Higher risk doesn’t improve chances of outperforming
There is a common misperception that investors will be rewarded for taking increased risk. However, the risk/return trade-off tells us that increased risk provides for the possibility of higher returns, but the outcome is not guaranteed. While higher risk means higher potential returns, it can also lead to higher losses.
While some may believe that taking on much higher risk in an equity portfolio improves the chances of outperforming the market (or a benchmark or index), statistics show that this is not necessarily the case: outperformance is not more common among equity portfolios positioned with higher risk.
In fact, SA performance data over the past three years show that higher-risk portfolios have a greater probability of underperforming than outperforming their benchmarks.
However, generating above-benchmark returns does require taking on appropriate risk, which involves carefully deciding how to position a portfolio with overweight and underweight holdings versus a benchmark so that it does beat the benchmark. This is called “taking an active position”. At Prudential, we manage all of our portfolios actively in order to provide above-benchmark returns for our clients.
The man-on-the-street may be tempted into thinking that portfolio managers who take large positions in companies versus a benchmark have a better chance of beating that benchmark’s returns because of their expertise in selectively choosing stocks and constructing a portfolio that should outperform. However, many of these “high-conviction” bets resulting in highly concentrated portfolios are risky, often proving unsuccessful for various unforeseen reasons.
How do you measure risk or “activeness” in a portfolio?
There are several ways to measure risk in a portfolio, two of which are active share and tracking error. Active share measures the total difference between the weightings of companies in a portfolio compared to their weightings in its benchmark at a point in time. It indicates how actively a portfolio is positioned right now. A 0% active share would mean that the portfolio is weighted exactly like its benchmark (such as an index tracking fund), while a 100% active share means that the portfolio has no holdings in common with the index.
Tracking error, meanwhile, shows the difference in returns between a portfolio and its benchmark. It measures the degree of variation of the active returns of the portfolio over time. Active return is the difference between the portfolio return and the benchmark return. Tracking error is calculated using standard deviation, using a series (list) of monthly active returns. A portfolio with 0% tracking error would match the index return perfectly, with no deviation (something not even the best index tracking funds can do), while a high tracking error would mean the portfolio’s returns are very different from its benchmark (not highly correlated).
Measuring activeness among SA domestic equity funds
The graph below shows that, among the domestic equity unit trust funds in South Africa, there are large differences in tracking error and active share – the most highly concentrated portfolios are reflected in the top right quadrant and index trackers in the bottom left quadrant. This demonstrates that South African fund managers have many different approaches to active management.
As expected, there is a relationship between the two measures: in general, the higher the active share, the higher the tracking error of the fund. Higher tracking error is a function of an investment manager taking larger, more concentrated active positions, resulting in the portfolio holding fewer stocks. Lower tracking error results from building a more diversified portfolio of stocks. The majority of funds have an active share of between 40% and 60%, while the average tracking error is around 5%.
The Prudential Equity Fund is currently positioned at just under 40% active share and just under 4% tracking error. This reflects our prudent approach towards constructing portfolios – because of the way we take risk (which is a topic for another article), and aim for an active share of between 40%-60%, our funds tend to have a tracking error that is lower than the average fund in the category. The level of the Prudential Equity Fund’s active share is currently at the lower boundary of our range because of market conditions – with share valuations relatively high, there are few real value opportunities for us to take advantage of.
Does increased activeness improve the probability of outperformance?
The above graph demonstrates that as funds’ active share increases, so does the dispersion of returns so that active return may be either more positive or more negative. In other words, taking more risk or larger active positions in a portfolio does not guarantee greater returns. In fact, the graph shows that among high active share funds, more tend to underperform (their active returns fall below the 0% line in the graph) than outperform. Research by Vanguard shows that this holds true in the US asset management market as well.*
Using tracking error as the risk measure instead of active share, the same trend again holds true. In the graph above it is clear from the SA equity fund performance that the higher the tracking error, the more likely a fund is to have underperformed over the last three years – again, among the higher tracking error funds there are more below the 0% active return line than above the line.
The Prudential Equity Fund: Consistent outperformance with below-average risk
The above graph illustrates the Prudential Equity Fund’s tracking error, active share and active return (how much outperformance the fund has generated versus its benchmark of the average return of the general equity unit trust sector) over time.
The fund’s outperformance has consistently exceeded the risk taken (tracking error) over time. This measure (active return/tracking error) is known as the “information ratio” (IR) of a fund, and is an accepted indicator of a fund manager’s skill. The information ratio is an example of a “risk-adjusted return” measure, because it indicates how much active return a manager has delivered per unit of risk taken. The higher the information ratio, the more consistent the fund manager.
Consistency is a useful characteristic to measure, because a more consistent past performance can offer some indication of future performance. However, IR, like other performance measures, is a retrospective measure, and past performance is not always indicative of future performance. It is considered that an IR of 0.5 reflects a “good” performance, 0.75 very good and 1 outstanding, although these levels differ depending on investment styles and asset categories, as well as certain time periods. The information ratio of the Prudential Equity Fund is 1 since its inception (to 30 June 2014).
From the above, it is easy to see why Prudential focuses on constructing well-diversified portfolios with prudent active positions in order to outperform our benchmarks. Maintaining the correct risk/return balance is a continual challenge for all fund managers, and what makes a manager consistent. Consistency is at the heart of Prudential’s investment process.
*Vanguard Research, “The search for outperformance: Evaluating ‘active share’”, by Todd Schlanger, Christopher B. Philips and Karin Peterson LaBarge, May 2012