Low Volatility Funds: Lost Decade or Flawed Design?
Long-only vs long-short low volatility strategies
February 2026. Reading Time: 10 Minutes. Author: Nicolas Rabener.
SUMMARY
- Low-volatility funds have generated poor absolute and risk-adjusted returns recently
- However, the long-term evidence still supports the defensive use case
- Significant difference between long-short and long-only low-volatility strategies
INTRODUCTION
Low-volatility strategies are relatively easy to market. There is almost always some economic or market risk that appears to justify reduced equity exposure. Adding to their appeal, academic research has long documented that low-risk stocks have outperformed the broader equity market on a risk-adjusted basis. On the surface, the case seems compelling.
Yet the experience during the COVID-19 crisis in 2020 challenged this narrative. Invesco’s S&P 500 Low Volatility ETF (SPLV) – one of the largest low-volatility ETFs by assets under management – experienced a slightly larger maximum drawdown than the S&P 500 itself, surprising and disappointing many investors expecting defensive characteristics. The fund’s significant exposure to traditionally anti-cyclical sectors, such as real estate, proved problematic as the pandemic abruptly altered consumer behavior and workplace dynamics, turning perceived low-risk sectors into sources of elevated risk.
In the years since, investor skepticism toward low-volatility strategies has increased. But is this skepticism justified? In this research article, we examine whether low-volatility strategies have merely suffered through a poor decade – or whether recent outcomes point to deeper structural shortcomings.
THE CASE FOR LOW VOLATILITY STRATEGIES
A substantial body of academic research documents that low-risk stocks tend to outperform high-risk stocks, with risk most commonly measured by a stock’s volatility or its beta relative to the broader equity market. AQR provides long-term evidence for this phenomenon across the United States, Europe, and the Pacific region, with data extending back to 1989. Their approach constructs portfolios that are long low-beta stocks and short high-beta stocks.
To isolate the low-risk effect, these portfolios are made beta-neutral by leveraging the long positions and deleveraging the short positions. The resulting long-s

