Choice of Collateral Currency in Differential Swaps

This paper extends existing valuation frameworks for differential swaps by deriving explicit pricing and hedging strategies for contracts with collateral denominated in a currency different from the cash flows, demonstrating that such foreign-currency collateral introduces significant non-trivial valuation adjustments and risk exposures that must be incorporated into modern multi-currency models.

The Gist

Imagine you are a landlord (a bank) who rents out a house (a financial contract) to a tenant. The rent is paid in US Dollars. Usually, the tenant puts up a security deposit in US Dollars as well. This is simple: if the tenant owes money, you take the deposit; if you owe them interest on the deposit, you pay it in dollars.

But what happens if the tenant says, "I'll pay my rent in dollars, but I want to leave my security deposit in Euros in a European bank account"?

This is the core puzzle solved in the paper "Choice of Collateral Currency in Differential Swaps" by Ding, Liu, and Rutkowski.

Here is the story of the paper, broken down into simple concepts, analogies, and metaphors.

1. The Setting: The "Security Deposit" Game

In the world of finance, when two parties trade complex bets on interest rates (called Swaps), they don't just trust each other. They exchange Collateral (security deposits) daily to make sure no one runs away with the money.

• The Old Way: If you trade a US Dollar deal, you put up US Dollars as collateral. The interest you earn on that deposit is based on US rates (like SOFR).
• The New Reality: After the 2008 financial crisis, rules changed. Sometimes, a US Dollar deal is collateralized with Euros (or other foreign currencies). The deposit sits in a European bank earning European interest rates (like eSTR).

The Paper's Big Question: Does it matter if the security deposit is in a different currency than the rent?
The Answer: Yes, absolutely. It changes the price of the deal and how you protect yourself from risk.

2. The "Mixed-Flavor" Smoothie (Valuation)

Imagine you are making a smoothie.

• The Fruit (The Contract): You are buying a smoothie made of US Apples (US Dollar interest rates).
• The Blender (The Funding): Usually, you blend it with US Water.
• The Twist: Now, you are forced to blend your US Apples with European Milk (Euro interest rates) because that's what your security deposit is made of.

Even though the final drink is still "Apple Smoothie" (US Dollars), the taste (the price) changes because the liquid you blended it with is different.

• The Paper's Finding: If the European interest rate is lower than the US rate, the "European Milk" makes the smoothie cheaper to produce. This means the fixed price you agree to pay for the swap changes.
• The "Cheapest-to-Deliver" Option: Imagine the contract says, "You can put up either Dollars or Euros as a deposit." The person posting the deposit will naturally choose the currency that costs them the least (the "Cheapest-to-Deliver"). The paper shows how to calculate the price when this choice exists.

3. The "Leaky Bucket" (Hedging/Risk)

This is the most surprising part of the paper.

Imagine you are trying to keep a bucket of water (your profit/loss) perfectly full.

• Scenario A (All Dollars): You have a leak caused by US interest rates. You fix it by holding a "US Interest Rate Shield." The bucket stays full. Perfect.
• Scenario B (Euro Deposit): You still have the US interest rate leak. BUT, because your security deposit is in Euros, the bucket now has a second, tiny leak caused by the Euro interest rate and the exchange rate between the Dollar and the Euro.

The Mistake: If you only use your "US Interest Rate Shield" to fix the bucket, you will miss the second leak. Your bucket will slowly lose water (money) over time.

The Paper's Solution:
The authors show that to keep the bucket perfectly full, you need two shields:

1. A shield for US rates (SOFR futures).
2. A shield for Euro rates (eSTR futures).

Even though the contract is entirely in US Dollars, the fact that the deposit is in Euros forces you to hedge against Euro risks. If you ignore this, you leave about 5% of your risk exposed (a "leak" that is small but significant for big banks).

4. The "Time Travel" Analogy (Backward-Looking Rates)

The paper deals with "Backward-Looking" rates.

• Forward-Looking (Old Way): "I promise to pay you interest based on what the rate will be next month." (Like guessing the weather).
• Backward-Looking (New Way): "I promise to pay you interest based on what the rate actually was over the last month." (Like checking the weather report after the storm).

Because the rate is only known at the very end of the month, it's harder to predict the exact value of the deal today. The paper provides a mathematical "map" to navigate this uncertainty, especially when mixing US and European currencies.

5. Why Should You Care?

You might think, "I'm not a bank trader, why does this matter?"

1. It's Everywhere: Almost all modern financial contracts (loans, mortgages, corporate bonds) use these rules.
2. Hidden Costs: If a bank doesn't account for the "Euro deposit" effect, they might price a loan too low or too high. This leads to unexpected losses or missed profits.
3. Risk Management: If a company borrows money and doesn't realize they need to hedge against two currencies (even if they only borrow in one), they could lose money when exchange rates or foreign interest rates move unexpectedly.

Summary in One Sentence

This paper proves that where you park your security deposit matters just as much as the deal itself, and if you park it in a foreign currency, you must build a special "double-layer" shield to protect your money, or you will slowly leak cash.

The Takeaway: In the modern financial world, you can't just look at the currency of the deal; you must also look at the currency of the safety net.

Technical Summary

Here is a detailed technical summary of the paper "Choice of Collateral Currency in Differential Swaps" by Ding, Liu, and Rutkowski.

1. Problem Statement

The paper addresses a critical gap in post-LIBOR derivative pricing: the impact of collateral currency choice on the valuation and hedging of interest rate derivatives, specifically differential swaps referencing backward-looking overnight rates (e.g., SOFR).

• Context: Following the 2008 financial crisis, collateralization became standard, and the industry shifted from LIBOR to Risk-Free Rates (RFRs) like SOFR (USD) and eSTR (EUR).
• The Issue: While it is common to assume a USD product is collateralized in USD, Credit Support Annexes (CSAs) often allow for collateral in foreign currencies (e.g., a USD SOFR swap collateralized in EUR cash remunerated at eSTR).
• The Core Question: How does the choice of a foreign collateral currency affect the valuation and risk management of a contract whose cash flows are entirely denominated in the domestic currency?
• Key Insight: The authors argue that even when contractual cash flows are domestic, foreign collateralization introduces a non-trivial stochastic risk exposure via the effective funding rate and discounting mechanism. Ignoring this leads to valuation errors and incomplete hedging (P&L leakage).

2. Methodology

The authors develop a rigorous mathematical framework combining collateralized trading theory with a Gaussian cross-currency term structure model.

A. Market Setup and Trading Strategies

• Two-Currency Market: Domestic (USD/SOFR) and Foreign (EUR/eSTR).
• Proportional Collateralization: The model assumes a fraction $\beta \in [0, 1]$ of the mark-to-market value is collateralized (earning the collateral rate $r_c$), while the remainder $(1-\beta)$ is uncollateralized (funded at the hedger's rate $r_h$).
• Effective Funding Rate ($r_\beta$): The wealth dynamics are governed by an interpolated rate:
  $$r_\beta = (1-\beta)r_h + \beta r_c$$
• Futures-Based Replication: The authors utilize exchange-traded futures (SOFR and eSTR futures) as the primary hedging instruments. They define a self-financing strategy where the gains from foreign futures are converted to domestic currency, accounting for the quadratic covariation between the FX rate and the foreign futures process (due to daily settlement).

B. Pricing Measure and Model

• Pricing Martingale Measure (PMM): They establish a pricing measure $\tilde{\mathbb{Q}}$ associated with the effective funding rate $r_\beta$. Under a specific Gaussian specification (Vasicek dynamics for rates and Garman-Kohlhagen for FX), they prove that $\tilde{\mathbb{Q}}$ coincides with the domestic martingale measure $\mathbb{Q}$.
• Term Structure Model:
  • Domestic ($r_d$) and Foreign ($r_f$) rates follow Vasicek dynamics.
  • FX rate ($Q$) follows a stochastic differential equation driven by correlated Brownian motions.
  • This allows for closed-form solutions involving deterministic "convexity corrections" and correlation adjustments.

C. Valuation Framework

The paper derives explicit pricing formulas for Differential Swaps (payoff: $\delta(R_d - \gamma R_f - \kappa)$).

• Decomposition: The payoff is decomposed into domestic floating, foreign floating, and fixed legs.
• Convexity Corrections: Using Lemma 4.1, the authors derive factorization identities for conditional expectations of exponentials of integrated rates. These include:
  • Self-adjustments (volatility).
  • Maturity adjustments (different time ranges).
  • Correlation adjustments (between domestic and foreign rates).
  • Drift adjustments (measure changes).
• Key Result: The price of a USD-referencing swap with foreign collateral depends on foreign rate futures and FX dynamics, even if $\gamma=0$ (purely domestic payoff).

3. Key Contributions

1. Explicit Valuation under Foreign Collateral:
   The paper provides the first closed-form pricing formulas for USD-referencing swaps under proportional foreign collateralization. It demonstrates that the price is not merely a discounted cash flow but involves a complex interaction between domestic rates, foreign rates, and FX volatility through the effective discount rate $r_\beta$.

2. Identification of Hidden Risk Exposure:
   The authors prove that foreign-currency collateral introduces a systematic risk exposure even for purely domestic payoffs.

   • When $\beta > 0$, the swap value is driven by the foreign discount factor.
   • Hedging solely with domestic SOFR futures leaves a residual exposure to the foreign rate and FX risk.

3. Futures-Based Hedging Strategy:
   They derive explicit hedge ratios ($\phi_d, \phi_f$) for both domestic (SOFR) and foreign (eSTR) futures.

   • The hedge ratio for foreign futures is non-zero even when the payoff is purely domestic, provided $\beta > 0$.
   • The strategy is shown to be a perfect replication in the continuous-time limit.

4. Cheapest-to-Deliver (CTD) Interpretation:
   The paper formalizes the "cheapest-to-deliver" option inherent in multi-currency CSAs. It shows that the choice of collateral currency acts as an option that shifts the par swap rate, creating a spread between USD-collateralized and EUR-collateralized valuations.

4. Results and Numerical Findings

The authors conduct numerical experiments using synthetic parameters representative of USD and EUR markets.

• Valuation Impact (Par Rates):

  • Foreign collateralization shifts the par swap rate curve.
  • In their baseline (where the foreign rate is lower than the domestic rate), increasing the collateralization level $\beta$ increases the par swap rate.
  • The effect is maturity-dependent: for a 10-year swap with full foreign collateralization ($\beta=1$), the par rate shifts by approximately 7.15 basis points compared to the uncollateralized benchmark.
  • Sensitivity analysis shows the shift is most driven by FX volatility and the correlation between the domestic rate and FX.

• Hedging Effectiveness:

  • Domestic-Only Hedging: Hedging a foreign-collateralized swap using only domestic SOFR futures eliminates the majority of risk but leaves a systematic residual error.
  • Quantitative Impact: In their baseline scenario ($T=5$ years, $\beta=1$), domestic-only hedging leaves approximately 5% of the terminal variance unhedged.
  • Full Hedging: Including eSTR futures in the hedge portfolio reduces the residual variance to negligible levels, confirming that the foreign collateral introduces a distinct, hedgeable risk factor.

5. Significance and Implications

• Pricing Accuracy: The study demonstrates that using a purely domestic discount curve (OIS) for derivatives collateralized in a foreign currency leads to systematic pricing biases. Modern multi-currency modeling frameworks must incorporate the effective funding rate $r_\beta$.
• Risk Management: Risk managers cannot rely solely on domestic rate hedges for collateralized products if the collateral currency differs from the payoff currency. Failure to hedge the foreign rate/FX component induced by collateralization results in P&L leakage.
• Post-LIBOR Relevance: As markets transition to RFRs (SOFR, eSTR), the standardization of collateralized trading makes these cross-currency effects more prevalent. The paper provides the necessary tools to price and hedge these instruments accurately.
• CTD Mechanism: The work clarifies how the "cheapest-to-deliver" option in CSAs translates into a quantifiable spread in par rates, offering a transparent metric for traders to assess the value of collateral flexibility.

In summary, Ding, Liu, and Rutkowski provide a mathematically rigorous and practically significant framework showing that collateral currency is a fundamental risk factor, not just a credit mitigation tool, requiring specific adjustments in both valuation and hedging strategies for modern interest rate derivatives.