The Firm: The Story of McKinsey and Its Secret Influence on American Business

by Duff McDonald

  One such engagement was a 1980 project for AT&T. AT&T had invented an early version of wireless technology but was worried about the return on its investment in radio towers, among other things. McKinsey, as always, carefully studied the issue and came up with an estimate that was, to say the least, laughably off the mark: In the year 2000, McKinsey deduced, the total wireless market would have less than one million subscribers.

  It’s worth remembering that at the time, handsets were so clunky and expensive and bandwidth was so limited that the only customers were very wealthy people and corporate executives. It’s also worth remembering that predicting the future is hard. However, this was clearly a blunder, and a very costly one. AT&T dropped the project, dooming the company to playing catch-up in wireless and necessitating its eventual sale to SBC Communications in 2005. That’s the consultant’s equivalent of a malpractice case, in which the patient dies an awful, avoidable death.

  Sometimes McKinsey has been accused of selling not just faulty research but defunct strategy, such as when it worked for the Continental Illinois Bank, then one of the nation’s hottest banks, in the 1970s. Continental saw itself on a par with Chase Manhattan and Citibank, two well-known McKinsey clients, and wanted a new organizational structure. The consultants told Continental the same thing they had told numerous other banks: It needed to move away from its hierarchical management structure toward something called “matrix management.” The idea, in simple form: In a matrix system, different people may work in one department and may technically report to one manager, but they may also be assigned to a different project and a different manager at any point in time. The result is a theoretical increase in flexibility in return for a diffusion of responsibility and accountability. At first the reorganization seemed brilliant. Between 1976 and 1981, Continental had the fastest asset growth of any major bank, a remarkable 110 percent. Lending grew at an ever-faster rate, 180 percent. The bank’s return on equity during that time—14.4 percent—was second only to Morgan Guaranty.

  But James McCollom, a former employee of the bank and author of The Continental Affair, claimed the advice doomed Continental. The bank’s loan-to-asset ratio also increased to 69 percent in 1981, the highest in the industry, arguably making the bank the riskiest of its peers. McCollom pointed to the findings of Peters and Waterman, who later wrote in In Search of Excellence that executives “shared our disquiet about conventional approaches [of organizational design]. All were uncomfortable with the limitations of the usual structural solutions, especially the latest aberration, the complex matrix form.” In other words, McKinsey was pushing on Continental a structure that its own employees had concluded didn’t work. “McKinsey had sold us matrix management, the very snake oil that excellent companies avoided,” wrote McCollom. “They had sold us lawyers, secrecy . . . [and] buzzwords. . . . McKinsey had sold the Continental Bank its obsolete equipment.”10 A few years later, Continental earned the dubious distinction of being the biggest bank failure in U.S. history to that point.

  In the consulting the London office did for the National Health Service, McKinsey failed to move the stultified British bureaucracy an inch. (The consultants were still in there in 2012, with critics still questioning the benefits of their work, as apt an example of the controversial nature of the McKinsey transformational relationship as any.) McKinsey was in the BBC for years with similar effect, at one point pushing the idea of an internal market in which BBC staffers bought and sold services to one another. This was another consulting theme that looked good in theory but was to be poor in practice, one that forced producers to engage in endless negotiations internally just to do things as simple as reserving a studio or finding airtime for a project. In other words, the consultants pushed excessively complex management systems on an oblivious client. One project reportedly had the consultants working with BBC staffers by cutting out paper frogs and pretending to sell them to one another.11 A scathing June 1995 newspaper investigation had already put the lie to the notion that McKinsey had helped cut costs in return for its extravagant fees: “The BBC now costs more to run while employing less people than ever before,” the investigation concluded. “The ‘savings’ have turned out to be an extra staff cost of 140 million pounds compared to four years ago.”12

  And McKinsey never really left the building. A decade later, when John Birt, then director-general of the BBC, stepped down from his job, McKinsey brought him on as a part-time consultant. In 2005 Birt severed all ties with the firm after critics suggested that working for 10 Downing Street in London and McKinsey posed a conflict of interest—as if hiring him shortly after extracting millions of pounds out of the BBC hadn’t been. Birt’s replacement at the BBC’s helm—Greg Dyke—immediately slashed spending on outside consultants by 75 percent.

  Few clients are going to hire McKinsey and then say the consultants weren’t worth it. In a sense, the firm is a spiritual relative of “La Belle” Otero, a courtesan living in Paris in the early part of the twentieth century who was once considered the most sought-after woman in the world. Carolina Otero was very selective about her clientele and charged outrageous fees that reportedly topped $1 million in 2012 dollars. Her customers included Prince Albert I of Monaco and the king of Serbia, and it was widely argued that everybody who had the means had to have her at least once. And once you’d done that, what could you say? Once you’d paid a million francs for a roll in the hay, you weren’t about to admit it wasn’t worth the price paid.

  It’s actually quite difficult to make the case that McKinsey has had any lasting—or fundamental—effect on the way businesses are managed since a few signal accomplishments such as reorganizing American conglomerates in the 1940s. Silicon Valley has surely had a far larger impact on the way businesses are run in the last thirty years than McKinsey might claim to have had. “What have they fixed?” asked former McKinsey consultant Michael Lanning. “What have they changed? Did they take any voice in the way banking has evolved in the past thirty years? They did study after study at GM, and that place needed the most radical kind of change you can imagine. The place was dead, and it was just going to take a long time for the body to die unless they changed how they operated. McKinsey was in there with huge teams, charging huge fees, for several decades. And look where GM came out.”13 In the end, all the GM work did was provide a revenue stream to enrich a group of McKinsey partners, especially those working with the automaker.

  The last time McKinsey was influential at Apple Computer was when John Sculley was there, and that’s because he’d had a brand-marketing heritage from Pepsi. And Sculley was a disaster. Did McKinsey do anything to help the great companies of today become what they are? Amazon, Microsoft, Google? In short, no.

  Still, McKinsey’s high self-regard survives even in the face of evidence to the contrary. McKinsey consultant Tom Steiner recalled a strategy study done for the New York office by another partner, Chuck Farr. “He had two slides. The first was the top clients of the New York office, by billings—companies like AT&T, American Express, and Manufacturers Hanover. All the partners got up to talk about what special thing McKinsey had done to become so vital to those clients. Before we knew it, there were only fifteen minutes left of what was supposed to be a two-hour meeting. Someone said, ‘What’s on the second slide?’ It was Booz Allen’s top clients. And they were pretty much the same companies.”14 McKinsey may have been earning more than Booz at the time, but it was from a client base that was clearly willing to pay for advice from everyone. There’s nothing special about that kind of product.

  A General Disaster

  McKinsey’s work for General Motors in the early 1980s showed quite clearly that consultants can sometimes do far more harm than good to a company, even if it’s one of the most important clients they might ever have. At the time, General Motors—the Titanic of American business—was being pummeled by Japanese carmakers like Toyota. Chairman Roger Smith decided in 1984 to embark on the most massive reorganization of a company in Ameri
can history. He hired McKinsey to devise it with him. The carmaker soon accounted for more fees than the rest of McKinsey’s entire U.S. manufacturing practice. According to one report, at one point GM was paying McKinsey as much as $2 million a month.15

  McKinsey interviewed the top sixty-five executives in the company and eight hundred employees, asking them what organizational problems they saw. With their poll results in hand, the consultants proposed a new structure. Instead of organizing the company by brand—Chevy, Cadillac, Buick, et cetera—the consultants said it should be organized by type of car—large, small, truck. This was a basic McKinsey maneuver—don’t organize around this, organize around that—and it had worked in the past. The consultants had successfully helped AT&T organize around its markets as opposed to its technologies, for example. But it was not the right prescription for GM.

  “At the time, neither McKinsey nor General Motors understood the true nature of Japanese competitiveness,” said Maryann Keller, author of Rude Awakening: The Rise, Fall and Struggle for Recovery of General Motors.16 “Only later did people begin to realize that it wasn’t the fact that the Japanese had a compliant low-paid workforce that got up every day and sang the Toyota hymn. It was that they built cars with fewer parts, fewer defects, continuous improvement processes, and took man-hours out of making cars without compromising productivity. I’m 100 percent sure McKinsey had no clue about these revelations.”

  Instead, McKinsey just moved people around, and this had the effect of destroying institutional knowledge and the informal networks of people who knew how to get things done inside the massive company. “They were flying perpendicular to where they should have been,” said Keller. “It was the thousands of little things Toyota did well that mattered, not the organizational design of General Motors.”

  The reorganization was such an unmitigated disaster—it ran up huge costs with no measurable improvements in output or efficiencies—that Keller claimed it planted the seeds for GM’s bankruptcy in 2009. “But is it such a surprise?” she asked. “These are not people who have ever run anything. These are people who have spent their lives talking to high-level executives. And what do high-level executives know about making things? Not much, usually. Do you think Roger Smith knew anything about making a car?”

  Whose fault was it? Not McKinsey’s, said McKinsey. Despite refusing to discuss the work in specific, one senior director told Fortune writer John Huey that the fault lay squarely with the General Motors management. “We told them like it was. We weren’t passive at all. We told them to take their medicine,” he said. “It’s like being a doctor. You do the best you can, but if the patient won’t quit smoking, he still dies. This is a problem the world over. Corporate executives are not risk takers. They don’t see trouble clearly until they’re going down the drain.”17

  General Motors offers one of the great lessons of strategy consulting. As Matthew Stewart so elegantly pointed out, “The idea of strategy, like the owl of Minerva, typically arises just as the sun is setting on an organization. An old saw has it that strategy is when you’re running out of ammo but you keep firing on all guns so that the enemy won’t know. As a rule, corporations turn to strategy when they can’t justify their existence in any other way, and they start planning when they don’t really know where they are going.”18

  It Depends

  The most valued possession of a McKinsey consultant is a long-term client relationship. That’s not so different from any business, but at McKinsey, with no particular products to speak of, the relationship is all. In any given year, some 85 percent of McKinsey’s revenues come from existing clients. “We insinuate ourselves,” Ron Daniel told Forbes.19 It is a talent that bedevils the firm’s rivals and amazes even its alumni. “May we all get there, when clients just keep coming back to you,” said Alan Kantrow, former editor of the McKinsey Quarterly. “They have follow-on work not just because they’re good at what they do, but because they are trained in how to manage these kinds of client relationships. They understand that the core reality is the relationship and the conversation, and that any particular engagement is merely epiphenomenal.”20

  Smart clients say that the best way to use McKinsey is not to let them insinuate themselves—to prohibit walking the halls of the client’s offices looking for new business. Jamie Dimon of JPMorgan Chase, for example, will hire McKinsey, but for one-off projects in which the entire body of knowledge generated is transferred to JPMorgan Chase at the end of the project. The firm’s operating committee has to approve any consulting engagement, and the JPMorgan Chase executives don’t take just any consultants; they pick and choose the specific people they want on the project.

  “We’re not selling time and answers, like law or accounting firms,” Ron Daniel told Forbes in 1987.21 “We’re selling a benefit called change. Change is where the value is.” Daniel told Forbes that clients must “trust McKinsey” to set a fair fee. But clients don’t always comply. When Jamie Dimon was CEO of Bank One, he hired McKinsey to look at the bank’s credit rules in 2002, something no one had done for years. He told the consultants they would get 50 percent of their fee up front and 50 percent based on performance. McKinsey earned its full fee. A year later, on a project overseen by Dimon’s deputy Heidi Miller, Bank One decided McKinsey hadn’t earned its performance fee and refused to pay the second installment. The consultants were flabbergasted, but McKinsey partner Clay Deutsch conceded the argument, and the firm took home just half its “fair” fee. Despite Daniel’s lofty rhetoric, every now and then McKinsey is confronted with the simple fact that for some customers, consulting is little more than a packaged goods business. At the end of the day, the dogs have to eat the dog food.

  And that brings the focus back to the difference between the elusive promise of big-picture, blue-sky thinking and the true value that a consultant can almost always provide. Real consulting, writer Kevin Mellyn has argued, isn’t about change or leadership or vision. It’s about helping people manage what one might call Industrial Prussianism—organizing activities through highly rational bureaucratic routines that promote effectiveness, efficiency, and honesty. The Prussians beat the French in 1871, but not because they had inspired leadership. They won because their system was based on rules, orders, and norms that allowed their army to run efficiently and without the need for heroes. Likewise, Industrial Prussianism skips the heroes and focuses on efficiency and frugality, with training and accountability. The only point of strategic planning is to think through all contingencies in advance so as to make fewer tactical execution errors; he who makes the fewest and has the best-organized and best-trained troops wins. The companies that survive, like the best armies, can recover from unexpected blows of fate and competitor breakthroughs.

  Bower and his contemporaries understood this. They understood that when businesses become big, they need efficiency experts to help hone their internal processes. McKinsey was part of that wave of consulting engineers. Even today, the best consultants are ex-engineers, not perfectly designed social operators with a Harvard MBA. Engineering teaches you to define your solution space, determine the relevant levers you can pull or push, and then find your solution. Strategize all you want, but if your processes aren’t well oiled, you’re a goner.

  Hiring McKinsey Just to Make a Point

  By the end of the 1980s, McKinsey’s brand had moved into a whole new realm. Observers had long been comparing it to IBM in the context of “you never got fired for hiring” either one of them, but now companies had begun to hire McKinsey just to make a point. The mere announcement of the hiring of the firm now had its own legitimizing effect: the press release as cattle prod of the corporate sphere. Every consulting firm does this. McKinsey just does it better than the rest of them. More than forty years ago, London’s Sunday Times put it thus: “Calling in McKinsey has proved to be a highly effective way of nailing the red rag of revolution to the masthead.”22

  When you hired IBM, you were hiring it to do something specific and
real—revamp your entire technology infrastructure, say, or outsource your payroll and personnel processing. With McKinsey, the hiring of the consultants could be a mere tactic. If, as CEO, you felt you needed to cut 10 percent of costs but didn’t feel you were getting buy-in from your employees, the hiring of McKinsey generally got the point across quite clearly. Companies still hire McKinsey to send this type of message.

  In 2009 publisher Condé Nast famously hired the firm to ram the message through to a staff used to spending extravagantly that 30 percent needed to be removed from the company’s cost structure. Likewise, Warner Amex Cable Communications hired McKinsey in 1984 because of infighting that was getting in the way of executing a turnaround. “We had to go outside for a view that wasn’t biased,” said then-CEO Drew Lewis. “And McKinsey did a fine job.”23

  In 1993 Delta Air Lines was bleeding money as the result of some newly acquired European operations. Shareholders were in revolt. What did management do? It hired McKinsey, and told the world about it. “Such an announcement sends out several messages,” John Huey wrote in Fortune at the time. “ ‘We know we have a problem. We’re doing something about it. We hired the most expensive help we could find. Give us some time, okay?’ ”24 This brought McKinsey a whole lot closer to the bond rating-agency model than it liked—the McKinsey engagement as corollary to the AAA rating. But AAA-rated bonds can (and do) fail. So do some recommendations made by McKinsey.

  McKinsey claims that it tries to avoid this kind of situation, going so far as to build into its contract language that prevents clients from mentioning that the firm has been hired: “McKinsey’s work for the Client is confidential and intended for the Client’s internal use only. McKinsey does not make public client names, client materials or reports prepared for clients without their prior written permission. Similarly, the Client agrees that it will not use McKinsey’s name, refer to McKinsey’s work, or make the Deliverables or the existence or terms of this agreement available outside its organization without McKinsey’s prior written permission.”