Thank Fathom it's Friday

On measuring political beliefs, and why the electorate may be less divided than you think

26 April 2019|

A sideways look at economics

As an undergraduate student of economics in the early 1990s I was made to take a course in comparative politics. For the most part, I found it rather dull. “I’ve come here to study economics. Why do I have sit through all this politics stuff? It’s so boring”, was pretty much my attitude at the time. But there was one exception. I do recall listening with interest to a lecture concerned with the measurement, and the classification of political opinion. Over a two-hour period we were presented with a critique of the traditional Left-Right spectrum. The idea that all political views could be represented as being along a straight line, with those looking to maintain the status quo on the ‘Right’, and those favouring reform on the ‘Left’, came from the layout of the National Assembly in revolutionary France.

Taking things to their logical conclusion, advocates of this approach might place Lenin and Stalin at the extreme left and Hitler and Mussolini at the extreme right. The difficulty, of course, is that in one important respect communism and fascism are really very similar. Both are extremely authoritarian, favouring the state over the individual. Writing in the 1950s, psychologist Hans Eysenck offered a solution. Retaining the split between what he called ‘Radicalism’ and ‘Conservatism’ on the horizontal axis, Eysenck added a second, vertical axis, which he used to measure an individual’s preferences with regard to state control. With that innovation political beliefs could be placed on a two-dimensional grid.

I was reminded of my undergraduate lecture last week when a colleague sent me a link to the Political Compass. By visiting their website, and either agreeing (strongly or not) or disagreeing (strongly or not) with a range of often moral propositions, the site produces a visual representation of your political beliefs, on a variant of Eysenck’s two-dimensional grid. Needless to say, I took the free five-minute test, and was moderately surprised by the result — of which more in a moment. But the creators of the Political Compass website have gone a step further and used their framework to classify the political beliefs of major parties, both across countries, and through time. Based on their own assessment of the beliefs of party leaders, inferred from a combination of political manifestos and parliamentary voting records, they find that, at the time of the 2017 General Election, the beliefs of the UK’s two main political parties were further apart than they have ever been, with the Conservatives well up in the north-eastern ‘Right-wing/Authoritarian’ quadrant, and the Labour Party moderately into the south-western ‘Left-wing/Libertarian’ quadrant. This chimes with a message we have been giving our clients for more than a year now. Following the next General Election, which according to bookies’ odds is more likely to occur this year than it is in any other, we risk seeing a shift in the way this country is run that far exceeds anything seen in more than a generation.

There are good grounds to doubt the absolute positioning of the dots on the chart above — in general, tests like these extrapolate from particular examples, or examples that have a particular resonance at a given point in time, to make inferences about a general position. This awkwardness can’t be avoided and can be felt when answering the questions: the mind wrestles with the question and very often wants to respond, as you would in a normal conversation: “well, it depends…”. That response isn’t permitted, because then the extrapolation from particular to general wouldn’t be valid.

So, the points should come with a rider along the lines: “assuming your answers can be generalised into areas not addressed in this quiz”, or something similar. Economists love assumptions, so this isn’t a problem — it’s just important to be clear about it.

What can one infer about the views of the Great British Public from the ebbs and flows of political parties within Eysenck’s two-dimensional grid? The answer, sadly, is not a lot. Political science has very little concrete to say about where parties should position themselves, except in the very special case where individuals are voting on a single issue, preferences are single-peaked, and there are only two parties. Let’s assume the issue at hand is redistribution, and that preferred outcomes are uniformly distributed across individuals between no redistribution at one end, and total redistribution at the other. By construction, the median voter wants to halve the gap between the richest and the poorest in the country. Theory tells us that, if there are two political parties, both will align themselves perfectly with the median voter. Any party offering to redistribute by more, or by less, is guaranteed to lose.

What if we now introduce a third political party? Then it gets complicated. Even in this simple example, a third party is likely to upset the apple cart. A third party might, for example, offer to close the gap between richest and poorest by just 40%, assuring themselves of the votes of at least the 40% of the population who favour less redistribution, leaving the two more established parties to fight it out over the remaining votes. A game then ensues where each party tries to shift towards one of the extremes, and to avoid staying in the middle at all costs.[1] Does the fact that the UK’s two main parties appear further apart than at any time since at least the early 1970s tell us that political opinion has become polarised across the Great British Public? Not necessarily. It may be that political beliefs are largely fixed across the electorate, but political parties are constantly reinventing and repositioning themselves in an attempt to outflank their rivals and capture a larger share of the vote.

When the link to the questionnaire was shared around the office most of us couldn’t resist completing it. How did I score? Well, in common with most of my colleagues, I was placed firmly in the Libertarian rather than the Authoritarian camp. That may reflect, in part, my background as an economist. Students of economics, in the UK at least, are often taught first to imagine a world where there is no state, no rule-maker of any kind (not as any kind of ideal, you understand, but rather as something of a straw man). A role for government is then introduced, bit by bit, in response to real-world imperfections that prevent an efficient market outcome. What surprised me is that I also came out to the left of centre. My views in that regard have almost certainly shifted over time. Had I taken the test while still a student I would almost certainly have found myself to the right of centre.

It has reportedly been said: “Any man who is not a socialist at age 20 has no heart. Any man who is still a socialist at age 40 has no head.” It seems I have neither.



Fathom prides itself on offering consultancy services to a diverse set of organisations who respect and value our independence. Fathom staff may hold a range of opinions, political and otherwise, but those opinions have no bearing on the nature of the independent advice that Fathom offers to its clients.


[1] This isn’t quite true. If both parties move sufficiently far towards the extremes, it can again make sense for the third party to position themselves in the middle.

Thank Fathom it's Friday

Our free, weekly blog taking a sideways look at economics