[Library](https://stochastly.com/library) / [Risk and portfolio](https://stochastly.com/library/risk-and-portfolio)

# Risk Parity (Equal Risk Contribution)

Does equal risk contribution prevent concentration in a portfolio?

Estimate each asset's contribution to portfolio risk under a stated covariance matrix and choose weights that equalize those ex ante contributions. Recheck them after correlations change; equal estimated contributions do not ensure equal realized losses or superior returns.

Equal risk contribution is an ex ante allocation construction, not a guaranteed return improvement.

Volatility per point for USDJPY, AUDUSD and XAUUSD, with bubble area for each one's P&L contribution; XAUUSD is the most volatile.

## Evidence map

| Aspect | Finding
| Idea | Allocate so each asset contributes the same amount to total portfolio risk, rather than the same capital.
| Key result / formula | Total risk splits additively (Euler's theorem): sigma_p = sum_i RC_i, with risk contribution RC_i = w_i*(Sigma w)_i / sigma_p and marginal risk MRC_i = d sigma_p / d w_i.
| Practical rule | Use ERC/inverse-vol when you trust covariances but not expected returns; add modest leverage to reach a target return (levered risk parity).

## Idea

This stops the highest-vol asset (usually equities) from dominating portfolio risk and maximizes ex-ante risk diversification.

## Key result / formula

Equal-Risk-Contribution (ERC) sets w_i*(Sigma w)_i equal for all i. For uncorrelated assets this reduces to inverse-volatility weighting, w_i proportional to 1/sigma_i. Maillard, Roncalli & Teiletche (2010) prove the ERC portfolio's volatility lies between the minimum-variance and 1/N portfolios, and it requires no expected-return input.

## Practical rule

Weakness: it can over-weight low-vol/crowded assets (bonds) and ignores valuation — pair with a vol target and leverage limits.

## Worked weights

For two uncorrelated assets with annual volatilities of 10% and 20%, inverse-volatility weights are proportional to 1/0.10 and 1/0.20, giving about 67% and 33%. Their standalone volatility contributions then match under the zero-correlation assumption. If their correlation changes, equal risk contributions require the full covariance matrix; the simple inverse-vol rule is only a special case. Re-estimate covariance on past data at each rebalance, calculate realized risk contributions and include the turnover cost. Leveraging the lower-volatility sleeve to meet a return target increases financing and liquidation risk. Risk parity equalizes estimated risk, not economic valuation or expected returns.

## Additional check

For the two-asset example, the 67/33 weights produce equal approximate risk contributions only with zero correlation; report the covariance estimate and both realized contributions after each rebalance.

## Source

Maillard, Roncalli & Teiletche, "The Properties of Equally-Weighted Risk Contribution Portfolios," Journal of Portfolio Management 36(4) 2010, SSRN 1271972; Qian, "Risk Parity Portfolios" (PanAgora, 2005). [Primary source](https://papers.ssrn.com/sol3/papers.cfm?abstract_id=1271972)

## Related in the library

[Betting Against Beta: Leverage Aversion as a Premium](https://stochastly.com/library/risk-and-portfolio/betting-against-beta)

[Extreme Value Theory for Tail Risk](https://stochastly.com/library/statistics/extreme-value-theory-for-tails)

[How the assistant cites the library](https://stochastly.com/product/assistant) · [The checks behind a verdict](https://stochastly.com/product/guards)

---
Source: https://stochastly.com/library/risk-and-portfolio/risk-parity
