The Silo Effect Read online

  Thank you for downloading this Simon & Schuster eBook.

  * * *

  Join our mailing list and get updates on new releases, deals, bonus content and other great books from Simon & Schuster.

  CLICK HERE TO SIGN UP

  or visit us online to sign up at

  eBookNews.SimonandSchuster.com

  For Romaine Joy Carley Tett, in grateful, loving memory

  Contents

  Note from the Author

  Introduction: Bloomberg’s Skunkworks

  PART ONE: SILOS

  1. The Nondancers: How Anthropology Can Illuminate Silos

  2. Octopus Pots: How Silos Crush Innovation

  3. When Gnomes Go Blind: How Silos Conceal Risks

  4. Russian Dolls: How Silos Create Tunnel Vision

  PART TWO: SILO BUSTERS

  5. Gun-Toting Geeks: How Individuals Can Silo-Bust Their Lives

  6. (Re)writing Social Code: How to Keep Silos Fluid

  7. Flipping the Lens: How Doctors Tried Not to Behave Like Economists

  8. Bucket-Busting: How Breaking Down Silos Can Produce Profits

  Conclusion: Connecting the Dots

  Acknowledgments

  About Gillian Tett

  Notes

  Index

  Note from the Author

  THIS BOOK STARTED DURING THE Great Financial Crisis of 2008. But it is not a book about finance. Far from it. Instead, it asks a basic question: Why do humans working in modern institutions collectively act in ways that sometimes seem stupid? Why do normally clever people fail to see risks and opportunities that are subsequently blindingly obvious? Why, as Daniel Kahneman, the psychologist, put it, are we sometimes so “blind to our own blindness”?1

  It was a question I often asked myself in 2007 and 2008. Back then, I was working as a journalist in London, running the markets team of the Financial Times. When the financial crisis erupted, we threw ourselves into trying to understand why the disaster had come about. There were lots of potential reasons. Before 2008 bankers had taken some crazy risks with mortgages and other financial assets, creating a gigantic bubble. Regulators had failed to spot the dangers, because they misunderstood how the modern financial system worked. Central bankers and other policymakers had given the wrong economic incentives to financiers. Consumers had been dangerously complacent, running up huge credit card debts and mortgage loans without asking whether they could be repaid. Ratings agencies misread risks. And so on.

  But as I dug into the story of the Great Financial Crisis as a journalist (and later wrote a book about it, Fool’s Gold2) I became convinced that there was another reason for the disaster: the modern financial system was surprisingly fragmented, in terms of how people organized themselves, interacted with each other, and imagined the world. In theory, pundits often like to say that globalization and the Internet are creating a seamless, interlinked world, where markets, economies, and people are connected more closely than ever before. In some senses, integration is under way. But as I dug into the 2008 crisis I also saw a world where different teams of financial traders at the big banks did not know what each other was doing, even inside the same (supposedly integrated) institution. I heard how government officials were hamstrung by the fact that the big regulatory agencies and central banks were crazily fragmented, not just in terms of their bureaucratic structures, but also their worldview. Politicians were no better. Nor were the credit rating agencies, or parts of the media. Indeed, almost everywhere I looked in the financial crisis it seemed that tunnel vision and tribalism had contributed to the disaster. People were trapped inside their little specialist departments, social groups, teams, or pockets of knowledge. Or, it might be said, inside their silos.

  That was striking. But as the 2008 crisis slowly ebbed from view, I realized that this silo effect—as I came to call it—was not just a problem at banks. On the contrary, it crops up in almost every corner of modern life. In 2010, I moved from London to New York, to run the American operations of the Financial Times, and when I looked at the corporate and government world from that perch, I saw a fragmented pattern there too. The silo syndrome cropped up at gigantic companies such as BP, Microsoft, and (later on) General Motors. It plagued the White House and Washington agencies. Large universities were often beset with tribalism. So were many media groups. The paradox of the modern age, I realized, is that we live in a world that is closely integrated in some ways, but fragmented in others. Shocks are increasingly contagious. But we continue to behave and think in tiny silos.

  So this book sets out to answer two questions: Why do silos arise? And is there anything we can do to master our silos, before these silos master us? I tackle this partly from the perspective of someone who has spent two decades working as a financial journalist, observing global business, economics, and politics. That career has trained me to use stories to illustrate my ideas. So in this book you will hear eight different tales about the silo effect, ranging from Michael Bloomberg’s City Hall in New York to the Bank of England in London, Cleveland Clinic hospital in Ohio, UBS bank in Switzerland, Facebook in California, Sony in Tokyo, BlueMountain hedge fund in New York, and the Chicago police. Some of these narratives illustrate how foolishly people can behave when they are mastered by silos. Others, however, show how institutions and individuals can master their silos. Some of these are stories of failure. But there are also tales of success.

  But there is a second strand to this book. Before I became a journalist (in 1993), I did a PhD in the field of cultural anthropology,I or the study of human culture, at Cambridge University. As part of this academic work, I conducted fieldwork, first in Tibet, and then down on the southern rim of the former Soviet Union, in Soviet Tajikistan, where I partly lived between 1989 and 1991 in a small village. My research was focused on marriage practices, which I studied as a tool to understand how the Tajik had retained their Islamic identity in a (supposedly atheist) communist state.3

  When I first became a financial journalist, I was often wary about revealing my peculiar past. The type of academic qualifications that usually command respect on Wall Street, or the City of London, are MBAs or advanced degrees in economics, finance, astrophysics, or another quantitative science. Knowing about the wedding customs of the Tajiks does not seem an obvious training to write about the global economy or banking system. But if there is one thing that the Great Financial Crisis showed it is that finance and economics are not just about numbers. Culture matters too. The way that people organize institutions, define social networks, and classify the world has a crucial impact on how the government, business, and economy function (or sometimes do not function, as in 2008). Studying these cultural aspects is thus important. And this is where anthropology can help. What anthropologists have to say is not just relevant for far-flung non-Western cultures, but can shed light on Western cultures. The methods I used to analyze Tajik weddings, in other words, can be helpful in making sense of Wall Street bankers or government bureaucrats. The lens of anthropology is also useful if you want to make sense of silos. After all, silos are cultural phenomena, which arise out of the systems we use to classify and organize the world. Telling stories about the silo effect as an anthropologist-cum-journalist can thus shed light on the problem. These tales may even offer some answers about how to deal with silos, not just for the bankers, but government bureaucrats, business leaders, politicians, philanthropists, academics, and journalists too.

  Or that, at least, is my hope.

  * * *

  I. In the United States this is known as cultural anthropology. In the U.K. it is known as social anthropology. Either way, cultural/social anthropology is about the study of how people live, think, and interact, not how they function in a biological sense. The study of hum
an biology and evolution is normally called physical anthropology. These branches of anthropology often blur. But cultural anthropology does not usually place great emphasis on science.

  Introduction

  BLOOMBERG’S SKUNKWORKS

  “We can be blind to the obvious, and we are also blind to our blindness.”

  —Daniel Kahneman, Thinking, Fast and Slow1

  IN THE EARLY HOURS OF April 25, 2011, a brutal fire broke out in a poor neighborhood in the Bronx, New York.2 Within minutes, flames engulfed the building, at 2321 Prospect Avenue. Dozens of firefighters rushed to the scene. But it was too late for Juan Lopez, thirty-six, a soccer-loving Mexican construction worker, his wife, Christina Garcia, forty-three, and her son, Christian, twelve.3 As the flames spread, they were trapped by a maze of illegally constructed walls inside their tiny apartment. On the street outside, firemen and bystanders could hear their desperate screams.4 Nobody could rescue them.

  In the days after their deaths, the press angrily hunted for scapegoats. Some pointed the finger at the New York government in City Hall. The building at 2321 Prospect Avenue had been illegally subdivided many times,5 to enable the landlord to extract more rent; but though the neighbors had reported the dangerous conditions to city officials, nothing had been done. Other observers blamed a gang of local drug dealers,6 who were using the basement of the building as a den. Still others pointed to bankers.7 The registered owner of 2321 Prospect Avenue—Dominic Cedano—had taken out a subprime loan to buy the building during the credit boom, but later defaulted.8 The house had duly been tossed into legal limbo by the banks, and the local utility cut off the electricity. Relatives of the Garcia family begged them to move. But Juan Lopez was finding it tough to get construction work, and since a room at 2321 Prospect Avenue was only $100 a week, the family stayed on, using candles for light. “We don’t know what happened, we’re really sad,” Katia Garcia, a relative, told the newspaper. Rosemary Pagan, a neighbor, said: “Katia kept telling Christina, ‘You have to move from there.’ [But] financially, they didn’t have the money to move.”9

  For a few days recriminations flared. Then they died down, as the media’s attention moved to the next scandal. But a few miles away, inside the majestic City Hall building in downtown Manhattan, the tragedy sparked debate. When the news of the Garcia fire first broke, the mayor, Michael Bloomberg, asked some of his staff if there was anything that could be done to prevent this type of fire. At first glance, it seemed not. One of New York’s dark secrets is that house fires have been tragically commonplace: in the decade before 2011, about 2,700 structural fires broke out each year,10 on average killing about eighty-five people,11 and these typically occurred in subdivided buildings in poor neighborhoods, or places where poor immigrants, such as Juan Lopez, tend to live, crammed together. In theory, New York had teams of inspectors who were supposed to combat this fire risk. But the inspectors faced fearsome odds: each year the Department of Buildings received about 20,000 complaints about dangerous housing, which the building and fire inspectors were both supposed to chase up.12 But New York only employed 200-odd inspectors, who were supposed to monitor four million properties in one million buildings. The city did not have more funds to expand the department.13 The odds were hopelessly stacked against anybody hunting for firetraps. Even when the inspectors looked at buildings that had attracted specific complaints, they only found unsafe conditions 13 percent of the time.

  But though the problem seemed daunting, two men on Bloomberg’s team, Mike Flowers and John Feinblatt, thought they might, just possibly, have a solution. This did not involve anything to do with firefighting equipment, but something else: thinking about “silos.” City Hall sat atop a very large organization, which employed roughly 150,000 people.14 Like most government bodies, the New York government was run as a bureaucracy, subdivided into more than three dozen agencies, which ran a wide variety of services, ranging from firefighting to cultural affairs to urban planning to education. Most of these different agencies were independent in spirit and function.

  Communication between these teams was patchy, at best. The Fire Department was a case in point: the firemen worked in a unit that was revered by New Yorkers, particularly after the bravery that firemen displayed during the attack on the World Trade Center in 2001. But the department was also profoundly independent, semidetached from everyone else. So much so, that when emergency workers rushed to the World Trade Center in September 2001, they discovered that the radios and walkie-talkies used by the fire, policy, and health departments could not tune into the same communication channels.15 Nobody had noticed this before, precisely because these different teams were so disconnected.

  But what would happen, Flowers and Feinblatt asked, if somebody tried to break those specialist silos down? Was it possible to take a joined-up view of the problem of fire risk? Could silo-busting change how people imagined firetraps—and even save lives? It was a novel idea. Indeed, it seemed so alien to the City Hall culture that Feinblatt and Flowers initially kept their project secret, christening it “skunkworks” (a reference to the hidden military programs that the U.S. defense company Lockheed Martin has long organized to design military planes).

  But in the months after the Garcia deaths, this skunkworks ended up producing some surprising results. What Flowers and Feinblatt discovered is that if you start thinking about silos, and even breaking them down, this endeavor can produce big wins. And not just in relation to handling fire risks, but almost every other area of the modern world.

  MIKE FLOWERS NEVER EXPECTED to be a silo-buster. His journey toward that role started a long way from New York, in the unlikely place of Iraq. A burly, cheery man who grew up in Philadelphia, he initially trained as a lawyer, and during the 1990s he worked as a government prosecutor in the tough jungle of the Manhattan district attorney’s office.16 The role suited him: with his balding head and rapid-fire speech, Flowers has a passing resemblance to James Gandolfini, the actor who played Tony Soprano on the television show—albeit with a slimmer physique and more affable manner.

  But after a few years of the gritty Manhattan beat, Flowers tired of the daily battles and moved to Washington to work in a high-paid private legal job. Then he swerved again, after deciding that corporate law was just too dull for his taste, and signed up for a job in postwar Iraq. There he started working as government prosecutor for the American military, which had taken control of the country a year before and was starting to conduct trials of officials linked to the former regime of Saddam Hussein. One of his first tasks was to bring witnesses from across Baghdad into the courtroom in the military zone to testify against Hussein.

  It was a difficult job, since the Baghdad traffic was constantly being snarled by car bombs or roadblocks. “It was a war zone, then, obviously, and I was trying to get my witnesses in and out for the trial,” he recalled. “We had a problem moving them about, and not getting shot at.” Initially Flowers accepted this unpredictability. One day, however, he got chatting to a young marine, and realized that an innovative piece of research work was being conducted by a part of the military that went under the unwieldy name of the “Joint Improvised Explosive Device Defeat Organization,” or JIEDDO. This initiative pooled data from the traffic flows and matched it up with an analysis of where the roadside bombs were exploding. Nobody had tried to link these pools of data before. But when they were combined, they illustrated a pattern: whenever a car bomb was about to go off in a certain quarter of town, the traffic in that area would die down. So Flowers started watching the JEIDDO data to get clues about when violence might erupt—and when the traffic lulls suggested that he should not move his witnesses around. “I guess that was because there was local intelligence, people knew that an attack was planned and got out the way,” Flowers observed. “But to be honest I didn’t care about the cause—all I wanted to know was if I should bring my witnesses in on a Tuesday or Wednesday, say.” Either way, it taught him a simple lesson: sometimes it paid dividends to conn
ect seemingly unconnected bits of information.

  In 2009, Flowers returned to Washington to work with the Senate team conducting investigations into the 2008 financial crisis. But then officials at the City Office of New York offered him a job investigating financial fraud in New York. He was wary, not wanting to get sucked into the never-ending task of grappling with financial reform. So he presented an alternative idea: if he came to New York, could he conduct investigations by using some of the data-crunching—or connecting—techniques that he had witnessed in Iraq? “I am a lawyer by training, not a math geek, but my experience of Baghdad had shown me what you do with data,” he explained. “And I knew that New York City is the ultimate collector of information—it collects data on everything. Traffic tickets, building codes, tax liens, you name it! So I reckoned that if anyone could ever get all their arms around that information, it could completely change how we investigate fraud. And from that it’s a gradual step to say that the information is not just used for fraud, but for anything the city does.”

  Flowers’s timing was perfect. At the start of the decade, Bloomberg had been voted in as mayor after a career spent in finance and then as an entrepreneur running a successful financial data company. He had arrived in his post determined to change how City Hall worked, and had two particular obsessions. One was an interest in how organizations managed information flows; or more often mismanaged it. “If you cannot measure it, you cannot manage it,” was one of his favorite slogans.17 His second obsession was breaking down internal silos: he was convinced that the best way to run an office was with open plan office spaces that forced employees to intermingle. This was not how government was usually laid out: City Hall—which dates from 1812, making it the oldest municipal building in America18—had traditionally been divided into dozens of tiny offices, separated by thick walls, marble halls, and pillars. But when Bloomberg arrived he demanded that the officials move out of their marble rabbit hutches and into the only large space that available—the historic speakers’ hall. There he placed dozens of desks under the oil paintings and statues to create an open plan office. He called it the “bullpen.” “Everyone has the same size desk, the same size computer,” explained Robert Steel, deputy mayor. “The mayor sits in the middle, along with everyone else.”