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

by Steven G. Mandis

  In the press, Goldman’s conduct in the El Paso deal was portrayed as brazen, incestuous, and shameless. It was particularly shocking in that it occurred soon after Goldman issued its business standards report, a recommitment to its values and principles.22 It indicated that in some ways the organizational drift continued.

  After Strine’s opinion was read, Goldman continued to claim that it had been completely transparent with El Paso regarding Goldman’s conflicts, and El Paso chose to continue to work with Goldman. Once again, Goldman claimed it completely fulfilled its legal responsibilities. Many observers found it hard to believe that Goldman would even get involved in the deal, with its $4 billion investment in Kinder Morgan on one side of the negotiating table, and a $20 million fee on the other, especially considering the scrutiny Goldman had been under since the congressional hearings.23

  Lloyd Blankfein had clearly defined Goldman’s position on conflicts in an interview years before:

  The crucial differentiating advantage of Goldman Sachs would be one that outsiders might find surprising: Its complex variety of many businesses was sure to have lots of conflicts. Goldman Sachs, Blankfein said, should embrace the challenge of those conflicts. Like market risk, the risk of conflicts would keep most competitors away—but by engaging actively with clients, Goldman Sachs would understand these conflicts better and could manage them better. Blankfein (who spends a significant part of his time managing real or perceived conflicts) said, “If major clients—governments, institutional investors, corporations, and wealthy families—believe they can trust our judgment, we can invite them to partner with us and share in their success.”24

  Blankfein’s concept of the relationship between an adviser (that is also an investor for itself and with clients) and its clients (that are not coinvesting) can be at very least confusing and have unintended consequences. The competitive, organizational, technological, and regulatory pressures are there to maximize opportunities in this confusion and ambiguity.25 And so the organization drifts, accepting a legal standard as its ethical base.

  Conflict Clearance

  In my role as business unit manager of Goldman’s global M&A department in the late 1990s, I addressed strategy, business processes, organizational policy, business selection, and conflict clearance issues. Before Goldman could accept work with a client, it needed approval (clearance) from me or the people with whom I worked.

  This position was both an honor and a curse. Keep in mind that I did this job in the late 1990s, when there was a boom in investment banking, stock market, and tech industry activity and the firm was growing at a rapid pace internationally and in proprietary trading.

  I had two voicemail boxes that could accept some seventy to one hundred voicemails each. I regularly went home at 2 or 3 a.m. and woke up at 7 a.m.—and both message boxes would be full. It is hard to imagine, but e-mail was just starting to catch on and there were concerns about security and potential legal implications. More and more issues arose with Goldman’s growth as conflicts were becoming more complex and global.

  My predecessors in the job typically had to worry about issues such as, If two clients wanted to buy the same company, whom would we represent, and why? (Typically, we would represent the first to have raised the idea, particularly when we had done previous work for the client.) But now, the first question typically was, Have we worked in the past with another company analyzing the transaction, or have we advised the target company? We might have made commitments and possess confidential information, both of which could preclude us from working on the deal or would complicate our potential involvement. To determine whether we had worked on it or for the target previously, we consulted a huge Yellow Pages phone book in which we catalogued and cross-referenced each assignment by the target’s name and the acquirer’s name, so that if someone asked, we could look it up. Generally, senior partners had a good institutional memory and could help identify potential issues. But so many companies were merging and their names and legal structures changed, and the book could not necessarily keep up.

  Suddenly, companies were being bought, reincorporated, and renamed at an amazing pace. In addition, when Goldman worked on an assignment, a confidentiality agreement usually had to be signed. We filed those and made sure the companies involved were listed in the book, which was getting larger by the day. We had the added issues of international names and entities, and the regional offices sometimes were less forthcoming in sharing the information, concerned about who learned what information and about protecting their clients’ confidentiality. With all that was going on, and with the money being invested in systems for managing trading risk, we still had an essentially manual process. The complexity was far outpacing our methods.

  We began to push for more automation. We even developed a factory floor diagram, breaking the M&A and other advisory processes into parts, so that we could track the process as it moved down a virtual conveyer belt. In hindsight, our system had flaws, but we were reacting to the challenges we faced as investment bankers who had to process as much information as possible to facilitate deals or opportunities. We were neither compliance experts nor systems experts, and we did not understand all the possible consequences of our approaches. There was no way we could know whether every conversation that might have divulged confidential information had been entered into the system. We kept trying to hire more people to help us. We brought people in from compliance to help. One of them was so overwhelmed and overworked that, sadly, he had a clinical breakdown and had to take time off.

  The pressures were enormous, because everyone wanted to work for his own clients and lobbied for clearance—as well as for a speedy resolution, because a delay could be interpreted by the client as a problem and the client might start speaking to other firms. If we said no, some people took their views (and anger at us) to a higher court—meaning someone on the management committee. And, in the end, if we disagreed and thought there was a potential or perceived conflict, there was always the possibility of getting the client to agree to sign a big boy letter that indemnified us and represented that the client understood the situation and the risks.

  Some clients strongly wanted Goldman to work on the deal despite a conflict, actual or perceived, because they anticipated a reward that outweighed the risk. Then the issue became, Who are we to tell a smart, sophisticated CEO with her own advisers that she is not smart enough to agree to such a thing? We felt the pressure of our competitors doing it, and we felt that if any firm would hold itself to a higher standard, it would be Goldman. The internal thinking was, We are different from the other firms; we have higher standards and principles. As discussed later, it was this rationalization—a sense of higher purpose supported by Goldman’s public and community service and folklore—that in part helped create the unintended consequence of a blind spot.

  In my position, something else began to come up more often: the issue of our having a proprietary equity position in the publicly traded target or owning its bonds. The questions were becoming more frequent, just as in the El Paso case. The decision I found the most difficult, and one I did not like, was advising multiple private equity firms on a transaction. Traditionally, Goldman would only work with one private equity firm in a potential acquisition. In the early 1990s, there were not very many leverage buyout (LBO) firms. If a bank aligned with only one, it would risk missing out on advising the winner, but it was accepted practice to work with one. But by the late 1990s several private equity firms would come to us to help finance the same leveraged buyout transaction and the commercial banks adopted the practice of having separate teams within the bank work for separate private equity firms. The rationale was that there were only so many banks, so if each bank committed to only one firm, then one could argue that it would limit the number of buyers, which would hurt the seller. So in the end we changed our policy and we set up separate teams. At first we preferred they were physically on separate floors, but then there were not enough floors. This policy
change was another signal of business practices changing.

  Even more concerning to me was the policy change to allow our internal funds to bid for a company while providing financing advice to competing private equity firms. There is a real possibility in that situation that a client will lose to Goldman and will perceive that the firm misused information, such as the knowledge of what the client was going to pay. Even if all information is in fact completely protected, if Goldman wins the bidding in such a case, the perception may be bad. I’m not alone in my concern. In its 2011 business standards committee report, Goldman mentions that to “strengthen client relationships and reputational excellence … Goldman will carefully review requests to provide financing to competing bidders when a MBD [Goldman’s Merchant Banking Division] fund or other firm-managed private fund is pursuing an acquisition as a bidder.”

  The report states that when Goldman has “multiple roles in a particular transaction … the firm may be able to address potential conflicts by providing disclosure to its clients, obtaining appropriate consents, relying on Information Barriers, carefully defining its role and/or requesting that the client engage a co-advisor.” When I discussed this with a legal expert, he said that this policy basically covers most imaginable scenarios and would technically allow the firm to play multiple roles and follow different legal standards in serving clients, from the highest, that of having a fiduciary responsibility to them, to the lowest, that of being a market maker.

  More often, we had clients sign big boy letters. The phrase itself is telling. It neutralized what we were doing; it implied the client was sophisticated and that it was OK. If we were not convinced about a company—typically a smaller company that might be less sophisticated—we would not allow the banker to get a big boy letter from the client. Which clients qualified as big boys was subjective. Usually, it was an easy discussion, because we could determine that a smaller company with a smaller potential fee was not a big boy and therefore not worth the banker’s appeal to the management committee.

  All these decisions were highly vetted and discussed with senior management, appropriate committees, and the legal and compliance departments. We had whiteboard sessions trying to figure out what might be a problem or what might look bad. During the time of my role in the late 1990s, the gold standard for determining the proper behavior expected at Goldman was to imagine how we would feel if our actions were disclosed in the Wall Street Journal. This test was constantly drilled into our heads. In fact, if it was a decision that would even make the Wall Street Journal, we tried to err toward caution. But we couldn’t imagine all scenarios.

  Slowly, even during my years, the world incrementally changed from more “one off” instances—“No, we can’t do that; it could look bad”—to, “If both clients agree, or if the sophisticated client signs a big boy letter, then it is okay.” Because of Goldman’s market share and proprietary trading, and because Goldman’s growth coincided with a booming market, issues came up often. Eventually, the key senior person who reviewed conflicts and was seen as “more administrative or compliance-oriented” rose all the way to the management committee because of the importance of the job. No other firm had (or has) such a person at such a high level. It is a critical job, because it involves not only figuring out how to get to yes, how to please clients, and how to maximize revenue opportunities, but also ensuring that the firm is protected and following a process that will stand up in court. This is what Goldman refers to as “managing conflicts.” During our interview, one partner compared the chief conflict management person with the chief risk officer for the firm.

  Goldman still has its leading market share, in part because it effectively manages conflicts to its advantage—and it also faces reputational and legal questions and consequences, because conflict management is art and not science. If Goldman had not managed conflicts, its ability to grow and maintain market share would have been challenged. The management of conflicts maximizes opportunities to access scarce resources. The diligence and effort and thoughtfulness and angst that go into managing conflicts are immense, and it is inevitable that something will not be considered or given the proper weight as a possibility. And when that happens, Goldman most likely will be sued, and it will be on the front pages, with its reputation questioned. Such questions are the consequence of Goldman doing business and maximizing opportunities as well as the consequence of the limitations of human thought, processes, and systems.

  About the time of the IPO, a partner told me that there was a limit to the number of transactions, of capital flows, in the world. What Goldman was good at was getting involved in those transactions or discussions because of its relationships, smarts, and information. What Goldman was better at than any other bank was working to maximize the revenue from the transaction or flow—finding and managing “multiple roles in a particular transaction.” For example, if it were a cross-border M&A deal, then Goldman would provide M&A advice as well as involve its foreign exchange desk to handle the currency exchange for the purchase price. If Goldman missed the deal—meaning our bankers were not involved—then proprietary trading might possibly be involved in merger arbitrage (oftentimes, Goldman would make more money in proprietary merger arbitrage than if it had been hired to advise on the deal). Goldman ensured that we looked at each transaction and each flow and had some way to make money from it. The more roles we played in a transaction, the more opportunities we had to make money on a cost base that was essentially fixed, and thus the transaction would be much more profitable for us than if we had played only one role.

  Managing conflicts maximizes that opportunity. Because of this recognition, the firm has invested many more resources into conflict management.26 This is one of the keys to Goldman’s higher returns on equity compared with those of its peers. But it also relies heavily on an organizational culture, a “residual partnership mentality,” the partner told me, that emphasizes sharing information to maximize opportunity.

  Goldman has the same conflicts as most other Wall Street firms. However, unlike other firms, according to William Cohan, “Goldman has taken the decision that it is in the business of managing conflicts and the joke around Goldman was ‘If you have a conflict, we have an interest.’ While that has led to tremendous success it is also the firm’s Achilles heel—because of its unfailing belief that it can manage conflicts it gets itself into positions where other firms will just say no.”27

  When I raised this issue with a Goldman partner, his comment was that the other firms do not just say no. He implied that this argument was based on the premise that the other firms had some sort of higher values, which, he said, based on his experience working against and alongside them for decades, he believes is wrong. He said they most likely wished they had Goldman’s network of relationships (a network with leading market shares), which means Goldman must deal with these conflicts. It is easy not to have client conflicts, he stated, when you have fewer trusted clients and so many clients asking you to advise on or provide liquidity in so many situations. I pointed out that it’s not really the number of clients but rather Goldman’s various roles or proprietary investing businesses causing the conflicts.

  Management Committee Composition

  The changes in the membership of Goldman Sachs’ management committee corroborate the idea of a shifting emphasis toward the legal standard of compliance and away from the original interpretation of the first principle, as well as reflecting the increasing complexity of Goldman’s business. In 1999 the firm had twenty-two management committee members, two of whom were in legal functions (around 9 percent). In 2009, the management committee had grown to twenty-nine members. There are two members related to legal functions plus one individual who is known for political connections, one individual responsible for compliance, and one person responsible for conflicts (representing 17 percent of the committee). At the IPO, banking and trading/PIA members represented around one-third of the members each. In 2009, banking was around 20 percent and tr
ading/PIA/markets-oriented backgrounds represent close to half the committee.28 So there has been a shift at the management committee level as illustrated by an increase in legal and in trading/PIA/markets-oriented people. Remember, Sidney Weinberg had sought balance with a trader running the firm and banking having a large presence on the committee.

  The balance of composition between banking and trading in leadership has been an important organizational element (as was the balance between newer and older partners). Sidney Weinberg sought this balance when he selected Gus Levy, a trader, to become senior partner, but he created the management committee with a majority of banking partners to instill some balance. There’s been a shift in the management committee composition as well as at the very top, with the top two having trading backgrounds.

  Outside Board of Directors

  Some have argued that boards of directors of the banks must take more responsibility for oversight of conflicts, so it’s important to consider the role of the Goldman board here briefly.

  The Sarbanes–Oxley Act, which became law in 2002, sought to improve the oversight of boards by requiring the appointment of more outside directors. In 1999, Goldman’s board of directors had seven members, including four executives of the firm; two outside, independent directors; and John L. Weinberg. In 2001, Goldman’s board of directors had eight members: four executives at Goldman and four outside directors (John L. had decided not to seek reelection to the board after his initial three-year term at the IPO). After Sarbanes–Oxley passed, and in compliance with new requirements of the NYSE, Goldman added two outside directors. Thus, the majority of its board (60 percent) was comprised of outside, independent directors.