The risk premium does not exist and the scope of its failure is wide, says, Eric Falkenstein, Ph.D. and low volatility investing expert.
“I want to convince you that there is no risk premium,” said Eric Falkenstein, Ph.D., quantitative modeling expert and former hedge fund manager. “There are no intuitive metrics of risk that are robustly correlated with returns.” Proving this point was the basis for his recently published book, Finding Alpha: the Search for Alpha when Risk and Return Break Down, and the subject of his talk at the Robeco Minimum Volatility Investing seminar.
His argument that there is no relationship between high-risk stocks and higher return is based on a mountain of empirical research, his own as well as by other practitioners and academics. It all began with a study by the well-known financial economists, Eugene Fama and Kenneth French.
As Falkenstein described in his talk at Robeco, Fama and French published research as early as 1992 that rebutted the capital asset pricing model and showed a flat relationship between volatility and return. In an interview with the New York Times when the research was published, Fama said, "Beta as the sole variable explaining returns on stocks is dead."
This was a complete about-face for one of the original proponents of the capital asset pricing model, which fundamentally states that higher risk stocks can be expected to produce better returns than lower risk stocks.
Falkenstein, however, goes one step further from Fama's finding of a flat relationship between risk and return, and instead asserts "the real relationship is actually negative." This is visible when the performance of US equity high-beta and low-beta portfolios are compared against the S&P 500 Index. (The data for such a comparison is available on Falkenstein's web site.) Over a period of more than 50 years, low-beta stocks clearly outperformed both the market and high-beta stocks, with high-beta stocks turning in the worst performance of the three portfolios.
Falkenstein's extensive research reveals the lack of a connection between high-risk securities and high returns across a variety of investment categories, including equities, currencies, commodities, corporate bonds, private equity and even film investing and horse racing. While it is human nature to bet on long-shots at the race track, the higher risk bet is seldom rewarded, and the actual relationship between risk and return in horse racing is overwhelmingly negative.
"The best evidence against the risk premium is not one test," he says, "it is the scope of the failure. There are 20 asset classes where the risk premium fails to occur."
Falkenstein does note a positive association between risk and return in a few investment niches, including the short end of the yield curve and the spread between BBB-AAA corporate bonds. Efficient equity investing, such as indexing, should also yield a positive risk premium, he says, "But the average equity investor does not get the equity risk premium because of problems with taxes and expenses."
He points out that his back-tests of minimum variance portfolios, derived from a range of indexes, including the FTSE, MSCI Europe, Nikkei and S&P 500, have all outperformed the relevant index. "In every case, volatility is 30% lower and returns are 2-3% higher."
For Falkenstein, the take-away from this research is clear cut. "If risk and return are empirically uncorrelated, low-volatility investing is a rather straightforward normative implication of what a rational investor should do." In short, for equity investors, risk is not rewarded. Hence the rising popularity and success of low-volatility strategies.
1 Eugene Fama and Kenneth French, “The Cross-Section of Expected Stock Returns,” Journal of Finance, 47 (June 1992).