More pot luck than skill to picking a top fund
Even if you select and invest in a unit trust fund that performs consistently, you will most likely end up earning the average return over time, an independent researcher has found. Laura du Preez reports
It is better to invest in a local equity unit trust fund that produces average returns consistently than in a fund that at times perfor ms better and at times worse than average, an independent researcher says.
Ideally, you should invest in funds that achieve above-average returns consistently. The difficulty is that you cannot accurately predict which fund will do so, Daniel Wessels, a partner at Cape Town financial advisory company Martin Eksteen Jordaan Wessels, says.
His study of how consistently broad-based domestic equity unit trust funds perfor m from one year to the next found that, on average, funds have a 32-percent chance of maintaining their performance, a 34-percent chance of doing better and a 34-percent chance of perfor ming worse. (Broad-based domestic equity funds invest across all sectors of the JSE.)
However, a few of the funds did produce exceptionally consistent performance over five-or seven-year periods, Wessels says.
He looked at the relative perfor mance of all the funds in domestic equity general, growth and value sub-categories for each 12-month period from March 31, 2002 to March 31, 2009.
In particular, he considered how the funds fared when they were ranked on the basis of their performance and how, when the rankings were divided into four quartiles (quarters), the funds moved between these quartiles from one year to the next.
Based on his analysis, Wessels concluded that the relative performance and ranking of local equity funds exhibit randomness year-on-year.
But he found that funds in the top quartile (those among the top 25 percent of funds ranked by performance) on average were more likely to maintain their position among the top 25 percent than were funds that featured elsewhere in the rankings.
Wessels also found that if funds were awarded points depending on in which of the four quartiles their performance ranked, with higher points for better performance, it would demonstrate which funds perfor med well consistently. The PlexCrown Fund Ratings, which Personal Finance publishes on our unit trust prices page, take into account the consistency of a fund’s performance. The rating combines four different measures of perfor mance that take into account the risk the fund took to achieve its return over periods of up to five years.
RELATIVE TO THE INDICES
When Wessels looked at the performance of the equity funds relative to the two local equity market indices – the JSE All Share index (Alsi) and the Shareholder Weighted index (Swix) – he found that their perfor mance typified the active versus passive investing debate.
Local indices, particularly the Alsi, contain a sizeable portion of shares in the mining and resources sector, whereas local actively managed equity funds tend to be underweight in this sector and in individual shares that constitute a large portion of the market, such as BHP Billiton and Anglo American.
Wessels says it is no surprise therefore that only a minority of funds out-performed the Alsi during periods when the mining and resources sector did particularly well (between 2005 and 2008). But most actively managed funds outperfor med the Alsi when the resources sector lagged other sectors (as it did from 2002 to 2004 and from 2008 to 2009).
The Swix is therefore a fairer benchmark for local equity funds, Wessels says. The Swix downweights shares that are listed on both the JSE and a foreign stock exchange, because many of these shares are held by foreign investors and are not available to local investors.
Nevertheless, only a small portion of the funds reviewed outperformed the Swix in most years (that is, three out of five years), especially over the past five years.
USING PAST PERFORMANCE
Wessels also tested investment strategies that select funds based on their returns in the previous year. He tested the invest-andhold strategy, as well as strategies that involved switching funds throughout the seven-year period he reviewed.
For example, Wessels tested the results of investing in funds that were in the previous year’s top quartile and switching at the end of each year to funds that were then in the top quartile against investing in funds with a performance that put them in the second quartile each year.
On average, the strategy of investing in second-quartile funds produced better returns than the strategy of chasing the top-performing funds.
The results produced by switching to funds in the top quartile were also markedly worse than the actual annualised returns achieved by funds in the top quartile between 2002 and 2009. And the strategy only resulted in slightly better returns than those achieved by the Alsi, but the returns lagged the results of the Swix.
No switching strategy that Wessels tested produced results that outperformed the Swix.
To test the buyand-hold strategy, Wessels looked at the results of investing in and holding various types of funds based on their returns at the start of the period (March 31, 2002).
Funds that were in the top quartile at the start of the period did, on average, produce the best returns for the entire seven-year period to March 31, 2009.
The results were in line with the Swix but were less than what you would have earned had you been able to predict and invest in the unit trust funds that ended the period in the top 25 percent of funds on performance.
Wessels also tested 5 000 random portfolio selections for both a buy-and-hold strategy and one that involved annual switching.
The results of these simulations were very similar to the returns you would have achieved by investing in the Alsi and, on average, less than what you would have earned from the Swix.
TEND TO AVERAGE
Wessels concludes that whatever methodology you use to select a top-performing fund, most likely you will end up earning an average return over time.
Although some funds may have a great track record of producing above-average retur ns consistently, it is difficult to predict whether their future performance will also be consistent.
He says it is almost as difficult to select a fund that is capable of out-performing the market as it is to select individual shares.
Picking the right fund possibly involves more luck (as opposed to skill) than most of us are willing to concede, he says.
You should, he suggests, try to control what you can – namely, the cost of investing – by investing in a low-cost passive investment, such as an index-tracking unit trust fund or an exchange traded fund.
Wessels suggests you would do well to allocate a significant portion of the equity component of your portfolio (25 to 50 percent) to a low-cost passive investment.
When it comes to the rest of the equity component of your portfolio, Wessels suggests you look for an active fund manager whom you regard as a prudent guardian of your investment rather than simply investing in line with the manager’s past track record.
He says a fund with a consistent track record may instil confidence in the fund manager you choose, but you must consider the manager’s investment process and philosophy.
The quarterly unit trust performance figures for periods up to 30 years, the PlexCrown Fund Ratings and the consistency of performance results are published on the Personal Finance website. Go to www.persfin.co.za and then click on the “Unit trust results” link on the left-hand menu.