Tracking error leads to incorrect assumptions

The wrong assumptions are being made about fund managers and their approach to risk by those using tracking error, according to Russell Investments’ portfolio manager, Scott Bennett.

Speaking at the Portfolio Construction Conference in Sydney yesterday, Bennett suggested that ‘active money’ – a measure of how different the portfolio weights are from the benchmark weights – could be a useful and additional measure of a manager’s approach to risk, asserting that tracking error might not be the best measure used in isolation.

Based on Russell Investments research, Bennett said in environments of greater valuation dispersion, in which active managers would be expected to find greater opportunities, they actually spent less, whereas during periods of lowest opportunity they actually spent more. Bennett said this pointed to managers showing a risk aversion in periods of greatest opportunity and then being forced to spend money in periods of lowest opportunity.

Bennett noted that the research identified that in a volatile period when tracking error was climbing up, the measure of active money actually showed that they were not spending money. As such, he suggested that incorrect assumptions were being made about fund managers’ behaviour based on tracking error.

Bennett asserted that there is a misunderstanding of what tracking error actually means.

“People randomly use tracking error and don’t know what it means,” he said. “Be careful of tracking error as a measure. In terms of being a measure of managers’ intentions, it can mean something very different.”

Bennett added that providing managers with a tracking error range could lead to “strange behaviour”.

“Because they are managing tracking error, at times of opportunity they could actually pull back,” he said, adding that as such Russell would review its approach to active money as a useful and additional measure of risk.

He asserted that active money as a risk measure was useful in that it gave insight into the actual underlying portfolio and could enable a better informed decision in evaluating fund managers.

“Tracking error is still important, but like any measure you don’t look at it in isolation,” he said. “As an industry as a whole, we need to start asking more questions and not just accept the numbers.”