Covered Bonds: Structure and Relative Value

A covered bond gives investors recourse both to the issuing bank and to a segregated cover pool, creating a dual-recourse credit structure rather than a conventional unsecured claim. Its apparent simplicity can hide extension, asset-encumbrance, jurisdictional, and funding risks that matter when comparing its spread with senior bank debt or sovereign paper.

Structural credit map

Before issuer default, the bond is generally a senior unsecured obligation. After default, the investor has a preferential claim on cover-pool cash flows, often with contractual overcollateralization and liquidity provisions; any deficiency becomes a claim on the issuer estate. Exact protection depends on national law and program documentation.

LayerPrimary protectionRisk remaining
issuer recoursebank capital and earningsbank default
cover poolmortgage/public-sector assetsasset losses and timing
overcollateralizationcollateral value buffermarket-value decline
liquidity reservepayment continuityfinite horizon
regulatory frameworksegregation and supervisionlegal/jurisdiction risk

The cover pool is dynamic. Assets may be substituted, prepayments change amortization, and triggers can require top-ups. Model asset cash flows, liability cash flows, and collateral eligibility separately; a static collateral snapshot is only a starting point.

Relative-value measure

Compare instruments on an asset-swap spread or OAS basis using consistent curves, settlement, accrued interest, and optionality. A raw yield spread between a covered bond and a senior preferred bank bond can largely reflect different coupon, maturity, and curve exposures.

def asset_swap_pv01(discount_factors, accruals):
    return sum(df * alpha for df, alpha in zip(discount_factors, accruals))

def spread_pv(spread, asw_pv01):
    return spread * asw_pv01

def overcollateralization(eligible_assets, covered_bonds):
    return eligible_assets / covered_bonds - 1.0

The code does not model optional calls, negative yields, or discounting conventions; use it only to illustrate that a spread has value through an annuity. Sensitivity should be decomposed into rates DV01, issuer-credit spread DV01, cover-pool spread sensitivity, and basis to the selected hedge.

Stress the dual recourse

A useful scenario grid combines bank stress with cover-pool stress. The wrong assumption is that one automatically offsets the other: falling house prices, unemployment, and bank funding pressure are often correlated. Conversely, covered bonds may retain value in a bank stress if statutory segregation and collateral quality are credible.

ScenarioBank senior spreadCover poolKey question
rates sell-offwider funding spreadprepayment slowsasset-liability extension
housing downturncapital pressureLTV deteriorationOC adequacy
sovereign stressredenomination/legal riskpublic assets affectedgoverning-law protection
funding shockrepo haircut risesassets may be soundforced-sale liquidity

Apply prepayment and extension scenarios to mortgage cover pools using behavior consistent with mortgage prepayment risk. The bond’s contractual maturity extension can be valuable for payment continuity but changes duration and liquidity profile.

Trading implementation

A common relative-value trade is long a covered bond versus short a matched issuer senior bond, government future, or swap hedge. The hedge must neutralize duration and curve exposure before interpreting the residual as structural credit value. Fund the cash leg realistically: repo availability, haircuts, collateral eligibility, and specialness can exceed an expected spread pickup.

Monitor issuer CDS, senior and subordinated curves, cover-pool composition, OC disclosures, and primary-market supply. CDS is a useful bank default proxy but does not price cover-pool recourse directly; CDS hazard-rate pricing explains its mechanics.

Track both headline and adjusted overcollateralization. Eligibility haircuts, currency hedges, arrears assumptions, and substitution rights can make the published pool balance a poor estimate of protection available after issuer default. A robust relative-value signal penalizes that uncertainty rather than assuming every nominal asset is immediately monetizable.

Key takeaways

  • Covered bonds combine issuer recourse with a legally protected cover-pool claim.
  • Normalize relative value with matched curve, duration, settlement, and funding assumptions.
  • Dynamic cover pools require prepayment, substitution, and overcollateralization scenarios.
  • Bank and collateral stress are correlated, so dual recourse is not risk-free.
  • Include repo and liquidity economics before converting a spread differential into expected return.
#covered bonds #credit #relative value #fixed income