What is the validity of recommendations made not in notes from sell-side equity research, but in the financial pages of the UK’s broadsheet and tabloid newspapers?
What is the validity of recommendations made not in notes from sell-side equity research, but in the financial pages of the UK’s broadsheet and tabloid newspapers?
June 2022
“I think the people of [the UK] have had enough of experts from organisations with acronyms saying that they know what is best and getting it consistently wrong.”
– UK Secretary of State for Levelling Up, Housing and Communities Michael Gove, June 2016
Introduction
Michael Gove’s notorious soundbite, though reviled by the metropolitan elite, seemed to tap into a latent frustration in the wider British population, in the build-up to the Brexit referendum. As with politics, so with investment, there is an allure to the idea that specialist training is not required – and indeed may actually be detrimental – in arriving at sensible projections of future outcomes. Such sentiment was on display in early 2021, when a band of Redditors scored a series of major successes against blueblooded hedge funds, to much cheering from those looking on.
But, as they say, the plural of anecdote is not data. And we’ve been interested as to how this retail segment has performed over a much broader sample size. Others have already investigated this problem from a variety of perspectives1, but one, we think, that has yet to be studied is the performance of stock tips that appear in non-trade publications. In other words, what is the validity of recommendations made not in notes from sell-side equity research, but in the financial pages of the UK’s broadsheet and tabloid newspapers?
Methodology
As a shortcut to trawling through multiple publications, we turned to The Week magazine, of which we are avid readers. For those not familiar, it is, as the name suggests, a weekly which aims to summarise all the news of the previous seven days as reported by the full spectrum of – chiefly British – newspapers. One of its recurring pages lists a selection of stock tips, as made by those same papers.
We collected 11 years’ worth of data, from 2008-19, some 550 PDF files. Lacking an overkeen intern to spend a month going through them all and typing them into a spreadsheet, we instead wrote a python script that scraped the six ‘buy’ recommendations from each. After accounting for a few here and there which were unreadable, we were left with 2,774 company recommendations, made by 19 publications, across 1,001 companies.
Could we stop paying thousands to analysts at sell-side research firms and just pick up the papers instead?
Headline Results
So, could we stop paying thousands to analysts at sell-side research firms and just pick up the papers instead? Alternatively, the opposite could just as well turn out to be true: by the time the popular press are on to a particular stock, perhaps that is the sign that everyone now knows about it, and the right course to plot is the reverse. We were apathetic as to the outcome, given that we could trade either, as a momentum or contrarian signal respectively.
Sadly, neither turned out to be the case, as is shown in Figure 1. Relative to the FTSE All-Share, the median stock recommendation was fractionally negative over one and three months, and fractionally positive over 12 months. Meanwhile, the positive hit rates across all three timeframes were almost exactly 50% (to be precise 49.8%, 49.9% and 50.3%, respectively).
Figure 1. 'The Week' Stock Tips Page Example (9 July 2016)
Problems loading this infographic? - Please click here
Source: Bloomberg, Man Solutions; as of August 2016 (for one month), October 2016
(for three months) and July 2017 (for 12 months).
One interesting lens through which to view the data is to look at differences between publications.
Performance by Publication
So, in aggregate, the short answer to our title question is ‘no’ – there is no relationship between retail stock tips and subsequent performance. You might as well flip a coin. But is there anything to be learnt underneath that headline? One interesting lens through which to view the data is to look at differences between publications, which we show in Figure 2. We limit the analysis to those papers which have provided a minimum of 50 tips over our sample period, and anonymise so as to spare any blushes.
From this, we can see significant discrepancies between the various papers. ‘Paper 3’, on a 61% hit rate and well over six points of outperformance over the subsequent 12 months, has a ‘track record’ to be proud of. Others, somewhat less so…
Figure 2: Median Outperformance and Hit Rate by Publication
Problems loading this infographic? - Please click here
Source: Bloomberg, Man Solutions; 2008-2019.
Best- and Worst-Case Scenarios
In Figure 3, we illustrate the extremes – specifically the top and bottom five recommendations over the full dataset. We observe that while the best tips offer the opportunity to realise 3-6x the initial investment over a 12-month period, the worst generally result in a complete and permanent loss of capital.
Figure 3. Top (Blue) and Bottom (Yellow) Five Recommendations Performance Over 12 Months
Problems loading this infographic? - Please click here
Source: Bloomberg, Man Solutions; 2008-2019.
The organisations and/or financial instruments mentioned are for reference purposes only. The content of this material should not be construed as a recommendation for their purchase or sale.
A Random Walk Down Main Street?
“International affairs should never be run by gentlemen amateurs,” says Lewis in Ishiguro’s Remains of the Day. And perhaps the lesson of this analysis is that stockcraft – like statecraft – does indeed require professionals. Consolation for the gentlemen amateurs is twofold, however.
To take the extremes here, we can see that the track record in communications is vastly better than in technology.
First, some evidence suggests that the professional tipsters are little better. Indeed, Bloomberg’s 12-month forward EPS estimate aggregate for Wall Street has been wrong by 10% or more half of the time since 1990. And secondly, some investigation into the detail does suggest that there may be pockets of value. We have already discussed the relative performance of different newspapers, but a clearer example is seen in recommendations by industry sector, and we illustrate this in Figures 4-5.
To take the extremes here, we can see that the track record in communications is vastly better than in technology. In the former, the retail tipsters outperform the sector benchmark 59% of the time, and by a median eight percentage points.2 In the latter, they outperform 38% of the time, and are behind by a median 12 percentage points. In both sectors they are around two percentage points overweight, so little evidence of skill in sizing. There is, though, perhaps scope for identifying certain discrepancies such as this to be profited from. Indeed, this is a technique that has been used to some success by a number of hedge funds in the professional stock research space.
Figure 4: Relative Performance and Hit Rate of Retail Recommendations by Sector
Problems loading this infographic? - Please click here
Source: Bloomberg, Man Solutions; 2008-2019.
Figure 5. Sectors as Proportion of FTSE All Share and Retail Recommendations
Problems loading this infographic? - Please click here
Source: Bloomberg Man DNA.
Conclusion
But, to cut this story short, do retail stock picks contain an alpha signal? At an aggregate level, almost certainly not. Eddington’s monkey dancing on the typewriter may indeed come up with the complete works of Shakespeare, but it will take a jolly long time.3
1. See, for instance, T Buz and G de Melo – Should You Take Investment Advice From WallStreetBets? A Data-Driven Approach – Cornell University, 2021.
2. Sector benchmark is the FTSE All-Share Sector.
3. See the ‘infinite monkey problem’:
https://en.wikipedia.org/wiki/Infinite_monkey_theorem
You are now exiting our website
Please be aware that you are now exiting the Man Institute | Man Group website. Links to our social media pages are provided only as a reference and courtesy to our users. Man Institute | Man Group has no control over such pages, does not recommend or endorse any opinions or non-Man Institute | Man Group related information or content of such sites and makes no warranties as to their content. Man Institute | Man Group assumes no liability for non Man Institute | Man Group related information contained in social media pages. Please note that the social media sites may have different terms of use, privacy and/or security policy from Man Institute | Man Group.