Although active share has the popular appeal of being a simple proxy for portfolio active risk, it shouldn’t be relied on as the only risk measure when differentiating between funds. A deeper exploration of predicted tracking error can reveal significant differences – including allocation of risk – that should be understood when assessing and comparing portfolios.

In a previous post, DISCOVERING RISK DIFFERENCES BETWEEN COMPETING FUNDS, we compared the overall levels of portfolio diversification for two mutual funds, the Oakmark Global Fund (“Oakmark”) and the RBC Global Opportunities Fund (“RBC”). These results were shown against a backdrop of competing global equity products in the Morningstar World Stock universe of US domiciled funds, as at end March 2016.

We had noticed that, although both funds were both stock-concentrated and both had similarly high levels of active share, the predicted tracking errors (a.k.a. active risk) of the funds vs. the MSCI World index were significantly different. Oakmark had a predicted tracking error of 5.5% per annum (16th percentile vs. peers), while RBC had a predicted tracking error of 3.6% per annum (median vs. peers).

IDENTIFYING AND COMPARING RISK CONTRIBUTORS BETWEEN FUNDS

The charts below show how each fund’s tracking error can be decomposed into sources of risk. The red bars show the main sources of risk from currency, market, sector, style and equity (stock selection). The blue bars simply reflect the potential diversification or concentration arising from correlations between these risk sources.

Risk Attribution Oakmark RBC Global Opps

It’s clear from the charts that sector risk is a more significant source of risk for Oakmark, and indeed there is some extra reinforcement of risk between styles and sectors in that fund (the “Style XTerms” blue bar).

In fact, the active sector positions in RBC accounted for only 0.2% of the fund’s 3.6% tracking error, compared with 1.6% out of 5.5% for Oakmark. This is perhaps not surprising given that RBC was fairly sector-neutral with the largest active sector weight being only a +3.3% overweight in Consumer Discretionary. Oakmark had a +12% overweight in Information Technology, a -9.5% underweight in Healthcare, and a -9.3% underweight in Consumer Staples. The charts are therefore fairly intuitive, and convey important distinctions between these two funds.

Further decomposition of the funds (not shown here) revealed that RBC had a significant +18% overweighting in Large Growth and a +15% overweight in Mid Growth. This compared with Oakmark’s underweights of -14% and -4% in the same two style categories respectively. Oakmark’s style positioning within sectors was also risk-concentrating, with, for example, a +7% overweighting in Large Value stocks within Financials.

COMMUNICATING RISK DIFFERENCES BETWEEN FUNDS

So, although active share can offer a simple, handy estimate of a fund’s active risk, it doesn’t always provide the insights necessary to differentiate between funds. As we’ve seen here, further risk and style insights can often reveal essential differences between funds. Incorporating these objective comparisons can support further decision making and clearer portfolio communications, such as for quarterly portfolio reviews or positioning a fund against its peers.

To learn more, contact Style Analytics.


Leave a Reply