Cheating

The New Scientist takes a look at corruption

If you think you're incorruptible, think again. Nobody is immune, but understanding that may make dishonesty easier to counteract

Most of us think of ourselves as pretty honest, seeing corruption as something that involves other people. But new research shows that anyone can be corrupted at the drop of a hat. Indeed, when looked at in evolutionary terms, clinging to the moral high ground could be seen as an irrational position. If everyone else is cheating, then playing by the rules will leave you with the smallest haul – where the haul, whatever it is, translates sooner or later into reproductive success. So it makes perfect sense to be as devious as you can while at the same time exhorting everyone to be honest. “I think of hypocrisy as the background state,” says psychologist Rob Kurzban. Clinging to the moral high ground could be seen as an irrational position


If we are mostly honest, most of the time, that may just be down to a lack of opportunity to cheat. That is certainly what as-yet-unpublished research by Samuel Bendahan and colleagues at the Swiss Federal Institute of Technology in Lausanne indicates. They have devised a game in which players must distribute a sum of money between themselves and their “employees”. There are three options: raise your employees' salaries at your own expense, maintain both parties' rates, or reduce employees' salaries and take home a fatter pay packet – the “theft” option.

Asked beforehand how they should behave, barely 4 per cent of players condoned theft. Indeed, those given control of just one employee refrained from stealing during 10 successive rounds of the game, which was played for small amounts of real money. However, people allocated three employees – giving them more power and more to gain from underhand dealing – soon departed from their initial moral stance. After the first five rounds, 20 per cent had resorted to theft. And, if people were offered more ways to profit at their employees' expense, by round 10 the figure had increased to 45 per cent.

The corrupting influence of power has also been demonstrated by Joris Lammers at Tilburg University in the Netherlands and Adam Galinsky at Northwestern University in Chicago. They primed individuals to feel powerful or powerless by getting them to recall a past incident in which they had experienced one of these feelings. They then split each group in two, asking half to rate certain hypothetical acts on a scale of morality, and others to play a game of dice in an isolated cubicle, reporting their scores to a lab assistant. Possible scores for the game fell between 1 and 100, with higher scores bringing bigger rewards. Dice being a game of chance, the average score should have been around 50, but it was actually 70 for the group primed to feel powerful. Empowered individuals were prone to cheating, say Lammers and Galinsky, yet they were also harsher in their condemnation of immoral acts than were people primed to feel powerless. And they were hypocritical, judging such acts to be less blameworthy if carried out by themselves than by others.

The British historian and politician Lord Acton was clearly right when he said “all power tends to corrupt”. But power doesn't simply provide more opportunity for underhand dealings – it also influences the way we think. Lammers believes that it brings about a kind of moral myopia. “I liken the effects to those of alcohol. Alcohol narrows your focus and it also leads to behaviour which you could call hyper-self-confident or hyper-assertive.” He points out that colleagues at Tilburg, led by Maarten Boksem at Erasmus University in Rotterdam, the Netherlands, have monitored people's brain activity while they were being primed to feel powerful, and found that areas associated with disinhibition were activated.

Power is not the only thing that brings out our dishonest tendencies. Psychological distance also seems to make it easier to perform a corrupt act. Behavioural economist Dan Ariely, at MIT, has found that people cheat more readily for tokens which can be exchanged for money than for money itself. Acting via an intermediary is another way of distancing oneself. Danila Serra of Florida State University in Tallahassee, who is investigating the role of intermediaries in corruption, says there are several reasons for this. One is that the intermediary removes uncertainty by giving a tariff for a certain service, such as bribing a politician, and in doing so helps to normalise the act. What's more, even if intermediaries are merely hired guns, their existence reduces the risk of punishment for their clients and creates an illusion of shared responsibility. A good example is the recent phone hacking scandal in the UK, in which some journalists from the now defunct News of the World newspaper paid private detectives to access people's voicemail messages.

In fact, where there is a culture of it, corruption seems almost to be contagious. Each year, Transparency International publishes a list of nations ranked according to corruption levels. Based on surveys of analysts and people in business, its Corruption Perceptions Index (CPI) has been criticised for lack of objectivity, but Serra's research suggests it does reflect people's behaviour. Working with Abigail Barr at the University of Oxford, she conducted a series of experiments with Oxford undergraduates hailing from 34 countries covering a wide range of CPI rankings. Each person had to decide whether or not to bribe an official for a service, such as being moved up a hospital waiting list. The pair found that people from countries with the worst CPI scores were more likely to engage in bribery. Barr and Serra concluded that our propensity to engage in corruption is strongly cultural, reflecting the social norms of the country in which we live.

Given our tendency to baseness, anti-corruption campaigners have their work cut out. But the news from the lab is not all bad. One ray of hope lies in the discovery that individuals can become less corrupt. When Barr and Serra repeated their experiment, they found that a person's tendency to bribe declined the longer they had spent in the UK. The pair also made an intriguing observation that suggests some individuals might be less susceptible to corrupting cultural influences in the first place. While the corruptibility of undergraduates reflected the CPI ranking of their home countries, the same was not true of graduates – those from countries where corruption is higher tended to be more honest than undergraduate compatriots who had spent equal amounts of time in the UK. Barr and Serra see these people as non-conformers who could one day fight corruption back home. “We think of them as agents for change,” says Serra.

The problem is identifying such people. So far, the hallmarks of incorruptibility remain elusive. All Serra and Barr can say is that their graduate students are a self-selecting group whose decision to study abroad for a PhD makes them unusual. Similarly, Bendahan has found no characteristic that consistently protects people from the corrupting influence of power, even though over 300 students have played his game to date. An initially honest individual, for example, is not immune. Neither does an altruistic outlook or a strong desire to recognise the contribution of others help. The only clue comes from Ariely's research. He has found that the more creative a person, the more likely they are to cheat. “A lot of dishonesty is about being dishonest while telling yourself a story about why this is really okay,” he says, and creative people may be better at that.

A more promising way to reduce levels of corruption might be to beef up the deterrents. Kurzban believes that the only reason people ever consider not cheating is because others occasionally hold them to account. Punishment certainly works in laboratory experiments. When people play cooperative games for a reward, they are far more likely to resist cheating if they know that other players could fine them for trying to profit at the expense of the group. In real life, the punishment inflicted on a cheat tends to be social disapproval, ranging from ostracism to incarceration. In recent months, for example, five British politicians have served prison sentences for fiddling their expenses.

Of the many forms corruption can take, one of the most pernicious and hardest to eradicate is that perpetrated by elite groups against the majority of their fellow citizens, according to Peter Turchin, a population biologist at the University of Connecticut in Storrs. Using mathematical modelling to study human societies, he has found that throughout history this kind of corruption has been a good indicator of a mature state on the verge of collapse.

In a flourishing, nascent society, resources are shared more or less equitably, there is full employment and the population is in a growth phase, Turchin says. With time, population outgrows the demand for labour, the price of labour drops and employers grow rich. This causes inequalities to widen and makes the elites proliferate and compete with each other for power and patronage. If one elite succeeds in grabbing a bigger slice of the pie than others, trouble won't be far off.

History is full of such examples. In 16th-century France, four decades of civil war were precipitated in part when competition between aristocratic clans resulted in the House of Guise elbowing out other noble families. A more recent example of an elite group grabbing too much power, Turchin says, was the election of the former CEO of oil services provider Halliburton, Dick Cheney, as vice president to George Bush. He also notes that Egypt saw a quadrupling of graduates, a classic sign of a burgeoning elite, in the decade leading up to the Arab Spring.

There is evidence that mobilising social disapproval can reduce corruption. In Indonesia, Olken tested the impact of various anti-corruption measures on embezzlement from road-building projects. The most effective method was to increase the number of audits, though it didn't even have to go that far. “Just sending a letter saying the state audit agency is going to come and look at this project reduced that missing expenditure number by about a third,” he says. Grassroots measures were less successful. Olken found that holding accountability meetings at which officials explained to local people how they had disposed of the budget made little difference. However, lab experiments carried out by Serra suggest that anonymous complaints can reduce corruption provided they are logged centrally and trigger an official investigation once a threshold number has been reached.

Finally, on the grounds that powerful people are the ones we should be most worried about, Bendahan says that power should come with more built-in checks and balances. People seeking power must expect restrictions on that power in the form of democratic control, he says. Ariely agrees, but he adds that those at the top must first recognise their corruptibility, something they often fail to do – as illustrated by Nixon's famous “I'm not a crook” speech. It would appear that the first lesson we should all learn from the burgeoning field of corruption research is that nobody is immune.