Smart beta investing: Not so smart?

Jonathan Boyd
clock
Smart beta investing: Not so smart?

Smart beta and factor products are expected to reach $1.2trn in assets by 2019 according to a Citi survey from last year.

Both product types are based on factor investing, which simply groups stocks by similar attributes, for example by their valuation, compared to analysing stocks by sectors or countries. This type of investing is nothing new and has been practised by quantitative investment managers for decades; however, smart beta exchange traded funds (ETFs), which offer cheap and transparent exposure to factors, are a relatively new innovation. These have proven a huge commercial success for product providers, mainly at the expense of active fund managers who have been losing market share to these passive instruments.

However, not all is well in the smart beta land. First, the product idea itself might be challenged as the performance attribution is difficult. Most investors associate factor investing with the academic research from Eugene Fama and Kenneth French. These academics create factors by constructing long and short portfolios of stocks, but smart beta ETFs are long-only index products with tilts towards certain factors. It is therefore quite challenging for an investor to analyse the impact of the factor tilt. If the smart beta ETF gained today, was it the performance of the index or the factor? A better solution would be buying a plain vanilla ETF and the factor as a long-short portfolio, which would be much more transparent in terms of performance. Until a few years ago this was difficult, but long-short factors are now traded as futures in Europe and ETFs in the US.

Secondly, surveying the smart beta landscape highlights that a large part of the ETFs are not supported by academic research, for example smart beta ETFs focused on the growth factor, which tends to tilt towards companies with strong sales and earnings growth. There is very little evidence that this strategy shows structural positive excess returns. However, there is a significant amount of evidence that the value factor, which can be considered the opposite of the growth factor and tilts towards cheap companies, generates positive excess returns across time, countries and sectors. It’s somewhat ironic that investors seem to believe in factor investing, but then ignore the results when it comes to picking investments.

Thirdly, our research shows that there is a significant difference between expected and realized factor returns from smart beta products. The key reasons are likely portfolio construction, factor definitions, and transaction costs. Investors need to be aware that they should take a healthy discount when observing returns in academic papers and be cautious on what to expect from actual products.

Despite these shortcomings, smart beta products are still a great addition to the investment universe as they cost a fraction of an actively managed fund. Fees have a significant impact on returns, especially over the medium to long-term, and from that perspective these are a smart choice.

 

Nicolas Rabener is managing director & founder of FactorResearch