“Oh, people can come up with statistics to prove anything, Kent. 14% of people know that.” – Homer Simpson.
Could statistical lightening strike again?
The world holds its breath again this weekend as the French troop to the polls for round two of the French presidential elections. The polls still don’t give Marine Le Pen much of a chance, but who is going to listen to them after last year? Many pundits, still traumatized by last year’s upsets, have perhaps understandably opted for splinters over strong views. Many have talked about Le Pen’s path to victory as a simple function of a low turnout and a majority of the current ‘don’t knows’ unveiling themselves as shy Marine Le Pen voters. As we pointed out last week, the immediate aftermath of a Le Pen victory will be very messy from a capital markets perspective; all of our stores of composure will again be needed. Our sense is that calm would return and the still improving fundamental backdrop would eventually reassert itself (we would rather not find out of course). However, the point remains that we see the probability of a Le Pen victory as remaining insufficient for us to take pre-emptive action in portfolios; we remain overweight European equities and the euro in our current tactical asset allocation.
How likely is it?
Many have tried to put exact numbers on the probability of a Le Pen victory, however such precision is likely disingenuous. Before the US election, Trump trailed Hilary Clinton by around 2 percentage points in the average swing state. In the Brexit vote, the lead for remain was between 0.5 and 2 percentage points depending on your method of averaging the polls. So while both President Trump and Brexit seemed to go against the polling, they were utterly routine occurrences from a polling error perspective, well within normal error ranges. However, Le Pen would need to rely on a 20 plus percentage point error based on incoming polls. That suggests a far rarer statistical phenomenon would be needed to upset the applecart this weekend.
Some will counter that much of this will be eaten away by poll-shy Front National support on the day. Perhaps, though a recent study by American statistician Nate Silver found no evidence that right wing candidates such as Le Pen systematically outperform their polls. In fact, of the 47 European elections since 2012 that met thresholds of right wing representation and regular polling, right wing parties were predicted to win 13.5% of the vote and ended up with precisely that. In the five elections in 2012 in France, the Front National has averaged 21% in the polls and finished with 21% of the vote.
If the polls continue to hold firm at current levels, Le Pen could beat her polls by as much as Brexit and Trump combined and still lose to Macron by close to 20 percentage points.
It is this very low probability that has persuaded us to resist the urge to take tactical action within portfolios ahead of this weekend. If the unlikely, but of course not impossible, does happen then the wider strategic diversification of the portfolios should help absorb the worst excesses of the market, at least until the dust settles and investors refocus on the likelihood of cohabitation, a state of institutional coexistence, that will materially constrain Marine Le Pen’s ability to carry out much of her incendiary agenda.
We would not expect the ECB to stand idly by if markets become significantly unruly either. This leaves us still focusing on that firming fundamental backdrop for the world economy and its capital markets as the primary input into our investment thinking. The balance of probabilities still argues heavily for portfolios to lean towards assets that will benefit from an improving global growth backdrop, namely developed and emerging market equities.