Consensus rarely holds for ever. As we saw above in the examples of shaken baby syndrome and stomach ulcers, contrarians come along and shift opinions. As the philosopher Thomas Kuhn asserted, the evolution of knowledge is generally peaceful. But when learning and knowledge are partly formed by social interactions, not just supported by them, then knowledge can go down a very crooked path. If we were always conscious of this, it might not be such a problem; but academics and other experts may, consciously or unconsciously, follow a group consensus. It is lucky, then, that, as Kuhn observed, there are intermittent revolutionary phases accompanying paradigm shifts – when the consensus is suddenly jolted onto a new path.32
Amateur experts
Some of the lessons from experts can be applied to collective decision-making more widely. Experts are not the only ones susceptible to these individual pressures when forming opinions with consequences for others – juries are an example. The balance of positive and negative externalities will intensify as opinions spread from experts to amateurs, because amateur opinions have weaker foundations. By definition, amateurs do not have deep knowledge and expertise. They have less private information to use, and this makes them more strongly susceptible to social influences. Sometimes herding is the only obvious option when information is very patchy and uncertainty is endemic. On juries, judgements concerning guilt or innocence are often influenced by group dynamics and herd behaviour. Mock jury experiments illustrate that social influences can generate significant distortions.33 Mock jurors are susceptible to peer pressure and some studies have shown that the susceptibility of an individual juror to the opinion of others is affected by individual differences in personality. The fact that some types of juror are more conforming than others suggests that some jurors may be more easily influenced than others. Juries are not as impartial, objective and uniform as we need them to be.34
Another example of amateurs thrown into a quasi-expert role are the lay members of investment oversight committees. Dr Anna Tilba from Durham University and I were commissioned by the UK’s Financial Conduct Authority to explore some of these group influences. We focused on the impacts they might have in impeding competition in the UK’s asset management industry. This industry includes the large institutional investors – pension funds, insurance companies, charities and endowment trusts. Together they manage large portfolios of funds, and oversight committees are constituted to ensure that this job is done well. We focused particularly on pension fund oversight committees, which manage very large sums of money. Amateurs, such as employee representatives, are included on the investors’ decision-making committees. A key task for these committees is to appoint investment consultants, and they often do this by way of a ‘beauty parade’, whereby different investment consultants present what they could offer the investor, and the committee decides who to choose.35 Herding can have a strong influence during these parades. Imagine that one member of an oversight committee has superior private information. They may have done extra research into the options available and/or the track records of the investment consultants interviewed by the committee. But when this committee member sees how other members are deciding, especially more senior members and/or other members to whom they attribute superior decision-making capacities, then they will often defer to the group decision. Amateurs are often included on these committees because a diversity of opinions is valuable. If, however, these amateurs are too easily persuaded to go with the consensus view, then the value of their representation will be lost.36
All experts – whether scientists, academic researchers or expert witnesses such as doctors and lawyers – are generally keen to give their objective view, based on truth. We need people to interpret evidence for us. In reality, poor information, unreliable data and profound uncertainty mean that it is not so easy to untangle the evidence. We cannot always separate good hypotheses and theories from bad. When the lacunae in knowledge or understanding are large, experts become as susceptible to herding influences as anyone else. As we have seen, this social susceptibility can have profound implications, not only for the individual expert, but also for the path of knowledge and research. The wider social costs can be large.
Whether experts are copycats or contrarians, interpreting their opinions can be problematic. We cannot know whether a copycat or a contrarian opinion is better. Expert opinions may be distorted by consensual experts herding behind the consensus view because they find it easier, or by contrarian experts promulgating a divergent view for the sake of attention and career progression. We need contrarian experts, but we need them to be contrarian for good reasons. The challenge is to separate the motivations and incentives that might lead an expert mistakenly to agree with a herd consensus from a genuinely supported consensual view that is correct because the consensus is correct. Similarly, in interpreting a contrarian expert’s views, the challenge is to balance the extent to which contrarians’ personal motivations and incentives are driving their opinion against the extent to which their contrarian view is more likely to be correct because it is challenging a misplaced consensus. The herd consensus may be right, it may be wrong. In interpreting expert opinions, our Herculean challenge comes in telling the difference. When experts find it difficult to interpret evidence, then they will be less sure that they have the correct answer. And, as we have seen in previous chapters, herding is more likely to take hold when people are unsure. An expert with strong convictions may be less susceptible to blind conformity, but if their strong convictions reflect overconfidence then their dissent may be as destructive as being excessively susceptible to collective opinion.
So, what can we do about it? Some solutions may lie in developing some parallels with the literature on the management of common resources. Nobel Prize-winning economist Elinor Ostrom explained how close-knit communities manage common resources well, much better than many economists predict.37 Are there lessons for research and knowledge management, tailored to getting the best from experts whether they be copycats or contrarians? We cannot rely on individuals to manage knowledge and expertise because their incentives and biases can lead them, consciously or unconsciously, down a wrong-headed path. We need institutions to ensure the safe stewardship of expertise.
What concrete solutions could be introduced? Communities of researchers and experts should encourage the extensive replication of results. Academic communities could move away from the idea that only novelty and originality have value and are worthy of publication. Professional societies are already developing initiatives in this spirit, such as the Association for Psychological Science’s ‘Registered Replication Reports’ policy. Similarly, journals such as the Journal of Negative Results can play an important role in controlling the fads and fashions in academic research driven by an ephemeral preoccupation with attracting public attention. Journal editorial boards can limit the influence of social pressures by ensuring the anonymity of journal submissions and blind reviewing of submissions. We need more academic, scientific and professional institutions that encourage dissent. Even-handed monitoring of researchers, publishing the names of a paper’s reviewers alongside the paper, and requiring researchers to publish the data they have used to justify their conclusion can all help. Some journals and learned societies have instituted these solutions already. But it will be difficult to implement practical concrete initiatives if professional associations, expert groups, journals and publishers have too much invested in the status quo.
In this chapter, we have seen that social influences have more traction in uncertain situations. How do we judge our experts? They can be all combinations of good and bad, and right and wrong. Often, we can’t know which. We know that Marshall and Galileo were correct. Stapel and Wakefield were wrong. The problem remains that we cannot judge very easily if we do not know the truth. But, as noted above, we can implement practices to ensure that we get as close to the truth as possible as soon as possible. Better standards of analysis and reporting in academic research, better and mo
re transparent journal review protocols, better education so that lay people can more easily understand scientific arguments, clearer sanctions on experts who exploit their authority over others: all these solutions could help to ensure that distortions reflecting the unconscious biases of experts, whether copycats or contrarians, are minimised.
A feature we have seen in this exploration of experts and their opinions is that a forceful personality can often distort experts’ assessment of evidence. These distortions can be especially large if a vigorous, aggressive personality leads a group. Group members will, understandably, be reluctant to dissent from the views of the leader, either for psychological or economic reasons. The relationship between leaders and their followers illustrates again that copycats and contrarians do not exist in isolation, but are enmeshed in a symbiotic relationship. We shall turn to this relationship between contrarian leaders and copycat followers in the next chapter.
8
Following the leader
In the well-known fairy tale ‘The Pied Piper of Hamelin’, a rat-catcher is hired by the mayor of Hamelin to deal with the town’s rat infestation. Playing his magic pipe, the Pied Piper entices the rats away from the town and drowns them in the river. When the mayor refuses to pay the rat-catcher he punishes the town by luring its children away, following him and his music into the mountains. It is a strange and wonderful story, though there may also be some truth in the tale, with some accounts suggesting that it concerns the deaths of children during the plague.1 Whatever the case, it is an intriguing example of a leader’s power over their followers.
Another all-too-real example is provided by today’s global terrorism. On 11 September 2001, nineteen al-Qaeda terrorists led four coordinated aircraft attacks on New York, Washington and the Pentagon. The attacks caused the deaths of close on 3,000 civilians, with many others injured, as well as trillions of dollars of damage to property and infrastructure. This event is burnt more indelibly on our collective memory than any other in recent times. The motivations of al-Qaeda’s founder Osama bin Laden and his confederates seem to be straightforwardly apparent: they gained power and some gory glory from the event. The question that seems unanswerable to many of us is: What led those who directly perpetrated the attacks to obey their leaders in sacrificing their lives in such a spectacular way? This is not a phenomenon limited to the religiously fervent in today’s War on Terror. Pressure to participate in horrific acts – from discrimination to genocide and everything in between – is dispiritingly regular in human history. Not even major atrocities such as the Nazi Holocaust and Stalin’s purges are as uncommon as we might hope. Genocides are an enduring feature of our history, including those that happened in Rwanda, Bosnia and Darfur not so long ago, and in Iraq and Syria today.2
One of the most potentially sinister facets of herding is the relationship between a particular type of contrarian – a leader – and a particular type of copycat – a follower. The interactions between these leaders and followers can have large impacts, positive and negative. A leader’s influence can be detrimental on a catastrophic scale. Many brutal dictators have committed horrific crimes against humanity, demonstrating the terrible consequences that can emerge when people blindly obey a despot. And, on a lesser scale, we are surrounded by ambitious politicians manipulating voters in their personal pursuit of power. More edifyingly, some of us also have opportunities to follow benign, egalitarian and benevolent leaders such as Mahatma Gandhi, Martin Luther King and Nelson Mandela, to name a few of the most famous. When we are led by inspiring leaders, the consequences can be as positive as the consequences of following brutal leaders are bleak.
What motivates us to follow a leader? Why do some people demonstrate extreme manifestations of loyalty? For Sigmund Freud, whose analysis of group psychology we explored in chapter 2, these leaders are essential to our group relationships, especially in institutions such as the army and the Church.3 For Freud, leaders play a transformative role:
All the members [of a group] must be equal to one another, but they all want to be ruled by one person. Many equals, who can identify themselves with one another, and a single person superior to them all – that is the situation that we find realised in groups which are capable of subsisting. Let us venture, then, to correct [the assertion] that man is a herd animal and assert that he is rather a horde animal, an individual creature in a horde led by a chief.4
Business leaders and followers
The maverick entrepreneurs we explored in chapter 6 provide a simple example of leaders in the economy. They often lead the way in producing and distributing innovative products and services, with other businesses following along behind. Imitation is a common strategy in business, and it can be a good way to maximise profits. Joseph Schumpeter, whose ideas about innovation and entrepreneurship we also introduced in chapter 6, explored how businesses’ decisions to imitate each other play out in leader–follower relationships. For Schumpeter, these entrepreneurial leaders are essential to a thriving economy. Innovative, risk-seeking entrepreneurs lead swarms of imitators and so play an essential role in catalysing new waves of business activity.5 At a microeconomic level, leader–follower relationships are easy to explain in terms of self-interest and can be understood from the relatively simple perspective of rational choice theory. German economist Heinrich Freiherr von Stackelberg captured leader–follower relationships in his model of industry leadership of oligopolistic firms (a classic example in undergraduate economics textbooks).6 Stackelberg’s model is used to illustrate what economists call a first-mover advantage. If a new business produces something innovative, or perhaps just moves into a new area that is currently lacking a product or service, it takes the advantage of being first on the scene and mops up most of the potential customers. Latecomers are left with just the small number of customers remaining.
To illustrate, imagine a small town that is not yet connected to broadband. An internet provider spots the opportunity and wants to enter the market. To do so the provider needs to invest a lot in terms of start-up costs, new technology and new infrastructure – these are examples of what economists call barriers to entry. When barriers to entry are high and costly, it is hard for new businesses to enter a market because they have to spend so much to get started. For the first provider to enter the market, the revenue and profits may justify the costs of entry into the market. But if a second internet provider considers entering the market, they too would have to overcome the same barriers and invest in new technology and infrastructure – but for much less revenue if most potential customers have already signed up to the first provider. The business case for this second mover may not be strong, so they may decide not to bother. These first-mover advantages are one reason why monopolies and oligopolies face so little competitive pressure to bring their prices down to a level consistent with consumer welfare, and this is why these types of industries are often regulated.
But leading businesses do not always enjoy a first-mover advantage. In other situations, perhaps where the business model, product or service is more complex, a follower can learn from the leader and improve their business strategies accordingly. Then the follower will be able to enjoy a second-mover advantage. Here, the follower wins. Drug design by pharmaceutical companies is a contentious example. One business invests money in research and development to develop a new pharmaceutical. A follower can come along and free-ride on the investment and technological innovations of the first business, offering a generic medicine at a much lower price and thus capturing a good chunk of the market. Partly, this is a good thing for consumers, particularly in the developing world, where people are urgently in need of access to cheap pharmaceuticals. If followers can take away a good chunk of your profits, though, what incentive is there to be a leading innovator? The first movers therefore protect their innovations with patents. The general point is that either leaders or followers – first movers or second movers – can be winners in a simple economic world. Some successful entrepreneurs
and speculators will be aware of when it works to be a leader and when it works to be a follower. They will build these insights into their business strategies, swapping roles when there is a suitable opportunity.
Economic theorist Harold Hotelling presented another microeconomic perspective on copycats in the business world in his simple model to explain why businesses copy each other in deciding where to locate their premises. Imagine that there are two ice-cream sellers, Ben and Jerry, on Bondi Beach. You would think that each would locate themselves a long way away from each other so as not to be competing for customers. Hotelling’s model shows that the ice-cream sellers will in fact capture a much smaller chunk of the market if they are far away than if they are close together, and so both sellers will move until they are as close to each other as is possible. Let’s say that Ben has already set up his ice-cream stall in the middle of the kilometre-long Bondi Beach, and Jerry sees he’s doing a roaring trade. We’ll also assume that the potential customers are all lazy beach bums and will just go to their closest stall to buy their ice creams. Where should Jerry set up his business? If Jerry decides to set up shop 200 metres south of Ben, he will attract a total of 400 metres-worth of customers: 100 metres-worth to the north (i.e. half the customers between Ben and him, because the other half will still be closer to Ben’s stall and buy their ice creams there) plus all the customers south of his own stall – another 300 metres. Ben will do much better: he will get all the customers north of his stall (500 metres-worth) as well as the 100 metres-worth of customers between himself and Jerry – a total of 600 metres-worth of customers. But then Jerry thinks: what if I set up right next door to Ben? Jerry will then capture all the 500 metres-worth of customers to the south of the two stalls, and Ben will keep just 500 metres-worth of customers to the north. Jerry will maximise his profits, and take half of Ben’s, by locating himself as close as possible to Ben. Hotelling’s model helps to explain why we often see similar shops – takeaways, betting shops, clothes retailers, estate agents – all collected together in one area of our high streets. Businesses copy each other with their business location decisions and, in imitating other businesses, business leaders capture markets and customers.7 Political scientists have also borrowed this insight to formulate the median voter theorem, explaining why political parties will try to pitch their manifestos to the median average consensus view to gain the majority vote share – though this insight seems less enduring in today’s more polarised political landscape.
Copycats and Contrarians Page 21