What Happened to Goldman Sachs: An Insider's Story of Organizational Drift and Its Unintended Consequences (15 page)

Proprietary Trading Becomes a Larger Percentage of Revenues

When one looks at the business principles that John Whitehead wrote in 1979, it is a particular challenge to reconcile the goals of proprietary trading with putting clients’ interests first, or the goal of being the leading adviser. Proprietary trading has one client: Goldman. As proprietary traders, we were walled off from client activity. We were not there to provide liquidity to clients, manage funds for clients, or advise clients. From time to time, we were asked by banking to tell them or their clients how hedge funds would evaluate a situation. From time to time, we were also approached by banking or trading to help finance a transaction or buy something from a client or coinvest with a client—actions that, most of the time, raised all types of potential conflicts and died because we were too busy to have long conference calls to discuss it. But generally we were in our own silo.

Proprietary trading at Goldman did not start in the 1990s. Bob Rubin joined the risk arbitrage proprietary trading area in the equities division in 1966, and by the 1980s it was considered one of the most profitable and powerful areas in the firm. When Rubin and Steve Friedman took over as senior partners in the 1990s, Goldman accelerated its additional role of risking its own capital versus being a mere “market maker.” It was the size of the losses from proprietary trading in 1994 that caused many observers to think the firm would not survive.

How big and important are proprietary trading and principal investing activities at Goldman? Glenn Schorr, a Nomura Securities equity research analyst covering Goldman stock, estimated that the Volcker Rule, which is intended to restrict proprietary trading and principal investing at investment banks, would impact 48 percent of Goldman’s total consolidated revenue. To put this into context, he estimated the impact at 27 percent, 9 percent, and 8 percent of total consolidated revenues of Morgan Stanley, Bank of America, and J.P. Morgan, respectively.

Certain Goldman client-oriented sales and trading desks had “proprietary trading” operations. They got to see client order flow, but theoretically they existed to provide liquidity or “facilitate client trades.” This was prevalent in less liquid, more opaque products and desks, especially fixed-income securities like high-yield bonds, where it may not have been easy to immediately match a buyer and a seller. It was also prevalent in relatively lightly regulated markets such as foreign exchange. Generally, proprietary trading on client-oriented sales and trading desks was less frequent in highly transparent and highly regulated areas such as equities.

Merchant Banking and Private Equity Become a Larger Percentage of Revenues

GS Capital Partners was started in 1992 with about $1 billion in assets and grew to $1.75 billion by 1995 and $2.75 billion by 1998. Ten years later, in 2007, GS Capital Partners had $20 billion in assets, making it one of the largest private equity firms. Goldman’s Whitehall Real Estate Fund also grew to have multiple billions of assets under management.
⁴⁸
Although separate, the funds do leverage Goldman banking and other relationships. When I was at Goldman we were often reminded how much more money Goldman could make investing the deal versus advising on it. Goldman was competing against clients in making acquisitions of companies and real estate properties. Goldman also had internal funds for managing Goldman’s own money and trying to buy assets. (The corporate and Whitehall funds raised money from external sources with Goldman’s coinvestment.)

The private equity business had many synergies with other parts of the bank and wielded a lot of clout as a profit center. For example, the private equity group could take its companies public, the executives whom it backed could become private banking clients, and the companies that it controlled could use Goldman to hedge its risks or finance its debt. The private equity group was careful to maintain relationships with all the Wall Street firms (similar to an unaffiliated private equity firm), but there was no question it also was aware of its primary affiliation—and disclosed potential conflicts in its documents for investors in the funds.

An executive of a major real estate firm told me the story of a property his company was trying to buy in the 1990s. Goldman had always been the company’s primary banker, and it had paid Goldman millions in fees over the years. The executive told his Goldman banker about the deal. A short time later, Goldman’s Whitehall Fund purchased the property. He was furious (although he hadn’t hired Goldman to advise him). He asked the banker to arrange a meeting with a senior Goldman executive. According to the client, at the meeting were senior people of the Goldman Whitehall Real Estate Fund. The corporate executive explained what had happened and expressed his frustration as a client. Goldman’s senior executive assured him that there were Chinese walls so that Goldman Whitehall would not know what he had told his banker. The Whitehall executives swore that they did not know about the client’s interest in the property. Goldman’s CEO said that merchant banking and Whitehall were very important to Goldman and that the firm had a fiduciary responsibility to do good deals for the clients they raised money from.

Skeptical, the corporate executive told me that since that meeting, he had done business with Goldman only if he had to. He doubted Goldman’s ability to manage conflicts, its various roles, and confidential information.

Steve Friedman, however, expressed confidence that Goldman could manage any conflicts arising from the firm’s new ventures: “The culture was transformed with a new strategic dynamism; and the principal investment business was well launched—although the old guard continued to worry and fret over conflicts with clients even though we explained to them that there are always conflicts, but you could manage the conflicts.”
⁴⁹

International Business Becomes a Larger Percentage of Revenues

In 1984 Goldman was primarily an American firm, with one partner outside the United States: an American based in London. After the 1986 partnership meeting, the consensus was to grow the firm globally to better serve clients. Consequently, international revenues increased as a percentage of total revenues, growing at a faster rate than did the US revenues, and a rate much faster than did the rest of the business. At the time of the IPO vote in 1998, 39 of the firm’s 190 partners were based in London, compared to 1 partner in the early 1980s. In 2011, 48 percent of Goldman employees were outside the United States (compared with 35 percent in 1996), and international net revenues grew 30 percent faster than US revenues from 1996 to 2011. One of the challenges Goldman faced with the increase in international business was maintaining its culture and principles outside New York, because many organizational issues arise in dealing with such rapid internationalization and global expansion, including training, socialization into culture and values, communication, local cultural nuances, and more.

When I arrived in Hong Kong in 1993, Goldman’s office was small, employing fewer than two hundred people. The cultural and legal barriers, as well as low deal volumes and the existence of entrenched local competitors, were major challenges. At the time, Hong Kong was a high-risk posting for a seasoned banker. It offered an advantage to advance more quickly but had the disadvantage that as locals became socialized and trained, they would ultimately have more value than an expatriate. And it was difficult to transfer back home.

Sometimes, when a senior banker was transferred to an international office, the reality was that the banker had lost a struggle back home or had to agree to the move to make partner.
⁵⁰
However, sometimes a foreign posting was intended to train a professional for a larger role. No matter how much Goldman wanted to portray itself as globally important, at the time all major decisions were made in New York.

Although the office furniture, office sizes, and set-up of cubicles were the same, the culture at Goldman in Hong Kong was different from what I experienced in New York. The Hong Kong office operated in a separate world. At the time, very few senior bankers from New York came for an extended period of time. Senior partners would jet in and jet out. Because Goldman was concerned about quality of execution, any deal of meaningful importance typically had a New York or London banker assigned to it.
⁵¹

Because of the cost of supporting international offices like Hong Kong and because of Goldman’s lack of relationships with important local people at the time, the firm realized that merchant banking and proprietary investments were much more profitable there than were advisory businesses. A team of four or five bankers would work on a deal and get paid a one-time fee of millions, and then the team would look for another deal to work on. Investing with clients, in contrast, was viewed as a way to build closer relationships with them. Once an investment was made, hopefully it would make money each year as it produced returns. Eventually, the investment would need to be sold or financed, and Goldman was in a good position to work on the deal and collect a fee (typically at a rate comparable to those charged in Europe or the United States, and not comparable to fees in Asia, which were notoriously low). The Goldman bankers’ close relationship would also provide an easier entrée for private banking. All of this required teamwork and collaboration, supported by a social network of trust among the few partners in Asia and their financial interdependence with the other partners around the world. These partners also had to trust that it was for the greater good to have Americans and British flying in and out all of the time.

Proprietary trading internationally was simpler, and it was seen as a high-margin business, because it was easily scalable—it took the same time for a trader to research an opportunity large or small. The markets in Europe and Asia tended to be less liquid and efficient, creating many opportunities. As Goldman established contacts and access, there were even more opportunities in which it perceived it held an informational advantage. Interestingly, Goldman preferred that its international offices follow the New York model as closely as possible—from people having the same titles (each person would be head of something and have one employee reporting to him, whereas the counterpart in New York with the same title would have hundreds of reports) to the same style of office furniture—all to make the firm feel united and cohesive. But although similar, the local hires typically had one approach, and the New York expatriates had another, and, to add complexity, the London expatriates that came stood somewhere in the middle. Often, junior people like me were caught in between.

The internationalization of the firm also had consequences for the acculturation of new employees, which traditionally had occurred through an oral tradition. With the overseas expansion, this method of passing down the cultural legacy was more challenging.

Though the partners had strongly indicated at the time of the IPO that they didn’t want to undermine the firm’s core values, the changes in business practices and policies, as well as in the business mix, clearly illustrate Goldman’s organizational drift. The daily grind of competition, and the success these changes led to encouraged those at the firm, including the partners, to overlook or discount them, or sometimes purposely ignore them, in the interests of rapid growth, which was seen as vital to the firm’s success and survival. This pressure for growth significantly intensified after the IPO, which also brought a new set of changes to the firm.

Others’ Capital at Risk

When Goldman was private, partners’ finances were
interconnected. At partnership meetings, partners from any area could
question traders. A banking partner had every right to ask a trading partner
about risk, because it was his or her capital at risk. An executive at a
competing investment bank explained to me that at other banks, it was
unheard of for an investment banking MD to challenge a trading MD, and the
idea of collaborating—sharing ideas or information and
challenging each other—would be entirely foreign. Although the
performance of an MD at Goldman can have an impact on the performance of the
entire firm, it is not the MD’s entire capital at risk or his
personal liability, though he can be held accountable for behavior resulting
in fines.

Furthermore, several current Goldman partners explained to me
that in the absence of financial interdependence after the IPO, MDs had
little motivation to question traders about matters outside their own areas
of expertise. The increasing size of the firm and resulting specialization
and functional silos, combined with the lack of financial interdependence
after the IPO, changed the social network of trust and reduced the
opportunities for debate. One partner told me that he heard a story from one
of his fellow senior partners about when he decided to retire that speaks to
how siloed the firm became. He was in an elevator, and in trying to be
friendly he introduced himself to the other person in the elevator and
politely asked him what he did. The person replied that they had met before
because he too was a partner. He then said he ran what the senior partner
called a relatively important business, and yet he hadn’t
remembered meeting before. By the time the senior partner got off the
elevator, he had made up his mind it was time to get out and retire.

Working with other people’s money also coincided with
changing attitudes toward risk management. With the change to a bonus
culture, there was more incentive to take risks, and because the partners
were no longer personally liable for covering losses, the constraints on
risk-taking (not just financial but also reputational) were loosened. Those
in areas such as proprietary trading had the opportunity to make more money
than banking partners if they made the firm significant amounts of money.
The incentive was to ask for and to invest as much capital as possible,
because the more money you were given, the more you could potentially make
with your trades. Traders could argue that if they worked at a hedge fund
they would receive 10 to 20 percent of the profits they generated and if
they didn’t get paid correspondingly by Goldman, they could
leave. And with so many more hedge funds cropping up, if the trades
didn’t work out and you got fired, you could be almost sure
you’d get a job at another bank or a hedge fund. The attitude of
many was that the Goldman pedigree would get you another job somewhere for sure.
⁸

One might ask how the change in the attitude toward risk was
evaluated by the board of directors. After 2002, when the
Sarbanes–Oxley Act became law, Goldman’s board was
composed largely of independent directors, most of them prominent in
business and academia. However, according to interviews, none of them had
ever focused on trading for a living. None would probably have been
classified as an expert in risk management by most trading experts. The
directors owned very little Goldman stock (less than 0.1 percent of the
total company), and what they owned generally was not significant to their
net worth. An interviewee speculated that the fact that Goldman’s
traders were making enormous sums of money for the firm and themselves also
made it unlikely the board would question that success.

In fact, it could have created the opposite effect. One partner I
interviewed said that the directors were not likely to question people who
made tens of millions of dollars and whose returns on equity and profits
exceeded those of their peers. Another partner speculated that as trading
became more important after the IPO and risk management was more critical,
the board relied on Lloyd Blankfein and his number two, Gary Cohn, from
trading, instead of Hank Paulson, and that may have contributed to
Paulson’s decision to leave to become secretary of the
Treasury.

When I was in proprietary trading, one of the partners received a
voicemail from Paulson, CEO at the time, on which I was copied. It related
to risk. The partner forwarded the message to Blankfein, answering the
question and asking Blankfein to deal with it. I asked the partner why he
had not responded directly to Paulson. He seemed more than a little annoyed
at my curiosity, saying essentially that Paulson knew a lot about clients
but little about trading risk, and he did not have the time to explain it to
Paulson. Whether or not Paulson understood trading risk, the fact that a
partner did not want to deal with the CEO was surprising to me. This was in
stark contrast to when I was an analyst in the early 1990s, when the senior
partner was held with the deepest respect. I remember being told that when
one goes to see the senior partner of the firm, one must wear a suit jacket
to show respect. In hindsight, I think I intuitively felt that Hank would
probably not be around for much longer if traders didn’t have the
time for him, and I privately questioned if a banker would ever again be
head of Goldman. But I don’t remember giving it that much
thought. I just went back to my daily routine.

Misaligned Incentives

Goldman’s incentive structure, like those of other
banks, also evolved in response to the changing nature of the
firm’s business mix. Rob Kaplan, former vice chairman of Goldman,
said that banks’ visions changed as they placed emphasis on
trading (see
chapter 5
). It
became more about making money than about “the value-added vision.”
⁹

More Cultural Stress: Envy, Self-Interest, and Greed

One of the basic principles of the financial system is that risk
is rewarded. Exactly how well Goldman partners were rewarded—what
they earned or owned—had been a closely guarded secret, but it
became public information in the filings for the IPO. Some might even say
that it was in everyone’s face. When I joined Goldman as an
analyst, a list was published by a finance magazine of the one hundred most
highly paid people on Wall Street, and it was passed around among the junior
people in great secrecy. I was told that I would be in deep trouble if a
partner caught me with it. The list contained so many Goldman partner names
that, except for a few top partners, the names were listed at the bottom of
the page, with no bio or background, unlike the non-Goldman partners, each
of whom got a short description.

The general reaction of the public disclosure of wealth within
the firm was envy stoked by self-interest, and that, when coupled with
freedom from personal liability, translated into greed and lack of
restraint. (Bear in mind that this was during the dot-com and equity market
booms, when many people were becoming extremely wealthy. And even some of
the partners would privately question why people who they didn’t
feel were nearly as smart or hardworking or as committed to the long term
were making more money than they were.)

I cannot emphasize strongly enough the impact on the organization
of the resentment stemming from knowing who was gaining how much at the IPO;
there was a reason many partners did not like Goldman’s financial
information being disclosed, and Jimmy Weinberg argued in 1986 that this was
one of the reasons the firm should not go public.
¹⁰
The average partner received around $63 million at the IPO price, an
amount that became $84 million after the first day of trading. In the class
of 2000, of those who just missed making partner before the IPO, some
received a fraction of that amount. The discrepancy was enough to cause a
great deal of resentment, especially among those who were hired at the same
time as members of the class of 1998 but were not nominated for partnership
until after the IPO.
¹¹
(See
appendix D
for a
table showing percentages, shares, and value of partners’ shares
at the IPO.)

I remember working with an MD-lite who had just missed making
partner before the IPO. Upon finding out the difference between her payout
and that of those who were elected in 1998, she did not come to the office
or return calls or voicemail for days. It was like a “mini
strike,” and it worked: the shares she received were increased.
It sent a strong message about how one needed to act to get what one
believed was promised, fair, and/or justifiable.

Some of the tensions were eerily similar to many of those who
were around in 1994, when so many partners retired, when the prevailing
sentiment was that the retirees were “sellouts,”
leaving to save themselves. After 1999, MD-lites and VPs, like the partners
who remained in 1994, wondered whether they had been sold
“motherhood and apple pie” or principles of
brotherhood, only to realize that there was a limit to the values and the
bond. I was also surprised that some nonpartners mentioned they felt that
the retirees who stayed in 1994 but left before the IPO, and even those who
really built the firm but had retired, were being treated unfairly.

Weinberg and Whitehead’s ideas about being custodians
of the firm for future generations of partners seemed to be less of a
priority. For those who believe that greed was always prevalent at Goldman,
imagine that if the earlier partners had decided to go public
sooner—they would have received multiples on their capital
instead of book value when they retired. In particular, interviewees
estimated that John L. Weinberg and Whitehead each owned about 5 percent of
the firm, which would represent billions of dollars today.
¹²

As a Goldman partner explained to me in an interview, Goldman was
slowly losing its allure: the prestige of partnership, the mystique that had
always marked the difference between Goldman and its competitors.
¹³
At the time of the IPO, no one knew that Goldman would (or would have
to, as explained in some interviews with partners) become the highest-paying
firm on Wall Street. It had traditionally paid less than its peers, except
for partners, a business practice Whitehead felt reflected long-term greedy
and attracted the right people who had this perspective. Before the IPO,
Goldman partners made outsized returns, in part by pocketing the difference
in lower compensation for the nonpartners. One partner with whom I spoke
said that what made Goldman unique was that it found really smart and
dedicated people with certain values to “drink the
Kool-Aid” and buy into the culture instead of taking more money.
He felt, over time, people didn’t value the
“Kool-Aid” or buy into the culture enough anymore, and
Goldman raised the compensation level to be competitive—another
signal of the drift at the firm. Goldman was not special enough, its culture
not distinct enough, the value of the partnership not high enough for people
to be “long-term greedy” and accept lower pay for a
long period of time.

In addition, Goldman faced heated new competition for talent from
other firms as well as other opportunities. The firm reacted by
significantly increasing compensation, becoming the highest-paying firm on
Wall Street. Also, compensation per employee increased with the profits from
proprietary trading and growth—and the changes. For example, in
2004 the average compensation per employee at Goldman was $445,390, compared
with $279,755 and $199,230 at J.P. Morgan and Lazard, respectively. In 2007,
the numbers were $661,490, $311,827, and $466,003 for Goldman, J.P. Morgan,
and Lazard.
¹⁴
According to interviews, before Goldman went public, it typically
paid its nonpartners less than its peers paid their nonpartners.

The idea of making partner, and its social meaning and identity,
had been taken down a notch—or at least there was a market price
for it. Before the IPO it was highly unusual for retired Goldman partners to
work at other firms, but after the IPO this phenomenon increased. Many
partners who had just made their fortunes in the IPO were primed to retire,
and when they left, they took not only their money but also their expertise
and their knowledge and respect for the firm’s history and
traditions. Of 221 total partners at the IPO in 1999, only 39 (16 percent)
remained as of 2011.
¹⁵