Factor analysis reveals risks hidden across asset classes

Transcript

From the transcript How I Invest publishes with the episode.

1. 11:46

   as part of your investment philosophy, you avoid investing from buckets, instead you focus on facts or factor exposure. Tell me about that, and how do you do that from a bottoms up? Yeah. So the we actually I like to sometimes refer to it as purpose driven investing. So we do have, you know, we do live in a kind of conventional world, if you will, with, you know, you never really we can't pick our trustees, and just thinking through that and again having that experience of sitting across from all types of trustees over a decent number of years, Wanted to make sure that we were having good conversations with stakeholders as well.

2. 12:20

   So we do have categories and asset classes, but they're pretty broad and vague. And so for example, we have growth, and within growth, we have public and private equity. And for us, we're like, okay, that's the high risk, high return part of the portfolio. So for something to be in that portfolio, we don't really care what the structure is, what the vehicle is, what the liquidity is. It could be public.

3. 12:41

   It could be private. You know, obviously, very you know, like a venture fund traditionally structured wouldn't go in public equity, maybe a crossover hedge fund that has 25% private markets that could be in a public equity, an evergreen private equity fund. We could put that in public equity or private equity. But as as we try and keep that generalness or that vagueness that allows us to be really opportunistic or open minded about how we invest and what we invest in, we have pretty healthy ranges and and latitude that our trustees have afforded us.

4. 13:13

   But at the end of the day, we run a factor model that and we have we try to have straightforward benchmarks that also represent the factors. So, obviously, you have ACWI for public equity. For real assets, we use S and P real assets index. For income, we use high yield one to three because we don't want duration in that portfolio. We want credit risk. And so as you're as you're thinking about that and communicating that with stakeholders, that's pretty straightforward.

5. 13:37

   Then when we go behind the scenes, so to speak, we're looking at that benchmark, the set of factor exposures, and how we can add value from it. And we either want to have intentional overweights across factors, that we think will be rewarded, or we want to neutralize the factor exposure but have a high residual. We're not necessarily trying to have our factors beyond the benchmark. We just want to know that, for example, we have equity beta coming from high yield.

6. 14:04

   We if you desmooth and interpolate private market returns, you know, you have equity beta risk coming from private debt, from private real estate. You might have interest rate exposure coming from infrastructure or private real estate, etcetera. So we just wanna make sure that we have a holistic view of the portfolio, what risks we're bearing, and whether or not we want to bear them or whether or not we think they'll be rewarded.

7. 14:25

   And that's and a, you know, sophisticated regression type analysis with factors is cheaper and easier to do than implement than, say, a a risk system that looks at all of your holdings and then tries to, you know, give you the sector weights, geographic weights, etcetera. And even still, I'm like, I'm not sure how much my equity beta really actually is just seeing of, you know, how much tech exposure I have when, you know, I also have data centers in real estate, for example, and and are those correlated?

8. 14:52

   Well, the holdings based analysis doesn't tell you that, but a factor based analysis can. And what what tools are you using to to break it down on a factor basis? Originally, we were using MPI, Markov processes. I remember can't if it's international incorporated, but, that's a great system, very sophisticated, very robust. It's a little more technical for the average user. And given that we're a smaller team and some turnover at the smaller or, I mean, at the at the younger level where, you know, you would want people running the systems.

9. 15:23

   We switched from that system to Venn, which is, you might be familiar. It's it's a software company started by Two Sigma, the hedge fund group. And they're they have a really robust suite of factors and a user friendly interface based on the web, and it can connect with your custodian, for example, and just a just an easier way when you have, you know, interns and people with couple years experience kind of helping you with analysis.