Knowledge base · Market structure

Short locate, borrow, and fees

Educational reference from the platform knowledge base — written agent-readable first, rendered here for humans. Mechanics, not advice: nothing here is a recommendation to buy or sell any security.

Short locate, borrow, and fees

Definition

Before a short sale, US rules require a LOCATE — reasonable grounds that the shares can be borrowed for delivery (SEC Regulation SHO). The borrow itself carries a fee (annualized, from near-zero on liquid names to triple digits on squeezed ones), can be recalled by the lender, and owes any dividends through. This plumbing is the short side’s cost-of-carry stack — every decline thesis is priced against it.

How it works / structure

  • The pipeline: locate (pre-trade) → borrow (securities lending market; institutional supply from custodians and funds) → sale settles with borrowed shares (ms-settlement) → daily mark-to-market collateral → return/recall.
  • Fee formation: general-collateral names borrow near the funds rate minus a rebate; SPECIALS (heavy demand, thin lendable supply) command escalating fees — D’Avolio (2002) documented the market’s structure: most names cheap, a small hot tail expensive and unstable.
  • Engine-relevant data: borrow fee level and trend, and utilization (share of lendable supply on loan) — a stress gauge, with sent-short-interest the positioning companion; synthetic-implied borrow (from put-call parity deviations — strategy-synthetic-stock) cross-checks the quoted market.
  • Failures-to-deliver: Reg SHO threshold-list mechanics force close-outs of persistent fails — a regulatory forced- buy channel.

When it applies

Every equity short (strategy-short-selling carries the strategy view); relative-value structures with short legs (strategy-pairs-trading); reading crowding — fee spikes and utilization saturation are objective squeeze-fragility flags; options pricing on hard-to-borrow names (borrow cost embeds in put-call parity — deep-ITM call early exercise on HTB names is rational borrow-fee avoidance).

Risk profile & failure modes

  • Fee repricing mid-thesis: a 2% borrow becoming 40% converts a sound decline thesis into a race against carry; fee ceilings belong in short-thesis parameters.
  • Recall at the worst moment: lenders recall when supply tightens — which correlates with squeezes; forced buy-ins execute into rising prices.
  • Dividend pass-through: shorts owe declared dividends; special dividends land as surprise carry.
  • Data blind spots: exchange short interest is twice-monthly and lagged; daily fee/utilization feeds are vendor-sourced estimates — precision varies, the platform labels source and staleness.

Evidence & limits

Locate and close-out rules are SEC regulation. D’Avolio (2002) is the canonical map of the lending market (fee distribution, specials dynamics, recall behavior); subsequent literature links high borrow costs to overpricing (short-supply constraints slow correction) — the mechanism behind treating fee spikes as information. Fee data quality varies by vendor; platform theses quote the source.

Falsifiable-thesis examples

Illustrations only, not signals:

  • “X’s borrow fee will stay under 10% annualized while this short thesis is open” — falsified by the fee series.
  • “Y’s utilization above 95% this week will be followed by a 10%+ upside move within a month (squeeze-fragility flag)” — falsified at the mark.

Cross-references

  • Strategy layer: strategy-short-selling, strategy-pairs-trading
  • Positioning companion: sent-short-interest
  • Parity cross-check: strategy-synthetic-stock, opt-put-call-parity
  • Account plumbing: acct-margin-rules, ms-settlement

Sources

The agent cites this page.

Inside the platform, this entry is live context: the AI reasons from it, quotes it, and grades against it. Make your case.

Inquire about founding membership