How to Build a Diversified Portfolio with Factor Investing

During their latest episode of the VALUE: After Hours Podcast, Taylor, Carlisle, and Weber discussed How to Build a Diversified Portfolio with Factor Investing. Here’s an excerpt from the episode:

Tobias: You say that you’re diversified or potentially better diversification than the S&P 500. But how are you thinking about diversification and concentration, names and sectors, and how do you think about that?

Scott: Yeah. So, in our industry, we’re trained to think about it in terms of sectors, and I find that to be really blunt. So a couple of years ago, S&P breaks out REITs, separate and distinct from financials. They get their own label, but that doesn’t make them more or less interest rate sensitive, for example. The other side of that is you look at a company like Amazon, it’s a consumer discretionary. But realistically, it behaves as if it’s a technology stock. So, we use a proprietary model– I’ll really stressed at the beginning, we’re fundamentally driven bottoms up. But we’ve got phenomenal factor resources internally that are proprietary. They’re driven largely by data that we stem from an outside vendor, but we can ascertain– It’s essentially a big regression on what’s moved stocks at the end of the day.

So, if you can ascertain what’s really driving stocks, you can own less than 30 stocks and have a beta that’s similar or lower than the index, because you’re not stacking factors, because you’re blinding yourself with sector bias. By and large, sectors are a generally good tool. But like I said, they’re blunt. And if you look at the index for comparison, whether you want to use the Russell 3000 or the S&P 500, it turns out, again, because cap weighting, that top end is monolithic. So you’ve got Microsoft, Apple, Amazon, Alphabet with the two classes added together, Meta, Tesla. You get down a pretty good ways before you get to UnitedHealth or Berkshire Hathaway. And so that top end looks a lot alike.

We’ve got a chart in our deck, which I don’t know, if you want to link these things, but we can link it. Fancy names for all of these, by the way. The better the data, the funnier the name, and we call it the bubble chart. But literally, it is just a factor correlation of each of the two stocks, and we cap sort it. You can see when you look at the index, the way we draw it’s a bunch of blue circles. If you look at our portfolio, it’s fewer blue circles, more blue and red, indicating that the names are not highly correlated from a factor sense. And so, we think, if for no other reason, because we start with an engine that’s considering a little bit above 30 overall factors, some of them move together.

To the extent that three factors or four factors behave similarly, you don’t need to diversify across them. You just need to have exposure to one of those. Now those relationships change a bit over time. But to the extent that you’ve got exposure to one of those factors, the other two or three are going to behave similarly, we call them virtually independent. Right now, you can describe the behavior of the index, I guess, if you will, in a mid-single digit number groupings of factors. That’s taking 33 or 32 down to 6 to 8, just because even though Amazon, as I mentioned, is consumer discretionary, it behaves a lot like Microsoft. Whereas our portfolio has somewhere close to double the number of virtual independent instruments, and that’s going to take a huge asterisk and description, I’m sure, for compliance purposes to describe what that is. But the net effect is, if three things behave similarly, you don’t need to own all three.

You can find out more about the VALUE: After Hours Podcast here – VALUE: After Hours Podcast. You can also listen to the podcast on your favorite podcast platforms here:

Apple Podcasts