Two-Factor Hull-White Only Captures Yield-Curve De-Correlation Under Specific Conditions

If your portfolio mixes swaps with different maturities, the way rates move together can change the price of credit risk. That is the issue this paper tackles through credit valuation adjustment, or CVA, which prices counterparty risk using the expected positive value of a netting set over time. The author studies the two-factor Hull-White interest rate model, a simple Gaussian model often used because XVA calculations stay tractable only with that kind of structure. By comparing an approximation formula with Monte Carlo simulation, the paper examines the correlation of co-initial swap rates and asks when the model really reproduces yield-curve de-correlation. The answer is precise: the two-factor Hull-White model captures that de-correlation only when its volatilities and mean-reversion strengths satisfy certain relationships. That matters because the exposure in a netting set can depend on correlation, including cases with payer and receiver swaps and CMS spread options. In short, the paper shows that this popular model is useful for XVA only in the parameter regimes where it truly matches how the yield curve spreads apart.

A 20-year payer swap and a 10-year receiver swap can tell very different stories. That is the puzzle behind this paper. In credit valuation adjustment, or CVA, the expected positive value of a portfolio is fed into the risk charge for counterparty default. If rates on the yield curve move together, or drift apart, the exposure can change a lot. That matters even for plain swap books. It matters again for cross-currency XVA, where XVA means the family of valuation adjustments used for funding and credit risk. The surprise here is simple. A two-factor Hull-White model does not always reproduce that de-correlation. It does so only when its volatility and mean-reversion settings line up in certain ways.

When a swap book starts to depend on correlation

The paper focuses on co-initial swap rates, which are swap rates starting at the same time but with different maturities. That is where the hidden stress shows up. A 20-year payer swap and a 10-year receiver swap can partly offset each other, but only if the rate curve does not move in lockstep. The same logic applies to CMS spread options, which are bets on the gap between two swap rates. The abstract says this de-correlation can be important even for a swap portfolio. It also says the two-factor Hull-White model is only effective for XVA when it captures that pattern well. So the real question is not whether the model is elegant. It is whether it can mimic the way long and short rate bets pull apart.

How the paper tests the model

The study uses two tools side by side. One is an approximation formula, which is a fast math shortcut for estimating swap-rate links. The other is Monte Carlo simulation, a repeat of many random trials, which acts like a slower but clearer check. The paper compares both for the correlation of co-initial swap rates. That lets it see when the two-factor Hull-White setup follows the same curve as the simulated result. The model has two sources of rate movement. It also has mean reversion, which is the pull that pushes rates back toward a long-run level. The key test is whether those parts produce the right amount of separation across the yield curve.

Why this matters for XVA desks

For a desk that prices counterparty risk, the lesson is practical. A simple Gaussian rate model can keep XVA tractable. That is one reason the Hull-White family is so popular. But simplicity only helps if the curve shape is still believable. If the model misses de-correlation, it can misread exposure in portfolios that mix payer and receiver swaps or include spread-sensitive trades. The paper’s message is not that two-factor Hull-White is wrong. It is that the model earns its keep only in the parameter regions where its built-in rate moves really separate the yield curve in the right way. Outside those regions, the exposure story can become too smooth.

What this points to next

The next test is not a bigger slogan. It is a sharper calibration check. A two-factor Hull-White setup should be checked against the swap mix that sits inside a real netting set, not just against a generic curve shape. That means watching how payer and receiver swaps, plus spread-driven trades, react as volatilities and mean reversion shift together. The paper’s core result gives a clear filter. If those parameter links are wrong, the model may still run, but it will not tell the right exposure story. If they are right, the model stays simple and still tracks the curve split that CVA needs.