What Happened to Goldman Sachs: An Insider's Story of Organizational Drift and Its Unintended Consequences

by Steven G. Mandis

  Some may object to the argument that competitive and external pressures to grow were behind the change at Goldman as overly simplistic. The truth is indeed more complex. For one thing, the increased complexity in management and systems that often accompanies growth must be factored in as well.50 Charles Perrow notes that institutional complexity prevents people from understanding the consequences of their behavior.51 Certainly, complex systems pose greater control challenges than do simple ones. And the faster an organization grows, the more the pressures and reactions to them compound and become interconnected, leading to increasing complexity and difficulty in seeing change.

  In addition to accelerating certain changes, Goldman’s structural change into a public company initiated a cascade of significant changes to other aspects of the organization. Now it had a board of directors that included independent directors (people from outside Goldman); a new partnership compensation program, with voting rights, as spelled out in the shareholder agreement (to help replicate some elements of the partnership); additional layers of processes, controls, and management needed by a public company; new compensation processes and consideration (stock in addition to cash) and more.

  But there were also pressures from within to grow revenues. During my interviews with partners, one of the recurring themes was the need for Goldman to “increase the size of the pie.” This means that revenues, and consequently the firm, had to expand for the firm to be successful in recruiting and retaining talented people. If the pie stayed the same size, then as more people became partners, each partner would keep getting a smaller slice of the financial rewards, making partnership less attractive. In this way, the external competition for people, combined with financial incentives, pressured the firm to increase revenues—not only to stay competitive but also to shore up the partners’ wealth. As we’ll explore in more detail in the next chapter, the acceptance of a number of changes in practices at the firm that deviated from the traditional values, and the normalization of that deviance—or what could be described as rationalization that they weren’t really deviations from the original meaning of the principles—were also crucial in the process of continuing drift.

  Chapter 5

  Signs of Organizational Drift

  IN THEIR FIRST LETTER TO SHAREHOLDERS IN THE 1999 ANNUAL report, the top Goldman executives wrote, “As we begin the new century, we know that our success will depend on how well we change and manage the firm’s rapid growth. That requires a willingness to abandon old practices and discover new and innovative ways of conducting business. Everything is subject to change—everything but the values we live by and stand for: teamwork, putting clients’ interests first, integrity, entrepreneurship and excellence.”1

  Changes were happening, and the partners seemed to be making it clear that they wanted any changes carefully managed to maintain Goldman’s core values. But as the firm encountered decision-making dilemmas related to change and growth, managers found it difficult both to acknowledge what was happening and to confront the conflicting needs and desires that underlay the issues. The Goldman partners I interviewed, even some of those who initially said there had been no change in culture, conceded that some changes had occurred, but they described them as “one-offs” or “special circumstances,” and others said the industry was growing and changing so quickly and becoming so complex that they hadn’t seen the changes clearly.

  It is important to understand certain unique mechanisms at Goldman that formerly had helped slow the pace of change. The pressures on these mechanisms provide a lens into the firm’s policies and business practices and how they evolved.

  Constrained Capital

  One inescapable reality of a private partnership is that capital is somewhat constrained by its being contributed by the individual partners—and there are, after all, only so many partners. This constraint acts as a check on growth. A second factor is the personal liability of the partners, which requires the maintenance of large capital reserves to cover potential losses as well as the adoption of business practices that help the firm avoid large regulatory fines or lawsuits.

  For most of Goldman’s history, these capital constraints provided a measure of self-regulation that not only limited growth but also helped maintain cultural stability and led the firm to emphasize client relationships. John Whitehead observed that “limited capital forces an investment banking firm to be careful in deciding what kinds of business it should be involved in and to what extent.”2 With restricted capital, the firm had to emphasize client-oriented businesses, rather than trading, because they required less capital and often entailed relatively less risk. That is one of the reasons why the M&A department was highly valued.

  Whether or not the lack of capital was a drawback is debatable, and it was debated by the partners at the time. Not only did Whitehead believe that capital constraints forced the firm to make better decisions, but also many competitors thought Goldman was at no disadvantage. As one commented, “I wish I could find a business where Goldman is capital-constrained.”3 But the stark fact was that in 1998, Goldman had roughly half the capital base of Morgan Stanley or Merrill Lynch—this, after having twice the capital base of Morgan Stanley in 1986. The increasing pressure for growth led the partners to seek outside capital in order to remain competitive, even though acknowledged Goldman culture carriers voiced concerns, fearing the impact on the firm of outside investors’ goals as well as the impact that more capital would have on the firm’s business mix, tolerance for risk, culture, and management.

  Another key loosening of the constraints on organizational change came from the changes in liability for losses, the dangers of which many partners were also well aware of.

  Partners’ Liability

  Earlier investments—the $500 million private equity investment of Sumitomo Bank in 1986, and the $250 million private investment made four years later by the Kamehameha Schools/Bishop Estate, had loosened the constraints on growth, but they had not relieved partners of personal liability. But first with the transition to LLC and then with an IPO, they would be freed from this liability, which a number feared would lead to a stronger appetite for growth and risk and would quicken change. They had already seen evidence of greater risk-taking even before the IPO.

  In my interviews, many of the partners pointed to their personal liability as having been a constraining factor on “doing stupid things.” Some pointed to the risk Goldman took in bailing out the hedge fund Long-Term Capital Management (LTCM) in 1998 as an early case of taking on more risk than the firm had traditionally been comfortable with. LTCM was a speculative hedge fund that used a lot of leverage and faced failure that year. As LTCM teetered, Wall Street feared that its failure could have a ripple effect and cause catastrophic losses throughout the financial system. Unable to raise more money on its own, LTCM had its back against the wall.

  Goldman, AIG, and Warren Buffett’s Berkshire Hathaway offered to buy out the fund’s partners for $250 million, provide capital of $3.75 billion, and operate LTCM within Goldman’s own trading division. The situation was so serious and the pressures on LTCM were so intense that the offer was made with a one-hour deadline. But at the beginning of the year, LTCM had $4.7 billion in equity, so its partners regarded the offer as stunningly low. They did not accept Buffett’s offer within the required one-hour deadline, and the deal was off. Ultimately, the Federal Reserve Bank of New York organized a bailout of $3.625 billion by the major creditors to avoid widespread financial collapse. Goldman and the rest of the top-tier creditors each contributed $300 million.4

  For some Goldman partners, getting involved in the bailout subjected their personal capital in the firm to too much risk. The investment banking partners, in particular, according to my interviews, generally were not happy about the potential risks to their personal capital and were worried about Corzine’s aggressiveness in pursuing the deal. To some of them, that the firm went ahead was a sign of both the pressures and the changes.

  As t
he risk rose—when the firm grew and invested its own money more aggressively—it subsequently changed the way the partners interacted with the firm.

  The IPO Debate

  Even though many Goldman partners now say that the culture hasn’t changed, at the time the IPO was decided, some of the partners were quite vocal about their concerns about cultural change. In fact, when people at Goldman—not only partners but also employees—discussed among themselves the idea of the firm going public, most of the concerns they expressed were about how Goldman’s culture might change rather than about financial or personal repercussions. Partners talked about their responsibilities as stewards to leave the next generation a stronger business with smarter people than the one they had inherited. In discussions, most IPO opponents expressed the concern that going public could “destroy what makes Goldman Sachs Goldman Sachs.”5

  A contemporary report described the process of debating and then postponing the IPO as “a wrenching experience that has bruised the firm.”6 It created tensions between the firm’s active general partners and retired limited partners, between Goldman’s investment bankers and traders, and between Corzine and Paulson.7 The discussion was personal, because people had to consider their own self-interest as well as the interests of the firm.8 Partners screamed and cried during one meeting in what one observer described as a “cathartic experience.”9 Conflict arose between new partners and those who had been at the firm longer and had a much greater stake in the firm and were about to retire. Partners with longer tenure, whose Goldman equity and percentage had gained greatly in value and now would get multiples of their book value, had a greater personal financial incentive to favor an IPO.

  Some partners were rumored to believe they should be compensated, through the IPO, for staying in 1994, turning the firm around, and taking the risk when others left. But John L. Weinberg said, “I always felt there was a terrific risk and still do, that when you start going that way [an IPO] you are going to have one group of partners who are going to take what has been worked on for 127 years and get that two-for-one or three-for-one. Any of us who are partners at the time when you do that don’t deserve it. We let people in at book value, they should go out at book value.”10

  While I was at Goldman, one partner privately told me that he felt terrible that an IPO would make him worth more money than the management committee partner who had helped build the firm and helped get him promoted to partner. He estimated that this man, who had worked at the firm for more than twenty years and retired before the IPO, was worth an estimated $20–$40 million, compared with newer partners like him, some of whom had worked at the firm for less than ten years and would be worth more than $50 million.

  Understandably, dissatisfaction was greatest among the retired limited partners, and tension between them and the general partners (essentially, Goldman’s controlling owners) became heated. As the IPO was originally structured, limited partners would have received a 25 percent premium over the book value of their equity, whereas general partners would have seen premiums of nearly 300 percent.11 Whitehead predicted a major problem if this inequity were not resolved. Issues related to fairness and compensation were also raised by nonpartners because of Goldman’s long-standing policy of not paying high salaries to the “all-important junior executives—the ones who do the grunt work—holding out instead the brass ring of partnership and its potential for eight-figure incomes.”12

  Divisions also arose between the investment bankers and the traders. While in general, investment banking partners, especially the relatively new partners, did not support the IPO because of concerns about culture change and increased risk taking, most trading-oriented partners supported it, in part, because with the additional capital they could grow their businesses larger and faster.

  One investment banking partner who did support the IPO explained to me that he did so because of concerns about capital risk and liability that outweighed those about change to the culture. All his wealth from working at Goldman for more than twenty years was tied up in the firm, and the IPO offered a way to get it out. Getting out at a multiple of book value was greatly beneficial and probably sold him on the IPO, he explained (sheepishly admitting some self-interest), but he had also been genuinely worried about the risk involved in the LTCM deal and he feared massive trading losses. He felt that Corzine was willing to take bigger risks than he was, and he was worried that the big traders—many of whom he didn’t know well and some of whom were not partners—were risking lots of partners’ money. It spooked him. But the firm needed to grow, he said, to continue to be the best place to work, to attract the best people, and to survive. He rationalized the IPO as a necessary compromise that was a result of pressures and changes.

  Key Signs of Organizational Drift

  One sign of organizational drift is a change in policies and business practices associated with a firm’s principles. Even before the IPO, Goldman began embracing opportunities it had once shunned out of concern for preserving its reputation for ethical conduct and to reduce conflicting interests.

  When firms get into new businesses in which they lack expertise or that are at odds with the values and principles that made them successful, they become vulnerable to veering off course, adding incremental risk—financial and reputational. The pursuit of maximizing opportunities can lead to rationalizing the drift. Previous decisions made to protect the firm start to be considered too conservative or out of date. Even the fundamental business model may be challenged. Goldman was not exempt from this process. Once again, I’m not judging or evaluating the changes; I’m simply pointing them out.

  Representing Hostile Raiders

  Goldman had made its reputation in banking by defending companies in hostile raids or unsolicited takeovers, when a company bids for a target company despite the wishes of the target’s board and management, typically when the board decides it is not in the best interests of the shareholders for the company to be sold at the price offered. A “hostile” bidder makes its offer “public,” taking it directly to shareholders—implying that the board of directors and management are not acting in the best interests of the shareholders and are trying to hold on to their jobs. To the target’s board and management teams, the bankers who work against the takeover to protect them and shareholders are viewed very positively. A 1982 Wall Street Journal headline captured this positive image: “The Pacifist: Goldman Sachs Avoids Bitter Takeover Fights but Leads in Mergers.” The accompanying article praised Goldman’s policy of not representing corporate raiders in hostile deals, although it included the few obligatory criticisms from competitors.13

  Goldman had made a strategic decision not to represent companies initiating hostile bids, the only large investment banking firm to do so.14 When questioned about the wisdom of this policy, Whitehead responded, “We have to dissuade them from going forward with this and explain to them why our experience showed that it would be unlikely that this unfriendly tender offer would turn out to be successful for them a few years later.”15 As a result, CEOs were more comfortable revealing confidential information to Goldman than to other firms, because they trusted Goldman not to use the information in representing hostile raiders against them.

  This policy lost Goldman some business and restricted the profits and growth of the M&A department, but it was a sound business decision that contributed to the positive public perception of the company. It was long-term greedy, calculated to make the most money for the firm over the long term, and Goldman may well have ultimately made more money because of it. Many clients actually paid Goldman an annual retainer to be on call in case of a hostile bid. Although the policy was not strictly about integrity, it had the effect of reinforcing Goldman’s reputation for integrity among clients and the public. Even a Morgan Stanley banker once said clients viewed Goldman as “less mercenary and more trustworthy than Morgan Stanley.”16 Most partners told me they felt the policy reinforced the image and culture of Goldman.

  In the l
ate 1990s, this policy was challenged, as many huge hostile deals were announced by Goldman clients and coveted potential clients. A series of internal meetings was held to discuss changing the policy. I participated in some of them.

  At one meeting, the people in the room were evenly divided. I was with the group that advocated against representing hostile raiders, whether they were blue chip corporations or individuals financed by junk bonds. We felt that doing so would cause us to lose both our credibility with clients and our perceived moral high ground. We reminded the group of Goldman’s advertising slogan, “Who do you want in your corner?” and observed that Whitehead had also resisted serious challenges to the policy. John L. Weinberg had supported the policy, too, even though one of his largest clients had requested Goldman represent it in a hostile raid.

  Those arguing in favor of representing hostile raiders claimed that very good clients of the firm were asking for our help. By working with them, we could try to reduce the “hostility,” implying that other advisers would not have as much tactical or moral sway with clients. They also said that they feared clients would not hire us to advise them on buy-side transactions because we weren’t willing to advise them on a hostile approach.

  We countered that in the past, Goldman had stepped aside and other firms had stepped in, to which the others replied that this was disruptive and did not serve the client well. We pointed out that in these very large deals, it was not unusual to have co-advisers anyway, so we would simply recuse ourselves, and the client would not have to get someone up to speed from scratch.

  In the end, senior partners decided that Goldman would work on hostile raids “rarely and reluctantly.” We developed a series of questions, essentially a test, to determine whether we would advise a hostile raider. At the same time, we did get a minor victory in that there seemed to be a gentleman’s agreement not to do it in the United States, where Goldman’s market share and association with the policy were particularly strong. In hindsight, this compromise is a clear case of an incremental shift, and the compromise provisions imply that at least subconsciously we were aware that there could be adverse consequences of this change in policy.