In Table 1, I show the stock allocations for portfolios constructed at three separate points in time: the original 25 June 2014 portfolio is the one I actually invested one third of my SMSF assets in and wrote about on this and my site; the 1 March 2015 portfolio is the one I rebalanced into last month – as I wrote about in my March posting of the Switzer Super Report; and, finally, the third portfolio is the latest 1 April 2015 generation. The blue cross that lies between the Discretionary black diamond and the IT black square in Chart 1 represents the estimated risk-return of that April Hybrid Yield-Conviction portfolio.

Table 1: Three generations of the Hybrid Yield-Conviction portfolios

Source: Woodhall Investment Research

Note: ‘x’ indicates that the relevant stock was not present in the given portfolio (column)

From the first row of the Table, it can be noted that Santos (STO) was not present in the first two portfolios but it is in the third (two x’s followed by a blank). Indeed, no energy stocks were present in those first two portfolios. Telstra has been in all three portfolios (three blanks). Importantly, Property is in all three portfolios despite its relative poor positioning in the risk-return trade-off of Chart 1!

Some of these changes in composition from month to month come about because my criteria for the inclusion of a stock in a portfolio depend upon all of – expected yield, market capitalisation, consensus recommendation and the strength of the sector measured by the estimated risk-adjusted return of the sector. For these reasons, as well as the evolving sector allocations formed from the optimisations used to get the blue ‘+’ in Chart 1, the number of stocks in the final portfolio happen to have changed from 16 to 12 and now to 14.

However, there are only 22 different stocks in the universe of stocks across all three portfolios – suggesting that churning (or high turnover/trading) is not a major feature of this style of portfolio construction. Of course, this set of stocks will likely expand a little over time – but my expectation from my experience is that the universe expands only very slowly. The algorithm has been searching for about 25 stocks for each portfolio each month – but there haven’t been enough good stocks around to populate the sectoral allocations!

I do not use equal weights (or even close to!) for the stocks within a sector or across sectors – but that’s another story. The weights are produced from both the optimisation process for sector weights and the relative strengths of the stocks within each sector. The indicative weights for my current March portfolio are in my previous posting on this site.

So is all of this quantitative stuff worth the effort? I have been doing this sort of work with real money (mine and that from High Net Worth clients) for over a decade across many diverse sets of market conditions. The total return results (i.e. including reinvested dividends) for my particular Hybrid portfolio that I started in June and which was rebalanced in March are shown in Chart 2.
Source: Woodhall Investment Research

I have made a total return of about 21% in nine months against 15% for the benchmark. That comparison does not include franking credits, which would likely favour my portfolio because of my yield tilt. These results are not uncommon in the work I have done over the last 10 or more years. My constructed portfolio is currently winning by a comfortable margin (+6%) over the benchmark ASX 200 but, of course, outperformance is not guaranteed.

We are required by law to point out that past performance is not a reliable predictor of future performance.

Importantly, the relative performance in Chart 2 is not jumping about. No one should want a big win one day and a big loss the next. In my opinion, there has been a gentle gain in relative outperformance in my portfolio over the benchmark over the last nine months.

Readers might recall last month that I chose not to substitute Tatts (TTS) with TabCorp (TAH) in March as the algorithm recommended for a portfolio constructed from new cash. I deduced that the differences between these two particular stocks were so marginal that a substitution was not worth the effort and the transactions costs. But should I make more changes now that I have seen the April portfolio? With only new cash, I would have bought the April portfolio but is a churn worthwhile?

I could have substituted Westpac (WBC) for Macquarie (MQG) but I see no strong case for that. Moreover, MQG is looking to have strong growth characteristics and one day, the high yield play that currently favours WBC will be a thing of the past – or so I believe.

To me, six to 12 months is about a reasonable rebalancing period – unless something weird suddenly starts happening. That’s why I do this sort of analysis each month even though I don’t plan to make any changes. I just want to make sure I am not missing anything that is obviously avoidable. Indeed, I do daily checks on certain broker statistics.

I also could have justified adding Santos and CSR but again I ignored those additions as not being necessary. At least not yet! Since I am still in the middle of a 12–24 month transition of my old SMSF portfolio from pre-June 2014, I happen to already hold Westpac and Santos in my ‘Other’ part of my SMSF portfolio. I could do some cosmetic accounting and pretend that I had rebalanced into these stocks. I don’t find it that easy to fool myself so I did nothing on that front!

If some spare cash comes along after all of my dividends are collected, I might stick my toe in the water with a little CSR – just to show willing. All will be settled in my next big rebalance – probably around Christmas 2015.

Important: This content has been prepared without taking account of the objectives, financial situation or needs of any particular individual. It does not constitute formal advice. Consider the appropriateness of the information in regards to your circumstances.