Are Software Returns Really That Good?

At the start of a decade, bloggers often reflect on the best and worst predictions of the past one. Perhaps one of the most prescient claims was Marc Andreessen’s “software is eating the world” in 2011. Since then, innovation in software has transformed the enterprise. Entire new markets and job titles have been created. A vibrant support ecosystem has emerged, replete with tailored metrics and industry conferences. Investors continue to fund fast-growing, loss-making companies at lofty valuations. And countless fortunes have been minted in the process.

Public software equity returns tell an analogous story. Consider an equal-weight portfolio of 74 software stocks. From April 2015–2020, the software portfolio returned 210%, outperforming the S&P 500, tech-heavy NASDAQ, (and the average public or private equity fund 😉).

But making a judgment based only on return is an incomplete assessment of the quality of an investment strategy (you wouldn’t conclude that betting a single number on a roulette spin is optimal just because the potential reward is highest). Return is invariably related to risk, but this trade-off seems under-explored in the public discourse on software stocks, which is dominated by revenue multiples and metrics benchmarks. Perhaps on a risk-adjusted basis, software returns aren’t as good as they seem?

Beta is risk metric used to gauge how an asset moves in relation to a market benchmark. A stock with a beta of 2 is expected to move in the same direction as the market but with twice the magnitude. (Mathematically, beta is the covariance of an asset and the market benchmark, divided by the variance of the market benchmark.)

Calculating beta using the Nasdaq as the benchmark (beta of 1 by definition), the software portfolio has a higher beta of 1.5 and the S&P 500 has a lower beta of 0.82. This is an intuitive result. You’re compensated with higher returns for holding the riskier software portfolio. And conversely, you’ll suffer more during market contractions — from March 2–16, the software portfolio plunged 30% while the Nasdaq and S&P 500 both fell ~22%.

But software stocks outperform even after accounting for their higher risk quotient. The software portfolio generates beta-adjusted returns of 140% over 5 years, compared to 36% for the S&P and 89% for Nasdaq. You were disproportionately rewarded for investing in the higher beta software portfolio.

Another measure of risk is the Sharpe ratio, which grades an investment’s total return relative to its consistency of return (i.e. standard deviation). A higher Sharpe ratio is preferred because it reflects our desire to own investments that generate high returns while minimizing large, stress-inducing price swings. (Mathematically, the Sharpe ratio is the excess return a portfolio generates above the risk-free rate, divided by the standard deviation of the portfolio’s returns.)

Calculating the Sharpe ratio over 5 years, the software portfolio is again the clear winner. Its exceptional annualized return of 25% more than offsets its high standard deviation, resulting in a Sharpe ratio of 0.88, 60% above the Nasdaq’s.

As COVID-related movements in March and April materially impacted stock returns and standard deviations, we also calculated a “peacetime” Sharpe ratio excluding those trading days. Lower standard deviations and higher returns boost Sharpe ratios across the board, but the software portfolio still reigns supreme with a Sharpe ratio of 1.07, 55% above the Nasdaq’s.

While the software portfolio’s Sharpe ratio is far from exceptional compared to a finely-tuned professional trading strategy, keep in mind that it’s simply an equal-weight basket of stocks that requires limited active management and no fees. And basic but intelligent asset allocation within the basket could further boost performance (our next post).

Or in other words: yes, public software returns really are that good