Insight

Uncommon truths: The art of being vaguely right

Uncommon truths: The art of being vaguely right

There are good reasons why we seek out too much information and trade manically but improved performance is not one of them.  What can be done to improve investment and life outcomes? 

Full disclosure: much of this document was first written in 2017 (“All things come to he who waits” – 26 February 2017).  Then, personal circumstances caused a rethink of working patterns.  Now, the holiday period in the Northern Hemisphere gives the chance to recalibrate and slow things down again. 

Much of my life has been spent rushing from pillar to post (usually across borders), scouring data and news flows, scanning emails and Twitter feeds, while constantly checking market levels.  Does that sound familiar?  Does it improve performance?   

Figure 1 suggests we have collectively become more frantic in our investment processes.  In the immediate post-war period, the average holding period for stocks was commonly above six years and sometimes as high as ten years (whether we look at the US or Germany). 

Is it pure coincidence that average holding periods then steadily declined as the investment industry matured?  By the time I had started my career in the mid-1980s, average holding periods had fallen to less than two years in the US and below one year in Germany.  Unsurprisingly, the apex of activity in that decade came in 1987, when the average holding period for German stocks was around three months! 

We became ever more frantic as the financial crisis approached, with the post-war low in average holding periods coming in 2008 (around six months in the US and two months in Germany).  Looking at pre-WW2 US data, a similar low (nine months) was reached in 1928 (and we know what happened next).  However, the post-1900 low was the 3.76 months achieved in 1901 (stocks gained 32% in the two years to the end of 1901 and surrendered much of that over the next two years, based on index data provided by US academic Robert Shiller, as explained in the appendices). 

Average holding periods seem to have increased since the financial crisis.  This may be due to the reduction of prop-trading activity within banks but it may also be the normal reaction to a period of stress.  There is evidence that a bit of volatility may encourage risk taking behaviour due to an increase in the stress hormone cortisol but that prolonged volatility may lead to a chronic rise in cortisol and depression (see Kusev et al 2017: “Understanding risky behaviour: the influence of cognitive, emotional and hormonal factors on decision-making under risk”).   

There were signs of a renewed decline in average holding periods in 2018 (especially in the US), with a further dip in 2020 as Covid struck and lockdowns were enforced.  However, holding periods increased in 2022 (in Germany) and so far in 2023 (in the US and Germany).  Perhaps the frenetic trading activity of 2021 should have served as a warning that markets would struggle in 2022. 

Coming back to personal behaviours, what do we hope to gain by in-depth research activity?  Apart from the excuse that more information is now readily available and that we have promised miracles to our clients, there are important psychological factors behind the desire to be seen to be “doing something”. 

Figure 1 – Average holding period of equities (years)
Figure 1 – Average holding period of equities (years)

Note: Based on annual data from 1900 to 2023 (2023 data is based on annualised trading value data to April for Germany and to May for the US). Holding period is calculated as the inverse of the stock turnover ratio (value of trades divided by market capitalisation). 

Source: NYSE, World Federation of Exchanges, Global Financial Data and Invesco Global Market Strategy Office 

First, we suffer from the “illusion of control”.  Fenton-O’Creevy et al tested how much control 107 London traders believed they had over a computer simulated outcome (they had none).  The traders believed they had control and those suffering the biggest “illusion” tended to be the worst performers in their day jobs and were the lowest paid (see Trading on illusions: Unrealistic perceptions of control and trading performance in Journal of Occupational and Organizational Psychology (2003)).  

Second, we nourish this “illusion of control” by gathering as much information as possible, building “better” models and meeting as many companies and analysts as we can.  Unfortunately, this does not improve our performance; it simply boosts our confidence.  James Montier’s Behavioural Investing cites numerous studies to make this point.  Also, in an experiment conducted by Huber et al it was found that having more information did not boost trading profits, until the information advantage reached “insider” proportions (see Is more information always better? Experimental financial markets with cumulative information in Journal of Economic Behaviour and Organization (2008)).  It is better to be vaguely right than exactly wrong (to quote Carveth Read).        

Finally, we literally get a high from short term gains and the closer in time we get to those gains the more the emotional “self” dominates the rational “self”.  For example, offered a choice between $10 today and $11 tomorrow we may take the $10 now.  However, if given the choice between $10 in one year and $11 in one year and a day, we will choose the higher delayed payment (see Brain battles itself over short term rewards, long term goals which is a 2004 summary provided by Princeton University of a study conducted by Cohen et al).   

We seem hard wired to seek near term gains, which is unfortunate because Figures 2a and 2b suggest short term investing is a mugs game.  The range of possible outcomes in a quarter is enormous but narrows considerably as the time horizon extends. 

So, how can we live better lives and (hopefully) improve investment performance?  First, we should accept the difficulty of outperforming on a regular basis and set expectations accordingly.   

Second, it is important to lengthen the investment horizon to something more reasonable.  I would suggest something measured in years (preferably five), rather than quarters.  Though positions can be adjusted as prices move, I often find the views formed calmly at the start of the year are better than those reworked later in reaction to events.  Clients may not like it to start with but performance should also be commented upon less frequently.  Regular reporting is just noise, trying to explain noise (I now publish less). 

Finally, rather than chasing our tails trying to be the most informed on the street, we should choose a meaningful set of simple decision tools and stick with them.  My anchor has always been long term valuations and on that basis I continue to believe that Chinese equities will prove more remunerative than their US counterparts over the next five years (see the cyclically adjusted price-earnings ratios in the latest Big Picture document). 

Unless stated otherwise, all data as of 21 July 2023. 

Notes: Past performance is no guarantee of future results. Based on monthly data from January 1986 to June 2023. “ann. returns” is annualised total returns. Annualised total returns are calculated over successive overlapping horizons (ranging from three months to five years), using the MSCI World index to represent equities and the ICE BofA Global Government Index to represent government bonds.  Source: ICE BofA, MSCI, Refinitiv Datastream and Invesco Global Market Strategy. 

Related Articles