building a portfolio backwards

Long-time reader Chris’s comment on my post from yesterday prompted me to respond here about why I should be overweighting Consumer discretionary names.

One of the more surprising facts about professional equity management around the world is that almost no one is able to perform better than an index fund–even before subtracting fees from the results. In the US, where professionals keep up better with their benchmarks than elsewhere, two-thirds underperform before fees, six-sevenths underperform after fees, in a typical year. Hence, the phenomenal growth of index funds/ETFs.

Early in my career, I worked right beside the best-performing international manager at my then firm. He had a simple formula. He’d try to find five stocks that he thought had a good chance of doing better than the market and five that he thought would be clunkers. His portfolio would mirror the benchmark names and weightings, except for the ten. For those, he’d sell the prospective losers and put that money into the five prospective winners. Then he’d watch those ten like a hawk.

Not very glamorous, but extremely effective.

Another thing. It didn’t much matter to him whether his highest conviction ideas were the winners or the losers. So he would sometimes find the losers first and let them drive the portfolio.

As for myself, from studying performance attribution reports, I began to realize that about half of my outperformance vs. the S&P 500 would come from sector under/overweighting and half from individual stock selection. So I began to pay a lot of attention to sectors. And, taking a page from my former colleague’s book, I realized that avoiding bad sectors was just as effective a tool as anything else.

Current sectoral breakout of the S&P 500:

IT 27.6%

Healthcare 13.7%

Consumer discretionary 11.3%

Communication services 11.0%

Financials 10.4%

Industrials 8.7%

Staples 6.8%

Utilities 2.9%

Materials 2.7%

Real estate 2.5%

Energy 2.3%.

My back of the envelope thoughts, working from the bottom up:

–unless they’re going to double next year, the bottom four sectors are too small to matter. If I were running a $10 billion portfolio (I’m not) I’d probably look for alternative energy utilities, warehouses and left-for-dead office buildings, and maybe electric car battery materials to keep my head in the game. As a private individual, I can live without all of them

–Staples is a funny one, too. Big international exposure, which I think will be a plus in 2021, but relatively slow-growing and generally not keeping up with changing consumer tastes (although I’ve noticed that American cheese-injected hot dogs have disappeared from the local Weis market). Apart from special situations, no reason to get excited

–Industrials in the US market means mostly companies that make things for consumers, so I think of this sector as an extension of Consumer discretionary. If there’s hope of economic recovery, I think it will happen first in the stores, not in suppliers. Again, if I were still working I’d think I should have something here, if for no other reason than to act as an early warning indicator. I’ll probably begin to pay more attention as next year unfolds, but I see no need for Industrials at the moment

–the Financials story, as I see it, has two parts: fintech as a disruptive force for world-lagging US banks and rising interest rates as the driver for the sector to do well. The second isn’t happening soon and I own the ARK fintech ETF (my worry about that is that it has done so well already). So I regard Financials as another source of funds, not as an overweight

With that, I’m down to four sectors.

–I think of Communication services and IT as either complementary sectors or basically the same thing. My portfolio issue here is that I have a lot and my stocks have performed very well. So I want to reduce my overweight

By process of elimination, i.e. building backwards, I’m left with Healthcare (which I’ve mostly equal-weighted and have farmed out to mutual funds/ETFs) and Consumer discretionary.

One response

Leave a Reply