Multi-management has come a long way since Russell Investments helped pioneer the concept decades ago but some things never change.
Jon Eggins, an Australian Russell portfolio veteran now based in the group’s Seattle HQ, said even today trainee analysts must still pore over the famous (in-house, anyway) ‘Duncan’s Tome’ that lays out the multi-manager basics.
The lengthy document, authored in 1994 by Russell alumni Duncan Smith, details the firm’s investment research process designed to identify top-quality managers.
Eggins – in NZ last week along with implementation portfolio manager, Andrew Zenonos – said the almost 30 year-old document has not lost currency in the intervening years.
“What is old has become new again,” he said. “We make our new analysts read ‘Duncan’s Tome’ – it avoids reinventing the wheel.”
But while the Russell rules for selecting world-class investment managers – or “all killers, no fillers” as the firm’s informal motto has it – remain more or less unchanged since inception, the portfolio mechanics have seen a revolution during the last 10 years or so.
According to Eggins, the 2008/9 global financial crisis (GFC) highlighted the fact that simply bundling the best managers into a portfolio was not enough.
He said the GFC handed Russell a few lessons in the art of multi-management, including:
- static buy-and-hold strategies are insufficient;
- even small unintended risks can have huge impacts in market extremes; and,
- using managers alone does not allow precise risk management.
For instance, the GFC exposed some persistent biases across the Russell managers such as an underweight to US shares, high exposure to volatility and “meaningful growth tilt”
Post-GFC Russell has taken a more hands-on role in managing risk across its portfolios, Zenonos said.
He said previously the Russell process was composed of manager research and portfolio analytics. In the wake of the crisis, however, the firm overlaid new risk analysis tools (strategic beliefs and forward-looking market views) along with in-house implementation techniques.
The implementation updates allow Russell to tweak overall portfolio risk with strategies such as futures or factor bets to counter manager biases.
As well, Russell has shifted many of its equities multi-manager portfolios to a centralised trading system known as ‘enhanced portfolio implementation’ – or ‘emulation’.
Under the emulation approach, managers supply their daily portfolio buys-and-sells to Russell, which implements them via a central trading desk.
“Emulation makes our multi-manager [share] funds more efficient and more nimble,” Zenonos said.
Depending on the market, emulation can shave off between 20 to 85 basis points in implementation costs through savings in broker and custody fees as well as ‘netting-off’ trades between managers.
Centralising multi-manager trading also allows Russell to more easily express environmental, social and governance (ESG) screens.
Since introducing emulation about 10 years ago, Russell has transitioned a large part of its equities multi-manager suite to centralised trading. In Australia, for example, over 80 per cent of the Russell multi-manager share funds are emulated.
The Russell NZ shares fund, though, remains in the traditional model, where each manager – currently Harbour and Devon – trades under its own steam.
Nonetheless, in 2018 Russell did add an in-house factor overlay to the NZ fund to tilt the portfolio as required. At some point, too, emulation is likely.
“We’re still doing research on how it would work in the NZ market,” Zenonos said. “But we expect to move to emulation there in the next couple of years.
“We’re pragmatic about introducing change.”