五彩紙屑在舞池上空盤旋。一隊身穿緊身衣的侍者依次走到中央，其中一位拿著一瓶在黑暗中熠熠閃光的唐·朱利奧龍舌蘭酒，另一位高舉一塊寫著“D SOL”的廣告牌，以表達對蘇德巍（David Soloman）的敬意。然后，午夜十二點半，在位于邁阿密楓丹白露酒店地下室的LIV夜總會，高盛集團的首席執行官閃亮登場。蘇德巍拿起一對厚實的白色耳機，走到電唱機轉盤后面，旋即陶醉在他的音樂DJ世界中。

這是10月的最后一個星期六。為慶祝萬圣節，這家位于南海灘的夜店已經改造成為“51區”的模樣，洋溢著一股濃濃的世界末日風；進入夜店的人皆是一副天外來客打扮——惡魔角、兔子耳朵、吸血鬼披風，不一而足。（抵達這家夜店時，我在門口報出蘇德巍的名字，保鏢一臉茫然。“他在這里叫D-Sol？”他點點頭，解開了門口的絲絨繩子。）蘇德巍身穿黑色T恤衫、白色褲子和一雙粉色的低幫匡威鞋，皺著眉頭，看上去比較嚴肅。在他身后的DJ間，兩位女士打扮成“便便”表情符號，還特意用幾卷廁紙和一個柱塞來裝飾她們的服裝。在接下來的一個小時，蘇德巍將為現場的狂歡者助興。他將槍炮與玫瑰（Guns N’ Roses）和白色條紋樂隊（White Stripes）的經典搖滾樂段與20世紀90年代的嘻哈音樂和舞蹈節拍融合在一起，偶爾還會揮舞拳頭。置身于低溫霧炮和閃爍激光燈烘托的氛圍之下，你可以感受到一股低音脈沖穿過胸腔。蘇德巍雙手按著旋鈕，隨音樂節拍舞動。然后，他播放了一首Bingo Players創作的小調，其歌詞主要由一段非常厚臉皮的副歌組成：“每個人都想知道我多么不在乎，我多么不在乎。”

鑒于許多人都在質疑華爾街最大牌公司的首席執行官是否應該長時間待在這樣一處喧嘩之所（凌晨3點，當高盛董事會的其他成員都在睡覺的時候，他還在夜店喝龍舌蘭酒），這很難不被解讀為一種滑稽的反駁。2017年，蘇德巍的夜店DJ身份被《紐約時報》（New York Times）曝光，彼時的他還是兩位角逐高盛最高職位的聯席總裁之一。在高盛內部，蘇德巍向來被視為一位極其專業、進取心很強、對閑聊毫無耐心的管理者。他當時擔心，人們不再把他當回事了——錯誤地把這份兼職等同于某種中年危機。一些顧問建議他終止其DJ生涯。“我想了一會兒，我能這樣做嗎，我不能做這個嗎？” 蘇德巍回憶道。他把演出和Spotify音樂播放的收入捐給了致力于戒除毒癮的慈善機構。（他籌集的資金總額據說達到了六位數。）這些疑慮很快就煙消云散，尤其是在他當時的老板、時任首席執行官勞埃德·貝蘭克梵和其他人表態支持之后。“你知道嗎，這就是我，沒有人會告訴我不要打高爾夫。”蘇德巍現在說。“為什么我不能做夜店DJ，就因為我是首席執行官？”

You see the confetti before you see him: It swirls over the dance floor as a parade arrives in his honor—leotard-clad servers bearing a glow-in-the-dark bottle of Don Julio and a billboard that reads “D SOL.” It’s then, at half past midnight, at the LIV nightclub in the basement of Miami’s Fontainebleau hotel, that the chief executive of Goldman Sachs emerges, picks up a chunky pair of white headphones, and steps behind the turntables and into his disc jockey alter ego.

It’s the last Saturday in October, which means it’s Halloween in South Beach, and the club has been made over into an apocalyptic Area 51; the crowd has shown up in a mix of extraterrestrial outfits along with devil horns, bunny ears, and Dracula capes. (When I arrived, I’d given David Solomon’s name at the door, eliciting nothing but a blank look from the bouncer. “D-Sol?” He nodded and opened the velvet rope.) Solomon, brow furrowed, looks comparatively sober in a black T-shirt, white pants, and pink Converse low-top Chuck Taylors. Behind him in the DJ booth, a couple of women appear to be dressed as the poo emoji, a costume they’ve accessorized with rolls of toilet paper and a plunger. For the next hour, Solomon will entertain the revelers, occasionally pumping his fist as he mixes rock guitar riffs from Guns N’ Roses and the White Stripes with 1990s hip-hop and wait-for-the-drop dance beats. You can feel the bass pulse through your rib cage, punctuated by blasts from cryogenic fog cannons and strobing lasers. Solomon bops along, his hands on the knobs. Then he drops a Bingo Players ditty whose lyrics consist primarily of the cheeky refrain, “Everybody wants to know how little I care, how little I care.”

It’s hard not to interpret it as a droll retort to anyone who would question whether the CEO of Wall Street’s most storied institution—who sipped tequila at the club until almost 3 a.m.—belongs in such a hotspot while the rest of the Goldman Sachs board is presumably asleep. Solomon was outed as DJ D-Sol by the New York Times in 2017, when he was still one of two copresidents gunning for the top job at the firm. Known at the office as an über-professional, hard-charging manager with little patience for small talk, Solomon was anxious that people would no longer take him seriously—mistakenly equating the side gig with some sort of midlife crisis. Some advisers told him to hang it up. “I thought for a minute, Well, can I do this, can I not do this?” recalls Solomon, who donates the proceeds from his gigs and Spotify plays to drug-addiction-related philanthropies. (He says the total he has raised is in the six figures.) Those doubts soon dispersed, especially after his boss at the time, then-CEO Lloyd Blankfein, and others endorsed it. “You know what, it’s who I am, and nobody would tell me not to play golf,” Solomon says now. “And why shouldn’t I—because I’m a CEO?”

蘇德巍對傳統行事方式的不敬，甚至是蔑視——盡管這種態度可能會在高盛的老前輩中引起爭議，并會得罪其中一些人——也是他得以登上高盛之巔的決定性品質。去年10月，在他擔任首席執行官一周年后不久，該銀行迎來了其150歲生日；高盛是華爾街仍然以其原名稱運營的最古老的大銀行。在紐約一場億萬富翁云集的雞尾酒會上，蘇德巍略帶自嘲地贊揚了公司的傳統：“在此之前，公司領導總是在正確的時間選擇正確的人出任首席執行官。”

他的任期恰逢一場歷史未有之變局。華爾街正在經歷一場前所未有的技術革命，而在一些分析師看來，高盛可能需要一場比其他任何大銀行都更大的變革。在此之前，貝蘭克梵執掌高盛長達十數年之久，他引領這家銀行走向輝煌，但當他掛印離職時，高盛卻陷入前所未有的不確定性。在金融危機爆發前的幾年里，高盛憑借高風險交易斬獲天文數字般的回報——然后比其他銀行更早地出清了一些籌碼，并由此安然度過這場危機。高盛在2008年扭虧為盈，隨后在2009年錄得創紀錄盈利。這番翻云覆雨的操作給公司領導人留下了一定的回旋余地，使他們能夠從容應對股東的質疑。只要能夠持續帶來回報，投資者就會容忍高盛不透明的財務披露方式，以及巨額薪酬和福利支出。“高盛是毋庸置疑的華爾街之王，他們做得非常好。”富國銀行的分析師邁克·梅奧說。

但此后的十年則是另一番景象。金融監管和電子交易的興起蠶食了利潤。債券和大宗商品等資產的做市業務曾經是高盛的主要收入來源，但鑒于這些業務現在大體上實現了自動化，它們已經變得支離破碎——2018年占高盛凈收入的37%，低于2009年的72%。（盡管高盛如今在這類交易中占據更大份額，但它們的凈收入占比還是下跌了。）而且，由于高盛的老方法在很長一段時間都行之有效，它從未像競爭對手那樣在后衰退時期進行深刻反省。“高盛是自身成功的犧牲品。”梅奧說。“他們并不覺得自己需要像其他表現不佳的同行那樣不斷革新。”

現在表現不佳的是高盛。雖然該公司仍是并購和IPO交易領域的老大，但其收入自2010年以來下降了6%，2018年跌至366億美元，今年還將進一步下滑。在過去的五年中，其股價表現落后于金融板塊和標準普爾500指數，是美國主要銀行中回報率最低的。此外，在高盛因其“大到不能倒”的地位而獲得聯邦巨額紓困資金，進而被諷刺為“吸血烏賊”之后，這家華爾街大鱷再次面臨一場嚴峻的聲譽危機。兩名前高盛高管因為合謀盜竊馬來西亞投資基金1MDB數十億美元而被起訴，這一丑聞可能導致該行付出高達50億美元的和解費用。

蘇德巍身高6英尺2英寸（約1米9）——他就坐在貝蘭克梵曾經坐過的同一間辦公室里——比他的前任整整高出一頭。他身材魁梧，氣度不凡，有一種咄咄逼人、時而好斗的氣勢。他摒棄了貝蘭克梵特有的幽默，喜歡直言不諱，愿意反思貝蘭克梵不會觸及的高盛慣例。

他需要這些品質來完成接下來的任務。蘇德巍敏銳地意識到，他需要進軍高盛此前忽視的新業務（從大規模定量交易、管理企業支付，到為不太富裕的消費者提供銀行服務，再到與小公司做生意）來推動高盛的收入增長。他還需要從公司內部榨取更多的效率。這可能意味著高盛員工不得不更加努力地工作，才能夠賺取讓高盛上層變得極為富有的薪水，同時還要吸引年輕有為的金融精英加入公司人才儲備庫。私募股權公司Stone Point Capital的董事長，高盛前首席執行官史蒂夫·弗里德曼表示：“我不覺得他是那種安于現狀的人，這是一種恭維。”

在蘇德巍出任首席執行官之際，弗里德曼建議他保留“一小部分核心”文化價值觀，但“其他一切都必須得改變”。蘇德巍以這一信條為指引。今年3月，他撕毀了該公司由來已久的著裝規定，那是一份長達35頁的卷宗，充斥著許多關于西裝、領帶和鞋子顏色的過時規定。今年夏天，高盛悄無聲息地取消了員工背景調查中的毒品檢測要求。同時被取消的，還包括過去由大廈安保人員實施的辦公室拍照禁令；蘇德巍現在經常在他的Instagram賬戶上發布這類照片。“這個組織有很多我想簡化的官僚規定。”蘇德巍說（盡管他并沒有因為改變毒品檢測和拍照政策而受到贊揚）。“我認為我們可以做一些工作，讓世人對高盛多一點欽佩和尊重，少一點嫉妒和恐懼。”

除此之外，他還試圖破除高盛通過一種由來已久的神秘且等級森嚴的企業文化而建立起來的“象牙塔”聲譽。在鋪著桃花心木的行政樓層，每當來訪的同事被要求坐在等候室時，他總是非常惱火；他允許員工繞過兩層接待員，徑直敲門。他喜歡自己拿咖啡，不請自來地出現在部門會議上，并且拒絕聆聽事先準備好的談話要點。“他討厭任何形式的禁衛軍。”高盛的辦公室主任約翰·羅杰斯說。在這樣一個數十年來主要通過群發語音郵件與3.8萬名員工進行溝通的公司，蘇德巍要求他的助手想辦法在高盛總部11層“天空大廳”（在電梯銀行和自助餐廳之間）召集一個讓全球員工都參與其中的硅谷式市政廳會議，以方便他即興開會。（該公司最終放棄了這個想法，因為協調全球73個辦事處所需的后勤保障工作實在過于困難和昂貴。）

讓一些董事會成員感到震驚的是，去年賺了2300萬美元的蘇德巍習慣性地選擇乘坐地鐵在紐約穿行。“說真的，為什么不坐地鐵呢？這樣更快、更有效率。你知道的，紐約市市長可以坐地鐵。為什么高盛的首席執行官就不能呢？”他說。

曾經是一家私人合伙企業的高盛已經上市20年之久。蘇德巍屢屢表示，他希望最終把高盛經營得像一家上市公司。盡管貝蘭克梵從未參加過高盛的季度業績電話會議，但蘇德巍不想錯過任何一場。在明年1月29日的投資者日，高盛將首次邀請約300名股東、分析師和監管人員光臨其曼哈頓總部，并公布蘇德巍的新創新：一項時長為三至五年的戰略計劃。這似乎沒什么了不起的，但直到現在，這家深信市場不可預測的公司只編制每個日歷年的預算。蘇德巍說：“我們投資的方式就像是把樹枝揉搓在一起，然后在上面撒一點水。”換句話說，高盛的員工每年都得想辦法比上一年賺更多的錢，但任何需要長期投入才能在日后收獲回報的事情都是不予考慮的。“不能僅僅因為我們是高盛，就得跑得更快、跳得更高、獲得更多利潤。”首席運營官約翰·沃爾德倫指出，高盛必須專注于投資。“你必須把錢投到隨時間增長的土地上。”

這一切都意味著高盛迎來了一個關鍵時刻，而蘇德巍將在很大程度上決定這家公司是再度稱雄，還是逐漸衰退，淡出視野。作為一個在DJ身份被曝光前很少在媒體露面的人，他仍然在適應與高盛王座相伴而來的聚光燈；今年秋天接受采訪時，蘇德巍坦言他正在努力地“過正常人的生活”。他的合作者表示，這種務實、樸實無華的心態或許正是這家銀行目前所需要的。邁克爾·戴爾表示：“我認為他正在做正確的事情，以推動高盛不斷演變，并參與21世紀的競爭。”他領導的戴爾科技公司成為高盛客戶的時間比蘇德巍在高盛工作的時間還長。（他弟弟創辦的初創公司Clarity Money去年被高盛收購）。“超級油輪通常不會轉向。”

Solomon’s irreverence, even disdain, for The Way Things Have Always Been Done, feather-ruffling and polarizing though it may be for the bank’s old-timers, is also the defining quality that has put him atop the venerable Goldman Sachs. Shortly after his first anniversary as CEO in October, the bank celebrated its 150th birthday; it’s the oldest of the big Wall Street banks to still operate under its original name. At a billionaire--studded cocktail party in New York to commemorate the occasion, Solomon, with a shade of self-deprecation, praised the firm’s tradition of leaders who were “always, up to this point, the right people at the right time.”

His time is shaping up to be one characterized by historic transformation, as Wall Street undergoes an unprecedented technological revolution—and Goldman, in the view of some analysts, may undergo greater change than any other big bank. It comes on the heels of Blankfein’s dozen-year tenure, which started in glory and ended in unprecedented uncertainty. In the years leading up to the financial crisis, Goldman Sachs made astronomical returns on the back of high-risk trades—then reversed some of those bets sooner than others did, riding out the crash on top of Wall Street. Goldman turned a profit in 2008, then racked up record-high earnings in 2009. That rainmaking afforded its leaders a certain leeway with shareholders. So long as they delivered the returns, investors tolerated their opaque financial disclosures and lavish expenditures on pay and perks. “They were really the kings and queens of Wall Street, they did so well,” says Mike Mayo, a bank analyst with Wells Fargo.

But the decade since has been a different story. Financial regulation and the rise of electronic trading have dampened profits. Market-making in assets like bonds and commodities, once Goldman’s bread and butter, has turned to crumbs now that it is largely automated—representing 37% of the firm’s net revenues in 2018, down from 72% in ’09. (That drop has come even though Goldman now commands a bigger share of trading in those industries.) And because Goldman’s old ways worked so well for so long, it never went through the postrecession reckoning its rivals did. “Goldman Sachs was a victim of its own success,” Mayo says. “They didn’t feel the same need to evolve as their underperforming peers.”

Now Goldman is the one underperforming. Though it’s still No. 1 in mergers and acquisitions and IPO dealmaking, its revenue has declined 6% since 2010, to $36.6 billion in 2018, and is on track to decline further this year. Its stock has lagged both the financial sector and the S&P 500 over the past five years—yielding the worst returns of any major U.S. bank. It is also facing its biggest reputational crisis since the backlash against its too-big-to-fail bailout famously led to its caricature as a “vampire squid.” Two former Goldman Sachs executives have been indicted for conspiring in a multibillion-dollar theft from the Malaysian investment fund known as 1MDB, a scandal that could cost Goldman as much as $5 billion to settle.

Enter David Solomon. At six-feet-two, he stands—literally, in the same office where Blankfein used to sit—a full head taller than his predecessor. A broad-shouldered and imposing figure, Solomon has an aggressive, at times combative style that forgoes Blankfein’s characteristic humor in favor of blunt straight talk and a willingness to rethink Goldman conventions Blankfein wouldn’t touch.

He’ll need those qualities for the tasks ahead. Solomon is acutely aware he needs to jump-start Goldman’s revenue growth by forging into new businesses the firm had previously overlooked (from large-scale quant trading and managing corporate payments to being banker to less wealthy consumers and doing deals with smaller companies). He also needs to squeeze more efficiency from the firm—and that may mean people will have to work harder to earn the paychecks that have made Goldman’s upper echelons enormously wealthy while luring bright young financial sharks to the firm’s talent pool. “I don’t have any sense that he’s a status quo kind of guy, which I give as a compliment,” says Steve Friedman, who was once CEO of Goldman Sachs and now serves as chairman of Stone Point Capital, a private equity firm.

When Solomon became CEO, Friedman advised him to preserve a “small core” of cultural values, but that “everything else has to be subject to change.” Solomon has led by that creed. In March, he tore up the firm’s age-old dress code, a 35-page dossier with outdated stipulations about suits, ties, and shoe color. And this summer, Goldman quietly eliminated drug testing from its employee background checks. Gone, too, is the blanket ban on taking photos inside the office, once enforced by building security guards; Solomon now frequently posts such pics to his Instagram account. “The organization has a lot of bureaucracy I’d like to simplify,” says Solomon (though he doesn’t take credit for changing the drug or photo policies). “I think we can do some work to be more admired and respected, and a little less envied and feared.”

Short of knocking down walls, he’s trying to dispel the ivory-tower reputation that Goldman has cultivated through a historically secretive and hierarchical culture. On the mahogany-lined executive floor, Solomon chafes when visiting colleagues are asked to sit in the waiting room; he insists they be allowed to bypass two layers of receptionists and knock directly on his door. He prefers getting his own coffee and showing up unannounced at divisional meetings, and rejects prepared talking points. “He hates any sort of idea of Praetorian Guard, by any name,” says John F.W. Rogers, the firm’s chief of staff. At a firm that has for decades primarily communicated with its 38,000 employees via mass voicemail, Solomon asked his staff to find a way to convene a global all-hands Silicon Valley–style town hall in Goldman headquarters’ 11th-floor “sky lobby” (between the elevator banks and the cafeteria) so that he can hold them impromptu. (The company abandoned the idea after the logistics of coordinating its 73 offices proved too difficult and costly.)

To certain board members’ horror, Solomon, who made $23 million last year, habitually chooses to travel around New York City by subway. “I mean, why wouldn’t you take the subway? No, seriously,” he says. “It’s quicker and more efficient. You know, the mayor of New York can take the subway. Why can’t the CEO of Goldman Sachs?”

Goldman Sachs, once a private partnership, has been a public company for 20 years. Solomon has been saying he wants at long last to run it more like one. Whereas Blankfein never dialed in to the firm’s quarterly earnings calls, Solomon doesn’t plan to miss any. At its investor day on Jan. 29, Goldman Sachs will welcome some 300 shareholders, analysts, and regulators inside its Manhattan headquarters for the first time and unveil Solomon’s new innovation: a three- to five-year strategic plan. That may seem unremarkable, but until now the firm, believing the market too unpredictable, had only ever budgeted through each calendar year. “The way we made investments is we’d rub sticks together, we’d sprinkle a little bit on something,” says Solomon. In other words, each year Goldman’s people had to find a way to bring in more money than the last, but anything that required a longer-term commitment of resources to pay off later was off the table. “You don’t just get to run faster, jump higher, and have more profit just because we’re Goldman Sachs,” says COO John Waldron, reinforcing the focus on investing. “You’ve got to put money in the ground that grows over time.”

It all adds up to a pivotal moment for Goldman Sachs, and Solomon will help determine whether the firm leads again or fades out of relevance. As someone who was seldom in the press before the DJ D-Sol revelations, he’s still adjusting to the spotlight that comes with his powerful role; in interviews this fall, Solomon described a personal struggle to “live my life in a normal way.” His collaborators say that kind of practical, unpretentious mentality may be just what the bank needs now. “I think he’s doing the right things to evolve Goldman Sachs to compete in the 21st century,” says Michael Dell, whose company Dell Technologies has been a Goldman client longer than Solomon has worked there (and whose brother’s startup, Clarity Money, was acquired by Goldman last year). “Supertankers don’t just turn.”

2018年10月，蘇德巍出任首席執行官，當時不僅是高盛的多事之秋，也正值他個人生活的動蕩時期。他此前剛剛辦理完離婚手續。履新第二周，一位前助理被控從蘇德巍的美酒收藏中竊取了價值120萬美元的葡萄酒，隨即從曼哈頓酒店跳樓身亡。兩個20來歲的女兒也在那年夏天搬到美國其他地區——蘇德巍此前經常與她們在紐約共進周末晚餐。70多歲的老父親在幾年前離開人世。“我想他希望他的父母能夠活著看到他成為高盛的首席執行官。”蘇德巍在漢密爾頓學院讀本科時就認識他的一位朋友說。“我認為，這對他來說意義非凡。”

現年57歲的蘇德巍發現自己處于職業生涯的巔峰，單身，不再承擔以前的許多責任——除了管理高盛這個小差事。“如果我不工作，我基本上可以做任何我想做的事情。”他說。“要是我認定我對某件事情感興趣，我就會去做，我的意思是，我會盡我最大的能力去做這件事情，否則我不會去做。”他把這種熱情用在了像DJ和風箏沖浪這樣的愛好上，一些同事稱他為“重生的千禧一代”。

他也把這種熱情用在了高盛的工作上。在那里，他一直是個局外人。2016年年底，蘇德巍與哈維·施瓦茨開始角逐最高職位，當時兩人都被擢升為聯席總裁。蘇德巍一度被視為處于下風的那一位。他是一位投資銀行家，而他的競爭對手和老板貝蘭克梵都是交易員出身。在同行中，他還帶著“空降”的烙印——1999年，他被高盛從貝爾斯登公司挖來，受聘為合伙人，而不是從內部一步步晉升至高管職位的。

最重要的是，在高盛和諧融洽的工作氛圍中，他是一股令人不安的力量。蘇德巍曾經在麥當勞工作過，申請加入高盛的頭兩次都被拒絕了。另一位外來者指出，從華爾街其他銀行空降高盛后，你可能“感覺就像走進了歡樂谷”。在以共識為導向的公司文化中，蘇德巍是一位意志堅強、動輒敲桌子的果敢決策者。他會當面告訴員工當年沒有機會成為合伙人，并在有25人參加的會議上公開挑戰同事的想法。“最初的幾次談話有點殘酷，但人們隨后開始欣賞這種風格——沒有廢話。”目前擔任投行業務聯席主管（蘇德巍此前曾經擔任此職）的格雷格·萊姆考表示。萊姆考向一位同事建言，如果你需要有人告訴你，你做得很好，“去雇個人幫你做吧。”

不過，蘇德巍有足夠的情商來駕馭公司的政治，并相應地調整自己的語氣。他栽培下屬的能力給上司留下了深刻印象。“他可能會因為我這么說而生我的氣。”蘇德巍的密友，希爾頓酒店集團的首席執行官克里斯·納塞塔說。“但我想很多人都不知道，大衛其實是個特別可愛的家伙。”

蘇德巍并沒有彰顯這種特質，因為他試圖培養一種更加嚴苛的問責基調。蘇德巍正在實施一套戰略，其重中之重是更嚴格的績效評估視角——這是他在執掌投行業務的10年中率先實施的一個基于指標的系統。在那里，他會收到一些反映傳統優先事項（比如每位銀行家賺取的費用）和他個人優先事項（比如誰違反了他禁止初級分析師在周六工作的規定）的統計報告。在年終評估中，蘇德巍要求某些員工為一些非常細微的選擇辯護，此舉讓他們大吃一驚。作為一位嚴格的財政紀律執行者，他把一些薪酬飛漲的經理趕下臺，用一些低薪酬人選取而代之；除裁員之外，過去的員工透露稱，在獎金發放季“一無所獲”往往意味著你需要另謀出路了。與此同時，他要求員工忠誠，解雇了一些被發現在高盛合同到期前就在其他公司接受下一份工作的年輕分析師。蘇德巍覺得這是一種利益沖突。談到他的哲學時，他說：“你必須建立一種重視紀律、讓人們做出艱難決定的文化。”

蘇德巍著手解決高盛性別差距的方式，非常形象地展現了他的管理之道。他認為問題出在底層，隨即要求他的投行家從2017年開始聘用的分析師必須有一半是女性，然后將這一指令擴大到所有新招募的員工。如果哪位銀行家將女性分析師占比從大約40%提高到49%時，他不會獲得任何獎勵加分。“做得差不多足夠好，是不算數的。”萊姆考回憶說。“這是典型的蘇德巍風格。”今年履新的銀行部門分析師在該公司率先實現性別平等；高盛預計，2020年新晉員工的女性數量將超過男性。“他真的是在打破玻璃天花板，”高盛合伙人、前白宮顧問迪娜·鮑威爾去年10月在《財富》最具影響力的商界女性峰會上表示。

事實上，有幾個同事也使用“打破玻璃天花板”一詞來描述蘇德巍的方法——他好像在翻找祖母的古董，以決定典當什么，扔掉什么。上任第一天，他就發布了一份備忘錄，目的是清除公司組織結構上的“條塊分割”頑疾：現在，只有“一個高盛”，蘇德巍寫道。無論隸屬于哪個部門，只要能夠為公司帶來業務，任何員工就可以獲得額外獎勵。高盛的投行部門董事長艾莉森·馬斯透露稱，在這封郵件抵達收件箱的幾分鐘之后，她就“接到50個來自各個部門的員工電話，”其中包括后臺工作人員。這些來電都是給她兜售招攬客戶的線索。“他好像轉動鑰匙，打開一扇門，讓所有人都出去了。”馬斯說。

接下來，蘇德巍把高盛的投行業務手冊翻了個底朝天，從前到后詳詳細細地評估了一番，就像該公司為無數客戶做過的那樣。在一場長達90分鐘的會議上，蘇德巍要求副手們將各自的業務“精簡到最低限度”，并詢問了多達40個問題。“他不會夸夸其談地問問題。”首席財務官斯蒂芬·謝爾表示。蘇德巍的法令是：“沒有哪項業務是神圣不可侵犯的。”包括證券部門在內。他的前任，交易員出身的貝蘭克梵一直拒絕對該部門進行改革；到目前為止，蘇德巍已經批準了更多的自動化交易和后端業務整合，預計這將削減一些工作崗位。他還合并了高盛的高凈值和消費者業務，目的是用一個單一的網絡平臺為這兩類客戶服務。蘇德巍在貝爾斯登的前老板，Guggenheim Partners公司的執行董事長艾倫·施瓦茨表示：“在金融服務行業，很少有領導者既注重細節，又具有戰略眼光，而蘇德巍正是兼具這兩種能力的少數人之一。”

鑒于高盛正在斥巨資提升技術能力（迄今為止，僅在消費者業務上就投入了13億美元），這種犧牲應該有助于該公司提高回報率。但這也意味著員工享受的奢侈津貼會減少。一些資深銀行家抱怨稱，他們現在不得不乘坐經濟艙，坐商務艙出行的美好舊時光一去不復返了；如今，高盛不鼓勵任何人（除合伙人之外）乘坐停留在紐約總部外的黑色轎車參加會議，并建議他們改乘網約車。蘇德巍表示，與開支政策相關的決策太微不足道了，他通常不會直接參與，但這些政策反映出該公司對開支和生產率日益嚴苛的態度。“當我們談到‘運營效率’時，人人都聽到過成本運營效率，你就是一項成本。”該公司的首席戰略官斯蒂芬妮·科恩表示。“我們真正在做的是幫助員工更加有效地工作，有時是成本的問題，有時是時間的問題。”

Solomon became CEO, in October 2018, at a tumultuous time not only for the firm but in his own life. He’d recently finalized his divorce. His second week in the job, his former assistant, who’d been charged with stealing $1.2 million worth of wine from Solomon’s collection, jumped to his death from a Manhattan hotel window. Solomon’s two twentysomething daughters, with whom he’d long had a standing Sunday night dinner date in New York, had both moved across the country that summer. His father had passed away in his seventies a couple of years earlier. “I think he wishes his parents had been alive to see him become CEO of Goldman Sachs,” says a friend who’s known Solomon since his undergraduate years at Hamilton College. “I think that would have been incredibly meaningful to him.”

The 57-year-old Solomon found himself at the zenith of his career, single and no longer tethered to many of his former responsibilities—outside the small task of running Goldman Sachs. “If I’m not working, I can basically do whatever I want,” he says. “And as a personality, if I decide I’m interested in something and I’m going to do it, I’m going to do it. I mean, I’m going to try to really do it to the highest capacity of my ability, or I’m not going to do it.” He’s applied that intensity to hobbies like DJing and kite-surfing—some colleagues have termed him a “born-again millennial.”

He’s also applied it at the bank, where Solomon has always been something of an outlier. Competing with Harvey Schwartz for the top job—beginning in late 2016, when they were both named copresidents—Solomon was seen as the underdog. He was an investment banker; both his rival and boss Blankfein were traders. Among peers, he also bore the brand of a “lateral”—he had been hired from Bear Stearns as a partner in 1999, instead of working his way up to the title from within.

Above all Solomon, who’d once worked at McDonald’s and been rejected the first two times he applied to Goldman, was a jarring force in the firm’s tirelessly collegial atmosphere. Coming from elsewhere on the Street, another transplant notes, Goldman could “feel like you walked into Pleasantville.” In the firm’s consensus-driven culture, Solomon was a strong-willed, table-pounding decision-maker. He would tell employees to their face when they had no shot of making partner that year, and openly challenge colleagues’ ideas in meetings of 25 other people. “The first couple conversations were a little brutal, but people then appreciate it—there’s no BS,” says Gregg Lemkau, who has Solomon’s former job as cohead of investment banking. If you need someone to tell you you did a good job, Lemkau advised a colleague, “Go hire somebody to do that for you.”

Still, Solomon had enough emotional intelligence to navigate the firm’s politics and temper his tone accordingly, and he impressed his superiors with his ability to nurture and groom underlings. “He’d probably be pissed at me for saying this,” says Chris Nassetta, Hilton’s CEO and Solomon’s close friend, “but I think a lot of people don’t know that David is a sweetheart of a guy.”

Solomon doesn’t emphasize that side as he fosters a more rigid tone of accountability. Underpinning the strategy Solomon is bringing to Goldman Sachs now is a stricter lens for performance evaluation—a metrics-based system that he piloted in the decade he ran the investment banking division. There, he’d receive statistical reports on such factors that reflected priorities both traditional—such as the fees each banker collected—and personal, like which ones violated his rule against making junior analysts work on Saturdays. In year-end reviews, he surprised some people by asking them to justify granular choices. A fiscal disciplinarian, he unseated some managers whose pay ballooned out of whack and replaced them with cheaper people; besides layoffs, past employees say getting “zeroed” at bonus season was a signal to start job-hunting. At the same time, he demanded loyalty, dismissing young analysts who were caught accepting their next job elsewhere before their Goldman contract was up. Solomon felt it was a conflict of interest. Of his philosophy, he says, “You have to set a culture where discipline matters, and people make hard decisions.”

The ultimate manifestation of Solomon’s method is the way he set out to tackle Goldman’s gender gap. Concluding that the problem began at the bottom, Solomon instructed his investment bankers to hire half-female analyst classes starting in 2017, then expanded the mandate to all new recruits. Bankers received no bonus points when they increased the share of women from roughly 40% to as much as 49%. “Like, doing almost well enough doesn’t count,” Lemkau recalls. “That is quintessential David.” The banking division’s analysts that started this year are the firm’s first to achieve gender parity; firm-wide, Goldman expects its 2020 crop to include more women than men. “He is really breaking glass,” Dina Powell, a Goldman partner and former White House adviser, said at Fortune’s Most Powerful Women Summit in October.

Indeed, several colleagues use the phrase “breaking glass” to describe Solomon’s approach—often in the sense of someone who set upon Goldman’s old ways as though he was rifling through Grandma’s antiques to decide what to pawn and what to toss. His first day as CEO, he put out a memo aimed at bulldozing the territorial silos in which the firm was organized: Now, there’d be just “One Goldman Sachs,” Solomon wrote. People would be compensated more for bringing in business anywhere in the firm, regardless of where they sat. Within minutes of the email hitting inboxes, says Alison Mass, chairman of the investment banking division, she had “50 calls from random people around the firm,” including back-office workers, with ideas to sell her clients. “It was literally like he turned a key and unlocked a door and let everyone out,” Mass says.

Next, Solomon turned Goldman’s investment banking playbook on itself, running the kind of front-to-back review it has done for innumerable clients. He asked his lieutenants to strip their businesses “down to the studs,” asking as many as 40 questions in the span of a 90-minute meeting. “He doesn’t ask a question rhetorically,” notes CFO Stephen Scherr. Solomon’s edict: “There are no sacred cows.” That included the securities division, where Blankfein, a trader by background, had resisted changes; so far, Solomon has green-lit more automated trading and the consolidation of back-end operations, which is expected to shave off some jobs. He also merged Goldman’s high-net-worth and consumer businesses, with the aim of serving both sets of customers with a single web platform. “There are very few leaders in the financial services business who are very detail-oriented and at the same time strategic,” says Alan Schwartz, once Solomon’s boss at Bear Stearns, now Guggenheim Partners’ executive chairman. “David is one of those who does both.”

The proverbial bovine sacrifice should help Goldman improve returns as it spends to scale up its technology ($1.3 billion on the consumer business alone, so far). But it also means fewer luxury perks for employees. Some senior bankers lament having to fly coach on trips where in the good old days they sat in business class; Goldman now discourages anyone but partners from gliding into the line of black cars waiting outside the bank’s New York headquarters to get to meetings, telling them to take Uber instead. Solomon says expense policy decisions are too small-potato to involve him directly, but they reflect an increasing rigor toward expenses and productivity. “When we say ‘operating efficiency,’ everyone hears cost—operating efficiency, you’re a cost,” says Stephanie Cohen, the firm’s chief strategy officer. “What we’re really trying to do is help people do their job more efficiently, and sometimes that’s a cost thing, sometimes it’s a time thing.”

今年夏天，高盛與蘋果合作推出第一張信用卡。自那以來，蘇德巍一直在打磨一種新的派對戲法。他首先會說：“請拿出你的手機。”然后，他會花兩分鐘時間在手機上刷幾下，再添加對方社保號碼的后四位。實際上，蘇德巍已經為他的許多朋友和熟人開設了蘋果卡賬戶。“很容易，對吧？”他說。“還不是非常完美。”

蘋果卡是為iPhone手機錢包量身定做的，與之對應的是無數鈦制實體錢包。這項合作關系是由2016年首次亮相的高盛消費銀行馬庫斯（Marcus）發展而來的。這張卡在銀行內部的代號為“餅干計劃”； 蘋果公司本身也被稱為“土星”——也許是因為高盛作為其合作伙伴，被視為其軌道上的眾多衛星之一。高盛和萬事達卡的信用卡的標識最終出現在白色金屬矩形的背面（在虛擬版上看不到）。

這張信用卡的發展堪稱一個生動的寓言，從中可窺視這家銀行在成立一個半世紀后探索新業務的努力——以及有時候發現自己無法適應新環境時的尷尬。直到最近，高盛還從未把自己想象成一家零售銀行，即使在金融危機時代一項監管政策的改變允許它這樣做之后。“我認為我們是在否認。”萊姆考說。但隨著收入穩步縮水，高盛不得不接受它必須得嘗試新事物這一現實。高盛的目標是到2020年年底，每年從新業務中獲得50億美元收入，而馬庫斯有望創造超過10億美元的收入。

Since Goldman Sachs launched its first credit card, in partnership with Apple, over the summer, Solomon has been workshopping a new party trick. It starts with the phrase, “Take out your phone.” In a two-minute routine requiring a few swipes and the last four digits of your Social Security number, Solomon has been virtually opening Apple Card accounts for many of his friends and acquaintances. “It’s pretty easy, right?” he says, adding, “It’s not perfect.”

The Apple Card—tailor-made for the iPhone’s mobile wallet, with a numberless titanium physical counterpart—was a partnership that grew out of Marcus, the fledgling consumer bank within Goldman Sachs that made its debut in 2016. The card was code-named “Project Cookie” inside the bank; Apple itself was known as “Saturn”—perhaps because Goldman, as one of its partners, was treated like one of many moons in its orbit. Goldman’s logo, along with Mastercard’s, ended up on the back of the white metal rectangle (and is nowhere to be seen on the virtual version).

The credit card’s development has been a telling parable about a bank exploring a new business a century and a half into its existence, and at times finding itself out of its league. Goldman until recently never fancied itself a retail banker, even after a financial--crisis era change to its regulatory status allowed it to become one. “I think we were in denial,” says Lemkau. But as its revenue has steadily shrunk, it has had to come to terms with trying something new. Goldman is targeting $5 billion in annual revenue from new businesses by the end of 2020, with Marcus on track to generate more than $1 billion.

蘇德巍和高盛都不喜歡做任何他們無法做得出類拔萃的事情。高盛認為，它至少能夠像舊金山那些在車庫中創業的企業家那樣，開創一項消費科技業務。然而，在創建馬庫斯三年之后，它遇到了一家初創企業經常遭遇的許多問題，但并沒有催生同樣的文化氛圍。近年來，馬庫斯實施了一系列形形色色的收購交易（包括今年斥資7.5億美元收購財務顧問網絡United Capital，這是高盛近二十年最大一筆收購案）。它不僅聘用內部工程師，同時還把許多工作外包給四大咨詢公司中的三家。幾位前馬庫斯雇員表示，這是一段巴別塔式的經歷。讓工程師惱火的是，高盛電腦要求運行合規的系統，而這往往會限制其功能。他們將其戲稱為“庫什納”——這是指美國總統特朗普的女婿在蘋果Mac電腦上運行微軟Windows系統的做法。由于沒有自己的消費者數據支持測試，高盛起初為馬庫斯貸款承受的損失高于預期。蘋果卡的推出也出現了一些問題；紐約監管機構已經開始調查高盛的信用額度算法是否存在性別歧視，該公司對此予以否認。盡管如此，馬庫斯仍然沒有一款移動App，但知情者透露稱，它很快就會推出。

馬庫斯真正擁有的，是讓大多數金融科技初創公司相形見絀的規模——550億美元存款和50億美元貸款。“一旦開始冒險，你會犯錯誤，你會虧錢，事情會變糟，”蘇德巍說。“唯一不同的是，我們不會站在那里，（像Facebook公司那樣）用擴音器大肆宣揚‘快速突破，除舊立新’。”

今年從高盛董事會退休的哈佛商學院教授比爾·喬治表示，蘇德巍對“可能出現什么問題”的敏銳預期，是他被選為首席執行官的一個主要原因。“你知道有些職業籃球運動員的后腦勺長了眼睛嗎？他就擁有那種周邊視覺。”

Neither Solomon nor Goldman likes doing anything unless they can excel, and the firm determined it should be able to start a consumer tech business at least as well as any garage-full of entrepreneurs in San Francisco. Three years after launching Marcus, though, it has encountered many of the problems of a startup without engendering quite the same cultural feel. Between a patchwork of acquisitions—including its $750 million purchase of financial-adviser network United Capital this year, the firm’s largest in nearly two decades—and engineers both in-house and outsourced to three of the “Big Four” consulting firms simultaneously, former Marcus employees describe a Babel-like experience. Engineers chafed at the way Goldman computers require running compliance systems that restrict their functions, and dubbed them “Kushners”—a reference to the way presidential son-in-law Jared famously runs Microsoft Windows on a Mac. Without its own consumer data to back-test, Goldman initially took higher losses on Marcus loans than it planned. The Apple Card rollout has also been glitchy; New York regulators have opened an investigation into whether Goldman’s credit limit algorithm has a sexist bias, which the bank denies. For all that, Marcus still doesn’t have a mobile app—though people close to the bank say it’s coming soon.

What it does have is scale—$55 billion in deposits and $5 billion in loans, dwarfing most fintech startups. “When you take risks, you make mistakes, you lose money, things go wrong,” Solomon says. “The only difference is we don’t stand up there and trumpet publicly in an amplifier, ‘Go fast and break things.’?”

Solomon’s keen anticipation of “what possibly could go wrong” is a major reason he was chosen to be CEO, says Bill George, a Harvard Business School professor who retired this year from Goldman’s board. “You know how they talk about professional basketball players having eyes in the back of their head? He had that kind of peripheral vision.”

長期以來，高盛內部經常用一個短語來形容那些最有可能走上員工們夢寐以求的合伙人之路的人：經濟殺手。“這在過去是唯一重要的事情。”一位前董事總經理說。在通往頂峰的路上，你會吃掉你殺死的東西：獎金實際上完全是根據你當年為公司創造的收入來支付的。

隨著蘇德巍尋求激勵公司上下更加緊密地合作，他執掌的領導團隊正在重新調整高盛向員工支付報酬的方式——到目前為止，員工薪酬一直是最大一塊支出，大約占今年收入的35%。這意味著重新定義何為“殺手”。假期前后，當今年的薪酬政策討論開始的時候，高盛發布了一項新規則。它將把獎金更多地與公司的新優先事項聯系起來，包括三年的時間線目標、多樣化招聘、彰顯“一個高盛”精神，以及公司的整體回報。這一制度實際上使更多的員工薪酬置于他們能夠控制的范圍之外，并可能面臨風險，尤其是在一些人越來越確信牛市即將見頂的情況下。“現實地講，我認為未來不存在巨大的上升空間。”一位最近離職的合伙人表示。“我覺得再也不會像以前那樣了。”

蘇德巍和他的首席級高管們都沒有明確表示要實施大規模降薪計劃。但他們為2019年前9個月設定的薪酬池比去年同期低11%，比收入的降幅（7%）大得多。員工們預計，他們的收入會日益扁平化。上述前董事總經理表示：“現在的100萬美元就相當于過去的300萬美元。”他補充稱，所謂的董事總經理（比合伙人低一級）薪酬標準今年可能接近75萬美元，包括獎金在內。

這一趨勢或許是導致或影響（或兩者皆有）近期合伙人大量流失的原因。自蘇德巍上任以來，約有10%的合伙人已經宣布離職，更多的人可能會在年底前離開。這種被首席運營官沃爾德倫稱為“稀釋”的現象，為公司創造了一個推動更具前瞻性思維的年輕一代高管上位的機會。蘇德巍說，他唯一的意圖是“讓年輕人拾階而上，那些已經在這里待了一段時間的人不會待太久。”

這種態度使得蘇德巍在千禧一代和Z世代中廣受歡迎，他們現在占高盛員工總數的75%。這些年輕人普遍認為，他就是高盛一直在等待的那位變革推動者。與此同時，蘇德巍試圖將合伙人隊伍從如今的440人左右削減到400人，以期讓高盛員工的職業生涯巔峰（即成為合伙人）更加難以攀登。鑒于在貝蘭克梵任期內，盡管收入停滯不前，合伙人隊伍仍然增長了63%，這種做法是有道理的。合伙人身份仍然是這家公司的圣杯，這也是為什么低級別員工長期以來愿意接受相較于華爾街其他公司薪酬標準的“高盛折扣” 的原因所在——這是為了換取最終掙大錢的機會。此舉要傳達的信息是，“如果你真的在這里取得成功，金字塔的頂端是極其令人垂涎的。”蘇德巍說。

這種權衡的一個結果是，對大多數人來說，高盛可能沒有那么多令人趨之若鶩的奢侈待遇，也不再是一個精英避風港。“我不希望讓高盛變得更具排他性。我所希望的其實恰恰相反。”蘇德巍說。“我試圖讓它變得更加開放，更容易接近，更容易理解，更人性化。”畢竟，許多年輕的高盛人既不記得繁榮時期的好日子，也不記得紓困時期的艱難歲月；對于他們來說，沒有什么比在夜店跟綽號D-Sol的音樂DJ合照一張自拍更能代表這家銀行了。（財富中文網）

本文另一版本登載于《財富》雜志2019年12月刊，標題為《打破高盛內部的舊規則》。

譯者：任文科

There was long a phrase inside Goldman Sachs to describe those most likely to ascend the firm’s coveted partner track: an economic killer. “It used to be the only thing that mattered,” says an ex–managing director. And en route to the top, you ate what you killed: Bonuses were paid virtually entirely according to the revenue you made for the firm that year.

As Solomon seeks to spur the firm to work more cooperatively, he and his leadership team are recalibrating how Goldman pays its people—by far the firm’s biggest expense, accounting for about 35% of revenue this year. And that means redefining who’s a killer. Around holiday time, when this year’s compensation discussions begin, a new rubric will tie bonuses more to the company’s new priorities, including three-year timeline goals, diverse hiring, participating in the “One Goldman Sachs” spirit—and the overall returns of the firm. The system effectively puts more of employees’ compensation out of their own control and potentially at risk—especially as some grow more convinced that the bull market is poised to top out soon. “I don’t think realistically there’s a tremendous amount of upside in the future,” says a recently departed partner. “I don’t think it will ever be what it used to be.”

Neither Solomon nor his C-suite have explicitly professed an intention to give paychecks a haircut. But the compensation pool they earmarked for the first nine months of 2019 is 11% lower than it was for that period last year, a steeper drop than the 7% decline in revenue. Employees anticipate a flattening phenomenon: “$1 million is the new $3 million,” says the former managing director, adding that this year’s norm for pay for the so-called MD tier—one rung below partner—may be closer to $750,000, including bonus.

That trend could be cause or effect—or both—of a recent exodus of partners. About 10% of them have announced their departures since Solomon took over, and more could leave by the end of the year. The “thinning out,” as COO Waldron calls it, has created an opening into which the firm is actively promoting a younger, forward-thinking generation of executives. Solomon says his only intention is “that young people get to move up, and people that have been here for a while don’t stay too long.”

That’s made Solomon popular with the millennial and Gen Z cohorts who now make up 75% of Goldman Sachs’s workforce—and there’s a mood among them that Solomon is the change the bank has been waiting for. At the same time, he aims to make the culmination of their Goldman Sachs career—the partnership—just that much harder to attain, by paring partner ranks even further toward 400, from about 440 or so today. It also makes sense when you consider that the partnership ranks grew by some 63% over Blankfein’s tenure, even as revenue stagnated. Partnership remains the firm’s holy grail, and the reason employees at lower levels have long been willing to accept the “Goldman discount” compared with other Wall Street salaries—in exchange for the chance to eventually make far more. The message, Solomon says, is, “If you really succeed here, the top of the pyramid is super-aspirational.”

The tradeoff is that for most, Goldman Sachs may have fewer of the trappings of a haven for the elite. “I don’t want to make Goldman Sachs more exclusive. In fact the opposite,” says Solomon. “I’m trying to make it more open, more approachable, more understood, more human. After all, many younger Goldmanites remember neither the boom times nor bailout days; to them, nothing symbolizes the bank more than a selfie at the club with D-Sol.

A version of this article appears in the December 2019 issue of Fortune with the headline “Ripping Up the Rules at Goldman Sachs.”