The unequal struggle between facts and conviction
When we argue on the basis of a conviction, we should be seeking to verify our position, rather than to confirm it
“When the facts change, I change my mind. What do you do, Sir?” This statement has been attributed to many great people — most frequently to the economist John Maynard Keynes. That is probably incorrect, but let us not be distracted by the question who might the originator of such a pithy and powerful symbol of evidence-based practice in everyday life.
It’s the kind of thing many of us wish we could say ourselves, or even that we believe we could actually legitimately claim to do. Who wants to stubbornly stick to a conviction in the face of facts that contradict it?
And — you can probably see where we’re going — yet, what people think they do and what they actually do are often two different things. In a recent episode of the BBC radio show More or Less, one of the topics was the overconsumption of sugar as a public health concern. Have the people of Britain really been eating and drinking too much of the sweet, sweet stuff? And does that warrant new regulation and taxes? Sugar consumption is calculated in different ways, but whichever way you look at it, the trend seems to be firmly downwards. According to the Family Food Survey consumption per capita has fallen from 92 to 71 g/day since 2001, the UN Food and Agriculture organization calculates it dropped from 45kg annually per person in the early 1960s to less than 35 today.
The key measure that is used in public health is the proportion of our energy intake derived from sugar (as it is deemed to be linked to obesity). This too has steadily dropped — the National Diet and Nutrition Survey suggests adults get just over 11% of their energy from sugar, down from close to 15%, but still way above the guideline of 5%. However, that was halved from 10% in 2015 — a move of the goalposts, cynics might observe. The justification for this cut is intriguing, as the committee responsible found insufficient evidence for a link between sugar consumption and key bad health outcomes, like cardiovascular disease, colorectal cancer and type 2 diabetes. The main reason seems to have been that, apparently, if we reduce our sugar consumption, we do not compensate by consuming more calories in a different form. By that mechanism, cutting down on sugar will reduce overall energy intake, and hence obesity.
All interesting facts. But the reasoning of the Head of Nutrition Science at Public Health England, the body that lowered the guideline, appears to be motivated more by convictions than by these facts. (It is worth listening to the interview snippet in question — it is about 4 minutes long, starting at 7:20). Despite multiple valiant attempts to bring the facts into the discussion, show host Tim Harford is met with the impenetrable stubbornness of the people who are convinced they are right.
Another example comes from a recent Wiki Man column by Rory Sutherland in the British weekly The Spectator. Perhaps the gilets jaunes, the Five Star movement in Italy, and even the Brexit or Trump supporters are not — as the media keep maintaining — evidence of resurgent nationalism and populism, he argues. Maybe they represent a sensible, overdue reaction against the multitude of changes that ideologically motivated elites have imposed on the world: economic globalization, free movement of people, the euro, large-scale immigration — too much, too fast. And the economists, who usually make such decisions, with little regard for either time or scale, get the brunt of the blame.
The economists’ central problem is the apparent neglect of the concept of identity. Economists pay too much attention to material wealth, and not enough to the fact that people seem to instinctively value their collective identity, their affiliation with groups (nationality, race, social class etc) just as much as the acquisition of wealth. And while the latter scales easily, the former doesn’t — “There is no logical reason why people cannot say ‘I support all the football teams in the north-west of England’ but nobody ever does.”
This is an interesting argument (I would of course say so, as I have written before about both the importance of identity and the effect of it on voting). An experiment from nearly 30 years ago by Charles Perdue and colleagues illustrates the strength of the sense of identity and belonging. The participants in their study (who were told it was about verbal skills) received a list of nonsense syllables (like xeh or yof) paired with either in-group words (us, we, ours), out-group words (them, they, theirs), or neither (he, she, his, me etc). The subjects then had to assign a ‘pleasantness’ score to each nonsense syllable. The result? People for whom a given syllable had been paired with an in-group word rated it as significantly more pleasant than the neutrally-paired syllables; and syllables paired with an out-group word were deemed the least pleasant.
So how come people overlook the importance of identity? We generally believe our own argument for or against something to be rational, so we tend to selectively support it with factual evidence. Those who disagree (and who must therefore reject our rational argument) are evidently irrational. This definitely was, and to some degree still is, a common Remainer narrative following the Brexit referendum. There was no strong pro-European identity on that side at the time, and the principal case for staying in the EU had a material foundation: the economic advantages of remaining were overwhelming, and the downsides of leaving were dramatic. How could reasonable people vote Leave? If they did so, how could it be explained other than by nationalism and populism, by xenophobia and nostalgia for a time when Britannia ruled the waves? Identity didn’t figure as a legitimate reason for wanting to leave.
Rory Sutherland even states that this is a principled argument for a hard Brexit. A provocative counterfactual claim like his can be rather useful: challenging longstanding and deep-rooted beliefs needs a proper shake-up, not a timid tap on the shoulder.
But when we confront blind partisanship, confirmation bias, selectivity and so on in this way, we need to be careful that we do not succumb to the same problem as the people we have in mind. When we observe what we judge to be faulty reasoning — especially if we feel strongly about it, we can start to believe our own reasoning is inherently and objectively correct. And that overconfidence can make us lose rigour in arguing our case.
There are undoubtedly economists who have no interest in social preferences like identity, but when the entire profession is presented as the villains, is that actually true, or just a stereotypical conviction that we no longer think needs to be verified? Unsurprisingly one economist took issue with such a broad-brush accusation:
Overstating the case by understating the facts
Identity may not be a mainstream topic in conventional economics, but overstating the case (by claiming it doesn’t figure at all) risks weakening its credibility. And of course, when — rightly –identity is introduced in the decision-making process, it needs to be weighed up against other elements, not least the economic consequences. Identity is only a principled argument if it is valued more highly than the corresponding economic loss. Such trade-offs between the immaterial (social and emotional preferences) and the material (economic outcomes) are very much the domain of economics and economists. (And arguably, not all proponents of Brexit motivated by identity had come to their decision with full recognition of all the facts.)
So we have the official being dismissive of the facts that question the reasons for cutting the sugar intake guideline by 50%, we have a simplistic Remainer narrative that ignores the fact that identity can play an important role in decision-making, we have a simplistic Leaver narrative that only looks at identity and brushes off the economic facts… And we have a critique of the Remainer narrative which, while essentially valid, is perhaps a bit quick to condemn a group of people, and that sounds a little uncritical to the simplistic Leaver narrative.
Ironically, among the better ways to avoid bias in such reasoning is a core instrument in the toolkit of the reviled economists: the economic way of thinking — the good old ‘on the one hand’ and ‘on the other hand’. If we want to avoid countering a one-sided argument with a one-sided argument of our own, we have to be critical of ourselves. It is not because others are wrong and we can point out why, that what we say is right just because we say it.
We still need to make sure we’re not falling in the same trap of believing that our argument is entirely rational, when our reasoning is mostly motivated by our own preferences, convictions and beliefs.
And perhaps most importantly we should always check our own facts. (I hope I did.)
Originally published at koenfucius.wordpress.com on January 18, 2019.
My gratitude goes to you for reading this piece — I hope you enjoyed it. If so, you can let me know by clapping (the clapping hands 👏 icon is nearby). Do share it too — use the Facebook or Twitter icons, use this link to post it on LinkedIn, or just copy and paste this link. If you want to read more stories like this, check out the entire list of well over a 100 posts here — you will find a new one every Friday around noon GMT.