What price democracy? 

November 6th, 2018

In late 2018, as Britain wanders down the path marked Brexit, the route ahead still looks murky, thorny and pot-holed. The country is still divided almost equally between those who think Britain was right to vote Leave in 2016 and those who think it was a mistake. An increasing number of hardcore Remainers are calling for a second referendum, while many hardcore Leavers argue that would make the first referendum meaningless, undermining democracy. 

So the question arises: is democracy a means or an end?  Let’s take an extreme example. Until sanctions bit in the late 1970s, Rhodesia was a fairly successful economy in southern Africa. There’s no particular reason to ascribe this to any great talent on the part of the white minority government: in common with its neighbours Zambia and Botswana, the country had excellent mineral resources and it also had a strong agricultural sector (though the label “the breadbasket of Africa” seems to have been bestowed retrospectively). 

After that, Rhodesia became Zimbabwe and it became a proper democracy with black majority rule in 1980. The government for many years enjoyed genuine popular support. 

Measured in economic terms, however, the government’s performance was lacklustre from the outset.  Despite a post-sanctions growth spurt, GDP per head grew just 11.5% in the 1980s (the equivalent figure over the same period for the USA was 38%, with Britain’s growth per head in the same period not far behind).  Things got progressively worse in the 1990s and 2000s.

Despite mounting concerns, democracy only really went off the rails in the late 1990s, post-dating Zimbabwe’s economic collapse. The policies that were so damaging, including the institutional weakening of civic structures that enabled the corrupting of the democratic process, appear to have been popular enough. Was it worth it? Is, in fact, vox populi vox dei? 

If, like Churchill, you take the view that democracy is the worst form of government except for all of the others then by implication you believe that democracy is a means not an end. Many have fought and died on the opposite assumption.  But let’s for now assume you take the utilitarian view. Is democracy a good idea?

Nothing is beyond the reach of mathematics. In the eighteenth century, the Marquis de Condorcet applied probability techniques to assess the value of democracy. Best known in geeky political circles for the importance of his work on electoral systems, part of his work has implications for democracy in general.  

So here is Condorcet’s Jury Theorem. Imagine that a group wishes to reach a decision by majority vote. Each voter has an independent probability of voting for the correct decision. If voters are more likely than not going to vote correctly, then adding more voters increases the probability that the majority decision is correct. On the other hand, if each voter is individually more likely to vote incorrectly, then adding more voters makes things worse, making the probability of a bad decision more and more likely. You would do best to let one person decide.  

Perhaps we understand this intuitively. On Who Wants To Be A Millionaire, contestants correctly “ask the audience” relatively early on, when they can be expected to be more likely than not to know the right answer, and save the “phone a friend” option for as late as possible.

If you run with this logic, one implication of Condorcet’s Jury Theorem is that we should not ask the public for their views on complex problems because they will get it wrong. The shade of Condorcet might agree – though a liberal, he died in the French Revolution.    

Remarkably, Britain’s electoral system is actually set up to take that into account. We elect people not policies. I think it was Matthew Parris who first noted that at successive general elections the public had chosen the best option available to them. This is to be expected.  If we believe that individuals are more likely than not to make the right choice between candidates, we can strongly expect them to elect the “right” government. Condorcet’s Jury Theorem works for us. 

Election over (having asked the audience), governments are then responsible for getting the policies and implementation right. Again, this is in accordance with Condorcet’s Jury Theorem. If the general public don’t have the skills to direct these, it is better to leave this to a non-democratic cadre of skilled people who do. As you would expect from the Theorem, the success rate is much lower here (but still as good as we can hope for). 

Democracy in Britain has always been tempered with Condorcettian common sense by this means. In recent years, however, Britain has started to hold referendums. The public is invited to weigh not people but ideas. According to Condorcet’s Jury Theorem, this is only a good idea if individual members of the public can be expected to give the right answer. We can ask the public questions that turn on value judgements. But we should not be asking technocratic referendum questions (and where referendum questions involve technical detail, the public needs to be briefed on that detail). When Clement Attlee and Margaret Thatcher referred to referendums being devices for demagogues and dictators, they were referring to referendums where strongmen secured spurious mandates for policies by force of personality. They had a point. 

The EU referendum is an edge case, judged by this test.  It raised questions of broad principle (what kind of nation should Britain be? how open to immigrants should Britain be?) and questions of technical detail (what kind of trading and societal relationships are feasibly capable of being secured with Britain’s closest neighbours? how would those interact with Britain’s other trading and societal relationships if changed?). Your view on whether the EU referendum result needs to be respected may well depend on whether you consider the referendum question to have been one about values or technical details. This is a question on which there can be honest disagreement.  That disagreement is part of what has made the aftermath of the referendum so bitter. 

How do we proceed from here? We can at least set out some principles for calling referendums. We might not be able to agree on a policy decision but we should at least be able to agree on the categories of questions on which the public can be expected to be able to set direction. Referendums on Scottish independence are proper subjects for referendums, since they are centred on questions of values. Referendums on voting systems look much more dubious, since they require technical judgements. 

If you believe that it was a bad idea to hold a referendum on Britain’s EU membership, however, you have a different problem: how to unravel this mistake. Unfortunately, it’s easier to make the mistake, if mistake it is, than correct it. If you think that it was wrong to hold a referendum on such a technical matter in the first place, then it is hard to see why holding a fresh referendum would be any better. A different means of resolution would be advisable.   

So that leads onto some fresh questions. Does it matter if democracy is undermined if that produces a better policy result? Or should the country make its choices by majority, even if they are objectively bad choices? I don’t have final answers to those questions. If Brexit is deemed to have gone wrong, however, they are going to be much-discussed. 

Alastair Meeks