Is there such a thing as smart beta?

"Smart Beta" is a term that is cropping up more lately in professional investment management circles. It is essentially a way that professional investors tweak index funds in hopes of achieving higher returns than funds that simply mimic an index.

In April, none other than Goldman Sachs Asset Management acquired a firm that specializes in smart beta strategies called Westpeak Global Advisors. In a June 13 research report, Goldman estimated that there are now $355 billion in assets in exchange-traded investment products that employ smart beta strategies, up from $65 billion at the start of 2009.

Given the strategy's growing influence, I thought it would be useful to explain what it is and provide my take on the concept.

There is some genuine controversy about whether or not it even exists. For example, Portfolio Solutions' Rick Ferri and Research Affiliates' Rob Arnott debated that issue last September. To me, the issue is clear. There's nothing smart about smart beta.

Beta is just beta

To understand smart beta, first you have to understand the term "beta." Nobel-prize winning economist Bill Sharpe coined the concept decades ago when he developed the Capital Asset Pricing Model (CAPM) of modern portfolio theory. As Sharpe explains, "beta is simply a portfolio's sensitivity to movements in the overall market."

Ferri argues: "How do you get smart out of that? It's neither smart, nor alternative, nor better. It just is." Is Ferri right?

The CAPM uses the single factor of beta to evaluate the risk of an individual stock, an index, a mutual fund, or a portfolio of funds. If the beta is greater than one, there's more risk and a higher expected return than the market. If it's less than one, there's less risk and a lower expected return than the market.

The corollary to beta is "alpha." Returns that aren't explained by beta are alpha -- which can be positive or negative depending on whether the portfolio's returns were better or worse than benchmarks.

That model was the operating model for about 30 years, until the early 1990s and the publication of work by professors Eugene Fama and Kenneth French. Their work led to a three-factor model becoming the standard used in analysis, adding the factors of size and value. Shortly thereafter momentum was added as a fourth factor.

Looks can be deceiving

An important result of the four-factor model was that we learned that what looked like alpha was very often "beta" -- or exposure to size, value, and momentum factors. In other words, active managers who tilted their portfolios away from a market-like portfolio to gain access to the small, value, and momentum factors outperformed the market because of their exposure to these factors, not because they were good stock pickers.

While multifactor models do a much better job of explaining returns than the original CAPM did, there are still remaining anomalies (remember, size, value, momentum were once anomalies) that the models cannot explain. Among the anomalies is that any asset with a lottery-like distribution has been shown to have poor-risk return characteristics -- exposure to these assets results in negative alphas (below benchmark returns).

That brings us back to the question -- is there such a thing as smart beta? In my view, while the answer is yes, it might just be a matter of semantics.

You say beta, I say alpha

To begin, there can be many portfolios that have the same loading or exposure to beta. Let's assume that we start with a mutual fund (Fund A) that owns the total U.S. market. By definition, it will have a beta of one. Manager of Fund B says we can create smarter beta by screening out all the stocks that have these lottery-like distributions (IPOs, penny stocks, stocks in bankruptcy, and extreme small growth stocks). Fund B will also have likely have a beta of one, but it can be expected to produce a higher return in the long-term.

Since the betas are the same, it seems perfectly appropriate to say that's smarter, or better, beta. Or you could say it's alpha. The difference is just semantics, not a real difference.

It's important to note that recent research by Robert Novy-Marx on profitability (highly profitable firms outperform despite having higher valuations) has further improved our ability to explain returns -- once again converting alpha (or smart beta) into beta (or exposure to a factor).

Additionally, some investment firms, such as Dimensional Fund Advisors and AQR, are now beginning to incorporate this new factor into the design of mutual funds. (Full disclosure: My firm Buckingham recommends Dimensional funds in constructing client portfolios) My view is that these funds are smarter and likely to produce superior risk-adjusted returns.

The bottom line

The bottom line is that the evidence demonstrates that incorporating the findings from academic research can result in the design of portfolios that produce superior results to total market portfolios and pure index funds.

To be confident that the academic findings are reliable, there should be a logical explanation (either risk based or behavioral) for the findings. The findings should also be persistent over long periods of time and across asset classes and markets; they should survive transactions costs. Whether you call the results better beta or alpha, the outcome is the same: superior risk-adjusted returns.