Executive summary
- Recovery from a drawdown requires a larger percentage gain than the loss itself, and the gap widens sharply beyond 20 percent.
- Volatility targeting reduces drawdowns caused by volatility clustering but not those caused by sudden gaps.
- Explicit tail hedges cost carry in every quiet period and must be budgeted like an insurance premium.
- Diversification across genuinely different risk premia is the cheapest form of tail protection, and the first to fail under correlated stress.
- Drawdown limits should be set against investor tolerance and financing terms, not against statistical comfort.
The arithmetic of recovery
The gain required to recover from a loss grows faster than the loss. A 10 percent drawdown needs an 11.1 percent gain; 25 percent needs 33.3 percent; 50 percent needs 100 percent. For a fund charging performance fees above a high-water mark, a deep drawdown also removes fee income for years, which affects staff retention long before it affects returns.
g = d / (1 − d)
- d
- — drawdown expressed as a decimal fraction of peak value
- g
- — gain required on the reduced capital to return to the prior peak
| Drawdown | Required gain | Years at 8% net to recover |
|---|---|---|
| 5% | 5.3% | 0.7 |
| 10% | 11.1% | 1.4 |
| 20% | 25.0% | 2.9 |
| 35% | 53.8% | 5.6 |
| 50% | 100.0% | 9.0 |
Volatility targeting: what it fixes and what it does not
Volatility targeting scales exposure inversely to a recent volatility estimate, so that risk taken remains roughly constant. Because volatility clusters, the approach systematically reduces exposure into turbulent periods and has historically improved risk-adjusted returns for trend and equity-like strategies.
w_t = σ_target / σ̂_t , capped at w_max
- w_t
- — scaling applied to the base portfolio at time t
- σ_target
- — the volatility the fund intends to run
- σ̂_t
- — estimated volatility from the recent window
- w_max
- — a hard cap preventing extreme leverage when volatility is low
The limitations are structural. The estimator is backward looking, so a gap move is taken at full exposure. Low measured volatility mechanically raises leverage, which is exactly the condition preceding many historical dislocations. And the scaling itself generates turnover, which costs money and, when many funds scale simultaneously, contributes to the move being hedged against.
Explicit tail hedging and the convexity budget
A tail hedge is a position that pays disproportionately in a severe move: deep out-of-the-money puts, payer swaptions, credit protection, or long-volatility strategies. Its defining property is negative expected return in most periods, which means it must be budgeted explicitly rather than judged on standalone performance.
- Set the annual convexity budget as a fixed percentage of net asset value, typically between 0.3 and 1.0 percent.
- Define the payoff target: the loss level at which the hedge should offset a stated share of portfolio loss.
- Measure the hedge on portfolio impact in the stressed scenario, never on standalone profit and loss.
- Review basis risk: the hedged index may not move with the portfolio in the scenario that matters.
- Pre-agree the monetisation rule, because an unmonetised hedge that round-trips has cost carry for nothing.
Diversification as first-line defence
Combining strategies with genuinely distinct return drivers reduces drawdown without an ongoing carry cost, which makes it the most efficient starting point. The qualification is that correlations estimated in calm periods understate co-movement in stress, and that leverage links otherwise unrelated strategies through a shared financing channel.
Practical portfolio construction therefore combines the three: diversify first, target volatility to stabilise risk taken, and purchase explicit convexity for the residual gap risk that neither of the first two addresses.
Practical example
A multi-strategy fund runs at a 10 percent volatility target with a 15 percent drawdown limit. Historical simulation suggests a 1-in-20-year loss of 22 percent, breaching the limit. The team allocates a 0.6 percent annual convexity budget to twelve-month put spreads struck 15 and 30 percent below spot on the dominant equity exposure.
In the stressed scenario the hedge returns approximately 5.5 percent of net asset value, reducing the modelled loss from 22 to 16.5 percent, and volatility scaling removes a further two points of exposure through the path. The residual 14.5 percent sits inside the limit. The cost in a quiet year is 0.6 percent of return — an explicit, board-approved price for keeping the fund inside its own constraint.
Limitations and caveats
- Drawdown statistics from a single historical path are a weak guide to the distribution of future paths.
- Option-based hedges carry basis risk and their cost varies sharply with the prevailing volatility surface.
- Volatility targeting can increase losses in sharp mean-reverting moves by de-risking at the low point.
- Correlation-based diversification benefits are estimated with error and are least reliable in stress.
Conclusion
Drawdown control is a portfolio-construction problem, not a trading decision taken during the loss. The three available instruments — diversification, volatility scaling and explicit convexity — address different failure modes and are complements rather than alternatives.
The discipline that matters is budgeting the cost of protection in advance and holding to it through the quiet years when the protection appears to be waste.
References
- Volatility Managed Portfolios — Journal of Finance, 2017
- Quantitative Risk Management: Concepts, Techniques and Tools — Princeton University Press, 2015
- Minimum Capital Requirements for Market Risk — Basel Committee on Banking Supervision
Author bio
Jonas Mohamed Osman Abdelghafour, known as Yonas Osman is an actuary, FRM and financial risk professional specialising in banking, insurance, model risk, capital modelling and quantitative risk management.