Chapter 5 revisits coordination problems, which were discussed in Chapter 2, but now explores how people can remain independent thinkers yet still arrive at a collective solution that conforms with the majority opinion. The chapter begins with an example of a coordination problem that is constantly present and constantly solved in daily life: pedestrian traffic. Surowiecki notes that crowds are relatively well coordinated because pedestrians trying to navigate packed roads will constantly anticipate the behavior of other people around them, even if they do so unconsciously. For instance, if person A accelerates to overtake person B, the latter may naturally slow down their rhythm to allow for a smooth transition. Pedestrian traffic is an embodiment of the coordination problem because there is not an objectively “correct” answer to how people should walk. Rather, the answer changes according to what everyone else in the collective does.

Thus, coordination problems are about finding the right solution for the individual and the collective. People have to think about the right answer in relation to what other people think the right answer is. For example, a worker might judge their salary’s appropriateness based on what other people with similar jobs are making.

Coordination problems are different from the cognitive problems discussed previously, as those have fixed answers. The answers to coordination problems are more complex, and crowds often arrive at good rather than optimal conclusions. Additionally, there is no independent decision-making, as everyone’s choice will depend on what everyone else does.

Social scientist Thomas C. Schelling created an experiment to measure the extent to which people can coordinate with each other. He gave his college students a problem, asking them when and where they would choose to wait for someone if they had forgotten to communicate exactly when and where in New York City they were scheduled to meet. Most of his students responded that they would choose the information booth at Grand Central Station at noon, a result that suggests that even when people have very little information, they have a good chance of selecting the same answer.

This idea is embodied in Schelling points, named after Schelling’s experiment, which describe the ability for people to arrive to collectively beneficial results without being directed and without talking to one another, so long as they think that others are also trying to match the group. Schelling points show that people’s experiences of the world are, to a large extent, similar. Most importantly, they help clarify why independent actors can still come to conclusions that benefit group coordination.

Culture plays a big role in shared experiences: Schelling’s New York students chose Grand Central Station because it represented the same thing to them, but the same is likely not true for people from abroad. Therefore, it’s possible to conclude that culture enables coordination.

Conventions also help regulate people’s behavior by establishing a pattern to follow. They help with coordination, such as when people naturally veer to one side of the stairwell to facilitate traffic from the opposite direction. Although some conventions are established by force, most of the time, they work best when they are collectively accepted and internalized. For example, the implicit rule of “first come first serve” for securing seating on a bus or train is so internalized that people take it for granted. When social psychologist Stanley Milgram tried to test this by asking his students to request strangers to give up their place without giving them an explanation, the hardest part was mustering the courage to ask. Conventions are internalized, accepted rules that guide the actions of most people, which is why they contribute to the ability for crowds to be wise.

The free market is a perfect example of a continuous coordination problem, as it is the process of finding the right balance of supply and demand, costs and profit. It involves producing the right thing in adequate quantity to distribute to the right consumers. Economist Vernon Smith conducted an experiment that proved that people can arrive at a collective solution that is near ideal even while working on imperfect information. He divided a group of students into buyers and sellers and then asked them to bid in a double auction where sellers were tasked with fetching a price above a certain margin, while buyers were tasked with purchasing at no higher than a specific price. By the end of the auction, most students had arrived at the market-clearing price, which is exactly where the demand and supply were expected to meet and where the group’s total gain was at its maximum. This showed that economic theory is possible in practice, even when the information and the traders are imperfect. It further showed that, even when every individual is attempting to maximize their own gains, they can still efficiently arrive at a near-optimal rate for the whole group, meaning that problems of coordination can be solved even when every individual acts according to their self-interests.

Smith’s experiment also demonstrated that market efficiency is not the same as evenly distributing wealth: Traders that were given less funds at the start of the experiment earned the same absolute amount as their rich counterparts, which meant that they were no better off by the end. This chapter thus demonstrates that individuals, when attempting to match their answers to the group, are often correct in their interpretations, thereby helping solve coordination problems.

This chapter looks at cooperation problems, the third and last type of problem that crowd wisdom can solve. Cooperation problems are similar to coordination problems in that they have no fixed answers, and arriving at the optimal solution requires everyone to account for everyone else’s actions. However, whereas coordination problems can be solved even when everyone pursues their self-interests, cooperation problems require people to adopt a broader understanding of self-interest to include the group and a longer timeframe. Critically, they require mutual trust.

Corruption is the greatest challenge to crowds solving cooperation problems that require everyone to chip in for the benefit of the group. However, if a very small group of people do not contribute, they can still benefit from the result of everyone else’s efforts through freeloading. If they are caught but not punished, honest contributors will lose trust in the system and question whether they should continue to buy in.

Psychologically speaking, when faced with a choice, people will tend to choose the option that benefits them personally, and this choice will not depend on the actions of others. However, in practice, there is a prominent exception to this rule: People will often forego their own benefit if they get to punish people they perceive as freeloaders. This behavior is called strong reciprocity and is defined as the willingness to punish bad behavior even when there is nothing in it for personal gain. It is this innate desire to maintain justice that allows the group to solve cooperation problems even without meaning to.

One of the most well-known psychological experiments is the ultimatum game, which proves that most people value maintaining cooperation, even with strangers, above maintaining their self-interests if they perceive cooperation to benefit the group and themselves down the line. The experiment pairs two people, where person A is given $10 to divide with person B. Person A can propose the terms of the exchange, and person B can decide to accept or reject their offer. If person B accepts, they each pocket their side of the deal, but if person B refuses, nobody gets the $10. If both players were economically rational, person A would offer to give person B $1 and keep the rest. Person B would accept the deal because leaving with $1 is still a gain. However, in practice, the most common offer is $5, and any offer below $2 is promptly rejected. Person B is therefore more willing to punish person A for being unfair than gaining a share that they believe is below their worth, and person A, anticipating this, often chooses not to split the money. This ultimatum experiment shows that players on both sides do not choose the materially superior option; their actions are influenced by what they perceive others will do, and that leads them to the cooperative solution.

One experiment changed the conditions of the ultimatum game: Experimenters asked the two people to take a test first and gave the $10 to the person who performed better on the test. They then repeated the negotiations, and this time, person A, who did better on the test, was more likely to offer less of the $10, while person B accepted the lowball offers every time. This conclusively demonstrated that people seek a reasonable deal between accomplishment and reward rather than looking for the rationally and economically correct decision.

This concept of strong reciprocity guides people’s actions in practice: For example, most people tip at restaurants that they will never frequent again, despite not gaining anything from doing so and not losing anything from opting out or freeloading. Strong reciprocity is actually a fundamental prerequisite to capitalism because fairness is essential to maintaining prosperity in the long run. Businesses rely on building relationships with suppliers, partners, and customers to maintain efficient operations, beat competitors, and establish trust in their brand. Older societies traded with people inside their communities; for example, Quakers became reliable traders because they had a reputation as trustworthy. Modern capitalism, however, expanded globally because it stretched this trust beyond ethnic, religious, or exclusive circles. This explains why modern businesses are much more ready to deal with strangers. It is also why modern consumers are relatively certain that they are not being scammed when buying products.

At the foundation of the modern capitalist system is self-interest in the context of strong reciprocity. If everyone, including competing businesses, protects their own interests and keeps others in check, then the system becomes reliable and cooperative overall.

However, this is not always the case. The more trustworthy an existing system is, the less incentive there is to improve it, as nobody wants to contribute money to changing the way things are, especially if they cannot convince others to contribute and there is no way to prevent non-contributors from enjoying the new system. Surowiecki uses the example of television surveys as an instance where individual smart players are incapable of cooperating due to nearsightedness and the desire to freeload. The company Nielsen Media Research installed “people meters” on 5,000 American televisions, which track what TV shows are being watched in a household. This helps advertising companies better understand how and when to run ads. The only alternative type of data collection is from a polling system called “sweeps,” where surveys are sent out to households four times a year asking them what they watched. Although “people meters” are proven to be more accurate, nobody in the television industry has adopted Nielsen’s people meters because they are exponentially more costly than sending out surveys. Even with the possibility of earning exponentially more in ad revenues, no players in the television business want to shoulder the costs for everybody else to freeload. Similarly, it is difficult to coordinate a collective donation, as everyone is waiting on the possibility of freeloading. Surowiecki concludes that the freeloader problem stops otherwise intelligent people from making the cooperative choice, as this kind of enlightened judgment requires the ability to think of the long term, a major problem for corporations that are looking to maximize profit immediately.

Paying taxes is another big cooperation problem, and it perfectly demonstrates all the previously established behavioral patterns. There is the freeloader problem: People reap the benefits even if they evade paying their taxes, so long as the majority are paying. However, if no one pays, the system falls apart. Most people choose to not evade because they believe in strong reciprocity, or the idea that other people will also chip in and will be punished if they do not. If they suspect that others are cheating and are not punished, they will not cooperate. A social experiment demonstrated that 25% of people are freeloaders, that a very small minority are altruistic and contribute no matter what others do, and that the vast majority of people swing, meaning that they stop paying if they perceive others as freeloaders. Since it is economically rational to freeload, Surowiecki concludes it is important to create incentives to reinforce cooperation.