For smarter portfolios, look for better beta

Dr. Lionel MartelliniThe EDHEC Risk and Asset Management Research Centre and the CFA Institute held an annual three-day seminar on advances in asset allocation in New York in early May. One of the main themes of the seminar was how investors align their long-term time horizons within short term constraints.

Professor of finance at EDHEC Business School, and scientific director of the EDHEC Risk and Asset Management Research Centre, Lionel Martellini, spoke to Amanda White about the challenges to better practical portfolio construction.

In the past 30 years, developments in asset allocation have focused on alpha chasing, and placed much emphasis on security selection. But according to the EDHEC Risk and Asset Management Research Centre’s scientific director and host of its advances in asset allocation seminar, Lionel Martellini, the newest wave of thinking focuses on better beta management.

More than 50 chief investment officers and portfolio managers of sovereign wealth funds and pension funds around the globe attending the seminar, which discussed how the gap between modern portfolio theory and practical portfolio construction can be bridged, and how integrating liability and risk management constraints into portfolio construction completes the picture.

“In the past few years all the things that have been discussed in asset allocation have stemmed from security selection, like alpha/beta separation,” Martellini says. “Our view now is this is a half-story, the tip of the iceberg – there is a more significant change in paradigm on the way.”

Staged in two parts, the seminar firstly discussed the inefficiencies of cap-weighted indices as an investable product, and looked at how to build more efficient portfolios by distinguishing between indices and benchmarks.

“In beta management investing, the core portfolios are in market-cap weighted indices, it is the most important decision you make, but it doesn’t get enough attention,” he says.

Instead, what EDHEC is going back to the roots of portfolio theory and revising risk/return trade off expected from indices.

“Indices are not well diversified because cap-weighted means you have higher percentages allocated to fewer stocks. The alternative is equally weighted benchmarks, which are well diversified but are kind of frustrating.

“The real challenge is figuring out how we can deviate from equally weighted indices. And we are exploring advanced techniques to nail down a better portfolio combining statistical analysis, common sense and economics.”

Diversification allows investors to build portfolios targeting an expected return with less concentrated risks, but according to Martellini, the next step is to realise that diversification is only a building block.

“Diversification fails us when we need it most. In 2008, if you invested in market cap indices you would have returned -40 per cent. If you invested in equally weighted indices you would have performed better, but still pretty badly, say -35 per cent. What you have to recognise is that while diversification is important, it is a building block, and you will fall down when it does.”

This leads to the second theme of the seminar, which probed the effectiveness of using either hedges and insurance to perform risk budgeting.

“You need to put the building blocks together with other ingredients,” he says. “LDI solutions are okay, but they are very static. We believe in dynamic management of these building blocks.”

“We believe most investors – sovereign wealth funds, pension fund chief investment officers, have long-term time horizons but short-term constraints. Until now portfolio management been very static – with buy and hold strategies the norm. But the only way to handle the short term constraints is to be dynamic.”

According to Martellini, investors need to accept they have short-term constraints – including regulatory, accounting and self-imposed constraints – and manage them with their long-term time horizons in mind.

“Dynamic asset allocation decisions are a tremendous value-add, they allow you to incorporate long-term horizons, but target date funds don’t make sense. [Investing] should be a function of the market and economy, not a date.

“Investors need to accept they have those constraints, and implement dynamic risk techniques.”