Home / Wiki / C / Credit Valuation Adjustment (CVA)

Credit Valuation Adjustment (CVA)

Economics Markets and Financial TheoryUpdated 16 Nov 2026

Credit valuation adjustment (CVA) is the amount by which the value of a derivative, or a portfolio of derivatives with one counterparty, is reduced to reflect the chance that the counterparty will default while the position is in the firm's favour. A swap valued without regard to credit risk assumes every payment will be made. CVA is the market price of the possibility that some will not, and it equals the expected loss from the counterparty's default, discounted to today.

CVA is the best-known member of the family of derivative valuation adjustments (XVA). It is paired with the debit valuation adjustment (DVA), which reflects the firm's own default risk. Banks report CVA in their accounts under fair value rules and hold regulatory capital against changes in it, and most large dealers run a dedicated desk that prices and hedges it for OTC derivatives. Under the original Basel III rules, banks with approved internal models calculated that capital charge with the advanced credit valuation adjustment (A-CVA) method, later replaced by the SA-CVA and BA-CVA approaches.

01

Formula

The standard unilateral formula divides the life of the portfolio into time steps and sums the expected loss in each:

CVA≈LGD×∑i=1nEE(ti)×PD(ti−1,ti)×DF(ti)\text{CVA} \approx \text{LGD} \times \sum_{i=1}^{n} \text{EE}(t_i) \times \text{PD}(t_{i-1}, t_i) \times \text{DF}(t_i)
  • LGD\text{LGD} is the loss given default (LGD), the share of the exposure lost if the counterparty defaults, equal to one minus the recovery rate.
  • EE(ti)\text{EE}(t_i) is the expected exposure at time tit_i: the average of the portfolio's positive values across simulated market scenarios, after netting and collateral.
  • PD(ti−1,ti)\text{PD}(t_{i-1}, t_i) is the probability of default (PD) of the counterparty between ti−1t_{i-1} and tit_i, usually implied from market prices.
  • DF(ti)\text{DF}(t_i) is the risk-free discount factor to tit_i.

Default probabilities are normally taken from the counterparty's credit default swap (CDS) spread, or a proxy spread where none trades. A rough approximation for the annual default intensity is λ≈s÷LGD\lambda \approx s \div \text{LGD}, so a credit spread of 1.20% with an LGD of 60% implies about 2% a year. Expected exposure comes from Monte Carlo simulation of the whole netting set.

02

Worked example

Take a swap with three annual time steps, an LGD of 60% and a default probability of 1% in each year. Expected exposures are 10 million, 8 million and 5 million, and discount factors are 0.97, 0.94 and 0.91. The three terms are 10×0.01×0.97=0.097010 \times 0.01 \times 0.97 = 0.0970, 8×0.01×0.94=0.07528 \times 0.01 \times 0.94 = 0.0752 and 5×0.01×0.91=0.04555 \times 0.01 \times 0.91 = 0.0455 million, which sum to 0.2177 million. Multiplying by the LGD gives a CVA of 0.60×0.2177=0.13060.60 \times 0.2177 = 0.1306 million, or about 130,600. The fair value of the swap is its risk-free value less this amount.

03

Accounting treatment

Under IFRS 13, fair value is the price at which an asset would be sold to a market participant, and that price reflects the counterparty's credit risk. Banks reporting under International Financial Reporting Standards (IFRS) therefore include CVA in the carrying value of derivative assets, and US GAAP has an equivalent requirement. Changes in CVA flow through profit or loss, so a widening of counterparty spreads produces a loss even if no counterparty defaults. This is a core part of fair value accounting for derivatives.

04

Regulatory capital for CVA risk

During the 2007 to 2009 crisis, the Basel Committee found that most counterparty credit losses on derivatives came from falls in the market value of exposures, through CVA, rather than from actual defaults. Basel III therefore introduced a separate capital charge for CVA risk, on top of the default charge for counterparty risk. The final Basel III reforms replaced it with a revised framework offering a standardised approach, based on sensitivities to credit spreads and market risk factors, and a simpler basic approach, with implementation set by the Basel Committee from 1 January 2023 and timing varying by jurisdiction. Trades with qualifying central counterparties are excluded, and the EU exempts some counterparties, such as non-financial companies, from the charge.

05

Managing CVA

Firms reduce CVA by legal and contractual means: close-out netting lowers exposure to the net amount, and a credit support annex (CSA) requiring daily margin reduces it further. Central clearing replaces the bilateral exposure with one to a clearing house. CVA desks hedge the remaining risk with single-name and index CDS and with interest rate and currency hedges for the exposure itself. A particular concern is wrong-way risk, where exposure rises as the counterparty's credit worsens, as with a currency trade against a bank in the same country; standard models that treat exposure and default as independent understate CVA in such cases.

CVA also enters pricing. When a dealer quotes a new trade, it calculates the incremental CVA, the change in the counterparty's total CVA caused by adding the trade to the existing netting set, and includes it in the price. A trade that offsets existing exposure can reduce CVA and so be priced more keenly than the same trade with a new counterparty.

Last updated 16 Nov 2026