Securities Lending Settlement

The Two-Leg Structure: Market Context, Participants, and FoP Mechanics

The global securities lending market routinely supports over $2 trillion of securities on loan across equity and fixed income markets. The economic rationale is structural: short sellers borrow securities to execute short positions, market makers borrow to meet client delivery obligations, and settlement participants borrow to avoid fails. On the supply side, institutional beneficial owners — pension funds, mutual funds, sovereign wealth funds, and insurance companies — hold long-term portfolios with substantial lendable inventory. Most beneficial owners do not lend directly; they delegate to agent lenders — State Street, BNY Mellon, JPMorgan, Northern Trust — who manage lending programs, negotiate rates, select borrowers, and handle the full operational lifecycle on the owner's behalf. Two loan types govern the market: open-term loans, recallable at any time by the lender, constitute the majority of outstanding volume; fixed-term loans carry a scheduled maturity and typically require borrower consent and a breakage fee for early termination.

A securities lending transaction settles in two separate legs that are economically linked but mechanically independent across the relevant settlement infrastructures — typically DTC for US equity lending and the applicable CSD for fixed income or cross-border equity. The loan leg typically settles Free of Payment: the lender delivers the securities to the borrower's account without a simultaneous cash movement, because the economic exchange occurs through the collateral leg rather than a purchase price. FoP also enables rapid inventory mobilization for short covering without requiring a simultaneous funding movement — an operational advantage when lenders are sourcing inventory across multiple custodial locations or recycling intraday positions.

The collateral leg takes one of two forms. In US equity lending, cash collateral is most common: the borrower delivers cash — at market convention approximately 102% of the current market value of the loaned securities — via Fedwire or the lending agent's cash accounts. The lender reinvests this cash, typically in short-duration instruments, and the economics are captured in the rebate rate: the spread between the reinvestment return and the rebate paid back to the borrower is the lender's net compensation. In European and cross-border lending, non-cash collateral is more common: the borrower pledges government bonds, agency debt, supranational bonds, corporate bonds, or (with higher haircuts) equities. Haircuts are applied based on credit quality and liquidity — typically 2% for government bonds, higher for equities or lower-rated debt. The collateral is delivered FoP or managed through a tri-party agent. Two further settlement events occur during the life of the loan: manufactured dividend payments (classified as Payment-in-Lieu, or PIL, income) when a stock goes ex-dividend while on loan, and collateral substitutions when the borrower replaces one pledged asset with another eligible security.

The Recall Process, Intermediated Chains, and the T+1 Return Window

The recall is the mechanism by which a lender terminates an open-term loan before scheduled maturity. Recall events are triggered by the lender's need — the beneficial owner has sold the position and needs the securities for the resulting settlement obligation; a corporate action requires the lender to hold the shares (proxy record date, rights entitlement, dividend capture); or an ESG mandate requires institutional shareholders to recall loaned shares to vote on board elections or contested corporate resolutions. Pension funds commonly run seasonal recall programs around proxy record dates. Partial recalls — requesting return of only part of the loaned position — and recall cancellations are routine operational events alongside full recalls.

In practice, recall workflows involve intermediated chains rather than a direct lender-borrower relationship: the beneficial owner instructs the agent lender to recall, the agent lender notifies the borrower, and if the borrower has on-lent the securities to a downstream borrower, it must complete its own downstream recall before it can return. Under standard GMSLA terms, the borrower has one settlement cycle to return. Recall instructions may be transmitted using ISO 20022 messages such as sese.034 or via proprietary agent-lender platforms — ISO 20022 adoption is growing but implementation varies across agent systems and counterparty configurations.

Under T+1, the recall window is operationally severe. If the lender does not issue the recall notice by the morning cutoff — typically around 10:00 to 11:00 AM ET — the borrower may have an additional business day to return, during which the lender's own pending sale may fail. If the recall is issued on time but the borrower cannot source the securities by approximately 11:59 PM ET on the trade date, a settlement failure on T+1 is almost certain. Persistent fails in threshold securities may trigger mandatory close-out requirements under Regulation SHO. For the lender, a recall failure that prevents timely delivery on a sale creates a cascading chain of settlement failures downstream.

Daily Mark-to-Market Margining, the Rebate Rate, and Three Operational Risks

Mark-to-market margining ensures the collateral held against a loan maintains its required ratio to the loan value throughout the loan's life. Closing prices are captured at the end of each trading session; the margin call calculation runs overnight; and margin calls are issued early the following morning for settlement on that business day. Market convention for US equity lending is approximately 102% of loan value for domestic securities and 105% for foreign securities — these are industry standards rather than hard regulatory thresholds. If collateral has fallen below the required level, a margin call is issued requiring additional collateral by the following business day. Excess collateral is returned when collateral value exceeds the required level by the configured tolerance.

Collateral haircuts calibrate the buffer against rapid market moves. Government bonds typically carry a 2% haircut; agency debt and supranational bonds carry slightly higher haircuts; equity collateral carries haircuts of 10% or more depending on volatility and liquidity. Three risks dominate securities lending: counterparty risk (borrower default before collateral can be liquidated to cover the position), collateral reinvestment risk (reinvestment pool declines below the amount owed back to the borrower on cash collateral return), and recall failure risk (borrower cannot return securities in time to meet the lender's settlement obligation). Daily margining is the operational control for counterparty risk; haircuts address market risk on collateral value; open loan tracking and recall monitoring address recall failure risk.

Tri-Party Collateral, Eligibility Validation, and Bilateral Substitution Risk

Tri-party collateral arrangements delegate the operational management of the collateral leg to a specialist agent — BNY Mellon, JPMorgan, Euroclear, or Clearstream — sitting between the lender and the borrower. The tri-party agent selects assets from the borrower's pledged pool that meet the lender's eligibility schedule (issuer type, currency, credit rating, remaining maturity, and concentration limits), validates each asset against the eligibility criteria, applies the applicable haircut, and runs automated daily margining across the entire pool. Concentration limit monitoring ensures no single issuer or asset class exceeds the lender's maximum permitted exposure. The agent calls for additional collateral or returns excess automatically without manual intervention from either party.

Collateral substitution — replacing one pledged security with a different eligible security — is managed automatically by the tri-party agent when the replacement meets the eligibility schedule. In bilateral lending arrangements, substitution is a manual coordination challenge with meaningful operational risk. The borrower must deliver the new collateral FoP and wait for the lender to release the original, or vice versa — creating a window of either temporary dual-collateral hold or, if the release precedes the delivery, a period of under-collateralization. Operations teams must track both legs simultaneously: the pending inbound (new collateral expected) and the pending outbound (original collateral to be returned). Substitution latency — the delay between the two FoP movements — must be monitored to ensure collateral coverage does not fall below the required threshold during the swap.

Position Management, SFTR Reporting, and Three-Dimensional Reconciliation

Open securities loan positions must be tracked and reconciled continuously across three dimensions: the securities position (loaned securities quantity — the lender records the securities as "out on loan," reducing the long position by the loaned quantity); the collateral position (cash or securities received as collateral, tracked as both an asset and a corresponding liability); and the mark-to-market margin position (net margin call and return balances outstanding each day). Position breaks in the loaned securities quantity — where internal records disagree with the CSD's confirmed position — carry the same operational risk as any settlement fail.

SFTR (Securities Financing Transactions Regulation) in the EU additionally requires both sides of each securities lending transaction to report to a Trade Repository by T+1, creating a parallel regulatory reporting obligation that depends on the same position data maintained in the open loan ledger. SFTR reporting requires a Unique Transaction Identifier (UTI) and Legal Entity Identifier (LEI) for each transaction — the same identifier infrastructure used in EMIR and MiFID II transaction reporting.