On November 8, the US votes for a president in an election that can best be described as bizarre. Both candidates are disliked and it is alleged that this makes the election difficult to forecast. Complicating projections further is the likelihood of a considerable presence of the “lying factor” in opinion polls. It is believed that some not insignificant percentage of whites (about 70 per cent of the electorate) are too embarrassed to admit that they will vote for the openly racist and misogynist Trump. In a race that is considered too close to call (for over three months now — see table) the “lying factor” can make any forecast of a Clinton victory go wrong.
Adding to the possibility of a Trump win is the fear of the Brexit shadow. For long it has been feared that the Brexit vote was a pointer to a worldwide trend. Hence, the widespread fear that the anti-globalisation Brexit virus was contagious and would spread across the Atlantic. However, there is a major problem with this facile “correlation”. In a referendum, you vote for one issue; in elections, you vote for multiple issues embodied in a candidate. The probability of reaching the same “outcome” in a referendum as in a presidential election is very small.
In two previous articles, I had forecast that after accounting for male-female differences in voting behaviour (‘Hillary Clinton by a landslide,’ IE, August 6) and ethnic differences in propensities to vote (‘A double-digit win for Hillary Clinton,’ IE, September 27), Clinton would win with close to a double digit margin. But that was before the FBI director James Comey decided to further complicate forecasts by (illegally?) re-opening the e-mail investigation against Clinton. It is popularly assumed that this will reduce Clinton’s advantage, if not eliminate it. We examine the effects of these special circumstances on the US election.
First, the Comey factor. The table reports three spreads in the vote margins of several opinion polls — before the first debate (September 26), after the leak of Trump’s lewd tapes but before the final debate, and over the last six days. Comey has not had an effect on Clinton’s vote share (steady at 44 per cent-45 per cent), but has increased Trump’s vote share by two percentage points.
Second, accounting for lying. There is a widespread belief about a considerable amount of lying in the opinion polls, that is, the polls are understating Trump’s strength because many voters are hesitant to admit that they will vote for him. Many believe that missing out on this lying factor got pollsters, and others, mistaken about Brexit, and Trump in the primaries. Note that the polls predicting the results of the primary elections — the state-level contests to decide the party nominee— always had Trump performing more or less in line with the actual primary election results. So the polls well-captured Trump’s support — not so many ‘liars’ in the primaries. But the political establishment was deeply sceptical that Trump would not self-destruct or assumed that the other more conventional Republican candidates would find a way to actually dominate him — and they were wrong. This contributed to the view that Trump has been underestimated. He was underestimated, but not by the polls, and not by “lying”.
Opinion polls ask several indirect questions relating to political views, in addition to a direct question about voting choices. The former can be assessed for their revelation about “true” choices. Towards this end, 61 questions (for example, favourability of a candidate, ability to handle an international crisis and trustworthiness) from five different pollsters were pooled. The striking result: The opinion questions reveal the same result as the presidential-poll questions, that is, lying may not be a factor in this election. Clinton is ahead by nine percentage points (ppt) for the larger question set, and ahead by six ppt for a set of five specific opinions.
A noteworthy feature about Trump’s popularity is that no matter what the criteria or question, he finds it very difficult to cross 42 per cent of the vote. Thus, the evidence overwhelmingly converges to this simple reality — Trump has a core support of only 35 per cent of the electorate.
Third, a large fraction of Trump’s core support is expected to come from the white, less educated population. The latest data on educational attainment (US Census Bureau) indicates that there are 180 million individuals above the age of 18 with zero college education. Of these, 80 million are whites; with a voter turnout of 60 per cent, that is 48 million potentially disgruntled white voters (men and women). Let us assume 70 per cent of these potentially disgruntled 48 million white voters are certain to vote for Trump (the reader can insert her own preference). Hence Trump receives 34 million votes from this demographic, or a 20 million voter advantage.
However, this vote advantage only covers three-fourths of the 27 million vote advantage that Clinton receives from non-white voters. Even if educated whites vote equally for Clinton and Trump, that still leaves Clinton with a seven million (out of 135 million) voter advantage. That is a five per cent victory margin with very extreme assumptions.
In no analysis do we get a Clinton victory of less than five percentage points, a result far away from the opinion polls. The reality will be known on November 8, so hold the brickbats (or compliments!) till then. If I am right, then it is useful to remember history for parallels.
There have been only 13 presidential elections since 1828 where a candidate has received less than 41 per cent of the vote. In five elections, third-party candidacy was a major factor affecting vote shares of the two major parties. Since 1948, there have been three elections without significant third party presence. This list has many honourable politicians and none said that he may not accept the result of the election. The three honourable losers: Barry Goldwater, 1964 with a 38.7 per cent vote share; George Stanley McGovern, 1972 with a 38.2 per cent vote share and Walter Mondale, 1984, with a 40.8 per cent vote share.
A 43 per cent vote for Trump will mean that he will exceed the maximum of the above three honourable losers; that is close to the upper bound that Trump is likely to obtain. It would also make Trump ahead of the kinder and gentler Mondale. The third party candidates are unlikely to reach double digits — say, seven per cent. This leaves Clinton with a 50 per cent vote.
There is a lot of international stake in this election. If it turns out as predicted, the election will likely provide a much-needed boost to those who believe in trade, open borders, and the traditional “American way”. The anti-trade, anti-immigration wave around the Western world will at least be halted, if not reversed, with a large Clinton victory. The recent high court judgment in England (parliament must vote on Brexit) has already provided the basis for the beginning of the retreat of anti-globalisation forces.
(This article first appeared in the print edition under the headline ‘No proof required: Victory amid controversy’)