Aberdeen Standard’s Millington on avoiding the ‘factor zoo’

Tom Eckett

a group of flamingos in a pond

Nick Millington, co-head of quantitative equity at Aberdeen Standard Investments, has said there are a number of ways investors can avoid getting lost in the “factor zoo”.

The huge number of factors backed by academic evidence is a major issue for the investment industry. According to a 2019 study conducted by Duke University’s Professor of Finance Campbell Harvey, there are over 400 factors published in top journals.

These factors have been created by inflated backtests and cause investors to “develop exaggerated [return] expectations”, Harvey said.

“Like any investment strategy, factor investing is not immune to data mining,” Millington added.

ETF providers in smart beta catch 22

In order to avoid this, Millington stressed there are five factors that have displayed long-term success; value, quality, momentum, small size and low volatility.

For each of the five factors, Aberdeen Standard Investments takes only three or four key metrics so as not to “overcomplicate” its strategies.

Furthermore, Millington said he prefers adopting a multi-factor approach in order to reduce the downside risk of investing in a single factor, which can underperform at certain points in the market cycle.

“It is very important to avoid overengineering the factor ingredients. Using three or four metrics for each factor provides enough diversification.”

Another area of growing importance is environmental, social and governance (ESG) investing. However, the reason why it has not become a proven factor, Millington explained, was due to the lack of available data.

ESG is a data miners’ dream

ESG data has improved dramatically in recent years, however, there is a big issue around the lack of correlated data between the providers.

For example, MSCI and Sustainalytics, the two most widely used ESG data providers, have a correlation of just 0.53 on their ESG scores.

“ESG is an emerging risk factor,” he said. “However, there is simply not enough data in comparison to other factors.”

Featured in this article


No ETFs to show.