As we head further into 2019, some of last year’s concerns, including market volatility and interest-rate uncertainty, continue to occupy investors’ minds. With the assumption that rates-related concerns continue and uncertainty looms in the global equity markets, the question is how minimum volatility indexes behaved in an environment dominated by these two opposing forces.
Our research finds that, over the last 20 years, minimum volatility indexes outperformed broad market-cap indexes in times of high volatility and heightened uncertainty, as well as when rates have been in decline.1 Moreover, while minimum volatility indexes have shown historical sensitivity to interest rates, market and macro-economic conditions had a greater impact on performance.2
For example, although 2018 was a year where both rising interest rates and heightened market volatility affected the market, the exhibit below shows that the MSCI Minimum Volatility Indexes across different regions outperformed their corresponding broad market indexes by 6 percentage points or more. And they showed lower volatility, resulting in higher risk-adjusted returns. These numbers suggest that market volatility was the dominating force behind the MSCI Minimum Volatility Indexes’ outperformance.
MSCI Minimum Volatility Indexes outperformed market-cap indexes in 2018
|MSCI World||World Min Vol||MSCI USA||USA Min Vol||MSCI EM||EM Min Vol||MSCI Europe||Europe Min Vol|
Period: Dec. 29, 2017 to Dec. 31, 2018.
Measuring the effects of market conditions vs. interest rate environment
We first examined the average annual active return of the MSCI USA Minimum Volatility Index for two market environments, or regimes.3 During market downturns, the index significantly outperformed the broad market index, while it underperformed in up market environments.
12-month average active returns for the MSCI USA Minimum Volatility Index
Data from Nov. 30, 1975 to Dec. 31, 2018
We then broke down the environment further to rising- and declining-rate environments. The exhibit below shows the performance of the MSCI USA Minimum Volatility Index relative to the market over four different market performance/interest rate regimes.
Most notable is the top left quadrant, where the market was down but rates were rising (2018 fits into this category). In this environment, the MSCI USA Minimum Volatility Index outperformed the broad market by about 2 percentage points annually on average. This suggests that while minimum volatility performance was sensitive to changes in rates, the health of the market played a more important role in its behavior. Alternatively, when interest rates were rising and the market return was positive, the MSCI USA Minimum Volatility Index underperformed on average.
Market performance had a bigger effect than interest rates On 12m USA Min Vol Average Active Returns
Data from Nov. 30, 1975 to Dec. 31, 2018
Measuring the effects of macroeconomics vs. interest rates environment
We also looked at the state of the economy and its effect on the MSCI Minimum Volatility Indexes in conjunction with the direction of interest rate changes. We use regression analyses to gauge the effect of these two parameters on the indexes’ performance relative to the market. We use the Institute for Supply Management® Purchasing Managers’ Index (ISM PMI) as the macro-economic indicator and the U.S. 10-year Treasury yield to measure rates.4 The analysis is done on the MSCI USA Minimum Volatility Index for the December 1975 to December 2018 time period.
As we can see in the exhibit below, both interest-rate changes and changes in PMI have negative coefficients. This means a positive change in interest rates or PMI led to index underperformance. However, the higher t-statistic (measure of a statistical estimates’ “reasonableness” or “extremeness”) for PMI suggests a higher explanatory power for this parameter of the index’s performance. Also, using the coefficients, we can calculate that one standard deviation change in PMI has historically resulted in a larger change in active return for the MSCI USA Minimum Volatility Index than one standard deviation change in rates (2.0% vs. 1.6%).5
PMI changes had a bigger effect than interest-rate changes on min vol active returns
|10Y Yield Change||-0.0128||-7.3836|
|Change in PMI||-0.0025||-8.9566|
Understanding what drove performance
These analyses show that changes in interest rates and market and macro-economic conditions all have affected the performance of the MSCI USA Minimum Volatility Index. However, looking at longer history, and also more specifically at 2018, the market and macro-economic conditions had a higher impact than the changes in rates.
1 Alighanbari, M. and S. Sharma. (2018). “The MSCI Minimum Volatility Indexes: 10 years on.” MSCI Research Insight.
2This report may contain analysis of historical data, which may include hypothetical, backtested or simulated performance results. The analysis and observations in this report are limited solely to the period of the relevant historical data, backtest or simulation. Past performance — whether actual, backtested or simulated — is no indication or guarantee of future performance, which may differ materially.
3Market regime is determined by the trailing 12-month returns of the MSCI USA Index, while the rates regime is determined by the trailing 12-month change in yield of U.S. 10-year Treasury Note. In both cases, a particular regime is ascertained only if it lasts for at least three months (current month and the following two months), to ensure that the data remain smooth.
4The Purchasing Managers' Index (PMI) is an indicator of economic health for manufacturing and service sectors. The PMI is compiled and released monthly by the Institute for Supply Management (ISM). The PMI is based on a monthly survey sent to senior executives at more than 400 companies and based on five major survey areas: new orders, inventory levels, production, supplier deliveries and employment. A positive change in PMI indicates a growing economy whereas a negative change indicates a slowdown.
5These numbers are calculated by multiplying the coefficients from the regression by the standard deviation of the relevant parameter over the analyses, which are 1.27 and 8.07 for rates and PMI, respectively.