worth everything to the Bakers, the company — with $70 million in revenue and 400 employees — was small beer on Wall Street. Dragon agreed to pay Goldman a flat fee of $5 million, less than some Goldman bankers were pulling down.
But who, if anyone, supervised these bankers — later called “the Goldman Four” in court documents — remains something of a mystery. One of the four, the most senior, testified later that their supervisor was Gene T. Sykes, a Goldman partner who at the time specialized in technology and who this year was promoted to head of M.& A. at the firm, one of the most powerful jobs on Wall Street. In a deposition, Mr. Sykes disavowed any involvement.
Most of the Goldman Four didn’t stay long at the bank. Richard Wayner, who was 32 when the Dragon deal was cut, struck out on his own in 2002 and eventually landed at the Keffi Group, an investment firm. T. Otey Smith, then 21, left Goldman in 2000 and now works for RLJ Equity Partners. Alexander Berzofsky, then 25, left Goldman at about the same time and is now a managing director at Warburg Pincus, the big private investment company. Chris Fine, then 42, was a Goldman information technology specialist who was enlisted on the deal and is still with Goldman. (None of the four agreed to be interviewed for this article.)
Before the engagement letter was signed in late 1999, Goldman sent Dragon a memo indicating that its first steps would include beginning to conduct due diligence — Wall Street-speak for kicking the tires — on L.& H. The memo included specific areas of concern, including L.& H.’s sources of revenue, its major customers, its license agreements and royalty agreements, its expected growth, its partnerships and its financial statements.
THAT December, Mr. Wayner of Goldman accompanied Ms. Baker and Dragon’s chief financial officer, Ellen Chamberlain, on a trip to Belgium to meet L.& H. executives. For the trip, another of the Goldman Four, Mr. Berzofsky, prepared a list of due diligence questions. Goldman also prepared a “merger analysis” that predicted the companies’ combined sales per share, earnings per share and total debt under three acquisition scenarios: all cash, all stock and half and half.
In its initial offer, L.& H. proposed paying $580 million, half in cash and half in stock. But the Bakers weren’t sure. News reports had questioned L.& H.’s revenue, particularly in fast-growing Asian markets, as well as some of the company’s licensing deals. Mr. Baker felt that L.& H. had inferior voice technology. But then, he reasoned, if L.& H. could generate so much revenue with lesser technology, imagine what it could do with Dragon.
By mid-February 2000, Ms. Chamberlain had sent an angry memo to Goldman. It urged the bank to move faster in its analysis of L.& H. Talks with the other companies had gone nowhere, and she expected Goldman to “drive” the due diligence process. Mr. Wayner testified later that the bank’s reaction to that memo was “to do as our client asked and to revisit all of our analyses.”
But on Feb. 29, Dragon received an odd memo from Goldman. It wasn’t addressed to anyone in particular at Dragon, and it wasn’t signed by anyone at Goldman. The Goldman Four testified later that they had no idea who had sent it. But the memo referred to many of the same due diligence issues that Ms. Chamberlain raised. The memo asserted, however, that Dragon’s accounting firm, Arthur Andersen, should do the work, not Goldman.
The memo shocked Ms. Chamberlain. She had come to Dragon from Seagate, where she had participated in similar deals. She believed that this sort of thing was generally done by investment bankers, not by accountants. But the moment passed. No one at Dragon or Goldman brought up the mystery memo again — at least not until the lawsuits began flying. (The Bakers also filed suit against other participants in the transaction.)
THE Dragon executives thought that Goldman was taking a hard look at L.& H. After all, Dragon was paying Goldman $5 million for its advice. If Goldman wasn’t conducting due diligence, what was it doing?
“They put items on and off the due diligence list,” Ms. Baker later testified. “We discussed the issues at — at basically every meeting that we were at, and we were meeting often in person or by phone, typically, several times a week in this time frame — sometimes multiple times a day, as we’ve seen. And so they knew what everybody was doing. And they were, they were directing it.”
One of the tasks was a conference call that Mr. Wayner arranged, at Ms. Baker’s request, between Dragon and Charles Elliott, a Goldman analyst in London. Dragon was wondering why L.& H.’s share price had been gyrating wildly. Mr. Wayner told Ms. Baker, he later testified, that Mr. Elliott was following L.& H.’s stock and was up to date on its fluctuations. And Mr. Elliott assured Ms. Baker that investors were worried about the market in general, rather than L.& H. in particular. He also said he expected the stock price of the combined companies to rise substantially once a merger was struck.
Years later, in his deposition, Mr. Elliott told a more complete story. He acknowledged that he actually had not been following L.& H.; that had been the responsibility of another Goldman analyst who had left the firm shortly after the Bakers retained Goldman. After the other analyst left, Mr. Elliott testified, Goldman terminated its coverage of L.& H. No one told the Bakers that Goldman was no longer covering the company they were about to bet their futures on.
Mr. Elliott also testified that he was unaware of press reports at the time that suggested L.& H. was claiming huge revenue gains in Asia. If he had been aware, he said, he would have been “very skeptical” of those gains, given the challenges that Asian languages present for speech recognition. He also acknowledged that it would not have been difficult for him to call up L.& H.’s customers and check the revenue claims.
As the Nasdaq composite index raced toward a record high that March, Dragon’s executives made fateful decisions. On March 8, the Bakers met with L.& H. executives and that company’s advisers from SG Cowen to try to reach a definitive agreement.
A few days before that meeting, Mr. Wayner of Goldman told Ms. Baker that he would be away on vacation and couldn’t make the session. He also said that he would be unable to call in and that it was pointless to send anybody else from Goldman because there wasn’t time to catch up on the deal. It was at this meeting that L.& H. proposed shifting the $580 million deal from half stock and half cash to all stock. The Bakers, with their high-priced investment bankers M.I.A., agreed.
Later, after L.& H. collapsed, Mr. Wayner testified that the bank “did not form a point of view” as to whether an all-stock deal would be risky or advisable for the Bakers. He said he could not remember if it had crossed his mind to warn the Bakers about potential issues with an all-stock deal.
Two weeks after the initial agreement was reached, Mr. Wayner told Ms. Baker that he would be leaving the next day for another vacation. He would not participate in a conference call with L.& H.’s accounting firm, KPMG, that was set up to discuss any open questions about accounting and due diligence. Mr. Berzofsky of Goldman did participate but later acknowledged that he did not raise any concerns. The Bakers say they believed that all issues had been addressed.
Mr. Wayner was still on vacation on March 27, when Dragon’s board met to take a final vote on the proposed acquisition. This time, Mr. Fine and Mr. Smith of Goldman attended the meeting, and Mr. Wayner called in from Argentina. No one from Goldman gave a presentation, but minutes from the meeting, taken by Dragon’s outside lawyers, indicate that the Goldman bankers expressed confidence that the combination of Dragon and L.& H. would produce a market leader. The board voted unanimously to accept the $580 million all-stock deal.
Years later, Mr. Wayner testified that lingering issues of due diligence had never been resolved to his satisfaction. He was asked if he had said as much that March day on the phone from his vacation.
“No, I don’t recall saying that,” he responded.
The deal closed on June 7. By Aug. 8, the merged companies were in crisis amid reports that L.& H. had cooked its books. Reporters for The Wall Street Journal did something the Goldman Four did not: they picked up the phone and called L.& H.’s supposed customers in Asia. They found that companies in South Korea and elsewhere that L.& H. had claimed were its customers weren’t doing any business with it at all. L.& H. had pulled sales figures out of thin air.
Although the Goldman Four never tried to call those customers, it emerged during litigation that other bankers at Goldman had done precisely that — about two years earlier, when Goldman itself considered investing in L.& H. in a plan known internally as Project Sermon. In it, Goldman’s merchant banking division took a closer look at L.& H. — but, apparently, never shared what it knew, and was never asked. Goldman was considering putting $30 million into L.& H., a step that, at the time, might have seemed conceivable, given the hype surrounding L.& H.
“Whenever we invest, we always want to talk to customers,” Luca Velussi, a Goldman analyst who worked on Project Sermon, later testified. Based on what Project Sermon’s team leader, Ramez Sousou, termed “preliminary” due diligence, Goldman declined to invest in L.& H.
By Nov. 29, L.& H. had plunged into bankruptcy. Indictments and convictions followed. L.& H.’s stock price sank to zero — and the Bakers lost everything.
Dragon Systems, the Bakers’ “third child,” was put up for sale at a bankruptcy auction. Visteon acquired some of Dragon’s technology. ScanSoft bought the bulk of it and went on to become a $7 billion giant, with a licensing deal with Apple. (The Bakers believe that some of their technology made its way into Siri.) ScanSoft later acquired — and assumed the name of — Nuance, another voice technology company.
Indeed, Nuance had gone public about the same time L.& H. bought Dragon, and Goldman handled the initial offering — a fact that still angers the Bakers. They say they had no idea Goldman was simultaneously representing their company and a rival.
It wasn’t until after the bankruptcy auction, the Bakers now say, that the full force of what had happened hit them. The money was one thing. But what they really wanted was the opportunity to complete the work they had started decades earlier. As part of the deal with L.& H., they had expected to continue their research. Once L.& H. collapsed, they had held out hope that they might get their technology back — either through litigation or through the bankruptcy auction. They now knew that it wasn’t going to happen.
“The door is closed,” Mr. Baker remembers thinking. “Not only do we not have the technology any more, but we have no chance of getting it back.”
THE Bakers’ case against Goldman is simple. Their lawyer, Alan K. Cotler of Philadelphia, captured it in a single sentence in a motion for summary judgment: “The Goldman Four were unsupervised, inexperienced, incompetent and lazy investment bankers who were put on a transaction that in the scheme of things was small potatoes for Goldman.”
Summarizing Goldman’s defense is more complicated. Based on the firm’s response to the complaint, its motion for summary judgment and testimony of the people it employed, most of that defense falls under one of three rubrics: The Bakers do not have standing to sue. Goldman had no obligation to do a financial analysis of L.& H. And Goldman’s bankers actually performed quite well. The firm released a statement that asserted, “Goldman Sachs was retained as a financial adviser by Dragon Systems, not its shareholders, and performed its assignment satisfactorily in all respects.”
Goldman’s lawyer, John D. Donovan of Ropes & Gray in Boston, has argued that under the terms of the engagement letter, only Dragon Systems had the right to sue, and Dragon no longer exists. Goldman has even filed a countersuit against Ms. Baker, contending that by suing Goldman she had breached the contract. Even though Ms. Baker lost everything in a deal Goldman orchestrated, the firm says Ms. Baker should now pay its legal fees.
To support the argument that Goldman was not obligated to perform due diligence, the firm points to that mystery memo of Feb. 29, 2000 — the memo that no one at Goldman has acknowledged sending — as establishing that Dragon Systems needed to push its accounting firm to explain any red flags or resolve outstanding worries.
Goldman’s lawyers have argued in a motion that while Goldman “strongly urged” Dragon to engage an international accounting firm to do a “forensic accounting analysis on L.& H.,” Ms. Baker “prevented” Dragon from following Goldman’s advice because she did not want to incur the expense of the due diligence and did not want to delay the transaction. The Bakers call this argument “complete fiction,” and even the Goldman Four seem dubious. They testified that Ms. Baker did nothing to block the performance of due diligence.
Goldman also hired an independent investment banker, Ian Fisher, who filed an expert report arguing that Goldman was not obligated to conduct due diligence because Dragon did not order what is known as a fairness opinion, an analysis of the acquisition price.
The Bakers have hired their own expert, Donna Hitscherich, a former investment banker with JPMorgan Chase who lectures at Columbia Business School. She wrote in her expert report that Goldman was obligated to perform due diligence with or without a fairness opinion.
If the case goes to trial in Boston, as scheduled, on Nov. 6, the final argument that Goldman can be expected to make is that the bankers, as Mr. Wayner testified, gave the Bakers “great advice.”
Mr. Berzofsky, too, testified in his deposition that the Goldman Four did a “great job.”
Even though Dragon lost everything?
“Yes,” Mr. Berzofsky said. He was given several opportunities to clarify. And then he was asked one more time — the fact that the Bakers and Dragon’s shareholders lost everything doesn’t affect your opinion?
“Correct,” Mr. Berzofsky responded. “We guided them to a completed transaction.”