The good, the bad and the ugly of factor investing
Factor investing resembles the 1966 epic Spaghetti Western film, “The Good, The Bad and The Ugly", with factors having historically performed differently in various economic conditions.
In the film, “The Good” is an intelligent and confident bounty hunter who in the end gets the reward. “The Bad” is a mercenary who just finishes the job he is paid for. “The Ugly” is a fast-talking bandit with long list of misdeeds and obnoxious behaviours.
In investing, factors display various risk and return characteristics in different economic and market environments. ‘The good’ is when a factor does well relative to the broader market benchmark, ‘the bad’ is when the factor is delivering similar risk-return profile as the broad market and ‘the ugly’ is when it severely underperforms.
Factors are cyclical — they tend to perform differently in various economic and market regimes. Both the business cycle and the market cycle go in phases, driven by emotions, markets and structural changes in economic activity. During the high economic growth phase people feel optimistic about the future. We often see a boost in aggregate consumption – e.g. buying cars or smart TVs – when prospects seem rosy; investors are confident in financial markets and businesses are hiring, alongside other sanguine stories.
During periods of low growth, consumer emotions are replaced by anxiety, fear, and in some unhealthy cases, by panic. Investment factor styles tend to reflect these emotions through behaviours such as taking risk off the table – either shifting into low risk stocks (low volatility factor), holding fewer value stocks, or simply buying more bonds.
Let's analyse the factor premia and behaviours in various regimes for a handful of economic and financial indicators. We focus mainly on three regimes: (i) high (high levels or changes); (ii) stable (unchanged), or (iii) low (low or decreasing).
The chart below provides a summary of the factor premia historically linked to growth, inflation, interest rates and other financial variables. Each dot or marker in the chart indicates the historic performance of the individual or the multi-factor strategy in the various economic regimes. For example, the yellow triangle shows how different factor strategies performed during various economic growth states (high, stable or low).
Note: We have focused on the academic factor definitions using Fama French data base. Δ stands for the year-on-year absolute changes in the underlying macro-variable to determine rising, no change or decreasing behaviour. For other macro variables we look at levels such as high, stable or low. For example, high growth is defined as real GDP in excess of 3.1% and stable GDP growth is between 3.1-1.7% and low GDP growth is below 1.7%. The cut-off of various states of each macro variable is determined by the 70th, mid and 30th percentile.
Momentum and quality star as ‘the good’ factors in most macro and financial conditions. The momentum factor seems to have a balanced premium across various economic regimes, while the quality premium is defensive, consistent and visible compared to other factors.
Low risk tends to be 'bad' or even 'ugly' in some economic scenarios such as a high economic growth or low inflationary environment.
However, the size premium (based on small capitalisation stocks) features 'the good, the bad and the ugly', with the highest dispersion across the macro variables and large sensitivities to short-term interest rates and oil prices. As the wisdom from the movie goes, “there are two kinds of spurs… Those that come in by the door and those that come in by the window.” The size factor shows a seemingly counterintuitive outperformance in a slow growth environment. Unlike many other ‘straightforward’ factors that do well during times of high economic growth, the size factor can perform better at the end of a one-to-two year economic slowdown, when markets bounce back ahead of visible growth.
While individual factors are 'good, bad and ugly' relative to each other in different environments, the multi-factor exposure has historically been ‘good’. It can be almost macro or financial regime-agnostic due to its ability to extract premia from various sources of investment styles.
Factor premia are driven by various micro and macro environments – ranging from behavioural, market structures and economic conditions. After all, the narrative of the 'good and the bad and the ugly' simply sums up the behaviours of the factor premia. Even though every factor plays its own tune, some factors are more consistent in various regimes compared to others.