Academic Insights on Investing

Weekly Academic Insight: Facts About Factors

TITLE: FACTS ABOUT FACTORS                   

Authors:        Paula Cocoma, Megan Czasonis, Mark Kritzman,  and David Turkington

Publication:    Journal of Portfolio Management, 2017, Special Issue

 Link:      Cocoma et al. (2017)


#What are the Research Questions? :

(1) Do factors offer superior diversification benefits relative to assets because factors are less correlated with each other?

(2) Does consolidating a larger set of assets into a smaller set of factors reduce noise?

(3) Are investors more skilled at relating current information to future factor behavior than to future asset behavior?

(4) Are factors and assets prone to the same types of errors that contribute to covariance instability?

(5) In summary, is it correct to use factors as building blocks for forming portfolios?

#What are the Academic Insights? : 

(1)  No, this claim is only true because factors include short exposure to the assets.

(2)  The authors find no evidence that factor groupings reduce noise more effectively than asset class groupings. In fact, they find the opposite.

(3) Investors who favor predicting factors face the additional challenge of mapping these factor predictions onto asset predictions, and they must also incur incremental trading costs to the extent that factor-mimicking portfolios change over time.

(4) Factors are less stable than assets mainly because, unlike assets, they are subject to mapping error

(5) Investors should not replace assets with factors as the building blocks for forming portfolios

#Why Does It Matter? :  

While the authors acknowledge that factors may offer risk premia, which are an important component to portfolio’s returns, they advise against using factors instead of asset classes when defying the asset allocation of a portfolio. In fact, after reviewing a series of claims that back this idea in the scientific community, they find no strong evidence that it adds value compared to traditional asset allocation approaches.