If we had to characterize the last five years in markets, we would call it a little bit rocky. It’s not only the moments of abject panic – like when the likely consequences of the initial wave of the pandemic crystallized, or like when Russia invaded Ukraine – though we have had a lot of those. It’s also the extreme swings in style and sector performance that have resulted from volatile and unpredictable macro and political developments. You really would have needed a crystal ball to see where a lot of single events, and some broad strategies, might go over time and invest accordingly. The only alternative path to out-performance over the past five years — and there’s no other — would have been to be diversified.
We launched our KTI Spider charts last year
What it means: These are designed to map the investment trust sector with some of the key characteristics used to position trusts against each other and one of those is diversification. In theory, selecting investments that are low-correlated to each other will enhance their risk-adjusted returns. We decided to see how this plays out in the data and what insights can be gained.
Methodological aside
For a summary (more detail here), we have categorised investment trusts into 13 super-sectors, merging AIC sectors where applicable. We have then selected a key benchmark among the peers for each super-sector to calculate a few useful performance indicators. The theory is that going in, most investors come up with a rough asset allocation and then choose funds to fill each bucket (that is, at least how they teach it to the professionals). In this context, it makes sense to compare, for investors, for example, all global equity trusts, whether generalist, sector, or thematic, against a basic global equity index. That’s a different way of looking at performance compared to poring over attribution relative to the benchmark the manager has chosen, more applicable for people trying to gauge manager skill. When we discuss trusts, it is worth keeping this in mind, as a trust may look different versus the peer group and benchmarks we have selected versus their official ones. All of our scores reflect performance of a trust versus the peer group rather than in absolute terms, so very much on the same theme. That is to say, a trust that scored nine on risk would be the tenth percentile of the super-sector (in the sense of relatively low volatility, for instance), and a trust showing an eight would be in the twentieth percentile of scores. Just like the underlying data, we should be cautious about over-interpreting results then: if all trusts in a sector are all very volatile, large differences in risk scores may just reflect small differences in volatility. Again, we’re trying to create a picture of where trusts sit about one another, not hand out prizes to managers for performance, risk or anything else.
Diversification scores
Diversification is one of the metrics we are monitoring. With our professional investor’s hat on, we thought a lot about what it might be most useful to be diversified into. Our score contains one factor drawn from the correlation to the sector benchmark. For instance, UK funds will be given a higher score for diversification depending on their level of correlation to the FTSE All-Share Index. Since most investors will have a globally-diversified portfolio, we also believe low correlation to the MSCI ACWI Index will raise the score on diversification. Finally, for most investors who use bonds to diversify their equity-heavy portfolio, we think a trust that is less correlated to bonds deserves to score a higher absolute rating for diversification as well. We have given equal weighting to all three correlations and normalised the scores to our super-sector peer groups, such that a lower correlation than the peer group earns a higher score.
Low correlation is appealing, but only when considered in conjunction with other traits. For instance, an asset that declined one percent every day would have almost no correlation to equity markets, but might not be attractive to all investors. A high correlation to a trust’s market, for instance, is consistent with high alpha scores, and our diversification scores, which incorporate the correlation to global equity and bond indices, are especially well-behaved in this sense. That said, diversification remains good information and can be very helpful when it comes to deciding how to choose your investments. Below, we flag up a few of the more interesting results thrown up by our spider charts. Specifically, combining a trust with a relatively high score for diversification in its super-sector, with one with a relatively low score, has delivered stable and attractive returns over the long term. This is especially interesting to think about for the past five years that have included extreme volatility in which any single-factor portfolio would have been crushed at least once.
