One topic I have not touched on in a while is portfolio construction, so I wanted to dedicate this post to the reasons why a sector-neutral portfolio makes sense, and to give investors some ideas for creating their own.
Here is why I favor sector-neutral positioning: I would argue that if you are deviating from the index by making active bets with your portfolio, and if you are subject to benchmarking, it makes sense to limit the number of things you have to “get right” in order to perform favorably relative to the benchmark. With that in mind, if I am going to craft a relatively concentrated but diverse portfolio with 30 positions I would rather take my active bets both intra-sector and intra-industry than by trying to pick which sectors will do best overall.
You can see from this performance quilt from Novel Investor that a sector’s performance can often go from worst to first and vice versa from one year to the next:
For example, financials were the bottom performers in 2007, 2008, and 2011, but they were in the top half of performers in 2016, 2017, and 2019, and second so far in 2021, and so on. You might think that is does not matter if you are skilled at security selection, but a recent paper from Standard & Poor’s suggests that around half of daily price movements for S&P 500 constituents could be explained by price movements within their respective sectors (graphic via S&P):
In other words, despite whatever fundamental qualities you might like about a given stock, if its sector is out of favor, there is a good chance that the stock you have selected will be as well, vice versa. It is difficult to quantify with certainty how much of a role the sector plays over the very long term, but in the short-term it does seem to make a difference.
So now that we have argued for a sector-neutral portfolio posture, what is the best way to structure and allocate the portfolio?
The first step is to decide how many positions you want to hold in the portfolio. There is a balance between diversifying and spreading positions too thinly, and so a number around 30 makes sense to me. As my friends at Ensemble Capital have demonstrated, the benefits of diversification start to diminish, and perhaps the quality of the portfolio starts to decay, as positions number more than 20-30 names (graphic via Ensemble):
The next step is to determine how to structure the portfolio in a sector- or market-neutral fashion. Of course this will depend on the market benchmark one uses, so for these purposes we will use the MSCI USA index. Here are the sector weightings of that index as of June 30th:
The second step is to re-base the portfolio according to how you want your positions to interact. For example, I do not see much purpose in distinguishing between information technology and communication services since the largest communications services stocks are no longer telecommunications names but platform companies such as Alphabet and Facebook, which were once part of the information technology sector, so I have combined those weightings in the re-based portfolio. Similarly, I have combined consumer staples and consumer discretionary because one can argue that a stock like Amazon, which is now the largest consumer discretionary stock by market value, has now taken on some staple-like characteristics, especially during the pandemic. The same can be said for a stock like O’Reilly Auto Parts, which is technically classified as discretionary, but has acted more defensively in the past as tough times economically mean people keep their cars longer and spend more on them. That being said, here is how I re-based the portfolio according to my personal criteria:
With the sector weightings decided, it is time to decide how to allocate within the sectors. In this case, I favor equal-weighting within sectors because it has been demonstrated that equal-weighting within sectors yields superior results, with the possible exceptions of energy and some other smaller industries.
A second consideration is whether you want to make your portfolio global in nature. I have written before that home bias may not necessarily be a bad thing for American investors; in fact, there is a reason why most global portfolios that focus on “quality” stocks have a heavy bias to American firms. But in some cases foreign stocks are just as high quality as their American peers, and in some cases they are superior, so I think it makes sense to consider including them when you see those opportunities.
Finally, we have the finished product, which is a globally diversified, 30-position portfolio that is both sector-neutral, and equal-weighted within each sector:
Or in pie-chart format:
The obvious caveats are that the end product will be determined by each person’s decisions as to how to classify and consider the 30 or stocks to be included in the portfolio – is Amazon a technology, consumer discretionary, or consumer staple company? – and so the end weightings will deviate a bit from the index weights, but in this way I feel an investor is balancing the desire to be diversified globally and across sectors, maintaining an agnostic view as to which sectors will do best, all while focusing most of the active security selection to the industry and firm levels, and not watering down the portfolio in the process.