Republican presidential candidate Donald Trump speaks at a campaign rally at the Greensboro Coliseum, North Carolina, 14th June. A Trump presidency combined with as Brexit and resurgent nationalism in Germany would harm our economy far more than the 2008 financial crisis, Kaletsky argues ©Chuck Burton/AP/Press Association Images

What is the biggest threat to the world economy in the year ahead? Forget about slumping industry in China, failing banks or unpredictable monetary policy. All these familiar economic problems pale into insignificance compared with the combined risks of Brexit, a Trump presidency and resurgent nationalism in Germany. If any of these threats are realised, the consequences will far outweigh the impact of the 2008 financial crisis. Just as in the 1930s, the upsurge of nationalism and protectionism turned out to be far more important than the 1929 stockmarket crash.

Business managers and investors usually downplay political risks. After declaring that their companies can succeed in any political environment, they often add, sotto voce, that politicians rarely carry out rash promises made in election campaigns. In today’s conditions, this is wishful thinking. The political insurgents will doubtless change their minds about many issues, but the promises to which they will have to commit themselves in order to win elections will be exactly the ones that are most economically destructive: a Brexit victory will pull Britain out of Europe; a Trump presidency will certainly start with a tariff war with China and an economic attack on Mexico; any involvement in the German government by the xenophobic Alternative for Germany (AfD) party will certainly mean expelling Greece from the euro, probably followed by Spain and Italy.

“A Trump presidency will start with a tariff war on China and an economic attack on Mexico”

In Britain, the likelihood of a deep economic slump if the nation chooses Brexit has at least been widely recognised and debated. Britons will have no-one to blame but themselves if they consciously vote for what George Osborne aptly calls “a DIY recession.” By contrast, Americans and Germans who vote for Trump or the AfD will expect to boost their living standards. When they discover that a trade war or break-up of the euro will have the opposite effect the blame will fall on foreigners and nationalism could turn into belligerent xenophobia or worse.

But what are the chances of such disasters? Financial markets, political analysts and computerised voting models have consistently put low odds on the insurgents winning—in late May, Brexit was at 22 per cent and Trump at 26 per cent—despite the fact that opinion polls continued to show Brexit and “Remain” running almost equal and now show Trump-Clinton close to a dead heat. But the low odds attached by experts to both Trump and Brexit are less reassuring than they seem.

One problem is whether to believe the polls or the experts. Another is that even a low-odds event has a high probability of happening if there are several independent tries. Suppose we accept the view of the betting markets that the odds of Brexit and Trump becoming president are each around 25 and also give similar odds to an AfD electoral breakthrough. Then a calculation shows that the probability of at least one of these events occurring is 58 per cent. In other words, a politically-induced disaster for the world economy this year is more likely than not.

Luckily, this calculation is only statistically valid if political events in Britain, America and Germany are completely independent, like three draws from separate packs of cards. If, on the other hand, the populist revolts in all these countries are to some degree manifestations of the same phenomena, a vote to “Remain” on 23rd June could substantially reduce the odds of populist victories in the US, Germany and other advanced democracies. This is not because Americans or Germans would be directly influenced by what happens in Britain, but because populist voters in all these countries come from broadly similar demographics and are responding to similar pressures.

The anti-establishment revolts are all driven by stagnant living standards, widening income inequalities and disillusionment with conventional politicians who have failed to respond adequately to a once-in-a-lifetime economic crisis. The revolts are occurring in strongly democratic societies with longstanding political traditions of two-party dominance and a clear electoral bias towards the status quo. And the polls in Britain, America and Germany are all distrusted for similar reasons: the breakdown of previous political allegiances means that the methods used by all pollsters to calibrate their samples can give wildly inaccurate results.

Economic conditions in Britain, the US and Germany are also quite similar. All are now back to more or less full employment, with unemployment rates of around 5 per cent, but many of the new jobs created pay low wages. Income inequality has become a big issue. House prices, consumer confidence and household finances have recovered above pre-recession levels, but feelings of economic insecurity are still widespread. Immigrants have replaced bankers as the scapegoats for all economic and social ills. Populist voters in these societies face similar warnings not to risk economic disaster by overthrowing the status quo—the campaign of admonitions often described as “Project Fear.”

If political pressures in Britain, America and Germany are all related, then 23rd June becomes even more significant. What Britain decides will provide the first credible evidence about the magnitude of the anti-establishment protests. And it will reveal whether a gulf really exists, as assumed by political pundits and financial markets, between what disillusioned voters tell pollsters and how they actually vote.

If the “Leave” camp loses, the low odds accorded by experts to the populist movements will gain credibility. In fact, statistical theory allows us to quantify how expectations for the US election should shift after a Brexit defeat.

Suppose for simplicity that we start by giving equal weight to opinion polls, which in mid-May suggested that Trump and Brexit were almost equal with their opponents, and to the views of experts, who have consistently given Trump and Brexit only a 25 per cent chance. Now suppose that Britain votes to “Remain”on 23rd June. A statistical formula called Bayes Theorem then tells us to change the initially equal weighting between expert views and polling data and instead give a two-to-one preference for the expert views. The broad message is clear: a vote against Brexit will sharply reduce the probability of success for anti-establishment campaigns in the US, Germany and other advanced democracies. If, on the other hand, Britain does choose Brexit, the world should brace itself for two even scarier prospects: President Trump and Deutschland uber Alles.