CHAPTER THREE

Inside and Outside Goldman Sachs

THE IMAGE OF my grandfather Samuel Seiderman’s life was very powerful in forming my own ideas of how I wanted to live mine. My grandfather was successful professionally and financially, and he had a recognized and respected place in his community. His broad range of involvements—especially in politics—made his life more varied and interesting.

Gus Levy was another model of a multifaceted, externally engaged professional life. Gus’s day job was running Goldman Sachs, but he had worlds outside the firm. Gus served as chairman of the board at Mount Sinai Medical Center, sat on the boards of cultural organizations such as Lincoln Center and the Museum of Modern Art, and was a major fund-raiser for and confidant of Republican politicians. His range of friends and contacts was enormous; Gus knew, or appeared to know, everyone who was anyone in business and politics, including President Richard Nixon and New York Governor Nelson Rockefeller. “There are really six Gus Levys,” the CEO of a major bank said. “At night they put on their tuxedos and fan out around the city.” I had no desire to emulate Gus’s peripatetic social life. But I was drawn to the kind of career that enabled Gus to contribute to causes that he cared about and earned him the same sort of place in his much larger world that my grandfather had in his. The various facets of Gus’s life were synergistic, with efforts in one area contributing to accomplishments in others. The variety of his involvements gave his career a meaning beyond the considerable achievements in his day job.

A third example for me was Bernard J. “Bunny” Lasker, a legendary arbitrageur of Gus’s generation. Despite an almost thirty-year age gap and vastly different political and social views, Bunny and I became close friends; perhaps, in some fundamental respects, we were on the same wavelength. A self-made man who had started on Wall Street as a messenger, Bunny was a vivacious, larger-than-life figure—one of those people who fill up a room just by entering it. Some people looked askance at Bunny because he lacked a certain polish, but that didn’t mean he wasn’t smart. Bunny never went to college, but he had a practical sense that Harvard Business School cannot teach. I used to say that if Bunny and our best MBA simultaneously started from the same point in a race that involved complicated practicalities, Bunny would run the course, have dinner, go to bed, and get up the next morning before his credentialed competitor reached the finish line.

Like his good friend Gus, Bunny knew everybody and was deeply involved in what he called the three great passions of his life—the Republican Party, the New York Stock Exchange, and West Point, where he served as a trustee despite having been rejected in his youth. He guided the New York Stock Exchange through hazardous times as chairman in 1970, when the investing public’s confidence was probably at its lowest level since the Depression. Richard Nixon publicly credited Bunny with saving his career by persuading him not to try for the presidency again in 1964, when Nixon almost certainly could not have won.

Another person I sometimes think of as a professional role model was someone I never met: Armand Erpf, a partner in the then well-established investment banking firm Loeb, Rhoades and Company, who was involved in an array of cultural and civic activities. I remember reading an article in The New York Times in 1967, shortly after I started at Goldman Sachs, about a chair being established in Erpf’s name at Columbia Business School. “My main interest is Loeb, Rhoades,” Erpf said. “After all, everything starts from there.”

That’s it, I thought when I saw the story, which crystallized a thought that had previously been inchoate in my mind. And in fact, Erpf’s line is not a bad description of my life over the next two decades, or at least what I tried to make it. Goldman Sachs was my fundamental and overriding involvement. From that strong base, I was able to reach out in various directions—nonprofit and charitable organizations, professional associations, politics, and, eventually, a second career in Washington.

Until the end of 1992, when I left Wall Street to join the Clinton administration, my inside and outside careers ran on parallel tracks. Inside Goldman Sachs, I was learning not only about markets, finance, and economics but also about management and human nature. Outside Goldman Sachs, I was learning about management and human nature in a variety of other contexts. These two careers developed side by side, each contributing to the other. Over many years, in a variety of disparate settings and institutions—Goldman Sachs, the White House, Treasury, Citigroup, Mount Sinai Medical Center, and the Harvard Corporation—what I learned in one place about working with people and how they think and react usually held true in another.

   

THE ANALYTIC MIND-SET I further developed doing arbitrage led me to look for inefficiencies or discrepancies elsewhere in the relative value of different related securities, and thus to the arcane business of stock options. Like arbitrage, options trading is a risk-reward, probability-based business, although more directly quantitative. At the time, even most people on Wall Street knew little about it. I read quite a bit on the topic, including a newsletter about opportunities in warrants. A warrant, similar to a “call option,” is a security that conveys the right to purchase a share of stock at a set price for some period of time, usually a certain number of years.

One article in the newsletter said that warrants in Phillips Petroleum were overpriced—based on valuation models—relative to the price of the Phillips stock. That created an opportunity in what is often called relative value arbitrage or relationship trading. Relative value arbitrage means going long one instrument—a security or a derivative—and short another related instrument, when one of these instruments is considered undervalued relative to the other. The bet you’re making is that the prices of the two instruments will return to their proper relationship and provide a profit from that movement. The instruments might be a common stock or bond and a “derivative”—an option or future that is convertible into the underlying stock or bond on some basis.

Not yet a partner at that point, I wrote a complicated memo recommending that we go short the Phillips warrants and long the common stock, betting that the price of the warrants would go down relative to the price of the stock. I gave the memo to L. Jay, who agreed with the recommendation. The warrants were overpriced relative to the stock, but the short position could still lose money if the stock price rose, so that the long position was an essential hedge. With that hedge, one would make a profit regardless of what happened to the price of the stock, as long as the discrepancy in relative value disappeared. L. Jay gave my memo to Gus, and Gus called me in.

“Ahh, I don’t want to do all that,” Gus said of my proposal to hedge. “Let’s just go short the warrants.”

“Gus,” I said, “you know we have to be hedged.”

Gus responded with a five-word sentence conveying that he didn’t care about hedging, didn’t care about my memo, and didn’t care about explaining the matter—because if I didn’t know this stuff, I shouldn’t be at the firm in the first place.

I went back to L. Jay, concerned about what to do. L. Jay said, “You better just go short the warrants, if that’s what Gus wants to do.” So we went short the Phillips warrants, and fortunately the stock didn’t run up while we were holding the position.

The same thinking that drew me to that transaction—and a lifelong tendency to restlessly reach into new areas—led me to other kinds of options. Stock options—instruments that allow, but do not require, an investor to buy shares at a prearranged price during a fixed period of time—had long existed but had always been extremely illiquid. To buy an option, you went to one of several small put-and-call houses, which ran ads in the newspapers and had a slightly questionable reputation. They traded options “over the counter,” which meant not listed on any exchange. Prices were nontransparent, to say the least. I thought that perhaps relative value arbitrage transactions could be done against these over-the-counter options. As I became increasingly involved, I saw that Goldman Sachs might well do what the put-and-call houses did—trade stock options directly with our clients, other brokerage firms, and the options dealers themselves.

My proposal met some resistance at the firm. During the Depression, stock options had led to vast losses on Wall Street. As a result, I was told, Sidney Weinberg had a rule against Goldman Sachs being involved with them. But by the time I first discussed this with Gus, Mr. Weinberg had died. At the end of our conversation, Gus said, in his gruff way, “If you want to get involved with options, go ahead,” and he got my proposal approved by the firm’s management committee.

Options are one type of derivative—a security, such as a warrant or a future, whose price depends on the price of an underlying instrument like a common stock or bond—and this was the beginning of Goldman Sachs’s trading in derivatives on securities. That business eventually became massive at our firm and across Wall Street as ever more new instruments developed that were based on equities, debt, and foreign exchange. But at the time, the options business was still at a primitive stage. Traders grasped that prices of options should reflect the volatility of the stock but as yet had no system for calculating values. However, an unpublished paper by Fischer Black and Myron Scholes was circulating that detailed a valuation formula based on volatility. For their work on option pricing, Scholes and another colleague, Robert Merton, won a Nobel Prize in 1997. Sadly, Fischer Black died too soon to share it.

The now famous Black-Scholes formula was my first experience with the application of mathematical models to trading, and I formed both an appreciation for and a skepticism about models that I have to this day. Financial models are useful tools. But they can also be dangerous because reality is always more complex than models. Models necessarily make assumptions. The Black-Scholes model, for example, assumes that future volatility in stock prices will resemble past volatility. I later recruited Fischer away from a full professorship at MIT to Goldman, and he subsequently told me that his Goldman experience caused him to develop a more complex view of both the value and the limitation of models. But a trader could easily lose sight of the limitations. Entranced by the model, a trader could easily forget that assumptions are involved and treat it as definitive. Years later, traders at Long-Term Capital Management, whose partners included Scholes and Merton themselves, got into trouble by using models without adequately allowing for their shortcomings and getting heavily overleveraged. When reality diverged from their model, they lost billions of dollars, and the stability of the global financial system might have been threatened.

What made options trading possible on a large scale was the Chicago Board Options Exchange, the first listed market for stock options, which opened in 1973. The key was the creation of standardized terms for the listed options and a clearing system, so that options could trade in a secondary market. I remember Joe Sullivan, the first head of the CBOE, coming to Goldman to tell me about his plans. I took Joe to meet Gus, who listened to him and said, with a twinkle in his eye, that this was just a new way to lose money, and then offered his support. I joined the founding board. Joe called the day before the CBOE first opened to say that he was afraid nobody would show up to trade. In fact, 911 contracts traded the first day, on 16 different stocks. Within a relatively short period, options trading turned into a genuinely liquid market and led to the creation of larger markets in listed futures on stock indices and debt.

I began as a Goldman partner during a period of ups and downs for Wall Street. The year 1973 was the first year the firm had lost money in many years, and 1974 wasn’t much better. Our chief financial officer, Hy Weinberg, told me that we junior partners would be unlikely to ever do as well financially as the older partners had because there would never be another period as good as the one that had just passed. That seemed highly plausible at the time, but turned out not to be the case.

During one particular bad stretch in 1973–74, the stock market fell 45 percent from its high. We had very large losses in risk arbitrage and block trading. We were still holding the acquiree stocks in several deals that broke up—contrary to our usual practice—because they seemed so badly undervalued. But as the market continued declining, these stocks’ prices kept falling. We thought our positions would eventually come back, and so we held on.

But sometimes, even if the market has gone way down and positions seem cheap, holding on may not make sense. I remember a customer who had a big position in a commodities arbitrage transaction. He bought soybean mash and sold soybeans because the mash was cheap relative to the soybeans. He expected to profit when the inefficiency corrected and prices converged. Instead, the spread widened and he had to put up more cash margin to creditors. As the spread kept widening, he ran out of cash and couldn’t put up more margin, so his positions were liquidated to meet obligations. Eventually the prices did converge. But by that time our client was bankrupt. As John Maynard Keynes once reportedly said, “Markets can remain irrational longer than you can remain solvent.” Psychological and other factors can create distortions that last a long time. You can be right in the long run and dead in the short run. Or you can be wrong in your judgments about value, for any of a whole host of possible reasons.

In that 1973–74 slump, we, like the trader with the soybean mash, were overextended relative to our staying power. In our case, the issue wasn’t solvency but the limits on our tolerance for loss. I finally went to Gus. We reexamined the merits of each of our holdings and how much risk we were willing to accept going forward, and we decided to sell about half of our positions.

Looking back at that episode, I realize that we hadn’t really been reevaluating our positions as the economic and market outlook changed. Holding an existing investment is the same as making it again. When markets turn sour, you have to forget your losses to date and do a fresh expected-value analysis based on the changed facts. Even if the expected values remain attractive, the size and risk of your portfolio must be at levels you can live with for a long time if conditions remain difficult. Praying over your positions—a frequent tendency in trading rooms during bad times—isn’t a sensible approach to coping with adversity.

   

AROUND THE TIME L. Jay Tenenbaum retired from Goldman Sachs in 1976, Ray Young, who was in charge of equities sales at the firm, gave me some advice. He said that now that L. Jay was leaving, I had to make a choice. I could continue to act in the manner I had developed working in a trading environment—focusing intently on my business, being short with people, and projecting an impersonal attitude. If so, Ray predicted, I would continue as a successful arbitrageur. But as an alternative, I could start thinking more about the people in the trading room and in sales—about their concerns and views—and how to enable them to be successful. In that case, Ray said, I wouldn’t be limited to arbitrage but could become more broadly involved in the life of the firm.

Ray Young’s advice pointed me toward a whole new world that I hadn’t thought much about. My tendency to be abrupt and peremptory—characteristic of Wall Street traders of that era—is exemplified by a typical episode: a colleague from investment banking came to ask me about the market impact of a deal she was working on. She had trouble explaining the deal to me, and I told her I was busy and didn’t understand how someone could work at Goldman Sachs and not understand some basic corporate finance. I dismissively suggested she go back upstairs and return when she was properly prepared. Ray made me understand that that kind of attitude limited how far I would go at the firm and how interesting my career there would be.

My general experience in life has been that most people can change only within a narrow range, if at all. Many people can acknowledge criticism and advice, but relatively few internalize it and alter their behavior in a significant way. Sometimes someone can change in one respect but not in another. I was involved in many discussions at Goldman over the years that centered on the question of whether a person who was highly capable professionally, but limited in some way, could grow to assume broader responsibilities. Often the limitations revolved around the ability to work effectively with colleagues and subordinates.

I’ve often asked myself why this advice affected me so much. Perhaps I simply responded when someone whom I respected, who clearly had my best interests at heart, raised a problem I hadn’t thought about and opened up new vistas. Judy’s view was that the harshness of manner Ray critiqued was a superficial attribute. Most likely, both reasons were true. In any case, my mind-set did change and I began to listen to people better, to try to understand their problems and concerns, and to more appropriately assess and value their views. And as I’ve since said to others, this not only had the effects in my business career that Ray had suggested but gave me something I hadn’t expected, a new satisfaction in the accomplishments of others.

I also connected what Ray told me with a comment that Richard Menschel, another more senior colleague who took a supportive interest in my career, had made a couple of years after I arrived at Goldman. Dick said that early in your career you may be concerned about bringing strong, younger people into your world. For some, that feeling remains; they continue to think that bright, more junior people threaten to outshine them in some way. But after a certain point, Dick predicted, I would become comfortable enough in my own position to eagerly seek out extremely capable young people for the arbitrage department.

He was right. Initially I felt uneasy about bringing a strong junior associate into the arbitrage department. But soon that changed and I wanted effective, aggressive people working with me in order to get the job done better. Moreover, as I found in everything I did thereafter, sharing credit with others didn’t mean less credit for me. To the contrary, I got credit not only for the results being better, but also for sharing the credit. I also enjoyed the recognition given to people I worked with. Dick was right that a lot of otherwise successful people never figure this out. They view smart junior associates as a threat rather than as a reflection of their own capabilities as managers.

I’d never given one second of thought to management as such. Once I began to think about these issues, however, I found them engrossing. How do you get people to work well with one another? How do you attract and keep strong people? How do you motivate them to do their best? How do you get a whole organization to be strategically dynamic and to act on difficult issues? I’d never been to business school or even read any books about management, but I developed views on all of this through experience.

   

JUDY CONTINUED in the theater and became a member of the actors’ union—Actors’ Equity Association—when she was cast in an Equity production. But she ended this nascent career when she became pregnant with our first child. Thanks to Judy we went to quite a bit of theater and dance, and I particularly enjoyed ballet. (Concerts and opera she did on her own.) Sherwin Goldman, who had been at Yale Law School with me, had become deeply involved with American Ballet Theatre (ABT) and had seen us at a number of dance performances. So he assumed I liked dance and asked me to join the ABT board. At that point I wasn’t in a position to raise much money, but I think Sherwin was trying to create a younger cadre on the board for the future.

Earlier, I said I thought you could draw a line from my election as fourth-grade president to the cabinet, showing how large a role chance and incident play in life. You can draw an even straighter line from my joining the board of ABT to subsequent opportunities, because being on the board of an arts organization caused people to view me as someone who was involved in civic activities. Soon thereafter, Gus asked me to help raise money for Lincoln Center, where he was treasurer. After Gus died in 1976, Bunny Lasker, who was always promoting my interests, got me on the board of Mount Sinai Medical Center. And so it went, with one involvement leading to another. The key was to get in motion to begin with. Also, at least for me at a number of critical junctures, some other person—for example, my fourth-grade teacher or Bunny—provided the critical impetus.

From the beginning, I had hoped to get involved outside the firm, and that desire never flagged. I wasn’t bored or awash in spare time, especially after our sons, James and Philip, were born in 1967 and 1971. But outside involvements added other dimensions to my life, providing a glimpse of what other people’s jobs and lives were like and an opportunity to contribute to purposes beyond my work. What’s more, outside involvements helped my Goldman Sachs career, as I met well-established people who were also clients or potential clients of our firm. Finally, these outside activities began to create a place in the community for me—though I may have set that back slightly at ABT when I suggested, at a time when the company was facing a 10 percent deficit, that nobody would notice if Swan Lake were performed with 10 percent fewer swans.

My real desire, however, was to get involved in politics. I’d done some work in a few New York campaigns, but that didn’t amount to much. An avenue for larger involvement opened in 1969, when Henry Fowler, who was Secretary of the Treasury under Lyndon B. Johnson, joined Goldman Sachs. “Joe,” as everyone called Fowler, was a courtly Virginia lawyer whose ancestors had come to America in the seventeenth century.

Many people at Goldman were not that interested in what Fowler had done in government. But to me, Joe was a fascinating figure—someone who had gone to Washington during the New Deal and served in every Democratic administration thereafter. I took every opportunity to talk politics with him and at some point mentioned that I would like to become involved. Joe called Robert Strauss, who had recently become treasurer of the Democratic National Committee.

Strauss had a different kind of charm from Fowler. Rather than flattering you, Bob insulted you. When I met him in New York in 1972, he told me that if I wanted to be involved in policy, I was of no use to him. But if I wanted to raise money, we should talk. He said that Nixon was going to be reelected, so he was focusing on senators and congressmen to make sure the Democrats retained control of Congress. Then Strauss said something I took to heart: in politics a lot of people promise to do something, but very few actually do it. If you don’t want to do what you’re asked, just say no. But if you say you’re going to do something, following through will set you apart. Then Strauss told me, “You know, you look good on paper. But now that I’ve met you, I don’t think you’ll amount to much. So you better work hard.” That seemed an odd way to get somebody to help you, but, in fact, it was the beginning of a great friendship.

I took off a week from work to call on people I knew and ask them to contribute to Bob’s effort. While I don’t have the personality ordinarily associated with fund-raising, I was determined to do it anyway. I called mainly on people in and around the arbitrage fraternity, only a few of whom were in any way sympathetic to the Democrats. I didn’t raise more than $25,000. But in those days, that wasn’t a bad start.

Even in his eighties, Bob has a magnetism that reminds me of Gus and Bunny. When he walks into a room, the effect is electric. With a twinkle in his eye, he continues to win people over through effrontery. “You must be the stupidest person alive,” began a letter he recently sent me about a business matter. He’s shrewd yet somehow the least cynical of cynics. I remember some of his early political advice: “Let me tell you about Washington, Bob. I could call President Carter once a week and just say anything—even talk about the weather. And after that, I could walk around town telling people that I had just been talking to the President today and, while it would mean nothing substantively, it would have meaning in Washington. That’s just the way this city works.”

From time to time others in the financial world have asked me for advice on how to get involved in politics. You can most readily acquire a place at the table by raising money, I tell them. And once you have a place at the table, you get to know people who work on campaigns. If those people think you have useful thoughts on politics, message, or policy, you can develop a broader involvement, at least informally. But there are actions and attitudes that can militate against crossing that line. One is if you appear to be trying to use politics to further your business or financial interests. Another is an attitude of self-importance. Many people assume that business success qualifies them to opine authoritatively about politics, but while business experience can be useful, politics and government differ in many ways.

Toward the end of the Carter presidency, Josh Gotbaum, who worked for Alfred Kahn, President Carter’s “inflation czar,” approached me about directing the Council on Wage and Price Stability, administering the price guideline program. Few notions were more appealing to me than seeing the world from inside the White House. But after going to Washington to meet the relevant people, I was left with the impression that that job wasn’t positioned to work, in terms of either its staffing and authority within the administration or its conceptual approach. In addition, I had a great deal to lose by leaving a partnership at Goldman Sachs.

In any case, the issue was academic. Shortly before this, I had been playing tennis in Westhampton when, for the first time in my life, my back started hurting. Instead of getting better, it got worse, and I went to see the chief of orthopedics at Mount Sinai, Robert Siffert. He looked and looked and couldn’t figure out my problem. Nor could others he referred me to. Nothing appeared on the X-rays. After a while the pain was so severe that I couldn’t sit up or stand for any length of time.

I was intent on not letting my back pain interfere with either Goldman or my outside activities, so I did everything I could to keep functioning. For many months, I’d have to lie down at the office on a couch. Some days, people at work would have to help me into a car to go home at two or three in the afternoon because the pain was so severe. I was in the hospital three times—for bed rest and diagnostic procedures—and each time I ran the arbitrage business from my bed. I was on the board of Studebaker-Worthington, and I participated in one meeting lying on the conference table. Once, the chief executive officer of the company, Derald Ruttenberg, called and asked me to meet him at his office on a Saturday to talk about selling the company. I thought, If I don’t go, he’ll hire Felix Rohatyn—the renowned investment banker from Lazard whom Ruttenberg had also mentioned. I couldn’t walk for more than a few yards at the time, or even sit, but I went to Ruttenberg’s office and lay on his window seat. We got the business, though much to my dismay, Ruttenberg gave Felix part of the fee. (It’s more than twenty-five years later, but I still remember the amount.) Ruttenberg said he wanted Felix to be satisfied, given his importance in the world. Since stress is not good for a back problem—as my doctors reminded me from time to time—trying not to miss a beat by working from a horizontal position probably wasn’t the ideal way to get better.

On the one hand, I was concerned the condition would remain undiagnosable and I wouldn’t get well. On the other hand, I woke up every morning hoping and almost expecting that that day would be better. The then chief of neurosurgery at Mount Sinai wanted to operate on what he thought was a disc problem. But Bob Siffert wasn’t sure he was right and told me to hold off. Months went by. Eventually, Siffert took another kind of X ray and found a barely discernible crack in a vertebra, the result of a genetic defect, which is apparently quite rare except in Eskimos. I had to have a spinal fusion, an operation in which bone from elsewhere in your body is used to mend the crack in your vertebra.

It was major surgery. I wore a brace for six months afterward and full recovery took about a year and a half, but the operation worked. Ever since then I have played tennis, fished, and done whatever I wanted to physically. But my back problem meant that I couldn’t even consider the Carter position. When I saw the article in The New York Times about the person chosen for the job, I thought, That could have been me—though I also recognized that I almost surely wouldn’t have taken the appointment anyway.

   

GUS LEVY SUFFERED a massive stroke in November 1976 and died shortly thereafter. He was sixty-six, young enough that he’d been able to ignore the issue of succession at the firm. However, shortly before the stroke, he had told George Doty, a senior member of the Management Committee, that he was going to name John Whitehead and John Weinberg co–vice chairmen of the Management Committee, and to George it was clear that Gus viewed them as his successors.

“The two Johns,” as we called them, were the logical heirs in any case. John Weinberg, Sidney’s son, was warm, not openly assertive, but highly effective and a great culture carrier in the firm. Born into the business, he had his father’s touch with clients, who loved him for his straightforward good sense. John Whitehead was extremely bright, lucid, and self-confident, and had a powerful strategic focus. He was central to converting the investment banking department from a highly individualistic operation, with Sidney Weinberg as the extraordinary star at the center, into an effective organization. Whitehead also began orienting the firm much more toward the world outside the United States. Our continued movement toward global involvement during my time at Goldman stood me in good stead during my years in Washington, when I looked at many of these same issues through the lens of public policy.

I learned a lot about management from both men. From Whitehead, I learned how to focus on being strategic despite the pressures of day-to-day business. He was always thinking in terms of where Goldman Sachs wanted to be years into the future. From Weinberg, I learned about working with people, both clients and colleagues, and how to treat the perspectives of others seriously and respectfully even when I disagreed with them. From both, I learned how to deal with joint management. There was initially much skepticism at Goldman about whether a co-CEO arrangement could work. When the Johns disagreed in a Management Committee meeting, there was no clear way to resolve the issue. But they figured out a way to function effectively that reflected their personalities. They talked to each other every weekend before the Monday morning Management Committee meeting and went through a list of issues. Weinberg was willing to let Whitehead take the lead on most, but if Weinberg felt strongly, he would assert himself. Though a problem would occasionally remain unresolved for an extended period because of their differences, their chairmanship showed me that a co-CEO structure can work—though I later came to believe that successful joint management is the infrequent exception, not the rule.

Ray Young represented all of the firm’s equity trading and sales activities on the Management Committee. In 1980, he retired, and I was one of three people who replaced him on the committee. Somewhat thereafter, I was asked to take on the problem of J. Aron & Company. A few years earlier, working with George Doty, I had tried to extend the arbitrage mind-set and our experience with trading in derivatives into building a commodities trading operation. Then, the year after I joined the Management Committee, Goldman bought J. Aron, its first acquisition since the 1930s. A hugely successful, family-owned commodities trading firm, Aron had sophisticated, well-established commodities operations, with connections all over the world. What neither we nor they realized, however, was that, because of various changes taking place, it didn’t have a viable business model for the future. Aron’s profits, which had been $60 million in 1981, fell to $30 million in 1982 and then to nothing in 1983.

Doty had responsibility for J. Aron and took the first difficult step of downsizing. With that done, the two Johns and George asked me to take charge of the problem. I could have said to myself that this might not work and could upend my position at Goldman Sachs. At the very least, I might have done some probabilistic analysis. But here, as in other major career changes that strongly attracted me, I didn’t calculate. I wasn’t at all cocky about my ability to turn Aron around, but neither was I anxious. Once I had the job, I just focused on trying to do what needed to be done. And I very much wanted the responsibility, because it was interesting and would enlarge my role at the firm. Moreover, Aron was a trading business with a strong arbitrage bent, so I felt suited to the task.

I walked around with my yellow pad for two or three months, just taking notes and trying to learn about the business before actually taking over. In the course of my inquiry, I found that the people doing the work had many thoughtful ideas about how to revise our strategy and move forward. After a while we changed the leadership, putting Mark Winkelman, who had been in the fixed-income department, in charge, reporting to me. Winkelman, who was born in Holland and worked at the World Bank before coming to Goldman, was extremely sophisticated about the developing business of relationship trading in bonds and foreign exchange. He had both the substantive background to understand Aron’s problems and the managerial skills to help set them right.

Together Mark and I worked with the Aron people to rethink the business model. Aron had been doing classic arbitrage, buying a currency or a commodity—such as gold—in one place and selling it as close to simultaneously as possible somewhere else. Aron was imaginative in crafting opportunities for classic arbitrage, for example by maintaining open phone lines to Saudi Arabia to trade gold and silver. This kind of trading had little risk, and Aron was intensely risk averse—so much so that when it lost track of its transactions, it would close in the middle of the day, to sort everything out and make sure it wasn’t holding an extra hundred ounces of gold on its books. The only exception to this model was coffee, where Aron acted as a large importer and trader.

One of the first conclusions others led me to see was that the relative stability of commodity prices, improved communications, and increased competition had eliminated the meaningful profit opportunity in Aron’s traditional business. Spreads were being squeezed, a reality that the Aron leadership seemed to have missed because of its great success in the past. I found over the years that the Aron experience was quite typical. Success often leads businesses and individuals to fail to notice change or to adapt to it, and so, eventually, to falter.

Mark and I determined that Aron needed to make several basic changes. The first was to focus on relative value arbitrage, or relationship trading, which Goldman was already doing in fixed income and equities. In the Aron context, that meant looking for distortions in the price relationship between different commodities or currencies, or between them and derivatives based on them, using interest rates and other factors to estimate the appropriate relative values. For example, short-term gold futures could be traded against long-term gold futures to profit when prices that seemed out of whack converged. That meant taking risks that the Aron people had always been proud of not taking, and with the firm’s own money. We decided to abandon the sure thing that no longer existed in favor of calculated risk taking. We also decided to greatly expand Aron’s foreign exchange trading, going from a pure arbitrage operation into relative value arbitrage, outright position taking, and increased client business—for example, helping businesses and individuals hedge against currency risk, which added to the services that Goldman could offer clients. Somewhat later, we went into trading oil and petroleum products, adding a vast new arena to our business.

These transformations at Aron required certain personnel changes, our most delicate undertaking. After extensive observation, we concluded that Aron had some extraordinarily capable people—who had already contributed greatly to rethinking the strategy—but that some others were so steeped in the old, risk-free way of doing business as to be unable to make the transition to a risk-based approach. We had to find places for those people elsewhere at Goldman or encourage them to move on. And we needed a process for recruiting. Hiring at the old Aron had been based on horse sense—somebody seemed as though he might make a good commodities trader. We formalized the process, looking for people whose experience and qualifications met our new needs. With all of these changes, we had reengineered Aron and the business started to work again—though in a very different way. We didn’t reach our ambitious $100 million goal the first year, but we exceeded it the second year and created a base from which Goldman earned enormous profits in the years after.

Like arbitrage, commodities and currency trading was an example of a good business based on calculated risk taking and that involved living with the large losses that sometimes—and inevitably—ensue. But the Aron transformation was also an illustration of how difficult change can be at big organizations. Michael Porter, a professor at Harvard Business School, argues that great institutions fail because, once successful, they become satisfied with themselves and stop changing, and the world passes them by. That was the case with Aron, which had lost its strategic dynamism. But even with an effective business model and a dynamic, strategic mind-set, a company needs a structure that works, a system for attracting capable people and putting them in the right jobs, and a culture in which people work together in a mutually supportive way.

But perhaps the most important management point about Aron was that the ideas for remaking its business came largely from the people who worked there, exemplifying my career’s experience that the people in the front lines of a business often have a better sense of what’s happening and what to do about it than the top executives. Our success at Aron was more evidence of Ray Young’s and Dick Menschel’s advice that to be most effective I was best off being surrounded by strong people, listening to them, and being sensitive to their concerns and quirks of personality—just as they had to be sensitive to mine.

   

AT THE SUGGESTION of Bob Strauss, the Democrats asked me to chair their 1982 congressional campaign dinner in Washington. Never having done anything like it, I wondered whether I would be able to raise enough money. Instead of saying yes or no immediately, I tried to get a better idea of how much I had to raise personally to be viewed as successful. People I spoke to named a figure of $100,000. So I called a family friend in Florida who had made a lot of money from my arbitrage advice. He and his partner said they’d each put up $20,000. With the $20,000 I could contribute under the legal limits and a few other ideas about where to raise money, I felt close enough to the $100,000, and I said okay. The dinner was successful: I raised much more than $100,000 on my own, and the dinner took in more than $1 million—large numbers by the standards of that era. Bob Strauss had told me that chairing that dinner put someone in a different position in the party. He was right. Soon after, both Walter Mondale’s and John Glenn’s campaigns sought my help for the 1984 election.

Substantively, I felt that the Reagan administration’s budget deficits created a serious threat to future economic conditions and that sooner or later we would pay the price. I remember speaking at a House Democratic Caucus meeting in the late 1980s. Congressman Barney Frank (D-MA) said, pointedly, that although I had been concerned about deficits for some time, the economy had continued to grow reasonably well. I replied that the laws of economics hadn’t been revoked. The timing of any market impact can be complicated, but the inevitable would surely occur at some point—as, indeed, happened not much later.

Many conservatives shared this concern. Martin Feldstein, the distinguished economist who was Ronald Reagan’s second chairman of the Council of Economic Advisers, argued strongly against large budget deficits, but his arguments were unsuccessful and he stepped down near the end of that administration’s first term. Gary Wenglowski, Goldman’s highly regarded chief economist, said he thought Reagan’s economic policy was the worst since Herbert Hoover’s. At that time, some conservatives argued that tax cuts should be accompanied by commensurate reductions in the cost of running the federal government. But that would have required program reductions that neither party was prepared to support, especially during a period when defense and entitlement spending were rising at a rapid rate. Another view was that tax cuts would generate sufficient additional growth to pay for themselves, which George H. W. Bush referred to in his 1980 presidential primary campaign against Reagan as “voodoo economics”—which seems to me about right.

The other aspect of Reagan’s policy that most concerned me was the failure to address the country’s social problems, many of which were clearly getting worse. Around that time, I read Ken Auletta’s book The Underclass. In vivid fashion, Auletta described the replication of poverty through generations. The book crystallized a lot of my thinking on the topic—though later, when I worked in a Democratic administration, I learned that the term “underclass” is no longer politically correct. (I was told that the acceptable alternative is “people who live in distressed areas, rural and urban.”)

I had some direct exposure to the problems of inner cities at meetings of a neighborhood group called the 28th Precinct Community Council in central Harlem in the 1970s. In searching for a way to get involved with these issues, I had met Warren Blake, an African-American police officer in charge of community relations for the precinct. Warren, a huge man with a personality to match, had strong ties to the community and cared deeply about what was happening to it. His wife’s family had run a prosperous funeral business in the neighborhood for many years, and the Blakes lived nearby in a large Victorian house that had once belonged to James A. Bailey of Barnum & Bailey Circus. When he was off duty, Warren sometimes drove a hearse. I remember a dinner Judy and I went to at their home. One of the other guests was a political activist from Papua New Guinea who proposed that Goldman Sachs finance a revolution in exchange for some of the country’s shrimp and timber concessions. I didn’t pursue this.

What I saw and heard at the 28th Precinct Community Council gave me a more personal feeling that it is just wrong that a country as wealthy as ours does not provide the resources to successfully address poverty that passes from generation to generation. The more I learned about these issues, the more I was convinced that there were approaches that would work if adequately supported. I also came to believe that the problems of the inner cities greatly affect all of us—no matter where we may live or what our incomes may be—through crime, the deterioration of public schools, the costs of social ills, and the lost productivity of a large group of people who are not being equipped to realize their potential. The belief that affluence can insulate is illusory.

And that helps explain why I am a Democrat. If you put all my views on public policy issues together, I wouldn’t fit neatly under any political label. In fact, many of my views, such as the importance of fiscal discipline to our country’s future growth and the centrality to our own well-being of American leadership on global issues such as trade liberalization, poverty, the environment, and terrorism, don’t really fit into any political camp. But when I look in both directions from the center, I find concerns that echo my own to a greater degree on the Democratic side, which has long seemed to me more committed to using government to meet the needs of middle- and lower-income people that markets by their nature will not adequately address.

However, I wasn’t necessarily in full agreement even with those who shared my concerns. Roughly twenty years ago, I heard Senator Ted Kennedy (D-MA) give a speech advocating a whole host of government programs that sounded worthwhile to me. The address was powerful and well delivered. But as much as I agreed with Kennedy’s goals and respected his commitment to them, I wondered, How are we going to pay for this? The focus by some social advocates on problems and programs—but not the means to fund those programs—bothered me. My deep involvement in the 1984 presidential campaign was largely driven by the conviction that we needed a President who combined Kennedy’s social concerns with a sense of fiscal responsibility.

Walter Mondale seemed to fit that bill. I met Mondale through his campaign chairman and former White House aide, Jim Johnson, who later became the CEO of Fannie Mae. Through Jim, I also met Mike Berman, the treasurer of Mondale’s campaign and a man wise in the ways of Washington. As I got to know Mondale better, he seemed to be very practical, with a well-known commitment to the plight of the poor, joined with a deep concern about our growing fiscal disarray. When the Mondale people asked me to be his New York State finance chairman, I accepted, after initially hesitating because once again I wanted to be sure I could raise enough money to be successful.

Though my place at the Mondale table came from fund-raising, my conversations with Jim Johnson, Mike Berman, and others in the campaign often shaded into economic policy and politics. Some people like opera. Some like basketball. I like policy and politics. Somewhere in the back of my mind, I also knew I wanted to work in the White House if the right opportunity should ever arise.

After Mondale’s defeat, gloom pervaded the Democrats. The party was widely seen as being in trouble and needing to reassess its direction. Some thought it was in thrall to the labor unions and interest groups, and that more centrist positions would be both better policy and more attractive to middle-class voters. Others felt just the opposite—that the focus needed to be on what they referred to as the “base.” I remember one dinner discussion at the house of Roger Altman, a fellow Wall Streeter who had served in the Carter administration and had worked with me on the Mondale campaign. Two other political strategists, Tom Donilon and the late Kirk O’Donnell, were also there, as was Jacob Goldfield, a colleague at Goldman Sachs. The focus of the discussion was that Reagan’s position was too simplistic to be serious policy but was easy to grasp and good politics: fight communism, cut taxes, and reduce government. How could views that reflected the true complexity of the issues be framed with enough political resonance to respond to such bumper-sticker simplifications? By the time dinner was over, no one had any very promising thoughts. In the years since, I’ve had many such conversations about this same conundrum.

   

THE ELECTION HAD consequences inside Goldman Sachs as well. In 1985, John Whitehead resigned and soon became George Shultz’s number two at the State Department, leaving John Weinberg as the sole senior partner. That same year, Steve Friedman and I became co-heads of the fixed-income division. We were roughly the same age and each of us had moved from a law firm to Goldman Sachs at about the same time, when that was rarely done. Steve and I became partners and joined the Management Committee within two years of each other, and we worked on many client assignments and firm matters together. Steve, a former national wrestling champion, was relatively conservative and a Republican. (Curiously enough, Steve assumed the same position in the White House at the beginning of the third year of the George W. Bush administration that I had at the outset of the Clinton administration.) Despite our differences—and perhaps despite our commonalities—we worked extremely well together for twenty-five years.

The fixed-income division at Goldman traded all kinds of interest-bearing instruments—government debt, corporate and high-yield bonds, mortgage-backed securities, fixed-income futures, options, and other derivatives. The business was big, with a lot of risk. And shortly after we arrived, the trading operation developed serious problems. Our traders had large, highly leveraged positions, many of them illiquid, meaning that they couldn’t be sold even at generous discounts to the price of the last trade. As losses mounted, Steve and I tried to figure out what to do.

I tend to think about fixed-income trading in three categories, although any single trade or position often involves two or even all three of them. The first is flow trading. You buy on the bid side of the market from clients and sell on the offer side, earning the spread between the two. The second is directional trading, based on judging the short-term direction of the market. You expect weaker economic numbers or a stock market dip to push bond prices higher. So you buy at $98¼, expecting to sell at $99 or even higher. The third is relative value trading, or fixed-income arbitrage. You decide that the relationship between two different securities is out of line and likely to return to form, based on valuation models, historical experience, and judgment; for example, the five-year Treasury bond is trading at an unusual discount to the ten-year. So you buy the relatively cheap bond and sell the relatively expensive bond, anticipating that their normal relationship will return.

Relationship trading was at the heart of the trouble that developed in the fixed-income department. Bonds and derivative products began to move in unexpected ways relative to each other because traders hadn’t focused on how these securities might behave under the extremely unlikely market conditions that were now occurring. Neither Steve nor I was an expert in this area, so our confusion was not surprising. But the people who traded these instruments didn’t fully understand these developments either, and that was unsettling. You’d come to work thinking We’ve lost a lot of money, but the worst is finally behind us. Now what do we do? And then a new problem would develop. We didn’t know how to stop the process.

We had lost about $100 million. Today that wouldn’t mean much, but in that world at that time, it was very meaningful. And it wasn’t just the losses to date but also a question of what those losses portended for the future. You have no way of knowing when a downward spiral will end or if the next period will be even worse. The fixed-income division had contributed greatly to Goldman’s overall profitability. Suddenly, our biggest trading operation had gone sour, and we didn’t understand why or what the future might bring.

Steve and I went into the trading room and said, “Let’s all sit down and try to understand what we’re holding. If we have positions we shouldn’t have, let’s get rid of them.” Leaving aside the psychological factors that incline most traders to resist taking losses, what soon became clear was that they hadn’t fully anticipated the behavioral characteristics these securities could have when conditions changed substantially. Many of these bonds had embedded or implicit options. As a simple example, the firm was trading mortgage-backed securities that represented the loans people take out to finance homes. As interest rates declined, people refinanced 13 and 14 percent mortgages at 9 and 10 percent. That meant that our bonds didn’t rise in line with the fall in interest rates and in some cases were paid off early—creating a loss on a position that had been hedged with Treasury bonds. A similar problem affected corporate bonds. As interest rates declined, bond prices rose to a point where the corporate issuers might exercise call provisions—the right to pay off an outstanding bond, usually to refinance at lower rates. When a bond was at $80, the borrower’s right to redeem at $102 was often ignored. Then, when the bond market had a massive rally, the call provisions did begin to reduce the otherwise expected premium, while the short position rose and created a net loss. What happened to us represents a seeming tendency in human nature not to give appropriate weight to what might occur under remote, but potentially very damaging, circumstances.

Another lesson in these 1986 bond market losses—which I had learned through mistakes made in arbitrage years earlier—was to think not only about the odds of making a profit on each trade, but the limits on tolerance for loss in the event market conditions became much worse then expected. The issue wasn’t simply financial staying power. A trader and his firm had to know the outer limits of what they were willing to lose relative to their earnings and balance sheet. A related problem was liquidity. Traders tend to assume that their positions will always be salable at very close to the last market price. When markets are doing reasonably well, they say, “Well, if I don’t like something I’ve bought, I’ll just kick it back out.” But when conditions deteriorate severely, liquidity diminishes enormously. Traders often can’t sell bad positions except at enormous discounts, and sometimes not at all. Then they may be forced to sell good positions to raise money. Thus, during periods of great market duress, investments can react in unexpected ways. Securities that have no logical relationship may suddenly move in tandem while securities that do have a logical relationship may diverge. Unexpected losses can develop rapidly and be huge.

Dealing with the 1986 meltdown in fixed income probably helped reinforce the position Steve and I had acquired as heirs apparent, and in 1987, John Weinberg appointed us co–chief operating officers of Goldman Sachs. Even after we assumed our new positions, however, I remained co-head of trading and arbitrage and the Management Committee member responsible for J. Aron—albeit with many fewer day-to-day responsibilities in each of those areas. Steve, on the other hand, chose to relinquish his position as co-head of the investment banking division, although he remained very involved in it.

Beginning in early 1987, we became embroiled in a crisis which—painfully—prepared me to help manage several other high-profile crises in subsequent years. A partner of ours, along with two people from another firm, was arrested for insider trading in a highly controversial case. Based on our counsel’s investigations, we believed our partner was innocent, and we stood by him through the entire ordeal. All the original charges were dropped. No new charges were ever brought against the other two. Our partner ultimately pled guilty to a single count unrelated to the original charges. His counsel advised him that he shouldn’t be found guilty but that, with a wealthy defendant in a jury trial in a matter of this sort, there were no guarantees.

For almost two years, Steve and I were engaged with this case and the issues it created for the firm. In dealing with this problem, we learned a number of important lessons. First, in managing a crisis, it is crucial that you not allow it to interfere significantly with conducting your business. One way to minimize this risk is to designate a small team to deal with the crisis. Steve and I told everyone else at Goldman that we’d keep them posted, but that their responsibility was to remain focused on their work. Second, we learned the importance of getting out and seeing your clients during a crisis. Clients have a lot of questions and the firm has to answer them. Third, you have to communicate with your own people frequently, especially since the media is likely to emphasize the negative. Fourth, regarding the media, my own experience over many years, in a number of crisis situations in both the public and private sectors, has convinced me that you simply have to grit your teeth and live through the rough early coverage—you ordinarily can’t affect the initial firestorm. What you can do is be candid about whatever the situation may be, affirmative with respect to your commitment to address your problems—if that is relevant in the particular situation—and confident about the future. Over time, your answers should begin to get more attention and the coverage should become more balanced. Finally, in all of this, you need to resist the tendency to become a little paranoid, thinking that people are looking at you differently when, in most cases, they’re not. It’s important not to be oversensitive, and to treat everyone normally. With all of the potential for something to go wrong in any big company, no matter how well run, and with the attendant media and political firestorms that can erupt in short order, knowing how to cope with a crisis can quickly become critically important for any senior manager.

At the end of 1990, John Weinberg stepped down, and Steve and I were named co-chairmen. Following the Weinberg-Whitehead example, we didn’t try to divide up responsibilities or subject areas. We told everyone to assume that either of us could speak for both and that touching one base was sufficient. This worked because we shared the same fundamental views about the firm, trusted each other totally, kept in close touch, and were both analytically minded in our approach to problems. When this structure does work—and that is a rarity—the advantages are substantial: there are two senior partners to call on clients and two people who can work together on issues with no hierarchical baggage, and who can reinforce each other in discussions with the rest of the organization. Also, when difficulties arise, having a partner reduces the feeling of loneliness at the top.

Absolutely key to our partnership was that when we disagreed, neither of us had his ego invested in winning. Steve and I had a rule that the one who felt more strongly would prevail, or at least have the decision more toward his direction. If one of us felt 80–20 and the other 60–40, the 60–40 one would say, “I sort of disagree, but if you’re eighty–twenty we’ll do it your way—or someplace in between, but more your way than mine.” In the rare cases when we disagreed and both felt strongly, we worked it out. Another proviso was that we usually—although not always—deferred to the more risk-averse position.

On maintaining a meritocracy, protecting the culture of the firm, and focusing on our customers, we agreed. But within those parameters, our views sometimes differed. As an example, Steve strongly favored greater differentiation in partnership shares, based on meaningful distinctions in performance. I, on the other hand, thought the possible ill will of those not favored outweighed the benefit of favoring the best performers—except where the difference was truly major. Over the years, I had seen partners who earned millions of dollars a year become deeply unhappy over tiny distinctions in partnership shares. (Bob Strauss once captured this dynamic when he said that a lawyer at his firm earning $90,000 a year—this was some time ago—and offered a $10,000 raise with the stipulation that a peer next door would get a $20,000 raise would prefer no raise at all to someone on his own level being paid even more.) Steve referred to my inclination to avoid conflict-provoking distinctions as “solving for maximum social harmony.” Because he felt more strongly than I did, we agreed to increase differentiation, although less than Steve would have done on his own.

My partnership with Steve was in some respects a forerunner of the relationship I had as Treasury Secretary with Larry Summers and Alan Greenspan. None of these people is a shrinking violet. But because of mutual respect, trust, and our analytic approach to issues, these relationships worked. Alan had somewhat different starting points on some issues, but through financial crises, G-7 meetings, currency interventions, and much else, we almost always analyzed our way through to agreement. Steve and I didn’t worry that someone might consider one of us weak because the other’s view had prevailed, nor did Larry, Alan, or I. The overriding drive in both relationships was to reach the best decision.

   

WHEN STEVE AND I first became co-chairmen, I found the difference between the senior position and what we had been doing far greater than I had expected. With John Weinberg as senior partner, the ultimate approval and responsibility were his, even if we had operating responsibility. Once Steve and I became senior partners, the ultimate responsibility was ours. Larry Summers said the same thing to me about becoming Treasury Secretary. When he was deputy secretary, he felt that the difference between his job and mine was small. After becoming Secretary, he realized that the difference was enormous.

In those days, Larry Tisch, the CEO of Loews Corporation and later of CBS, used to tell me, “Bob, you worry too much.” I’d say, “Larry, you don’t understand. There’s a lot to worry about.” One worrisome issue was Goldman’s lack of a permanent capital base. A firm like ours needed a lot of capital to support what had become a massive global trading operation, to withstand difficult times, and, later, to be competitive in investment banking with the commercial banks. As it was a private partnership, each new retiree could withdraw his share of the capital within a relatively brief period. In a public company, capital remains in the company, and a retiree simply sells his stock on the public market. If conditions were difficult and partners became nervous, Goldman could face a run on the bank. Building capital is thus far more difficult in a private firm. And finally, the partners were at risk not only for the money in the firm but for their entire net worth—a source of great concern in the Penn Central bankruptcy. For these same reasons—as well as the simple desire to cash out—all of our major competitors had already gone public or merged into larger concerns. Steve and I were convinced that the way to deal with these issues was to become a limited-liability corporation and sell stock to the public.

One argument against going public was the flexibility of being able to periodically adjust partnership percentages among the partners, to reflect performance and changes in seniority. Another was the mystique of being private—especially after all the other big Wall Street firms had gone public or merged. The initial public offering proposal presented in 1986 was rejected, largely because the younger partners wanted to preserve flexibility so their stakes could grow more easily. And that structure continued to work for more than a decade. Years after both of us had left, the inevitable eventually happened and Goldman finally did go public.

   

IN EARLY 1988, I met Governor Michael Dukakis a few times and was impressed with his intelligence. Although I shared the popular view that he was somewhat stiff as a candidate, I raised money and contributed a bit of advice to his campaign. At one point, Dukakis was way ahead of George Bush in the polls, and after his defeat many in the party felt bitterly toward him for the way he had handled his candidacy. Issues about his campaign aside, I still thought that expressing moderate Democratic views—or, for that matter, any sensible views that reflected the complexity of the underlying issues—in ways that resonated politically was extremely difficult. The political system’s bipartisan failure to address the growing deficit demonstrated the imperative need to figure out how to do so. The country remained in denial about serious social issues as well. Our public education system was deeply troubled, and life in the inner cities was getting worse. I wondered whether the country would muster the political will to address its problems. The alternative to facing up to these problems, as I discussed at a dinner Bob Strauss held for me in Washington after I became co-head of Goldman Sachs, was the risk of inexorable national decline.

I hadn’t decided among the Democrats who were considering running in 1992, but I was looking around. I hosted—along with David Sawyer, a well-known Democratic political consultant and Oscar-nominated documentary filmmaker who died at a young age—a series of small dinners at which roughly fifteen business and media people talked with candidates and potential candidates. Among others, we had Senators Tom Harkin (D-IA), Dale Bumpers (D-AR), Joe Biden (D-DE), and Bob Kerrey (D-NE). Bill Clinton was our guest at dinner in mid-1991 and was enormously impressive. I’ve been to many events where a candidate spends much of the time talking. For more than three hours, Clinton engaged in a real dialogue—a serious give-and-take—on the issues important to us. At the end of the dinner, I said to Lew Kaden, a New York lawyer and Columbia law professor deeply involved in Democratic politics, “This guy Clinton is amazing. It’s remarkable how well he understands this stuff.” But Clinton expressed uncertainty about running because of the effect a campaign might have on his family.

Almost a year later, in May 1992, when Clinton had not only decided to run but had pretty much locked up the nomination, he drew together a few so-called advisers in Little Rock to discuss economic issues. The group included several Wall Street investment bankers—Roger Altman and my fellow Goldman Sachs partners Ken Brody and Barrie Wigmore—as well as a centrist economist named Rob Shapiro and three friends of Clinton’s, Robert Reich, Ira Magaziner, and Derek Shearer, who had well-developed views on an active role for government. I had no illusions about our position as “advisers.” We were mainly surrogates intended to lend credibility to Clinton’s economic policies, which, in reality, were set inside the campaign organization. But at this meeting, Clinton took the extraordinary step of taking a day off from the campaign—with no media coverage—to assess his economic proposals and see whether positions developed under the pressure of the campaign made sense for governing. He wanted to stop running for a day in order to check his course.

Doing that showed remarkable seriousness of purpose for a candidate in the midst of a campaign. The group of us flew to Little Rock and spent a number of hours with the governor and Hillary, whom I had never met before. Our group had a range of opinions, but we agreed on the most important issues—the importance of deficit reduction, the need for greater investment in education and health care, and the benefits of trade liberalization. These remained the central components of Clinton’s economic strategy for his eight years in office. Within the context of that consensus, there were differences in emphasis. Ken Brody, Rob Shapiro, Roger Altman, and I emphasized reestablishing fiscal discipline more strongly. Reich, Ira Magaziner, and Derek Shearer tilted somewhat more toward investment in education and training.

Our group was asked to draft an economic statement that subsequently evolved into the economic section of the campaign platform, “Putting People First.” I suggested that Ken Brody, who shared my focus on deficit reduction, draft the document. But Gene Sperling, who had just joined the campaign and instantly became its economic engine, became the chief draftsman, with some input from the rest of us. I’d known Gene slightly from the Dukakis campaign, where he had played a more junior role. Gene was bright, knowledgeable about economics, extraordinarily productive, and highly adept at crafting a message. He was also slightly disheveled and almost impossible to get on the phone except in the middle of the night. When, or whether, he slept was a great mystery. But on a substantive, as opposed to a stylistic level, Gene was well ordered. He understood what an economic platform for a campaign should look like and how to meld economic policy, politics, and communication.

Gene would sometimes have the “outside advisers” talk to the press when Clinton discussed economic issues or announced a new proposal. And so I began learning how to engage with the media in a Washington context. Gene told me that it was crucial to get my points across in my response to questions—in effect, to be responsive but from my point of view. Throughout my years in Washington, I never lost my wariness of the media, developed from my earlier experience in crisis response at Goldman, but I did develop great respect for many of the people who covered us and tried to respond seriously to those who were serious with us. One of the ironies of my time in Washington is that by the time I left, I felt that some of the most knowledgeable and interesting people I had gotten to know there were journalists, while at the same time I continued to have reservations about the way the media as a whole functioned.

I spent election night in Little Rock, celebrating Clinton’s victory. A couple of weeks later, I was summoned back to meet with the President-elect. His mood was upbeat, and he jokingly said, “I’m the leader of the free world,” as I shook his hand. We talked for a couple of hours, hardly at all about economic policy, which I told Judy seemed rather peculiar. I wasn’t even quite sure what his purpose in seeing me had been. Later I realized that this had indeed been an interview and served a less obvious but important managerial purpose: Clinton was getting a sense of what I’d be like to work with and how I would work with others—a sensitivity about personalities and the interaction of administration members I would observe many times in the years ahead. I remember Clinton noting that despite being a senior partner at Goldman Sachs, I’d developed very comfortable working relationships with Gene and other younger people on the Little Rock campaign staff, such as Gene’s deputy Sylvia Mathews. In fact, I liked working with more junior people, who were often closer to the specifics of what was going on and had more time to speak with me. I thought—correctly—that I had much to learn from Gene and Sylvia about politics, campaigns, and much else.

I left the meeting more impressed than ever with the President-elect, but without any idea of what, if anything, would happen next. Clinton did ask who I thought should be Treasury Secretary. I felt I didn’t have the experience in dealing with Congress, the media, policy, or politics to handle the job at that point and I recommended Senator Lloyd Bentsen (D-TX), whom I knew pretty well and respected, and who as chairman of the Senate Finance Committee was well equipped for the job. I also remember talking to Clinton about coordinating the many offices and agencies that participate in economic policy. Clinton wanted to create an Economic Security Council—renamed the National Economic Council after the campaign—to do for economic policy what the National Security Council did to coordinate foreign policy.

Shortly thereafter, Warren Christopher, the head of Clinton’s transition team, called to sound me out about jobs. I wasn’t at all sure I would be offered a position, although I was certainly interested. Christopher, whom I already knew from being on the board of the Carnegie Corporation, which he chaired, said, “If you don’t become Treasury Secretary, would you be interested in running the National Economic Council at the White House?” I told Chris, as everyone calls him, that I would. Some time later, I was in Frankfurt, Germany, on a business trip, when the phone rang in my hotel room at 2:30 in the morning. Chris was calling to formally offer me the NEC job. Without any further deliberation, I said yes. Then I went back to sleep.

Christopher had seemed surprised that I’d been receptive when he first raised the NEC job. He may have thought that, as a senior partner at Goldman Sachs, I would consider a staff position at the White House, as opposed to a cabinet post, a step down, even though all economic policy was to be coordinated through the NEC. In retrospect, although not for reasons of status, I might have been wiser to think more seriously about the pluses and minuses of the NEC job before agreeing. For starters, Judy and I hadn’t fully discussed this—she was more surprised than Christopher when I called from Germany and told her I’d accepted the job. Beyond that, a rigorous weighing of all factors might have led me to conclude that the odds of succeeding were not so high. The NEC was a new idea, and I was untested in Washington. Many cabinet officers and senior White House staff could view the NEC as an added layer or a diminution of their authority.

But I just said yes. I’ve been asked a few times how this decision squared with my probabilistic approach to decision making. The answer, I think, is that I responded based on years of wanting to be part of an administration. However, a thorough probabilistic analysis would almost surely have come out the same way, because of the overwhelming positive weight I would have placed on this opportunity. My fascination with Washington and the political process—and my desire to get involved in issues I cared about—overrode all other considerations, and an unconscious probabilistic process may well have underlain my decision. As concerned as I was about the difficulty of making the NEC effective, I didn’t focus on the hazards; I focused instead on how much I wanted to do it.

Leaving Goldman Sachs after twenty-six years wasn’t easy—for me or for the firm. At a hastily convened partners’ breakfast meeting in our conference room on the thirtieth floor, I said good-bye. The company had had 650 people, all in the United States, when I had started in 1966 and had grown more than tenfold by the end of 1992 into a global institution. I said the firm was in extremely good hands—Steve, the others on the Management Committee, and the larger partnership. I also tried to explain something of what the firm had meant to me. I didn’t mention the story that had caught my attention in The New York Times all those years before, but it came to mind as I left. Goldman had given me a career filled with interest and challenge on the inside—with people I respected and whose values I shared—and a base for involvements on the outside. Armand Erpf—this man I’d never met—was right about what it meant to have that kind of base. Everything started from there.