The author discusses the origin and construction of traditional indexes and introduces the concept of dynamic indexes. These indexes are not necessarily market cap weighted or passive in the traditional sense. For example, strategy indexes embody a particular investment strategy. The author cautions that smart beta strategies may lead to unintended risks that will not earn a positive risk premium. Using the example of a volatility-adjusted dynamic index, the author elaborates further on the benefits and potential pitfalls of dynamic indexes.

Given technology-leveraged financial innovation, the author asserts a need for a new perspective on indexes and the distinction between active and passive investing. In particular, passive investing need not, and should not, imply passive risk taking.

Traditionally, there has been inertia around moving too far away from market-cap-weighted constructs because they have worked well in the past. Given the recent advances in computational, communication, and financial technology and the resultant technology-leveraged financial innovation, however, a new perspective is needed on indexes, indexation, and the distinction between active and passive investing.

In these times of high volatility, passive strategies may not be as appropriate as they were once considered. For example, the recent maximum drawdown (MDD) of the S&P 500 between 9 October 2007 and 9 March 2009 was 56.8%, which led to significant investor losses, especially for long-term investors who do not regularly monitor their portfolios.

One simple way to manage such volatility risk is to create a dynamic index that contains no alpha but actively manages the volatility to a target level. This strategy reduces the market exposure when short-term volatility is high. Given the inverse relationship between volatility and prices, the strategy effectively reduces equity exposure in declining markets and increases market exposure when volatility is lower. The author finds that, over 1926–2014, such a volatility-adjusted index has cumulative returns that are 4× higher than those of the benchmark; it also has lower MDD (–72% versus –84%) and less excess kurtosis (4.85% versus 16.97%). The results vary for various subperiods but are generally better for the volatility-adjusted index than for the raw index.

The author cautions that with increasing innovations and a wide range of available products, professional fund managers need a higher level of sophistication in terms of strategy evaluation. The backtest bias is especially problematic for smart beta strategies because the number of managers/models is rapidly growing, the signal-to-noise ratio is declining, and the performance measurement for these strategies depends on simulated statistics with no track record.

The author constructs a volatility-controlled index with target volatility and no alpha to evaluate and demonstrate the use of dynamic indexes. When the expected volatility is below the target volatility, the dynamic index uses leverage to increase exposure. When expected volatility is above target volatility, it invests a portion of the fund in cash. The leverage/deleverage factor is based on the target level of volatility relative to the estimated volatility of the index. The target level of volatility can be chosen with the investor’s risk tolerance in mind.

The author uses various parameters to compare the results of the volatility-controlled index with those of the CRSP NYSE Value-Weighted Market Index over the period, and rolling subperiods, between January 1926 and December 2014.

Smart beta and factor investing, quasi-active index strategies, recently have been extremely popular with fund managers. Historically, selected factors have outperformed the market indexes. These funds suffer from backtest bias and can carry higher risk as well as higher fees and higher valuations because of increased popularity. Although the arguments in favor of smart beta and factor investing are quite compelling, they require sophisticated evaluation and risk management overlays.