A Three Factor ETF Portfolio

As an average retail investor, let’s assume that we are convinced that passive investing is the most optimal investing strategy for us. Concept-wise, it is simple and easy to execute – just buy a ETF or low cost index fund. The results are also more reliable – you will gain slightly less than the market returns minus the cost.

But does it mean that by choosing the path of passive investing, we resign our fate and accept whatever the market bestows upon us? Not necessarily.

One strategy that has gained popularity recently propelled by latest research on financial markets, popularity of ETFs and movement towards low cost passive investing.

What is that strategy? Factor investing.

In factor investing, there has been quantifiable research that supports the concept that holding stocks with certain characteristics (a.k.a risk factors), provides higher returns than the market in the long term.

To put it simply, it promises to beat the market.

The First Factor – Market

We intuitively know that the more risk we take, the higher the returns we expect. But what exactly is the relation between risk and expected return?

In 1960, the capital asset pricing model (CAPM) was created to explain just that – expected return of a stock with respect to a single factor – the market risk premium.

Expected Return = Risk Free Rate + Beta ( Market Risk Premium)

The risk free rate is the rate of returns from holding risk free assets which are usually long term government bonds. You expect to get this return without taking any risk to your capital.

Beta is a risk measurement that measures the sensitivity of a stock with the overall movement of the market.

High beta stocks are more volatile than the market while low beta stocks are less volativle than the market.

The moment you decide to invest in publicly traded stocks, you are taking on the market risk – a systematic risk of the overall market.

Like it or not, the stock is going to be affected by the movements in the market. The recent market crash had just reminded us that the risk is real and stocks in the market cannot escape it.

The market risk premium is just the expected return of the market minus the risk free rate

According to CAPM, it is exposure to the market factor that drives the underlying risk and return of stocks. And fortunately, because market factor is a risk, investors expect to receive a premium taking on this risk, which is the market risk premium.

So, the higher the overall beta of the stocks in your portfolio, the higher the expected returns.

Three Factor Model

In 1992, Fama and French published a paper which introduces a three factor model, an extension to CAPM, to explain stock returns. The two factors other than market risk are size and value.

Expected Returns = Risk Free Rate + Market Risk Premium + SMB + HML

The actual formula looks like this

SMB is Small Minus Big, which is calculated by the average historical returns of portfolios containing small cap stocks minus the average historical returns of portfolios with large cap stocks

HML is High Minus Low, which is calculated by the average historical returns of portfolios containing stocks with high book-to-market ratios minus the average historical returns of portfolios with low book-to-market ratios. This is the definition of value stocks used.

Size Factor

They (Fama and French) showed that there is a risk premium holding smaller cap stocks. In the long run, based on empirical evidence from their paper, small cap stocks tend to outperform large cap stocks.

This is probably because smaller cap stocks are riskier (more volatile) and face higher risk of bankruptcy hence investors will demand a higher discount to compensate the risk of holding small cap stocks.

Hence, if the small companies do survive and grow, investors will be rewarded for holding these companies in their portfolio.

Value Factor

The other risk premium introduced is value, which are inexpensive stocks. The opposite of value stocks are growth stocks.

The definition of value stocks are stocks which are trading at a discount to their intrinsic value. Specifically stocks with high book value (If you add up all the net assets in the financial statements) but low market price are considered as value stocks.

Famous examples of value investors include Warren Buffett and his mentor Benjamin Graham who is also known as ‘the father of value investing’.

The price of a stock depends on the expectations of the investors. Investors tend to have very high expectations of high-growth stocks valuing them at a higher price. Examples include the tech FAANG stocks.

Conversely, investors tend to view slow growth stocks more pessimistically pricing them at a much lower price.

Hence, it is easier for value stocks to exceed the lower expectations of investors compared to growth stocks exceeding expectations of investors.

Constructing A Three Factor Portfolio

So, in theory, in order to increase expected returns, we need to move away from traditional market beta portfolios (or vanilla portfolios) to smart beta portfolios (smart beta is the name of factor based investing strategies).

How would an ETF portfolio with three factors look like?

Here is an example of my US portfolio which I had adapted from this article:

• US aggregate bond market ETF: AGG (30%)
• US total market ETF: ITOT (30%)
• US value ETF: IUSV (30%)
• US small value ETF – IJS (30%)
• Ex US ETF – IXUS (10%)

Note: The iShares ETFs here are just personal preference. There are also equivalent ETFs from Vanguard, SPDR, Schwab etc etc… for each of the ETFs I have listed. I often go to etf.com to compare various ETFs.

Are Multi-Factor Portfolios A Free Lunch?

It is not a free lunch. This is because you are taking on additional unique risks on top of market risk.

Just as an example, let us examine how the respective smart beta ETFs performed during the recent market crash compared to the S&P index. I will be using IVV (iShares S&P ETF) as a proxy for S&P.

Comparison of IVV, IUSV and IJS during the recent crash

From Feb 10 to Mar 20, IVV fell -32.69% while IUSV fell -36.72% and IJS fell -41.06%. The small and value ETF fell the most which is not surprising because smaller companies are more vulnerable to recessions compared to bigger companies with usually larger cash reserves.

The S&P consists of mainly large cap stocks. If we include ITOT which is a total market ETF which contains all large, mid, small cap stocks in the US market, you would see that ITOT also fell slightly more than IVV.

Comparison of IVV, ITOT, IUSV and IJS during the recent crash

ITOT fell -33.65% which is slightly more than IVV -32.65%.

My portfolio was hit harder compared to a vanilla index fund based portfolio. This is what it means to be exposed to more risk premiums – more potential downsides.

That said, if you are prepared to take on this additional risk, you may get rewarded in the long run.

Long Periods Of Underperformance

Exposure to factors does not always equate to market over performance. History has shown that certain factors also experienced long periods of underperformance.

The most recent is the underperformance of value stocks compared to growth stocks. Since 2009 to 2019, value stocks had annualized returns of -2.88% while the market has annualized returns of 11.78% according to data from Kenneth French’s website.

It is not uncommon for certain factors to underperform during certain periods. During those periods of underperformance, other factors may be overperforming the market. So it is important to diversify across different factors.

Since then, Fama and French had developed a five factor model that introduced profitability and investment. Also, many other factor discoveries had also been made namely momentum, quality and low volatility.

Further reading on the other factors in “An Overview of Factor Investing” from Fidelity

Costs And Tracking Error

In theory, constructing a multi factor portfolio is beneficial but in practice, the implementation is often very difficult.

Firstly, a factor based ETF is usually more costly to implement. For example, small cap stocks are more illiquid and will cost more to trade them. Sometimes, ETF uses derivatives to replicate these illiquid assets without having to own them physically. Whereas plain vanilla ETFs often consists of large cap stocks which are more liquid and cost less to trade.

The higher turnover ratios of certain factor ETFs will also cost more. Momentum factor is a rather difficult factor to capture which will involve more frequent and higher turnovers.

The more factors you add, the more complicated it becomes and the higher the cost is needed to implement it.

That does not mean that factor ETFs usually have high TER. There are even multifactor ETFs that have low expense ratios, take Goldman Sachs ActiveBeta US Large Cap Equity ETF (GLSC) for example with a low expense ratio of 0.09%. The tradeoff is you may have to accept higher tracking error.

Conclusion

Factor investing provides us passive investors the prospect of outperforming the market. It is also prudent to diversify across different factors to gain exposure to independent risks to gain a higher expected return.

Hence, factor investing is indeed an appealing investing strategy to pursue.

Despite all the drawbacks that I had listed, I am still betting on this strategy to pay out in the long run.

The last thing to note is that CAPM, the three factor model, the five factor model and many more models are just mathematical models created to help us understand and explain what is observed. It should not be the only source of truth nor should we treat these models as infallible.

Perhaps, this strategy might do well. Perhaps this strategy might be updated and improved upon in the future. Only time will tell.

Advertisement