By Ron Bewley*. Brought to you by Infocus
Just like with the ‘Brexit’ referendum, when the UK population voted on July 23rd to leave the European Union, Donald Trump came from nowhere in the polls to pip Hilary Clinton at the post and become President-elect of the United States of America. So what will a Trump administration mean for the US economy and the world? And how did the polls and commentators get it so wrong, again?
The market has already spoken. Trump was looking like a lost cause a couple of weeks before the election. Then the FBI announced it was re-opening the ‘email scandal’ concerning Clinton’s use of a private server. Wall Street fell for nine successive days making it the longest losing streak since 1980! The last 10-day losing streak was in July 1975. Over this run, Trump had been closing in on the polls which contributed to the slide – but then the gains in Trump’s perceived fortunes levelled out. A Clinton victory seemed done and dusted when the FBI concluded its findings with no further action required – so markets bounced strongly for the two days leading up to counting the votes.
But the important thing about this market slide is that it was orderly. The S&P 500 only fell 3% over this nine-day period of losses and it was only 5% off its all time high in mid-2016! The vote counts started to come in while our market was open on Wednesday 9th – and the US market was closed. After an initial solid rise at the open, our market started to tank when Trump seemed to have an outside chance. And then the Trump landslide gained momentum and our market fell from about 5,315 at noon to 5,050 at 2:30 pm (nearly 5% down) before bouncing back to close at 5,157.
At one point, the S&P 500 futures, which also trade while the cash market is closed, was down 5% around midnight New York time. At that point Carl Icahn, the legendary businessman and Trump supporter, is said to have left the Trump party and went home to take off some hedging trades! He claimed on Bloomberg TV afterwards that the fall in the futures had looked so overdone he had to take some financial advantage from it.
Prior to the elections, no leading forecaster had predicted more than a 3% to 5% fall on an unlikely Trump victory so Icahn’s move showed no special insight. The information was open to us all. When Wall Street opened at 1:30am AEDT, the market was mixed. It was up a little, then down a little at first. But the S&P 500 was then up strongly and closed up +1.1%. How can this be? A strong three-day winning streak straddled the election results!
Markets often over-react when the unexpected is thrown into the mix. It seems like it was Trump’s acceptance speech that did a lot to calm nerves and propel the markets higher. The divisive Trump from the campaign trail was not on the stage. Instead the calm, Presidential, Trump gave the speech and he started talking about a united future – one in which he will focus on infrastructure.
Roughly paraphrased, Trump said “The US will again have the best railroads, roads, airports etc. We will double growth to 4% per annum. We will create many new jobs.” Importantly, he will have the House of Representatives and the Senate onside. The Republicans won both chambers. Unlike Obama, who had to fight every step of the way to pass bills, Trump should find passing his bills a lot easier.
So are there two Trumps – one to fight tough to win against a formidable opponent and another conciliatory Trump to rule?
It is unlikely he will be able to stop his off-the-cuff one liners that make some cringe but one-liners do not make bills that go before the Congress. He will no doubt assemble an administration team of great quality and it will be those people who translate Trump’s thoughts into policy documents.
There may well also be overshooting in the markets on the upside. Trump does not take office until January and it takes a long time to appropriate infrastructure funding and even longer to build it. But markets are forward looking. They won’t wait for the infrastructure to be built before investors buy stocks. And different sectors will behave differently. Many stocks in companies that are aligned in the Trump infrastructure direction rose in double digits the day after the win. Some stocks lost.
His forecast of 4% economic growth per annum looks unachievable in the long run but in the odd year? Possibly. But a good infrastructure programme could make the end of the so-called ‘new normal’, with the old normal returning to centre stage.
No doubt a Trump victory will affect The Federal Reserve’s (Fed) interest rate policy. We – and many others – were thinking that there would be a very gradual increase in US rates – possibly only three hikes over the next two years to just over 1%. But if the economy is to rip, they will have to tighten much faster to contain inflation. The question now is when will the Trump effect start to emerge? The Fed would be foolish to pre-empt the Trump effect but market rates have already moved. There was a big surge in US 10 year government bonds in one day. There will be more to come.
Before the election, a sizeable portion of the market was factoring in a possible US recession in the next few years. The official Fed forecasts for growth were between 2% and 1.8% for each of the next two to three years. While Trump’s 4% growth may not be achievable in the long run, a recession anytime soon seems to be off the table.
Is there a downside? If we focus just on economic and not other policies, some degree of protectionism in trade is likely. The problem with economic theory (or at least one of them!) is that it talks about the representative person being better off with free trade. The average person might feel fine but those further down the food chain might be a lot worse off. And, as Clinton found out (and the pro-European marketeers in the UK) the representative person doesn’t vote. In countries where voting is not compulsory, people passionate for change are prepared to queue up at the polling booths to have their say.
So why did the media and the polls get it so wrong? People interviewed on TV are not drawn from the population with any reference to the differences in opinion over the whole population. Interviewees predominately have good jobs and nice suits. There are far more people who go to work not wearing a suit than those who do. And the unemployed are probably not usually wearing suits either. It is quite likely the people who feel worse off under current governments around the world probably watch the suits being interviewed – and that might make their disenchantment all the more stronger.
The polls are carried out by statisticians. It might surprise you to find the size of the population has little effect on the accuracy of the poll – in theory. So an opinion poll looking for political views in Tasmania would have the same sample size as one in New South Wales – or, indeed, in New York!
Some of the main US polls were based on a sample of 1,282 people to give a ‘margin of error’ of 2.7%. That means that the pollsters do not claim a lead is significant when it is only one or two per cent. To halve that margin of error, statisticians know that you need to quadruple the sample size. That means four times the cost and probably a lot longer in time to get the poll completed. But this aspect of sample size choice is not the main culprit in this polling inaccuracy story.
The statistical theory on which the sample size construction is based assumes that the population is homogeneous.
That is, there are no factors determining how people vote across the different demographics. If there are differences, the statistician should ‘stratify’ the sample. That is the proportion of women in the population should be reflected in the proportion of women in that sample. While the pollsters might pay some attention to gender and age, they can’t possibly take into account all of the important factors.
There is a classic case of a telephone poll many decades ago when the prediction turned out to be really bad. Somebody pointed out after the election that the majority of potential voters for the winning party didn’t have a telephone! But, even if modern pollsters took account of many factors, that sample of 1,282 would mean there would only be a handful of people in each classification.
If there were only seven categories of voter types (of equal proportions) such as gender, age, job, education, etc there are 128 combinations (two to the power of seven) of voting characteristics. So, in a country of over 350,000,000 people, what accuracy do you think you would get from a sample of 10 people (=1,282/128) in each cell!
And another important point concerns the secret ballot. Until the mid nineteenth century, ballots were not always secret and intimidation was often used to sway an outcome. How many times have you had to show your hand to vote at work and felt uncomfortable at stating your real point of view? The so-called ‘Chartists’ in England put a six-point plan to government to ‘clean up’ the voting process. We now have proper secrecy in expressing our political opinions both here and in the US, UK, etc.
Our experience has been that seemingly supporting Trump was a social no-no in professional circles. Many people didn’t like either candidate but discussion often seemed to be – well Clinton at least can do this and that. “You couldn’t vote for Trump, could you?” Apparently (from Bloomberg TV) there was a massive swing of 12% in what white collar, college educated voters from the polls to the booths in one survey.
But did Clinton offer the same purely economic view as Trump? Certainly not! Is Trump’s view better than Clinton’s? That depends on your politics and your circumstances – and the voters decided what they wanted in a democratic way. Will the US be better off with Trump? We can never know because we cannot run a parallel universe with a new President Clinton. Success is all about what bills are formed and which bills are passed through Congress. Just think back to the Obama administration.
It is very early days to evaluate a Trump administration as there is no clear policy document – as we would have. We don’t even know if some of his proposed ‘policies’ were purely for the theatre of it. But it would now seem that a Trump world could be good for Australia. Better growth in the US usually spills over to us and others. Isn’t it a shame we don’t have a big infrastructure program that passes through parliament to give us that injection of hope?
We plan to provide economic updates as actual policies become available.
*Ron Bewley (PhD,FASSA) – Director, Woodhall Investment Research
This information is the opinion of Infocus Securities Australia Pty Ltd ABN 47 097 797 049 AFSL and Australian Credit Licence No. 236523 trading as Infocus Financial Advice and may contain general advice that does not take into account the investment objectives, financial situation or needs of any person. Before making an investment decision, readers need to consider whether this information is appropriate to their circumstances.