Balancing all the Factors
Posted on January 4, 2019 under finance
As our clients are well aware, EBI uses "Factors" within our portfolios to "tilt" our holdings towards those areas that exhibit a premium over and above that of the market for exposure to various specific characteristics. All portfolios have a tilt towards Small Cap and Value shares, but for the World Portfolios, we also employ Momentum, via iShares and Vanguard managed ETFs. It begs the question as to why we don't use more "Factors", which we shall attempt to address here.
As the chart below describes, the growth of Factor Investing has been enormous, quadrupling in the last 6 years, as US investor interest has mushroomed.
An influential book, written in 2017 by Professors Feng, Giglio and Xiu  highlighted the fact that the advent of much higher computational power has allowed the "discovery" of nearly 250 factors, all of which appear to have some relationship with future asset returns; but many, being the product of data-mining (or spurious correlations) are either no longer useful or never actually were (as they failed to work in "out of sample" tests - i.e. they only worked in the period in which they were studied). The original intention was to try to understand how markets work, but over time it morphed into an attempt to sell financial products on the part of analysts.
Recall from Larry Swedroe's seminal book on Factor Investing that, to be suitable for inclusion in investment portfolios, a factor must satisfy 5 criteria - they should be
iv) Investable and
v) Have a logical, risk-based or behavioral finance-based explanation.
On this basis, a study from 2016 (which appeared in the Financial Analysts Journal in 2016) narrowed the field down to just 6, being Low Beta, Size, Value, Momentum, Illiquidity and Quality. As we already use Size (meaning Small Company shares), Value and Momentum, we shall address the others and explain why we don't use them in our portfolios, but we shall add Carry , Low Volatility and Profitability to this list for the sake of completeness.
The problem is that many of these are related in some way; Illiquidity is a generalised proxy for Small Companies (as the former tend to be, but are not exclusively) small companies; Low Beta shares tend to be Low (or Minimum) Volatility (as a firm with a Low Beta will tend to move less than the overall market, ceteris paribus). Firms with a High Dividend Yield (aka Carry), are often Value Firms, (as a high yield is often seen as a sign of relative cheapness) and returns are thus likely to be due to exposure to common factors; although not exactly alike, Quality is often a synonym for Profitability (a "High Quality" company would be expected to be highly Profitable).
Thus many of the Factors are at least partly correlated, reducing their diversification benefits. In many cases, there is no universal agreement on what constitutes the criteria for inclusion as a factor. In the case of Profitability, one could use EBITDA Yield , Return on Assets etc. but both of these are extremely dependent on accounting measures that are subject to Company Executive's discretion. Unfortunately, Firms have proven themselves all too keen to "gild the lily", when their own money (i.e. share option values) are at risk. In reality, many of the same firms could be in two or more universes, diluting the potential benefits of using them. Taken to extremes, tilting to multiple factors would end up with overall market exposure, potentially more expensively than just buying Index funds themselves. Introducing Profitability into a portfolio, for example, would tend to skew market cap averages upwards (as highly profitable firms - and those deemed high Quality - will likely be larger firms or they would not be as profitable). They would also be more growth orientated, thus mostly offsetting the Value factor exposure, blunting the positive effect of the latter (one in which we have far higher confidence).
Finally, to be useful, the factor must be investable - i.e. there must be a (cheap) product out there in which we can invest. For Value, Size, and Momentum this is not a problem, but Profitability, for example, is very hard to obtain (unless one were to pay the price premium associated with investing with fund managers such as AQR - a US Hedge Fund). Vanguard has launched a Quality ETF this year, but it is only available in the US (and invests only in US issues). Their Minimum Volatility ETF, launched in 2015 has thus far only attracted $29 million in AUM, which poses practical investment problems.
All these criteria create a high bar to success for Factor Investing. There is no certainty that ANY of them will persist forever (and as markets get ever more efficient, it is possible that even some of the stalwart factors will wax and wane over time, or even disappear altogether. Nobody ever said investing was easy...
 Carry is essentially Dividend Yield.
 Earnings Before Interest, Taxation, Depreciation, and Amortisation divided by the Firms Enterprise Value.