The weekend’s news has again illustrated the strong emotional responses produced by politics today and by Donald Trump in particular. It is no coincidence then that the current prevailing sentiment among investors is that returns and price behavior in the coming year will be determined by the actions of President Trump. But this is almost certainly wrong.
Much effort is spent in analysing cabinet appointments, poring over White House releases on new policy initiatives, examining the choice of initial Executive Orders, and trying to interpret provocative Presidential tweets in an effort to develop some kind of ‘view’ about what sort of administration Donald Trump will lead.
From a political or a social standpoint, this is fascinating stuff to explore and discuss. There are many aspects of what has unfolded in recent months that might help illuminate what it means to be an American citizen in today’s society. From purely an investment standpoint, however, it is less clear what further analysis of Trump can yield, other than a prematurely-formed opinion dangerously masquerading as knowledge.
As far as Trump goes, we should probably approach him in the same way we do other global policymakers, i.e. assess the direction and extent of actual policy decisions as and when they arise, without getting too caught up in the noise.
This is far from saying that the election was irrelevant. The fact that the US electorate took a political step to the right, with all branches of government now in Republican hands, is much more relevant, to financial markets, than the fact that Donald Trump is President. The overall structural environment for generating corporate earnings growth in the US is more likely to be favoured by a de-regulating Republican administration than a somewhat more restrictive Democratic one.
So when it comes to arriving at a ‘view’ on these things we should do our best to acknowledge that the genuine complexity and inherent unpredictability of the US and world economy – and its interaction with future investment returns – renders efforts to over-simplify and distil such complexity into digestible soundbites or tweets a total waste of time; no more than an entertaining, or depressing, distraction – depending on your viewpoint.
The dangers of data: Gross Domestic Product
This framework, or system, is evident elsewhere, notably in our fixation with economic data. To be sure, statistics and raw facts rarely provide such rich material for storytelling or outrage as that delivered by the new President, but the desire for some insight into the future encourages us to be constantly seeking new information or, even better, a prognostication from an expert.
This is despite the fact that we ‘know’ that knowing the future to any useful degree is impossible. And we also ‘know’ that even if we were in possession of future facts, our ability to translate this into asset price moves would be only very small or non-existent the vast majority of the time.
A brief analysis of GDP data provide an interesting example. Friday’s release that US GDP grew at an annual pace of 1.9% in Q4 2016 was “disappointing” for being a full 0.3% points (!) lower than consensus forecasts. Commentators rushed to explain this fact and to link it to movements in asset prices.
Such intense focus on individual data points is tough to comprehend when you consider that the revision-prone nature of this (and many other) series means that we can’t even be certain that its sign is correct. Since 2000, US GDP growth numbers have ended up being changed between initial and final release by 1.1% points on average; the onset and severity of the 2008 downturn weren’t picked up in the initial releases. The chart below shows the effect of this difference since the turn of the millennium over time (the lines on the chart) and each quarter (the bars).
Even if you knew the final number it’s unclear quite how informative this would be anyway. Data collection methods have failed to keep pace with the changing nature of the economy, leading to substantial GDP mismeasurement problems and rendering it a flawed indicator of economic progress. Despite all this, attention still oscillates from datapoint to datapoint, seeking to analyse and explain them to fit a particular narrative.
Conclusions: Characterising the world
Human beings allow over-simplification and emotion to drive decisions all the time. We take at face value the utterances of policymakers and of expert analysis and misuse it to our own ends – usually to justify an opinion already held or a decision already taken.
Investors have always had to grapple with this, but as Stuart wrote last year, the way much news is framed nowadays is increasingly geared to produce emotional and biased responses. There seem to be influences of this today: Are experts who personally dislike Trump more likely to emphasise negative implications of his policies? Are you more likely to listen if they do? Are you spending more time thinking about a Trump policy than for Xi Jinping’s or Vladimir Putin just because it dominates your news feed?
Ultimately we need to distance ourselves from unhelpful emotions. Abstracting from all noise, we believe there have been several key developments in recent months that remain the key areas of focus.
- The one-way bet that was bond prices (up) and expectations for the economy (down: be it earnings, GDP, inflation or rates) has turned: there is no longer such confidence in the ‘volatility dampening’ nature of bonds going forward, nor such universal use of stock alliteration such as ‘secular stagnation’ and ‘lower for longer’ to describe beliefs about the future.
- Reported economic activity (including corporate earnings growth) in various regions of the world has either resumed or continued the expansion that was, by and large, steadily unfolding for some time anyway, which markets only occasionally chose to acknowledge.
Given this environment, investors need to continue to focus on gaining exposure to areas that are attractively valued in their own right i.e. those that offer compensation for the risks of Trump policy, and for other risks besides. We should also be wary of putting faith in a continuation of historic correlation patterns and traditional safe havens. And finally we should be taking any opportunities afforded by market volatility, whether ‘Trump-related’ or not.
In a general sense, our mistake is not in seeking to understand and analyse Trump, GDP data or the behaviour of financial markets; it is in over-weighing the significance of our conclusions from such analysis and creating a sense of ‘knowledge’ or ‘certainty’ where there is little.