Monday, April 22, 2019

Factor Investing - Not as easy as many expect


Factor investing currently is at the forefront of asset management, yet like many hot topics in finance there has been exaggerated claims on what return and diversification benefits investors will receive. Factor investing is a significant advancement for decomposing risks and building unique portfolios of risk, but investors need to be aware of a number of potential pitfalls. A recent working paper describes some of these key pitfalls. See “Alice’s Adventures in Factorland: Three Blunders That Plague Factor Investing” by RobArnott, Campbell Harvey, Vitali Kalesnik, and Juhani Linnainmaa. The authors delineate three areas of concern.

Investors can form exaggerated expectations on what can be the potential return performance from factors. Some have described a “zoo” of factors. There are now hundreds of factors that have been analyzed and reported in the academic literature. Nevertheless, many have been hard to replicate, show performance return declines after being researched, and have failed when tested out of sample under realistic market conditions. These poor results may be from data mining, not accounting for transaction costs, poor design, and potential crowding. There are successful factors that have stood the test of time, yet even their performance has been time varying and may be less successful factor after accounting for all costs. Don’t be disappointed if actual returns are less than what is reported in academic studies.

Factor downside risks may be greater than what would be expected under assumptions of normality. Some strategies are subject to negative skew, and tail risks, in general, may be greater than what may be expected in a normal world. There may be large drawdowns that are somewhat masked when looking at long-term returns. A strategy that is focused on one factor will of course have more focused risks than what may exist with a market portfolio. This downside risks do not mean factors should be avoided. Rather there should be extra focus on tail events. Focused risk management is essential if you are going to focus on factor investing.

Factor may look like they are unique and have low correlation with market portfolios and with other factors; however, factors can see increasing correlation with different shocks. Some factors will be sensitive to volatility shock, market extremes, the business cycle, or inflation. Some factors show negative convexity while others have defensive properties and positive convexity. Stress testing factor portfolios is essential. 

Tempering expectations for factor investing will actually help investors build better portfolios and lower any potential investor disappointment. Knowledge of factor pitfalls will only strengthen their long-term use as an alternative method for adding diversification and tying returns to specific risk premia and financial drivers.


No comments: