All because of the Joneses
Imagine the scene. You have just taken delivery of a shiny new car. You are over the moon. At last you have replaced its clapped out predecessor, which excelled in unreliability as much as in lack of cool. This one totally looks the business, drives like a dream, and is more comfortable to sit in than your settee.
But what is this you spot from the corner of your eye? Your neighbours, the Joneses, also have a new car? And it’s bigger and shinier than yours? Way to ruin your day. Good thing you’re an animal lover, otherwise you’d definitely give the cat a good kick.
When we’re talking about the irrationality of us, people, then this kind of situation would seem to be a good example. Your new car met all your needs… except, the Joneses have a better one. To work out the value of a car we don’t just look at the practical, material characteristics like boot space, acceleration and reliability, or even at the emotional satisfaction we experience from placing our derrière in the leather seats or from the badge on the bonnet. We also want it to be a car that puts us at least on a par with to that of our colleagues, our friends and indeed our neighbours. The same often applies to homes, holidays, clothes and even education. We have to keep up with the Joneses.
Economists call these positional goods, a term coined by economist Fred Hirsch (coincidentally, I am writing this piece just a couple of miles from the place where he lived until his premature death, aged 46). Hirsch had observed that people are not satisfied with being better off than their parents and grandparents. If everyone is middle class, then nobody is. To be truly middle class, you need to be better off than your neighbours too, and that means acquiring scarce goods that they cannot afford.
This relativity with respect to our peers is quite likely an evolutionary phenomenon. It is not the absolute height of a flower that determines how much sunshine it gets: what matters is that it is taller than the ones surrounding it. To really impress a potential mate, a peacock needs to have a tail that is larger than that of its rivals; to become the leader of the herd, the male elk needs antlers that are more grandiose than that of his challengers.
Unsurprisingly, we also tend to consider our salaries in relation to that of our peers. In a 1991 book chapter, Amos Tversky and Dale Griffen describe an experiment in which they asked final year students to consider two one-year assignments after they graduated: one paying $35,000, at a company where similarly experienced and trained people earn $38,000, and another one paying $33,000, at a company where similar colleagues earn $30,000. When asked at which job they would be happier, for 62% of the respondents this was the lower paying job where they’d earn more than their peers. Sara Solnick and David Hemenway carried out a similar survey, and found that half of the respondents preferred a world in which they would earn $50,000 per annum while others earned $25,000 over one in which they would earn $100,000 but others earned $200,000.
This is not just a matter of laboratory surveys. The illusory superiority bias (also known as the Lake Wobegon effect, named after the fictional town from Garrison Keillor’s writing, where “all the women are strong, all the men good looking, all the children are above average”) means that we dislike being paid less than our peers. We are (convinced we are) better than most of them after all.
Pay arms race
That is likely why enforced transparency with the aim of lowering the pay of top executives has failed in that objective. Alexandre Mas looked at such a measure imposing disclosure in 1934, and found that average CEO pay actually increased. In a 2012 study Cornelius Schmidt investigated the evolution of executive compensation in Germany after a reform of corporate governance demanded disclosure of individual packages of key executives. He concludes that “enhanced disclosure can lead to a, likely unintended, effect of higher compensation levels and might explain recent excessive compensation.”
This kind of arms race is welfare-reducing, argues Robert Frank, an economist who has been studying positional goods for decades. In his book The Darwin Economy, he compares the evolution of speed in gazelles with that of antler size in elks. In the former species, becoming faster means a greater chance to outrun a cheetah. This confers an advantage to both the individual and the species as a whole. The primary purpose of a male elk’s antlers is to combat other males, and that means natural selection will drive the development of ever bigger, and ever more cumbersome antlers. So what is beneficial to the individual is actually detrimental to the species as a whole: a pack of wolves has an easier time catching elks carrying 18 kilos of bony protrusions. The big antlers are wasteful, and so is, for us people, the chasing of higher income and bigger cars.
But what if it is not individuals that feel hard done by, but whole population groups? Pay discrimination based on gender or ethnicity is a perennial issue. It hit the headlines in the UK again a few weeks ago, following the high profile resignation of Carrie Gracie as the BBC’s China editor because she was vastly underpaid compared to her peers.
A report by PwC found a gender pay gap of just under 7% among 824 BBC journalists. But does that necessarily imply a bias in setting pay? (The report claims to have found none.) One of the problems is indeed the difficulty in controlling for non-gender-related factors in the actual pay. The lack of clarity means it is hard to conduct a meaningful debate: are we talking about annual pay, or pay per hour? Are we mixing up part-time and full-time work? Are we really comparing like with like?
What’s behind the gap?
However, a very recent study carried out by five economists at the taxi platform company Uber, looking at over 740 million trips, shines some welcome light on the discussion. By coincidence, the gap in hourly wages it revealed between male and female Uber drivers is 7% — very close to that found among BBC-journalists. This surprised the authors, who had expected that the gender-blind algorithm allocating rides to drivers would have avoided any discrimination. One of them, the behavioural economist John List, had even predicted a slight advantage for women: since they work less, they could cherry pick the most productive hours during the week, and he also assumed riders would have a slight preference for female drivers.
But no. An advantage of 7% to male drivers… how was that possible? They found the gap is entirely explained by three factors. The first one is driver experience. This favours drivers who drive a lot, and stay in the job for longer: they know better which rides to accept and which ones to reject, for example. This factor explains about 1/3 of the pay gap. The second factor, accounting for around 20% of the difference, is the choice where to drive: men have a higher tendency to pick the most profitable locations. But the most significant factor, explaining half of the gap, is average speed: men prefer to drive faster.
What can we take away from this? The gender pay gap can exist entirely as a result of people’s preferences, as at Uber, without any bias in pay setting. But determining objectively what is behind differentials in wages for most other jobs is very hard to do. And that means unfairness, real or perceived, is likely to carry on influencing our attitude towards pay.
If we don’t know others’ situations, we may well be perfectly content with our lot. But once we find out they’re slightly better off than us, our innate drive to compare ourselves with, and keep up with the Joneses will ensure the arms race continues unabated.
Originally published at koenfucius.wordpress.com on February 16, 2018.
Thanks for reading this story. If you thought it was good, then do let me know: click or tap the 👏 (‘applause’) button (this will also help others find the piece). And do leave a comment — dialogue is a great source of wisdom. Maybe you can also share it in your network: there are handy Twitter and Facebook buttons hereabouts, or simply copy and past the link in LinkedIn or wherever you hang out. Thank you!