Risk Parity: Why Equal Dollars Are Not Equal Risk

Risk parity is a way of deciding how much of each asset to hold so that every part of a portfolio contributes a similar share of the risk, instead of a similar share of the money. It starts from one observation: equal amounts of money are not equal amounts of risk. A dollar in commodities swings far more than a dollar in Treasuries, so a portfolio split evenly by dollars ends up dominated by its most volatile holding.

Choosing which asset classes to own is only half the job. The weights decide what a portfolio really is, and that is where risk-based weighting comes in.

The 60/40 portfolio: 60% of the money, 86% of the risk

The classic 60/40 mix of stocks and bonds looks balanced. Measure where its ups and downs actually come from, though, and the picture changes.

Using weekly total returns for SPY (stocks) and TLT (long-term Treasuries) from March 2007 to June 2026, the lesson's historical analysis found:

• Stocks: about 86% of the portfolio's risk
• Bonds: about 14% of the portfolio's risk

Two things explain the gap. Bonds swung less than stocks, and over this period they often moved in the opposite direction. So in risk terms, a 60/40 portfolio was mostly a stock portfolio with a cushion attached. Past performance is not a guide to the future, but the pattern shows why a split by money says little about a split by risk.

Equal dollars vs equal risk: how risk parity changes the mix

The lesson's second example takes $100,000 and spreads it across four asset classes: stocks (SPY), Treasuries (TLT), gold (GLD) and commodities (DBO). The figures below are a hypothetical illustration based on weekly total returns from March 2007 to June 2026, not a recommendation.

Equal dollars, $25,000 each:

• Stocks: 24% of the risk
• Treasuries: 2% of the risk
• Gold: 21% of the risk
• Commodities: 52% of the risk

With equal money, commodities, the most volatile of the four, carried more than half of all the risk, while Treasuries barely registered.

Weighted by risk (inverse-volatility weights):

• Stocks: $25,337, carrying 27% of the risk
• Treasuries: $32,348, carrying 14% of the risk
• Gold: $27,080, carrying 32% of the risk
• Commodities: $15,235, carrying 28% of the risk

The calm asset gets more money (Treasuries rise to roughly 32% of the capital) and the wild one gets less (commodities drop to roughly 15%). The result: no single asset class carries more than about a third of the risk. Inverse-volatility weighting is the most basic form of the idea: divide 1 by each asset's volatility, then scale the results so they add up to 100%.

Going further: correlation-aware weights

Volatility tells you how much an asset swings on its own. It says nothing about how it moves alongside everything else. More advanced versions of risk-based weighting also account for correlation.

The logic works like this:

• Assets that move almost as one share a budget. The S&P 500 and the Nasdaq-100 tend to rise and fall together, so holding both adds less diversification than two separate allocations suggest. A correlation-aware approach treats them as one risk, not two.
• Assets that move differently earn more room. A holding that zigs when the rest zags lowers the risk of the whole portfolio, so it can justify a larger weight than its volatility alone would give it.

This matters because a portfolio of five markets that all react to the same news is, in practice, much less diversified than it looks.

Guard rails for any weighting scheme

Estimates of volatility and correlation are never exact. They come from past data, and markets change. A weighting method without limits can push half the money into one place on the strength of a noisy estimate. Sensible portfolios set guard rails.

Caps:

• A maximum weight per market
• A maximum weight per family of assets (for example, all equities together)
• A small, fixed cap for very volatile assets such as crypto

Checks:

• Re-estimate risk and correlation regularly
• Apply rebalancing on a fixed schedule rather than on impulse
• Never borrow to raise exposure, because leverage magnifies every estimation error

On top of the weights, some investors set an overall risk level, a volatility target, that determines how large the whole portfolio's exposure is. The weights decide how risk is shared. The target decides how much risk there is in total.

Rough weights often beat precise guesses

None of this requires a computer. The lesson's practical conclusion is that rough risk-based weights capture most of the benefit:

• Round weights to the nearest 5%
• Cap each asset family
• Check once a year

Precise weights built on precise forecasts usually do worse than rough weights built on honest estimates. The forecasts are the weak link, and extra decimal places only hide how uncertain they are.

What to do with this

1. List your holdings with their volatility. Annual volatility figures are widely published for most funds and indices. 2. Find what dominates your risk. Multiply each weight by its volatility and compare. The largest number is your main risk driver, which may not be your largest holding. 3. Draft an inverse-volatility version. Divide 1 by each volatility, scale to 100%, round to the nearest 5%, and compare it with your current mix. 4. Group holdings that move together. If two funds track similar markets, consider treating them as one budget. 5. Write down your caps and a review date. A maximum per family and a yearly check keep the weights from drifting into concentration.

This article is based on Lesson 9.3 of our course Mastering Investments, where Module 1 is free.

Frequently asked questions