Synthetic Securitisation
A synthetic securitisation is a transaction in which a bank keeps a portfolio of loans on its balance sheet but transfers the risk of their losses to investors through a credit derivative or guarantee — typically selling the first-loss tranche to a specialist fund. The loans stay, the customers notice nothing, and the bank's capital requirement on the portfolio falls. It is how European and, increasingly, US banks free capital for new lending.
- Also known as
- Significant risk transfer, SRT, Synthetic risk transfer, Capital relief trade
- Used at
- bank balance-sheet management on portfolios of SME, corporate, leveraged, mortgage and consumer loans; growing fast since 2020 as capital rules tightened
- Binding
- yes — documented as a financial guarantee or credit-linked note under ISDA or bespoke terms, and subject to the regulator's significant-risk-transfer (SRT) assessment before capital relief is recognised
- Typical structure
- a first-loss (junior) tranche of 5–12% of the portfolio sold to investors, sometimes a mezzanine tranche too, with the bank retaining the senior risk; the investor funds its tranche in cash held as collateral
- Typical pricing
- coupons of 8–14% a year on the junior tranche; a typical trade references €1–3bn of loans
- Parties
- the originating bank, the investors (credit funds, insurers, pension funds), the regulator that grants the relief, and the borrowers, who are never told
How synthetic securitisation works in practice
A bank with a €2bn portfolio of SME loans holds capital against all of it, at risk weights that reflect the loans' credit quality. It could sell the loans — a true-sale securitisation — but that means moving the customer relationships, the servicing and the pricing to someone else. Instead it buys protection on the portfolio's losses. It defines a reference pool, agrees with investors which losses they will bear and in what order — the first 8% of losses on the pool, say — and the investors put up that amount in cash, which the bank holds as collateral and pays a coupon on. When a loan in the pool defaults, the loss is charged against the investors' collateral; when the trade ends, the investors get back whatever is left.
Because the bank now bears only losses beyond the first 8%, the regulator lets it treat its retained risk as senior and hold less capital, provided the trade passes the significant-risk-transfer test: the risk really has moved, the tranche is thick enough, the bank has not kept it through a side arrangement. The capital freed can support new lending, which is what the bank tells its regulator it is for.
The investor is taking concentrated credit risk for a high coupon, and the market has grown into a specialist one: funds that underwrite bank portfolios loan by loan, price the expected loss and demand a premium above it. For the bank the calculation is the coupon paid against the cost of the equity capital released, and the regulatory review that could take the relief away.
Example
A bank has a €2bn SME portfolio with expected losses of 0.5% a year and risk-weighted assets of €1.5bn, requiring €180m of capital at 12%. It sells the 0–8% tranche — €160m — to a credit fund at a 10% coupon: €16m a year. The bank's retained senior risk is now weighted at a fraction of the original, taking RWAs to about €450m and the capital requirement to €54m: €126m of capital released, at a cost of €16m a year, or 12.7% — below the bank's 14% cost of equity, and the released capital supports €1bn of new lending. Over the trade's five years the fund expects about €50m of losses against €80m of coupons, and its return depends on the bank's underwriting being as good as its data room said.
Common mistakes
- Structuring a tranche too thin to pass the significant-risk-transfer test, and losing the relief on review. - Protection from an investor who has not funded it, so a downturn produces losses and a counterparty that cannot pay. - A reference pool defined loosely enough that the bank can substitute worse loans into it — the investor's fear — or that the regulator decides the bank never really left. - Reading the freed capital as profit rather than as lending capacity that costs a coupon every year. - In diligence on a bank, missing that part of the loan book's risk has been sold and the yield on it reduced.