Snap election time! Macron and Sunak: pure madness or the paradox of choice?

Two of Europe’s most powerful leaders have, in quick succession, called a snap parliamentary polls in the wake of their parties’ poor performance in another poll (EU elections for Macron, local elections for Sunak). The reaction in both cases was incredulity.  

What on earth are they doing? 

Both are centre right leaders facing a growing political threat from their right flank, the populist and popular Marine Le Pen and Nigel Farage. The immediate reaction was that both had made terrible mistakes in called snap elections, or at least that is a view held by at least some senior figures in their own parties. 

The paradox of choice 

Choices, choices. Stick or twist. Economists have grappled with the problem of choice under risk and uncertainty for a long time. And two economic phenomena might help shed some light on the apparent madness of Macron and Sunak. 

First of all, the basics of the choice both politicians faced. They were choosing between a surprise ‘go now’ option, versus waiting out to the end of the parliamentary term (or some other later date). Broadly speaking, the situation is thus: 

• Now: we know the polls look pretty bad, but they are where they are. 
• Wait: the trend has been getting worse over time, but anything can happen.  

Neither option is particularly good or particularly certain. And humans are not great at making choices under risk and uncertainty, let alone choosing between two very different gambles. 

Enter the paradoxes of choice. 

Paradox 1: the Ellsberg Paradox 

As is now well known, there are known unknowns and unknown unknowns. A famous economist Frank Knight first made this point as far back as the 1920s: in economics there is a difference between risk (a quantifiable, probabilistic assessment of chances) and uncertainty (the unquantifiable, we don’t-know-what-we-don’t-know). You can describe this difference as between a situation where the probabilities are known, versus when the probabilities aren’t known. 

Economist Daniel Ellsberg captured this difference in a famous experiment in the 1960s in which people are offered a choice between two urns. 

• Urn 1: a 50-50 mix of red and black balls 
• Urn 2: an unknown mix of red and black balls 

You win if you happen to pick a black ball out of the urn you have chosen. When this experiment is run, it consistently shows that people have a preference for taking quantified risk of Urn 1, rather than subjecting themselves to the more radical uncertainty of Urn 2. 

Perhaps we can see Macron and Sunak’s choices as being influenced by this preference. Yes, things are bad right now. But at least one knows how bad they are.  The decision to wait and see involves subjecting oneself instead to fundamental uncertainty. This is not just a choice between rolling two different looking dice. It is a choice between rolling a dice on the table, or rolling a dice in a sealed envelope that you can’t open for another year.  

The problem, that Sunak has already found out in the worst possible way, is that election campaigns (snap or not) are full of events and their own radical uncertainty. In that sense, perhaps Macron is the more rational to call a very radically short election campaign to minimise that uncertainty, and he does seem to have benefited from causing a bit of disarray amongst his right wing rivals. 

As Prime Minister Harold Macmillan supposedly once said, “events, dear boy, events”. 

Paradox 2: the Allais Paradox 

In theory, politicians are simple creatures. When it comes to parliamentary elections they have one objective: win as many seats as possible. They should make whatever choice maximises their expected number of seats. In economics, this fits rather well with what we call “expected utility theory”, a model of behaviour that (somewhat simplistically) assumes people choose whatever option is expected to generate the highest returns on average – recognising that some possible outcomes are more like than others. 

The real world is a more complex place. It’s illustrated nicely by another choice experiment popularised by Maurice Allias. This experiment involves choosing between two gambles. First, you get to choose one of these options: 

• $1 million for sure, or  
• 89% chance of $1 million plus a 10% chance of $5 million and a 1% chance of nothing 

Next, you get to choose one of these two options: 

• 89% chance of nothing, and a 11% chance of $1 million, or 
• 90% chance of nothing, and a 10% chance of $5 million 

On its face, both choices involve trading a 1% chance of nothing in exchange for a 10% chance of getting $5 million. But people seem much more reluctant to take that risk in the first scenario – when they know they are on course to win the $1 million. When the baseline expectation turns more negative, they tend to quite like rolling the dice on a small chance at a bigger upside. 

One of the explanations of why these choices are rational is the important role of regret. How will you feel when things go badly? If you’ve turned down an alternative that could have avoided the mess you’re now in, you’re going to feel like a prize idiot. It’s an important reason why people will trade quite a bit in return for eliminating risk. But if you’re already on course to lose, you become more risk loving because there is less to regret from radical action. In other words, we neither love nor loathe the risk itself, it all depends on the context for the choice. 

Both Macron and Sunak faced a baseline outcome that looked fairly unattractive. Both appeared to be heading, with high probability, towards a loss in parliamentary elections. This potentially changes their relationship to the gamble of a snap election. If they wait it out and accept the inevitable loss, it will be in the knowledge that they turned down the more radical option that could have turned it all around. The attractiveness of the radical snap election gamble can still be there even if the expected outcome looks worse. What matters to the snap election is the belief that the surprise throws everything up in the air, and introduces a small chance of turnaround.  

Bloody economists 

The curse of being an economist is, of course, to see absolutely everything through the lens of economics. And, is always the case, economists have a way of using complicated theories to tell us something we already knew.   

C’est la vie!