A CONKERS3 FEATURE : The Analytical Surveyor
Analytical Surveyor has worked for over 40 years in the City and West End of London as a real estate investment manager, an equity analyst, a multi-manager, a fund manager and as a researcher and strategist. In his area of expertise, he is highly regarded and currently works as a consultant.
There is an enormous amount of academic research that provides insights into the functioning of investment markets. Rarely will this help with stock selection, but much of it will help in understanding how and why the market functions the way it does. The Analytical Surveyor trawls through the work that is available in the public domain and summarises the work into usable notes for investors.
Different research methodologies lead to contrasting answers and investment returns:
Statistically, 99% of investment professionals will accept data at its face value, provided that its source is not questionable. They will not question the definitions used or the mathematical rendering of the raw data into the result.
How do I know this? Well, to be honest, it was just a guesstimate, and I may be a questionable source. What I do know from many years of experience of presenting the analysis of data (both raw data and data already processed to a greater or lesser extent) is that the definitions used in compiling the data are at least as important as the data itself. Indeed, much of my time in managing teams of researchers was in questioning the results of the analysis and, very often, discovering that the methodologies or the selection of data have been deliberately or accidentally biased towards the expected or desired result. Sometimes the exact opposite conclusion was more valid when different data sources or methodologies were used.
What was rather annoying was seeing similar distortions occurring in the work of other research houses, but with their fund manager clients gratefully accepting their results because they supported the story that they were providing to their investors. The easiest ‘facts’ to accept are those that we want to accept.
I am sure that many readers of this will be saying that (a) this sort of thing happens fairly rarely and only where the research is ‘cutting edge’, and not of standard metrics or (b) the distortion is only minor and would only matter to a pedant or (c) the significance is not very great and the main ‘message’ would not change if it was to be corrected. In my experience, none of these arguments are generally valid.
A recently published paper[1] on the calculations used for adjusting stock prices for cash dividends and stock splits by James Felton and Pawan Jain illustrates that even very basic standardised calculation methodologies, in use for many years, can be subject to debate and potentially errors.
The authors argue that return numbers are critical to academics and practitioners both as an explanatory variable (such as in deriving momentum and reversal trading strategies or estimating market risk factor, such as beta) or a dependent variable (as is the case with most of the asset pricing literature). Anything, therefore, that affects how the returns are calculated is of interest, and maybe concern, to all who use the data.
As an example as to how the incidence of stock splits[2] affects share returns, they use the stock return calculations for the US stock of Cisco. Where the company’s shares were split into two, the data providers revised the historic share prices and halved all of them before the split. That generates the ‘adjusted share prices’ for the stock and is a very simple uncontentious calculation, providing continuity in the pricing of the stock. For the early part of the period under consideration, Cisco was not paying dividends. When, however, it started paying dividends, an adjustment was also made for that, and the authors observed that the resultant post-split share price was slightly lower because of that[3].
A second example, the stock of AbbVie, has never had a stock split, but does pay quarterly dividends. We all know that, when a stock goes ex-dividend, the share price falls by an amount approximately equivalent to the dividend. In practice, the share price fall is typically less than the dividend, and there is a body of research seeking to explain why that should be, but the consensus seems to be that a fall of less than 100% is attributable to the difference between dividend tax and capital gains tax rates[4].
Using the methodology adopted by Yahoo!, the authors estimated that the return on AbbVie’s stock between 9 January 2013 and 11 July 2014 was 72.78%. An alternative method, however, would have produced a result of 76.95%. A third method results in 70.50%, a fourth in 72.42% and a fifth 7.51%. Each of these methods is perfectly valid mathematically, although one does have the problem that the initial share price becomes negative, but each treats the dividend in a different way. Some of the difference in returns is purely due to the mathematical formulae, and most of the rest is dependent on whether it is assumed that the dividend is held in cash or re-invested in the stock.
Felton and Jain make a good case that the methodology employed by Yahoo! (and by extension most of the other data providers) is inaccurate and results in a return number that is too high. Their view is that, if the tax adjustment (CGT v dividend tax) is ignored, the dividend adjustment should be the amount of the present value of the dividends. If tax differentials were to be taken into account – and they are in the market pricing – then the dividend deduction should be less than actual payment made.
Two things probably need to be said in conclusion. First, the returns of the AbbVie are high, whatever method is used, and the differences between the various methods of calculating the effect of dividend are not very great as a proportion of the total returns. But, over a long period, chain-linking these returns together and compounding them will have a significant effect, which will distort the returns between dividend-paying and growth stocks. Second, the value of the dividends is only one part of the maths used in the calculations. Even the time gap between a stock becoming ex-dividend and the dividend being received by the investor varies enormously between companies. Consider how benchmark returns are calculated, and there is another dimension to the problem.
In my experience, it is not just investors who are blind to the machinations (to use an extreme word) of the data providers, but the regulators have little understanding of the processes and, when real life investors cannot understand why their monetary returns seems less than the claimed returns, there is usually just an unwillingness to investigate further. But if the simplest of calculations are open to dispute, what about the further layers of calculations which are built on these for the analysis of more sophisticated metrics?
[1] True Returns: Adjusting Stock Prices for Cash Dividends and Stock Splits, James Felton and Pawan Jain, Accepted in May 2018 for publication in Advances in Financial Education
[2] Readers may question why listed companies should go to the not inconsiderable expense and trouble of splitting their shares. The received wisdom is that advisers believe that where a share price has risen very significantly in a short period, investors can psychologically believe that the share is now expensive and are reluctant to bid it higher. Splitting shares in, say, two, halves the share price and restores it to an ‘acceptable level.
[3] The method of adjustment was used was by Yahoo! Finance, but the calculation methodology is used by other data providers.
[4] It seems surprising that that this has not been arbitraged away by the different tax regimes of investor groups. While some investors take advantage of that to make a profit on the ‘inefficiency’, maybe there are not just enough able to do so, or the costs involved negate the advantages.
Thank you for reading this research article, we hope you enjoyed it. Are you an institutional investor or private investor? What are your current thoughts with regards to the markets, stocks, shares, macro-economics or possibility of a recession in any particular country? Please lets us know. We look forward to hearing from you. You can find more research articles, interviews, podcasts, videos and investment insights on our website and across some of our social media platforms that include Twitter, LinkedIn, AudioBoom, iTunes, YouTube and Facebook .
