Is Your Stock Portfolio Strategy Working? Here's How To Find Out.

As many of you may know, I prefer creating my own equity portfolios by purchasing individual stocks as opposed to going the ETF route. I like the flexibility of being able to weight different stocks as I see fit, the possibility of using tax deferral strategies such as tax-loss selling, and lowering my risk profile depending on the economic environment. For the last number of years, I have seen little value in investing in Canadian ETF’s as I’ve been able to outperform the indexes (and top mutual funds) with very little turnover.

But especially in these turbulent times, I have to be realistic. Some of the times, this strategy will outperform ETF’s while other times it will not. Just because it’s worked in the past is no guarantee that it will work in the future. So it’s a good idea to, at least annually, do a performance analysis similar to the type an investment advisor may provide individual clients for their separately managed accounts. This type of analysis, called performance attribution, first calculates your portfolio’s performance versus the benchmark, and then breaks it down further into asset allocation and security selection. In other words, it’s going to tell you if your portfolio outperformed or underperformed because of your decision to allocate different weightings to different sectors, or because of good or bad individual stock selections within those sectors.

You may find that the reason you’re outperforming the benchmark is because you simply have a better asset allocation strategy. This would not be unheard of, especially in Canada, where broad-based ETF’s such as XIC and XIU are not very well diversified.

Or the opposite could be true. The stocks you are picking are on average, good investments. But you’ve just weighted them incorrectly, and are being held back by those one or two highly weighted sectors which just haven’t performed well.

A worst-case scenario would be if you’re underperforming both on asset allocation and security selection. Best case, of course, is that you’re outperforming on both measures. But until you know what your individual strengths and weaknesses are, you’re likely just comparing your portfolio to the index and hoping it outperforms. It’s important to know why, so let’s jump into finding out how to do performance attribution.

Step 1: Calculate Portfolio and Benchmark Returns

The first step is to calculate the returns of both your portfolio and the benchmark portfolio. When choosing a benchmark portfolio, choose one which is a good match for the types of stocks you pick. For example, if the stocks in your portfolio are only large-cap stocks, the S&P/TSX 60 Total Return Index is probably better than the S&P/TSX Composite Total Return Index. If you hunt down small cap stocks, consider using the TSX Venture Exchange. Or maybe you invest in North American equities mostly, in which case there exists several suitable ETF’s that will meet your needs. You get the idea. In my case there are individual capped S&P/TSX indexes for each of the 11 sectors, so I’ll just use those and adjust for dividends later.

Step 2: Calculate Excess Returns For Suitable Time Period

Calculate the difference in returns between your two portfolios. Use a suitable time period - one that is around the same amount of time you’ve been practicing your individual stock selection strategies. For example, if you’ve been picking Canadian large cap stocks for ten years, gather the returns of the S&P/TSX 60 Total Return Index for that same time period, or better yet the individual sector ETF’s or indexes if available. You can add an extra layer of analysis by changing up the benchmark sector weights each year as they change, but for simplicity purposes we can just take the most recent weightings. For my purposes I used index weightings as of January 1, 2020.

Step 3: Set Up Table To Calculate Performance Attribution

A performance attribution table simply has each sector as rows, and weights and returns of both your portfolio and the benchmark as columns. With some straightforward math, you can determine what’s known as the “allocation effect” and “selection effect” for your portfolio. You can download my template here to get you started, as well as my 2020 Canadian Equity Portfolio’s attribution analysis up to April 3, 2020. Below are the results year-to-date.

*Index returns are understated as they do not include dividends, which would represent approximately 0.5% to 1% per quarter. To invest in a sector ETF tracking these indexes costs an approximate 0.60% MER. To invest in a broad-based market capitalization-weighted index where no selection of weights is possible costs less than a 0.15% MER.

Step 4: Analyze The Results

As you can see from the table above, I’m thankfully positive in both the areas of asset allocation and security selection, however most of my gains this year are because I’ve chosen to allocate away from the Energy sector. Specifically, it accounts for 5.14% of my total outperformance of 5.86%. For Financials it's 5.20%. The rest, I’ve had mixed results. Now while this is a small sample, I have had similar results in the past and it continues to reinforce the idea that asset allocation is the number one thing you want to have control over. Similar results have been observed when allocating portfolios between stocks, bonds and cash, with some research indicating as much as 95% of a portfolio’s return is due to asset allocation rather than security selection.

Does this mean ETF’s aren’t useful in my situation? It depends, really. It certainly would make my life easier as I’d just have 11 ETF holdings as opposed to 30 stocks. But as I stated earlier, I wouldn’t have the same control over my investments and it would be more difficult to practice tax-loss selling if required. But my security selection isn’t exactly very strong right now, and if it drops below zero then ETF’s would come into play. I could then switch to ETF’s and weight my portfolio in the same manner as I do currently, but improve my security selection effect and therefore improve my total return.

That day hasn’t come yet though. Theoretically the selection effect would have to drop below -0.60% for it to be beneficial, because that’s the approximate MER’s of the ETF’s. As it stands now, dealing with a discount brokerage and doing very little trading hasn’t cost me very much in fees, so I don’t see the need of paying these fees just yet. But it’s something I’m going to be keeping my eye on.

Some of you may suggest simply investing in a broad-based market capitalization-weighted index ETF such as iShares S&P/TSX Capped Composite Index (XIC). It’s true that this is certainly a competitor. Its year-to-date total return (including dividends) as of April 3, 2020 is -23.30%, which is about 3% less than my portfolio to date. My main problem though is that it’s not customizable - it’s based on market capitalization, and that has been a problem over the last decade as the Energy sector has consistently underperformed.

Conclusion

There’s another option I haven’t mentioned yet, and that is to use a combination of individual stocks and sector ETF’s. This would be applicable if a particular sector’s selection effect is negative. According to the table above, my biggest selection drag is in the Financial Services sector, responsible for a 1.70% loss. I could have instead purchased the ETF instead, paid the 0.60% MER and netted a 1.10% gain. I may just end up doing this if my results continue to show me my security selection abilities for Financial Services stocks is poor.

The main point I’d like to make is that everyone has their strengths and weaknesses, and performance attribution is an objective way to see where your portfolio is going right or wrong. ETF’s are great options for a lot of investors, especially those who want low-cost targeted exposure (either to a broad market or to specific sectors) and who don’t have the time or inclination to research individual stocks.

So which type of investor are you? Would love to hear about your strengths and weaknesses in the comments section, by email or on Twitter @SundayInvestor. Thanks for reading!