The Ann F Kaplan professor of business at Columbia Business School, Andrew Ang will teach a case study on the Canadian Pension Plan Investment Board’s (CPPIB) reference portfolio in the fall. While for the most part complimentary of the approach and process, he challenges the Canadian fund to consider a more dynamic reference portfolio.
The CPPIB is respected by investment practitioners and academics alike for its approach to investment implementation, using what it calls a total portfolio approach. The strategy, which looks through asset class labels and considers each asset in terms of its underlying factors, is the result of a clearly defined governance structure that sets clear responsibilities for the board and management.
At its core is the reference portfolio, which is set by the board, and which management uses as a benchmark for making active decisions.
The reference portfolio is a passive mix that could reasonably be expected to produce the long-term average-annual real return of 4 per cent that is necessary to sustain the CPPIB at its current 9-per-cent-contribution rate. The current composition of the reference portfolio benchmark is 55 per cent global equities, 30 per cent Canadian nominal bonds, 10 per cent Canadian equities, and 5 per cent foreign sovereign bonds.
Active investment decisions are made against this reference portfolio, with every investment decision needing to be justified, or funded, against the reference portfolio.
The concept is simplistic and clean, but the implementation is quite complicated.
“The CPPIB reference portfolio is a low cost, tradeable, implementable portfolio that will meet their requirements. That is hard to beat,” Ang says. “If they did that alone then they would be top quartile because it is so hard to outperform because of costs. That thinking permeates their entire organisation and those that would normally operate in silos, according to asset classes, forces them to think across the entire portfolio.”
While the CPPIB approach is complicated to execute, Ang says from a top-down trustee level, it is simple and intuitive.
“The board has said buy anything you want but here’s the benchmark,” he says.
“Implementing it requires discipline, technical expertise, competence and independence that few investors have.”
He describes CPPIB as a professional organisation built in a structure that allows a manager a great degree of independence.
There is defined responsibility between the board and management and it is highly transparent.
The factor approach
Ang believes factor investing is what every investor should be aiming to do, but most investors haven’t embraced it.
In fact Ang advises investors to have a three-pronged investment approach which starts with a low-cost transparent portfolio.
He says then investors can benchmark to that and take on active management in a hopefully contrarian way.
Then ultimately asset owners can go to a third point of more discretion, with the rules giving you a conservative lower boundary.
“For example, in 2009 the rules would say buy, because markets were at a low, but a discretionary manager would buy more,” he says.
“The reference portfolio is the foundation stone. Most institutions don’t start with that, building a bridge without the foundation. CPPIB has a strong anchor.”
Chief investment strategist at CPPIB, Don Raymond, will join Ang at the New York City campus in the fall to discuss with students the complicated process of implementation that the fund has adopted.
Testing the hypothesis
Ang will challenge students on a number of aspects of the factor approach, including its applicability to funds such as US public-pension plans.
But he also has some suggestions in how to make improvements to the approach.
For one, he says the factors – the choices the board has made to include in the reference portfolio – are long only.
“There are many factors – usually the purvey of active managers – that are dynamic that involve long/short, that are systematic and can be done simply, and are hugely diversified.” He points to style-based factors, illiquidity premium, credit, and carry-on foreign exchange.
“This would make the reference portfolio more dynamic. Now it is a long-only reference portfolio done every three years. But some risk premiums can be accessed systematically.”
He does recognise that perhaps CPPIB investment staff are accessing these risk premiums in their active decisions, but believes they can be accessed more cheaply because they are systematic. It is essentially making active management harder.
If more dynamic factors were included in the reference portfolio, however, it raises governance questions as to where those dynamic factors sit, that is, with the board or the management team.
The other point Ang makes is that the reference portfolio doesn’t have any illiquid assets.
“Active management for CPPIB is collecting an illiquidity premium, and the board granted that decision to the manager. But it is a relevant question to ask what a fair hurdle rate for holding that liquidity premium would be. Whether the reference portfolio hold that is a question for the board.”
CalPERS is Ang’s next case study. And his reaction to the initial work on the case is shock. “You can’t even see an expense ratio,” he says. “It has taken us a lot of work to construct that number.”