With Donald Trump confirmed as the Republican Party presidential candidate, and Hilary Clinton’s position as his Democrat rival all but sown up, investor attention will shift to the implications of the presidential campaign proper on US stock markets.
In doing so we will try and look a little deeper than President Obama’s reassuring comment that we should not worry about Trump’s foreign policy inexperience because he had in fact met leaders from many other countries such as Miss Sweden, Miss Argentina and Miss Azerbaijan!
Alongside this, thoughts are also going to whether this year will follow the traditional pattern when investors should supposedly “Sell in May and go away”. This an adage that refers to those happy days when New York Stockbrokers would pack their families off to their holiday houses and effectively down tools themselves for most of the northern summer before normal service was resumed around Labour Day, ie the first Monday in September!
Seasonality and election years are fascinating areas to study – but potentially dangerous to rely on.
As can be seen in this chart from financial blogsite Macro-Man there is a lot of history backing the “sell in May” part of the thesis, but rather less for the “then stay away” part (Figure 1).
Source: Macro Man
Like many of these seasonal trading patterns you have to question whether this isn’t a self-reinforcing trend, which works really well until it doesn’t, whereupon you get carted out, but it is hard to argue with the data.
However superficially the story is much more stark – and different – for election years, as per a final chart from Macro-man. The message from this is that investors should take their chips off the table in May, but make sure they are fully invested for the early part of the presidential election campaign. They then should de-risk into the final few months before re-loading as the dust is settled and the market gets comfortable that there is a winner even if it doesn’t much care for that particular winner.
This data set is more limited – but not totally to be dismissed…… It can perhaps be rationalised by saying that in May investors usually finally get clarity as to what the choice will be in the upcoming election, and frequently take the view that their preferred option is ‘none of the above’, before finally coming to terms with this until new uncertainty prevails in the heat of the election itself, only to be followed by a relief rally when it is all over.
This may in fact be the most important insight, and one with wider applicability. A study we made of Indian national elections some years back found a similar pattern. No matter whether an election produced a surprise outcome, one that the markets would have been expected to like, reappointed an incumbent or saw a new prime minister, in almost all elections over the last forty years the markets were up from their pre-poll level over the next few months.
Source: Macro Man
Like many of these seasonal trading patterns you have to question whether this isn’t a self-reinforcing trend, which works really well until it doesn’t, whereupon you get carted out, but it is hard to argue with the data.
However superficially the story is much more stark – and different – for election years, as per a final chart from Macro-man. The message from this is that investors should take their chips off the table in May, but make sure they are fully invested for the early part of the presidential election campaign. They then should de-risk into the final few months before re-loading as the dust is settled and the market gets comfortable that there is a winner even if it doesn’t much care for that particular winner.
This data set is more limited – but not totally to be dismissed… It can perhaps be rationalised by saying that in May investors usually finally get clarity as to what the choice will be in the upcoming election, and frequently take the view that their preferred option is ‘none of the above’, before finally coming to terms with this until new uncertainty prevails in the heat of the election itself, only to be followed by a relief rally when it is all over.
This may in fact be the most important insight, and one with wider applicability. A study we made of Indian national elections some years back found a similar pattern. No matter whether an election produced a surprise outcome, one that the markets would have been expected to like, reappointed an incumbent or saw a new prime minister, in almost all elections over the last forty years the markets were up from their pre-poll level over the next few months.
Source: Macro Man
But will this apply this year? Perhaps not, if you give any credence to the next data set from UBS, which is even more limited, but might be more relevant to 2016, showing the difference between all presidential elections and those in years after a president has served two full terms, of which this year will be an example.
Source: UBS
Assuming that this pattern is not just created by trend following herds of investors, digging ever deeper into well-established ruts, how might we explain it?
Arguably all these charts could just be illustrating a special sub-set of investor behaviour: aversion to uncertainty about the new or the unknown. Thus, if the UBS chart has any predictive capacity it is simply showing that investors perhaps get particularly nervous, as they prepare to say goodbye to any devil they have gotten to know more than usually well, without having any idea which new unknown devil they will then have to live with for the next four to eight years!
In this context, it is worth considering what the key differences might be from a stock perspective from the two candidates.
Again the picture is cloudy – perhaps enhancing the view that the key message, other things being equal, is that uncertainty => risk aversion => take some money off the table.
This is especially true this year when one of the candidates is Trump, whose past political track record is zero, and whose utterances on the campaign trail often seem to range from the farcical to the scary.
With Clinton the opposite is superficially true. We know a huge amount about her views, and the consensus is probably that she will try and build on Obama’s legacy, but also that she would be generally a leader so bound by donor obligations etc, that she would do little to rock any boats. This probably means no pick up in regulatory burdens for Wall Street etc, and also only token moves on energy policy and welfare expansion or reform.
Our main caveat would be to point to her age. Not only does raise the possibility that she might not last the full eight years, but we are also reminded of the adage that old popes tend to be reformist, and in a hurry, because they don’t know how long they have, whereas young popes tend to be cautious and incremental…
However that said some predictions do appear plausible: Trump would probably loosen regulation in a number of fields including taking some of the shackles over banks that many commentators now feel are suppressing the economy by making credit harder to find and more expensive. Clinton is very unlikely to follow a similar track, notwithstanding her much criticised links with Wall Street. Partly reflecting the fact that we believe her victory is more likely (see below) we have a small net underweight to US banks.
Trump is also likely to be favourable to US domestic manufacturers, while US companies which manufacture offshore would be likely to suffer, based on threats he has made during the primary campaign. Chinese companies exporting to the US are likely to suffer if he goes ahead with proposed tariffs. On the other hand, this, combined with a plan to re-label China a currency manipulator, may have the perverse effect of bringing forward the day, when China’s domestic problems cause another devaluation of the yuan.
Clinton’s election however is likely to be good for a number of sectors. She is almost certain to protect and enhance Obama’s health care policies, which will be good for both hospital chains like Hospital Corporation of America (HCA) and health insurers like Cigna (CI), both of which figure among our larger positions.
The following table from Washington DC based political risk consultants Alpha Capital sets out a plausible broader list of winners and losers by sector and stock code. There will always be the question of how much of a Clinton (or less likely at this stage, Trump) victory is already in the price for these stocks, but based on our comments above about the impact of uncertainty, our guess would be that even the BLUE (the colour of the Democrat Party) Clinton favouring stocks would not yet fully factor in the potential benefits of her victory.
Source: Alpha Capital
Finally what do we think will be the election outcome?
The good news is that we don’t have to make a call yet. However polls run over the last six months or longer do show the hard work Trump will have to do to win.
Source: RCP
That said it is clear that this is a volatile series, and that several times in the last half year the two candidates have been literally neck and neck. Also clear is that in the last month, Clinton’s once 12 point lead has virtually halved.
My personal view is that Trump has about as much chance of winning the election as a snow storm occurring in Bondi, but rather more than the chance of a mermaid hauling herself out of the sea on the beach.
The distinction is important.
Markets are good at positioning themselves for mermaids: they discount them completely.
However they tend to move late, with a flurry of panic when snow storms become a late breaking more serious possibility, even if still fairly remote.
This is rather akin to what we saw in terms of the panics over Brexit and the Scottish referendum, or on the other side Modi’s victory in India.
It is also worth noting that Clinton’s failed primary campaign in 2008 does suggest she has a less than perfect touch in dealing with insurgent rivals. This Blog from the creator of Dilbert gives an amusing road map of where she could already be going wrong.
Subscribe now to get the latest reports and insights from
Morphic Asset Management.
