What impact will US elections have on the stock market?

GLOBAL stock market performance from here to the end of the year will be heavily influenced by the US elections in November.

The prevailing view among investors in Asia seems to be that US President Donald Trump would lose, largely due to the government's bungling of the Covid-19 infection which has resulted in Democrat Joe Biden's current lead in the polls.

However the last few years have shown polls to be unreliable in many global political events; we highlight the metrics that investors should pay attention to that are more reliable than polls.

The analysis below concludes that the elections will be much closer than the polls suggest.

Stock market tendencies around elections

Stock markets hate uncertainty, which is why the historical tendency has been that an incumbent party win is the most positive for stock markets in an election year, while an incumbent party loss the most negative.

Irrespective of your political inclinations and views of Mr Trump's character, an investor's job is to objectively assess the likely impact of each event, and what would be the most positive outcome for stock markets versus the downside risks.

Below is the average historical stock performance in election years by outcome (1st and 4th are the relevant ones in this upcoming election):

* Incumbent Republican wins: Stocks +16 per cent * Incumbent Democrat wins: Stocks +6 per cent * Incumbent Democrat loses: Stocks -2 per cent * Incumbent Republican loses: Stocks -8 per cent

Selling equities in the expectation of a Trump loss can be an expensive short-term mistake if he wins instead.

So what is the most likely outcome currently, and what should investors pay attention to over the next six weeks?

The election year backdrop

The level and speed of this monetary expansion have offset much of the negative feelings most voters may have from a typical recession and loss of wealth from a bear market.

However his approval ratings since mid-year have started rising, as Mr Biden has been unable to capitalise on Mr Trump's sub-optimal handling of the pandemic. Interestingly the president with the lowest approval ratings to manage a re-election win was Barack Obama in 2012, with only 47 per cent of voters approving at mid-year.

Polls versus bookies' odds

Rather than keeping track of public opinion polls, investors should track the odds given by betting websites. Speculators putting real money on the line to bet on a specific outcome is a much stronger indicator than what a voter says when taking a poll.

While polls show Mr Biden currently with a 7 per cent lead, current bookies' odds give him just a 3 per cent lead. More importantly, the trend over the last two months is strongly in Mr Trump's favour. Current odds are converging on even, after being as low as 37 per cent in July-August during the height of the Covid-19 outbreak in the US.

Stock market performance from end-August to the elections

One of the most interesting indicators is the US equity market's performance from the end of the US national conventions (Aug 27) up until election day. This indicator has had a remarkable correlation in calling whether the incumbent political party won the re-election, and correctly predicted Hillary Clinton's loss four years ago.

If positive (historically on average +6 per cent in this period) the incumbent has won a majority of the time. If negative (-1.4 per cent on average) the incumbent has lost most of the time.

While this election year's economic backdrop does not favour a Trump win, the two indicators above can give investors a stronger sense of the more likely outcome as the election date gets closer.

The biggest risk for investors is an extremely close election concluding in a very narrow Biden win, whereby Mr Trump refuses to concede and tries to force vote recounts.

As a silver lining for longer term investors, the negative stock performance in an election year when incumbent Republicans have lost have historically been followed by the strongest gains on record in the post-election years.