The Irish "backstop", the legal clause that offends the sensibilities of hardline Tory Brexiteers - to the point where they would be prepared not only to immolate their own government but even possibly lose Brexit altogether to be rid of it - is often described as "insurance".
It's insurance, of course, for the people on both sides of the Irish border in that it means - if there's no trade deal between the UK and the EU by 2021 - a hard border on the island will not descend. But it's also insurance for people in the rest of the UK too, given it's not in their interests, either, to jeopardise the Good Friday Agreement. That latter benefit should really be more emphasised in the UK debate.
Yet why do we need insurance in life? The truth is that we often don't. Buying a warranty on a new microwave or vacuum cleaner is a waste of money given the cost of the insurance and the statistical likelihood of the product failing. These are immensely profitable contracts for retailers, which is why they push them so hard at the tills.
And insurance companies are so profitable, in part, because people have a tendency to over-insure against the wrong things. We like the peace of mind that comes from the possibility of relatively small claims, even if we would usually be better off foregoing them.
The better reason to buy insurance is to cover losses we wouldn't be able to cope with. "Insure only those things you cannot afford to lose," is the advice of the economist John Kay. "You need insurance against your house burning down, but not for replacing the bedroom carpet. You need insurance against being hospitalised in the US, but not for an extra night's accommodation because your plane is delayed."
Some Brexiteers don't actually care very much about the Irish border. A clean break with the EU matters more. Some, apparently, privately think the Irish are impudent for using the leverage with the EU to demand a backstop at all and should "know their place".
But let's assume those who have voiced such opinions are outliers rather than representative. Let's assume mainstream Brexiteers genuinely want to avoid a hard border in Ireland but believe that the backstop is unnecessary, that it's overkill.
They place a low probability on it being needed because they are confident we'll conclude a free trade deal by December 2021 and that this will include an unproven technological fix to the problem of Ireland being in the customs union and single market and Northern Ireland being outside both, something that would ordinarily require border checks on goods and agricultural produce passing between the two.
But consider who is calculating those probabilities. And consider their record. When he was Brexit secretary, David Davis predicted that the UK would establish a free trade deal with the EU, and a host of other countries, the "very next day" after we formally depart on 29 March 2019. We know now that's not going to happen.
Liam Fox, still the trade secretary, predicted that a post-Brexit free trade deal with the EU would be the "easiest in human history". It's certainly not looking that way now.
Yet, in fact, this isn't the right way to weight up the value of insurance. For insurance is not to cover eventualities we expect to happen with a high probability, but those that we fear. Even if one put a low probability on a trade deal not being ready in time, that does not imply a backstop would be a mistake.
A good insurance policy protects us against worst case outcomes. That's what the Bank of England was thinking about when it put together its no-deal Brexit scenario modelling. It wanted to test whether the UK banking system could cope in the face of an even an extreme economic and financial shock.
That's why, even if one is sceptical of the assumptions used in climate warming models showing rising temperatures as carbon dioxide levels in the atmosphere rise, one can still be in favour of radical decarbonisation efforts. The risk of a super-heated planet is too great to take (and the cost of shifting away from fossil fuels is eminently bearable).
"Skin in the game," is the solution favoured by Nassim Nicholas Taleb for powerful people who propose to run risks affecting others. He gives the example of monarchs who personally led their armies into battle, and sea captains who go down with their ships. "Skin in the game means consequences when you are wrong as much as when you are right," he says.
Creating a political system with such clear symmetric incentives is difficult. But it's nevertheless a useful way to think about those loudly insisting we don't need the Brexit insurance of an Irish backstop.