A synthetic securitisation is a transaction in which the credit risk of a defined pool of loans or other exposures is divided into layers and transferred, in whole or in part, to investors by means of a guarantee or credit derivative. The loans themselves are not sold. They stay on the originating bank's balance sheet, the bank continues to deal with the borrowers, and the investors agree to compensate the bank for losses on the pool in exchange for a fee. Banks use the technique mainly to reduce the capital they must hold against their lending.
It contrasts with a true sale securitisation, in which the assets are legally sold to a separate vehicle that funds the purchase by issuing notes. A traditional securitization of that kind provides both funding and risk transfer; a synthetic one provides risk transfer only. Transactions done by banks on their own loan books are often called significant risk transfer or balance-sheet synthetic securitisations.
Structure
The bank identifies a reference portfolio, for example loans to large companies, small businesses, or trade and project finance exposures. The risk of the portfolio is then split by tranching into at least two, and usually three, layers:
- First-loss tranche. Absorbs the initial losses up to a low threshold. It is often kept by the bank, in whole or in part, and corresponds to the equity tranche of a traditional deal.
- Mezzanine tranche. Absorbs losses above the first-loss layer up to a higher threshold. This mezzanine tranche is the part normally placed with investors.
- Senior tranche. Bears losses only if the layers below are exhausted. The bank usually retains the senior tranche, which attracts a low capital charge.
The investor, called the protection seller, receives a periodic premium. When a loan in the pool defaults and a loss is established, the bank claims the loss allocated to the protected tranche. Deals typically include a replenishment period during which repaid loans can be replaced by new ones meeting set criteria, and they run for several years.
Funded and unfunded protection
Protection can be delivered in two ways. In an unfunded transaction, the investor gives a guarantee or enters a credit default swap and pays only if losses occur. The bank is then exposed to the investor's ability to pay, so unfunded protection is generally taken only from highly rated counterparties such as multilateral development banks and insurers.
In a funded transaction, the investor pays the full amount of its tranche at the start. The cash is held as collateral, either by the bank or in a special purpose vehicle that issues credit-linked notes, and is returned at maturity less any losses. Funded structures are the norm when the investors are credit funds, pension funds and asset managers.
Capital treatment
The purpose for most originators is capital relief. If supervisors accept that a significant share of the portfolio's credit risk has passed to third parties, the bank may replace the capital requirement on the underlying loans with a lower requirement on the tranches it keeps, reducing its risk-weighted assets and supporting its Common Equity Tier 1 ratio. The purchased protection is recognised under the rules on credit risk mitigation.
Recognition is conditional. Supervisors examine whether the risk transferred is commensurate with the capital released, whether features such as early termination rights, premium arrangements or use of excess spread undermine the transfer, and whether the bank has given implicit support. Under EU rules the originator must retain a material net economic interest, and since 2021 on-balance-sheet synthetic transactions that meet set criteria can obtain the simple, transparent and standardised label, which lowers the capital charge on the retained senior tranche.
History
Synthetic techniques emerged in the late 1990s, when banks first used credit derivatives to shed the risk of corporate loan books. In the years before 2008 the same tools were used for a different purpose: arbitrage synthetic collateralized debt obligations that referenced mortgage-related securities the arranger did not own, allowing investors to take positions for and against the housing market. Losses on these, and the opacity of their exposures, gave synthetic securitisation a poor reputation and prompted much tighter rules.
The market that returned after the crisis is narrower. It consists largely of privately negotiated deals in which banks hedge loans they originated and continue to hold. Volumes have grown substantially in Europe, with public development institutions among the protection sellers for small-business portfolios, and have increased more recently in the United States.
Risks and criticism
For investors, the main risk is credit loss on a portfolio they cannot inspect loan by loan; they rely on eligibility criteria, the bank's underwriting and independent verification of losses. For banks, the risks are counterparty exposure in unfunded deals and loss of capital relief if supervisors later disallow a structure. Regulators have also questioned whether credit funds that sell protection finance themselves with bank loans, which would return part of the risk to the banking system, and have asked for closer monitoring as the market expands.