60 pages • 2-hour read
The Wisdom of Crowds: Why the Many Are Smarter Than the Few and How Collective Wisdom Shapes Business, Economies, Societies and Nations
A modern alternative to SparkNotes and CliffsNotes, SuperSummary offers high-quality Study Guides with detailed chapter summaries and analysis of major themes, characters, and more.
Part 2, Chapters 10-12Chapter Summaries & Analyses
Part 2, Chapter 10 Summary: “The Company: Meet the New Boss, Same as the Old Boss?”
This chapter explores the ways in which a large company with hundreds of employees can organize to maximize their productivity so that the value of the product that the group makes is greater than what can be achieved by the sum of its parts. It broadly surveys the history of American production from World War I until the present, arguing that although businesses discuss to decentralization, they do not actually achieve it in practice.
The chapter begins by using fashion clothing store Zara as an example of successful coordination and decentralization. Prior to Zara’s business model, the fashion industry was inefficient. There would typically be a lag of up to six months between artists completing designs and the clothing arriving in store. This meant that artists had to predict what would be trendy six months ahead. When they were wrong, the unpopular items became piles of unsold inventory.
Zara shifted the paradigm by equipping their store managers with devices that allowed them to immediately report what clothing customers preferred. Artists would receive this information and design similar clothing, which was then produced in Zara-owned factories. By allowing store managers to share information with suppliers, and by choosing not to outsource their production—as they must then pay the cost of spending time negotiating with subcontractors—Zara improved the speed and efficiency of their operations to keep up with the pace of changing trends. They solved the coordination problem by matching their production to the fancies of their customers. They decentralized power by allowing managers to make decisions based on their localized knowledge. Finally, they ensured speed and accuracy by providing all parts of the business operation to share their localized information.
Surowiecki emphasizes that though Zara was successful, no single organizational style is perfect for every situation. Luck and circumstances play important roles and are often underestimated factors when people evaluate success.
Surowiecki then groups business organization into three broad categories. The first type, top-down hierarchies, is exemplified by big corporations where every manager needs to report to their manager, until the message is relayed to the Chief Executive Officer (CEO), who makes final decisions. This organizational style allows for decisions to be made swiftly and carried out decisively, but localized information often gets lost, which means there is no reliance on crowd intelligence. If the CEO decides wrong—which is likely given how little they understand the daily realities of the average worker—everyone pays the price.
The second type is characterized by the small business, which comprises of a small group of highly professional individuals. This organizational structure is efficient if members have trust, specialization, and mutual awareness of each other’s abilities. If these conditions are met, individual contributors are less likely to slack off or free ride, contributing to the group’s efficiency. However, there is often a limit to their resources, and this lack of diversity means they pay a greater price of failing.
The final organizational style is characterized by its ephemeral nature: It gathers a group of people to complete a project and then disperses. This is often the preferred style for project-based work. It offers the advantage of hand-picking experts for their abilities and incentivizes everyone to contribute since the reward is project based. However, there may be a lack of trust if the individuals gathered do not know each other and suspect that their peers are in it for different reasons.
Historically, up until World War II, most American businesses believed in a top-down organizational model that valued integration, hierarchy, and the concentration of power in the hands of a few top executives. However, even when companies began realizing the advantages of decentralization, most did not implement it properly. They thought that democracy meant asking people their opinions and getting everyone to agree, when it actually meant distributing decision-making power to different parts of the company. For example, companies would send out polls to employees asking how they felt about the direction of the company but would not allow them any say in changing their circumstances.
Companies can look to Toyota as a success case of decentralization. The business realized that workers must be taught what their individual responsibilities mean in car production. Prior to this shift, a worker at any segment of the assembly line had no idea how the product of their labor would be used by the people down the line; as a result, they did not know that a small difference in their procedure could jeopardize the work of the next person.
Decentralization allows for greater flow of information, which was not possible in a strictly hierarchical organization. For maximum efficiency and accuracy, local problems should be solved by those close to the problem; the CEO should not come up with a solution for a printer jam at an office since they have more productive avenues for their time.
Beyond decentralization, the top-down approach also suffers because there is a lack of cognitive diversity among top managers. This means that they receive all the credit when things go right, which has allowed the US to develop a culture of viewing CEOs as superheroes. The problem is that people believe that putting the right person at the top is key to a business’s success, which leads to inflated bonuses. However, it also means that CEOs are fired at an unprecedented rate when they fail to prove themselves. There is no proof that individuals can consistently make good decisions when attempting to forecast the future in the face of true uncertainty. Therefore, it is no surprise to Surowiecki that CEOs are put on a pedestal just as often as they are removed from their positions.
Once businesses understand how decentralization works, they can reap its two benefits: It encourages employees to be more engaged by giving them control and responsibility over their environments, and it makes coordination easier by reducing the need for supervision, cutting transaction costs, and giving managers more time to pursue other tasks. However, decentralization must be managed in a way that does not promote loyalty to the division above the company, or it risks having unproductive teams fighting for their survival and taking resources away from productive ones. Similarly, decentralization must not promote so much competition among teams or individuals that it compromises mutual trust and incentivizes people to pursue private interests at the expense of the company. Surowiecki underlines that “decentralization only works properly if everyone is playing on the same team” (215).
Finally, Surowiecki defends the utility of decision markets as a means to increase the flow of information and prevent political infighting, sycophancy, and inflated egos. These markets do not need to have the final say, but they should be utilized for the information they provide: The more important a decision, the less it should be the responsibility of a single person.
Part 2, Chapter 11 Summary: “Markets: Beauty Contests, Bowling Alleys, and Stock Prices”
This chapter analyzes the ups and downs of the stock market as it relates to crowd wisdom. It argues that bubbles are examples of crowd wisdom failing because they are the result of the fraying of diversity, independence, and decentralization on the part of traders.
When it operates smoothly, the stock market is incredibly efficient at speculating about the future and predicting outcomes. This is because people buy and sell based on what they believe will happen in the future; if they predict that a company will grow, they will hold on to their shares, whereas if they believe a company to be on the decline, they will look to sell their shares. As explored in Chapter 1, decision markets, which use a stock-trading system to encourage crowds to predict a future outcome, can anticipate outcomes better than the most famous polling systems. This is not because the stock-trading system makes no mistakes but because it relies on crowd wisdom, which is often more accurate than any individual or system.
The stock-trading system functions even when individual traders are irrational and self-interested. This is because crowd wisdom aggregates information from a diverse crowd and allows for the two extremes to cancel themselves out. In other words, individual investors need not be smart for the aggregate of their choice—that is, the collective outcome—to be optimal.
Though the stock market is usually reliable, it remains imperfect. There are times when it fails to match the price of a stock to the real value of the company, such as when bubbles are created. Numerous factors cause this imperfection. First, compared to decision markets, people who buy stocks attempt to predict a distant future where the factors are extremely volatile. Second and most importantly, stock traders are working with a speculative system that is infinite in timeline and has an infinite amount of solutions.
This means that the stock market is trying to solve problems that are fundamentally different from those of the decision market or regular consumer market: They are not dealing with cognitive problems, such as who will win the football season this year or how much one’s computer is worth, which have clear answers and clear timelines at which these answers will become apparent. Rather, stock prices may soar beyond the company’s actual worth, but people may continue to speculate because they can convince themselves that something will happen in the future that will make the gamble of holding on worth it. There is no objective method to demonstrate that the stock market is wrong until the bubble has already burst.
Bubbles are examples of collective decision-making going wrong and are typically caused by a lack of independence, diversity, and private judgment, which are all essential to crowd wisdom. There is a lack of independence because, to a large extent, people appraise the value of a stock based on what everyone else thinks. This is because stocks do not typically decrease in value over time like consumer products (a new computer is valued much higher than an old and used one). Instead, stock prices typically rise over time. Therefore, when buying, people are worried not only about whether the price is fair given the company’s performance in the moment but also about whether everyone else thinks the price is fair. If they believe that others will continue to value the bond at a higher price, they are willing to hold on to sell later. This can cause the price of a stock to become inflated beyond its true worth.
For example, bowling stocks became inflated in the late 1950s because the invention of an automatic machine that would replace the pins made the sport more pleasant to play. Wall Street became infatuated with this, investing over $2 billion in capital into the business. However, there was no evidence that bowling’s initial growth spurt would be sustained, and when bowling companies failed to attract additional demand in the early 1960s, the stocks fell by 80%.
A study conducted at Caltech showed that people often do not recognize that they are in a bubble, and even when they do, most continue to trade bonds at inflated prices, holding on to the belief that they can still cash in in the short term because there is a greater fool out there. The students in the Caltech experiment still bought overpriced stocks in hopes of selling them at even higher prices. They did this even when they knew the real value of the stock, were aware of the exact time the experiment was going to end, and did not even egg each other on. This showed that the price of a bond and its actual value can become divorced in a bubble and that people’s judgment is affected not by the price but by what they think other people think of the price. This interdependence causes crowds to make suboptimal or even disastrous collective decisions.
Bubbles also cause a lack of cognitive diversity and private judgment by encouraging people to rely on information cascades and buying into the popular option without referring to their private information. This is often exacerbated by the lack of diversified information and a fear of missing out. For example, when a CNBC program began to cover financial news, they were the first to share information outside of financial circles. They gained tremendous popularity among veteran speculators and amateur investors alike, to the point that what they reported began influencing the market. This was not because they were extremely reliable or had insider information. Rather, by virtue of knowing that other people would be watching and referring to them, people started buying and selling stocks seconds after CNBC announced anything; some hoped to move before the group and make short-term gains, while others followed the herd.
A lack of diversified information and independence can cause overreactions. In the case of CNBC, the way news was delivered made people assume either the worst or the best. Reporting on a slight dip in value made everyone else lose confidence and caused a frenzy of selling. Meanwhile, a slight gain meant that everyone else would soon look to invest, which caused prices to soar as everyone tried to get in first.
Surowiecki concedes that following others is not always a bad idea, as that has helped the human race survive, and crowds are often wise. However, when everyone piggybacks on this wisdom, they add nothing to it and easily cause information bias. Surowiecki ends the chapter by advocating for more diversity to balance the stock market, such as through short selling (the practice of borrowing and selling a stock predicted to decrease in value in the hopes of repurchasing it later at a reduced price) and creating more diverse channels of information.
Part 2, Chapter 12 Summary: “Democracy: Dreams of the Common Good”
This chapter explores how group wisdom can inform the democratic process. Political scientist James Fishkin is afraid that the American democratic system is fraying as voters become more polarized and less informed. He created a system called deliberative polling, which gathers hundreds of voters to deliberate on relevant political topics ahead of the elections. He believes that only through encouraging open and fair discussions can people become informed voters. Without this, he is afraid that people will vote based on their self-interests and mistaken beliefs.
Surowiecki proposes that it is not useful to think of voting as purely driven by self-interest. Instead, studies have consistently shown that people’s voting patterns are more swayed by ideology (conservatives are more likely to vote against national health insurance) than personal circumstance. This is why so many vote against their own economic benefits. Statistically, most Americans will never be wealthy, but that does not translate into them voting for raising taxes for the rich.
Surowiecki also suggests that it is not necessarily bad that some people vote on mistaken beliefs. According to the wisdom of crowds, even if some individual voters are uninformed, as long as the crowd is diverse enough, this might not prevent their aggregate vote from being optimal.
If democracy requires neither informed voters nor the pursuit of self-interest or group interest, its value must be understood in other ways. Surowiecki argues that democracy’s purpose lies in how it is defined. For some, it provides individuals with a sense of involvement and control; for others, it represents the ability to rule themselves; for still others, it is simply the most effective way to arrive at a collectively optimal solution.
Surowiecki does not provide a specific answer. Instead, he points out that studies have repeatedly and convincingly demonstrated that the average voter is little informed, which is why it is unreasonable to believe that they will make sensible policy choices. However, if the purpose of the representative democratic system is not to get the average voter to make policy choices but for them to elect the right official to help them make the right decision, then crowd wisdom dictates that this is more likely than not to happen. After all, the democratic system allows for a division of labor that helps society remain productive.
For this reason, Surowiecki is skeptical of implementing a technocratic system, which functions by putting into power a small group of elites who will decide everyone’s future. Although the technocratic system is often cited as a good alternative to democracy, like every small, insular group of elites, its decision-making risks falling into the pitfalls of groupthink, information cascades, herding, and a lack of diversity.
If democracy is about the common good, it is unclear whether there is an objective answer to what constitutes the common good. Politicians all claim to work for the common good, yet their visions might still clash. Crowd wisdom might not be as useful here, as it is not a cognition problem with a clear correct answer. However, Surowiecki defends democracy for helping solve problems of coordination and cooperation. Though they do not have definite answers like cognition problems, by working together, an independent and diverse group of people can continuously choose to adapt their own decisions to that of the group.
Crowd wisdom cannot “solve” coordination and cooperation problems, but they work by bringing forth an answer from the group—one that was not imposed on them from above. The decision to go out to vote and upkeep a democratic system is a demonstration and a product of crowd wisdom.
Part 2, Chapters 10-12 Analysis
This final section explores the multiple ways in which crowd intelligence manifests under a capitalist and democratic system. It defends the idea that the consumer and financial markets and the democratic system, by nature of relying on crowd judgment, are fairly optimized compared to alternative organizational systems. Although they are far from perfect, they mostly operate through crowd intelligence and, more often than not, arrive at an optimized solution for the collective. Surowiecki uses these domains to test how institutions translate the core conditions—diversity, independence, decentralization—into practice and to probe the themes of The Limits of Individual Expertise and The Fine Line Between Crowd Wisdom and Herd Mentality at scale. By tying abstract principles to institutions that people encounter every day—workplaces, financial markets, elections—he shows how systemic design either amplifies or undermines the same dynamics introduced in Part 1.
Chapters 10, 11, and 12 all individually spend considerable time discussing instances where the modern business, financial, and democratic systems break down. For example, in Chapter 10, Surowiecki argues that despite paying lip service to decentralization, modern businesses often still operate under a strictly hierarchical system. They mistake the meaning of decentralization as asking employees to arrive at the same opinion when decentralization is about redistributing decision-making rights to different levels of the business. This mistake fuels the “superhero CEO” myth, which equates organizational success with a single leader’s brilliance and failure with a single leader’s flaws. The narrative not only obscures the role of structure and incentives but also sustains inflated executive compensation while discouraging reforms that could genuinely broaden decision-making power.
In Chapter 11, Surowiecki underlines how easily bubbles can form in the financial market if people do not implement measures that deter uncontrolled speculation. In Chapter 12, he points out that groupthink or information cascades can potentially cloud crowd judgment. These failures illustrate the fine line between crowd wisdom and herd mentality. When organizations suppress dissent or overamplify a single signal, imitation replaces information and correlated error grows.
These counterexamples have the specific purpose of warning readers against believing in crowd intelligence as a panacea to the wails of society. Although groups have repeatedly proven that they can make collectively wise decisions, this outcome can only reliably happen when the proper conditions of independence, diversity, and decentralization are met. At the same time, the section continues to challenge the limits of individual expertise by showing that crises often worsen when firms or governments centralize judgment in a few celebrated leaders rather than building systems that aggregate many informed perspectives. The Columbia disaster from Chapter 9 echoes here as an institutional cautionary tale: When dissent is silenced and expertise treated as unquestionable, small-group errors scale up into systemic failures.
Therefore, after tracing the limits of crowd wisdom, each chapter turns to observe cases where businesses, the financial market, and societies have properly leveraged group intelligence. In Chapter 10, Surowiecki uses the example of Toyota to illustrate how decentralization can be properly implemented in big businesses. In Chapter 11, he presents short selling as a means to increase a diversity of knowledge and perspectives in the financial market. Rather than treating contrarian practices like short selling as destabilizing, Surowiecki reframes them as essential design features that keep markets diverse and resistant to herd cascades. Without these correctives, financial systems resemble poorly structured committees that amplify optimism until collapse. In Chapter 12, he defends democracy as a much more optimized organizational system compared to technocracies, which rely on experts rather than group decision-making. These examples also intersect with The Consumer and Stock Markets as Aggregators of Knowledge and Prediction Systems: Prices, internal prediction markets, and reputation systems act as information channels that compress many local judgments into usable signals. In each domain, the key question is not whether the crowd is always right but whether the system is built to preserve the crowd’s conditions for being wise.
Surowiecki concludes that, given that crowd wisdom leads to optimized conclusions more often and more accurately than the decisions of individual experts, people ought the find more ways to make use of it. The section’s closing emphasis is pragmatic rather than utopian. Institutions that preserve independence, broaden diversity, and create credible aggregation should be designed; the limits of individual expertise will become less constraining, while the risks that can come about when toeing the fine line between crowd wisdom and herd mentality will be contained. Democracy, in this light, is less about the rationality of individual voters than about the structural resilience of distributing judgment across millions of ballots; markets, similarly, work not because each trader is rational but because diverse, decentralized signals are aggregated into prices. By embedding this logic into institutions, Surowiecki suggests that societies can turn fragile group tendencies into reliable engines of collective intelligence.
Unlock all 60 pages of this Study Guide
Get in-depth, chapter-by-chapter summaries and analysis from our literary experts.
- Grasp challenging concepts with clear, comprehensive explanations
- Revisit key plot points and ideas without rereading the book
- Share impressive insights in classes and book clubs