Why CVA Miscalculation Happens - 6 Types

1. Wrong PD Curve:
Using historical PD instead of market-implied PD from CDS. Basel requires market CDS spreads. Many emerging market banks use internal rating PD, leading to understatement.

2. Wrong-Way Risk [WWR] Ignored:
When exposure increases as counterparty credit worsens. Classic: Bank sells USD to an oil company, oil crashes, company credit worsens and exposure spikes. CVA must be adjusted for WWR. Failure = huge loss.

3. No DVA / FVA Separation:

Entrepreneurs must understand that changes in their own credit risk (DVA) can also affect earnings, as seen in case of JPMorgan Chase in 2012, where a significant DVA gain was reported due to bank's widened credit spreads

Banks netting DVA gain against CVA loss illegally inflates P&L.

4. Collateral and Netting Not Modeled:
If CSA with daily margin, EE drops drastically. Miscalculation is applying uncollateralized EE to collateralized trade.

5. CVA Hedge Mismatch - Regulatory vs Accounting:
Biggest source of loss. Bank buys CDS to hedge CVA for capital, but accounting does not recognize CDS hedge for P&L, creating volatility.

6. Data Gap for Illiquid Counterparties:

Traders shocked by $712m CVA loss at StanChart... StanChart's predominantly emerging markets portfolio lacked the data that would have allowed it to change its approach to CVA

Standard Chartered had no observable CDS for its emerging market clients, used proxy spreads that were too tight.

Landmark Cases / Enforcement

Case 1: Deutsche Bank €94m CVA Mismatch Loss 2013

Deutsche Bank lost €94 million in the first half of 2013 on CDS positions that were being used to reduce Basel III capital requirements – the result of a disparity in the way regulators and accountants view the hedges

What happened: DB bought CDS to reduce CVA RWA under Basel III. Under accounting IFRS 13, CDS hedge was not designated as CVA hedge, so loss on CDS went to P&L, but CVA RWA benefit went to capital only. Result: €110m DVA charge, €149m CVA charge reported in Q4 2013 results. Case taught banks to align regulatory CVA hedge with accounting CVA desk.

Case 2: Standard Chartered $712m CVA Revaluation Loss 2012

Derivatives traders have been shocked by a $712 million revaluation loss at Standard Chartered after the bank changed the way it calculates counterparty risk – a switch that lagged the rest of the industry by as much as a decade

Bank finally moved from historical to market-implied CVA. Emerging market portfolio had no CDS quotes, had to use proxy + add-on. One-time hit $712m. Lesson: Using stale curves = miscalculation.

Case 3: JPMorgan London Whale 2012 - CVA / DVA Manipulation

JPMorgan's Chief Investment Office misvalued CVA on CDX tranches. Internal models used overly optimistic correlation. DVA gain of $1.5bn was reported as profit when bank's own credit spread widened. SEC fined $200m, required restatement. Principle: DVA gain cannot be monetized, CVA must be unilateral first.

Case 4: Basel II to Basel III Transition - Regulatory Case

Under Basel II, the risk of counterparty default and credit migration risk were addressed but mark-to-market losses due to CVA were not. During the financial crisis, roughly two-thirds of losses attributed to counterparty credit risk were due to CVA losses and only about one-third were due to actual defaults

This is why BCBS introduced CVA capital charge in Basel III. Banks that did not calculate CVA VaR were forced to hold extra capital - CVA risk = 10x CCR default risk in crisis.

Case 5: ISDA Disputes - CVA Close-Out Cases UK

Lehman Brothers International v. ExxonMobil EWHC: Court held CVA must be included in close-out amount calculation under ISDA. Exxon failed to include CVA in termination valuation, underpaid Lehman by $12m. Court: Fair market value includes CVA.[2016]

CFH Clearing v. Merrill Lynch UK Commercial Court: Claim of CVA miscalculation on FX options - client argued bank inflated CVA by using own credit spread. Court held bank must disclose CVA methodology under MIFID best execution.[2020]

India - RBI Context:

RBI requires CVA calculation for banks under Basel III standardized approach - CVA RWA = 10 x CVA VaR. Indian banks mostly use standardized CVA because CDS market illiquid. Miscalculation often found in inspection:

• Using sovereign PD for corporate counterparty [understates CVA]
• Not including CVA for client-cleared trades
• Failing to update EE after large market moves

If your CVA desk shows large P&L day-on-day volatility, 90% of times it is EE vs PD mismatch.

How to Audit Your CVA Model:

1. Check PD source - Is it Markit CDS or internal rating? Market PD > Historical PD = under-provision.
2. Check netting set - Is exposure aggregated per ISDA netting or trade by trade?
3. Check collateral - Does EE drop to zero after margin? If not, model wrong.
4. Check WWR flag - Any commodity / FX trades with emerging market counterparty flagged as WWR?