ERI Scientific Beta raises alarm over popular factor analysis tools
Popular tools for measuring the factor exposure and allocation of portfolios could be dangerous for some investors, according to a report from smart beta index provider ERI Scientific Beta.
Published in collaboration with a new venture, Scientific Analytics, the report argued that the use of some factor analysis tools could lead to a “serious” misalignment between investors’ factor diversification objectives and the measured and realised allocation.
In a research paper, the groups addressed two topics: factor definitions used by popular tools offered to investors – by providers such as Style Analytics, Bloomberg, MSCI and S&P – and the way in which measurement of factor proxies was implemented.
Introducing the analysis, Noël Amenc, CEO of ERI Scientific Beta, said some people would find them “slightly too controversial”, but that the message needed to be strong due to “the danger of transforming very important research concepts and results into poor and dangerous investment practices”.
Factor definitions used by popular tools were not supported by serious academic research, the report argued. This meant that most factors used in commercially available analytic tools were probably false.
Another conclusion, Amenc said, was that ‘non-standard’ factors could lead to incorrect measurement of exposures, or capture exposure to redundant factors.
“Analytic tools for investors distort the key idea of factor investing because they lack transparency and expose investors to providers’ conflicts of interests,” he added.
Factor ‘scores’ versus factor ‘betas’
The research paper also argued that using factor “scores” instead of beta data to measure portfolio factor exposures was “a cause for concern” because it was a departure from factor investing literature.
The major drawback with factor scores, according to the report’s authors, was “double counting” of exposures, which was due to their disregard for “the correlation structure of factors” and made them a “very poor proxy for factor betas”.
Overall, the organisations claimed the limitations of popular factor analysis tools could lead to investors being “unable to translate their risk allocation choices into a consistent allocation, with fairly severe financial consequences”.
Scientific Analytics is a new EDHEC Business School venture dedicated to factor analysis and allocation. It aims to provide institutional and private investors with tools that allow them to analyse the factor exposure of their portfolios and construct “completeness portfolios” that correspond to a search for better diversification. It plans to launch a freely accessible online tool at the end of next year.
The research paper can be found here [link corrected]. See also the EDHEC Research Insights supplement distributed alongside IPE’s November issue.