There is a mathematical truth about volatile investments that most investors never learn, and that the financial industry rarely explains clearly: volatility itself destroys wealth, independently of the direction of returns. This effect — called volatility drag, variance drain, or the volatility tax — means that two portfolios with identical average annual returns can produce dramatically different terminal wealth, depending on how smooth or volatile those returns are. It is one of the most counterintuitive and important concepts in portfolio mathematics.
The arithmetic of loss and recovery
The clearest way to see volatility drag is through the asymmetry of losses and gains. If you lose 50%, you need a 100% gain just to break even. If you lose 20%, you need a 25% gain to recover. If you lose 10%, you need an 11.1% gain. The gain required to recover from a loss is always larger than the loss itself — and this asymmetry means that volatile portfolios compound worse than smooth ones with the same arithmetic average return.
The formula and its implications
The relationship between arithmetic and geometric returns is approximately: Geometric return ≈ Arithmetic return − σ²/2, where σ² is the variance of returns. For a portfolio with 10% arithmetic return and 30% volatility: geometric return ≈ 10% − (30%²)/2 = 10% − 4.5% = 5.5%. The difference (4.5%) is the volatility drag. This explains why leveraged ETFs that rebalance daily consistently underperform their stated leverage multiple over longer periods: daily rebalancing into volatile assets accumulates volatility drag rapidly. A 2× leveraged S&P 500 ETF does not return twice the S&P 500 over five years — it returns significantly less, because the volatility drag compounds against you each day.
Volatility drag and portfolio construction
Volatility drag has direct implications for portfolio construction. Reducing volatility — through diversification, hedging, or simply holding less risky assets — increases the geometric return of the portfolio relative to its arithmetic return. Two portfolios with the same arithmetic expected return but different volatilities will produce different long-term wealth outcomes; the lower-volatility portfolio always compounding to a higher terminal value. This is sometimes cited as a mathematical justification for diversification beyond simply risk reduction: a diversified portfolio with lower correlation between assets reduces overall volatility, and therefore reduces volatility drag, even holding the arithmetic expected return constant. The geometric return is what actually matters to long-term investors — and it is always lower than the arithmetic return when returns are volatile.