AIMCo: Better Performance Attribution = Better Performance

The era of Excel and the Brinson-Fachler formula is over, the heads of Alberta’s sovereign wealth fund stressed in a white paper.

(July 15, 2013) — Effective feedback from modern data systems is key to producing strong investment performances, according to new research from the Alberta Investment Management Company (AIMCo).

AIMCo started its life in 2008 with roughly $70 billion and only spreadsheets to keep track of it.

CEO/CIO Leo de Bever and his deputy Jagdeep Bachher began rebuilding the fund’s IT infrastructure in 2009, but have only recently begun enjoying its full capacities.

One of those abilities is sophisticated and timely performance attribution. De Bever can see exactly which managers are making–or losing–the fund’s sovereign wealth and pension assets, and just how they’re doing it.

Never miss a story — sign up for CIO newsletters to stay up-to-date on the latest institutional investment industry news.

“We have found over the past four years that strong investment performance requires providing good people with continuous performance evaluation feedback from good business systems,” de Bever wrote in a white paper about AIMCo’s experience, co-authored by Bachher, IT consultant Roman Chuyan, and institutional academic Ashby Monk.

“This, in turn, drives informed investment decision-making and better performance. There is always room for improvement, but it is already clear that better data, systems, benchmarks, and decision-based attribution are having a measurable impact on our ability to target superior results.”

AIMCo’s investment heads found the traditional formulas for measuring alpha contributions to be “somewhat flawed”. The Brinson-Fachler daily decomposition is broadly used for active management performance attribution, but according to the paper, it does not work well with illiquid assets.

Furthermore, AIMCo’s managers make allocation choices via a “decision tree”, which accounts for the asset type, region, source of capital, and specific selection of every investment the fund makes. Not only did the Brinson-Fachler fail to attribute properly given this method, so did existing digital performance systems.

The C$68.6 billion ($66 billion) fund partnered with Boston-based vendor Opturo for a semi-bespoke system in September 2011, and the new platform was fully in place by August 2011.

Optura President Ken D’Silva said the company is “proud” of its contribution to AIMCo, which is a major client.

De Bever and Bachher have been outspoken in their support of high-tech data systems for asset owners. Upon AIMCo’s founding, the paper said, the fund not only lacked a sophisticated portfolio management system, but also clean and timely data. 

“While shocking, this is actually not uncommon for investment organizations; even those managing billions or tens of billions of dollars,” the authors wrote, “but that does not make the practice acceptable.”

Read de Bever, Bachher, Chuyan, and Monk’s entire paper, “Towards the Next Generatiuon of Performance Attribution for Institutional Investment,” here.

Related Article: Asset Owners on the Cutting-Edge of Data Management  

«