Monday 13 August 2018

Diversification


Diversification

Modern Portfolio Theory (Harry Markowitz 1952 – not that modern), suggests that portfolio diversification could “reduce risk and increase returns for investors.”

The conclusion is that a diversified portfolio of imperfectly correlated stocks reduces the overall risk to less than the average risk of the individual securities. The downside risk of one investment would be offset by the upside potential of another investment.

Unsystematic risk (i.e. the risk of a particular company performing badly), can largely be eliminated by holding a diversified portfolio. (NB Systematic risk – the risk of a general market fall – cannot be protected against by diversifying within the same asset class i.e. stocks)
Various academic studies suggest that 15 to 20 securities selected at random are sufficient to eliminate most investment-specific risk in a portfolio, however the more securities are added, the more the risk diminishes. (source – Investment Principles and Risk Textbook – CII)

I wholeheartedly agree with the premise that holding more stocks does help to protect against the unknown ‘black swan’ events that can come out of the blue and smash a company’s share price into smithereens, but does holding too many stocks throttle performance by diluting one’s best ideas?
In consideration of this, firstly I’d like to dwell on some personal experience.

Although I have dabbled with shares since the late 90s, my first serious stock pick was in 2009. I’d held a FTSE All Share tracker fund (obviously reasonably diversified) since 2005. I watched it rise for a couple of years, and then subsequently crash monumentally when the Credit Crisis hit. During that crash I decided to take a more active approach and disinvested the tracker fund to buy a handful of stocks that I perceived to be trading at significant discount to their intrinsic value.
Luckily one of those stock picks was Barclays. I bought into them in the early spring of 2009, and subsequently sold in the autumn making over 200% profit. Because this holding represented a relatively large portion of my portfolio, this pushed my total returns for the year to 75% which to date has been my best year in terms of percentage gain.

As time has progressed, the number of my portfolio constituents has grown steadily, and now it stands at 32 individual companies, 4 investment trusts and 11 OEIC/Unit Trusts (which I evaluate separately.)
Excluding the Unit Trusts, which have been a relatively recent addition, my average annual return (after all trading fees but including dividends reinvested) over the 9 years from 2009 to 2017 has been 16.95%. Although when you strip out that great year in 2009 my average annual performance falls dramatically to 11.2%. The All Share Total Return index also had a decent year in 2009 delivering 30.12%. It’s annualised return over the last 9 years has been 11.23% which falls to 9.1% per annum with 2009 stripped out.

To summarise, my personal experience thus far has been that higher returns were generated with a more concentrated portfolio, albeit this experience is heavily skewed by good fortune in one particular stock pick, during a particularly good year to be stock picking. If Barclays had gone to the wall during the crisis my results would look remarkably different. With this in mind, going forward would I be better off ditching some of my holdings and using a more concentrated approach? 

Most money managers with great track records appear to agree that a more concentrated portfolio is beneficial for delivering higher returns. Even Peter Lynch, the US fund manager who averaged an annualised 29.2% during his tenure at the Fidelity Magellan fund, which towards the end of his reign held as many as 1400 stocks, advises the small-time investor to concentrate on 8 to 12 stocks. He explains that his high number of holdings was strictly a result of the size that the fund had mushroomed to. (Source – The Great Investors by Glen Arnold)
Buffett is also quoted as saying that “Diversification is protection against ignorance. It makes little sense if you know what you are doing.” He walked the talk back in the 60s when he invested 40% of his partnership’s money in American Express stock after they’d tanked because of involvement in a scandal. This turned out to be a very lucrative move, and his portfolio returns were turbo-charged by his high concentration in this one stock.

I’d need an incredibly high level of conviction to consider investing 40% of my capital into one stock. I am undoubtedly more ignorant than Buffett as for starters I don’t do as much research as him due to having a day job, therefore I feel I need some amount of protection against bad decisions. My capital will ultimately fund my retirement as well as pay off an interest only mortgage, so my capacity for loss is finite.
This might seem paradoxical, i.e. why have more holdings if you have less time, but it does make sense to me to a degree. The reason is much of my research is done via software packages such as Sharepad and Stockopedia plus publications such as the Investors Chronicle. Using formats such as these you can quickly do a reasonable amount of research on a stock, but to get that extra layer of depth, i.e. go to AGMs, meet management and do in-depth research on competition etc, I just don’t currently have the time or resources to do this effectively. 

There is plenty of proof out there, Mr Lynch included, that demonstrates that you can achieve admirable returns with larger portfolios, especially if you are using a factor-based approach. 
Personally, I have no immediate plans to offload any holdings to meet a goal for a set portfolio size, but as an exercise I have ranked my holdings in order of conviction (easier said than done.) I have made a note of the order and I will evaluate performance of the top half Vs the bottom half going forward to see if there is much difference between my self-proclaimed ‘best’ ideas and my ‘worst’ ones.

The ranking exercise was certainly thought provoking. For instance, the stock that I ranked number 1 was Starbucks (SBUX), yet it only currently constitutes 2% of my portfolio, which means it isn’t even a top 10 holding. Why haven’t I bought more? I am wary that it is a consumer cyclical and could suffer a heavy pullback during a recession, but I do believe it will perform well in the long run, so in theory I should increase my position.

To conclude I think the ideal portfolio size will depend on personal preferences and levels of conviction. If you have the capacity (both mentally and financially), to stomach a large hit on a concentrated portfolio, and have conviction enough to entrust a large proportion of your savings into one organisation, then a concentrated portfolio might be for you. If you are right with your best ideas this should supercharge your returns. Otherwise I do not see that there is significant harm in holding a larger number of stocks provided you have the time to do enough due-diligence on each holding. If it worked for Peter Lynch, maybe it can work for us all.