Letter of Credit Discounting
Turning an accepted promise into cash today — the one facility where the risk moves off your client entirely.
// 01 — THE PROBLEM IT SOLVES
An exporter ships goods offshore under a usance letter of credit. The buyer's bank examines the documents and accepts: payment in ninety days. The exporter now holds something close to a bank-grade IOU — but the next production run, the wage bill, and the freight on the following shipment will not wait ninety days. Work completed, cash deferred: that is the gap, and it recurs with every shipment.
LC discounting closes it. The exporter's bank advances the accepted amount, less a discount charge, the day acceptance lands; at maturity the issuing bank pays the discounting bank and the loop closes. For the client it converts certain-but-later money into usable-now money. For the bank it does something rarer: it moves the credit risk off the client altogether. After acceptance, repayment depends on the issuing bank, not on the exporter's trading — which is the root of everything distinctive about how this product is priced, read, and broken.
// 02 — THE PARTIES, AND WHERE THE RISK SITS
Four parties, one of which matters most. The exporter — beneficiary of the LC, the discounting bank's client, and after acceptance not the source of repayment. The buyer — applicant for the LC; the discounting bank never faces them directly, though their standing shaped whether the LC was opened at all. The issuing bank — the buyer's bank, and once it accepts, the true obligor: its rating, its country, and its behaviour at maturity are the deal. The discounting bank — advising on the LC, checking the documents, advancing the funds, and holding issuing-bank risk until maturity.
The lifecycle runs: sales contract specifying payment by usance LC; the buyer's bank issues it; goods ship; the exporter presents the documents the LC demands — invoice, bill of lading, insurance, certificates; the documents are checked and forwarded; the issuing bank accepts, committing to a dated payment; then and only then the discount is advanced. Before acceptance there is no debt to discount — only conditions that might become one — and an exporter needing cash before shipment is shopping for a different product: a trade loan underwritten on the trade itself.
// 03 — WHAT IT REALLY COSTS
The discount rate is built in layers, and reading the build-up tells you what the product really is. A constructed illustration, every figure invented for teaching:
+ Country / transfer risk ≈ 0.40%
+ Issuing-bank risk ≈ 0.60%
+ Bank margin ≈ 1.20%
= All-in discount rate ≈ 6.70% p.a.
Notice what is absent: a premium for the exporter's own credit. The rate prices a bank in a country, not a client in a business — which is why an exporter with a modest balance sheet can sometimes fund more cheaply through discounting than through any facility priced on themselves. On top of the rate sit the handling fees of the documentary machine — advising, examination, negotiation, discrepancy charges if documents need repair, courier and messaging costs — small individually, real in total, and the honest quote is the all-in cost of the transaction rather than the rate alone.
One structural choice moves the price more than any other: recourse. With recourse, the exporter remains liable if the issuing bank fails to pay — cheaper, and the norm for most flows. Without recourse, the discounting bank keeps that tail — priced wider, reserved for stronger issuing banks, and the only version that takes the receivable cleanly off the exporter's balance sheet. Which one is on the table belongs in the first conversation, not the fine print.
// 04 — HOW THE DESK READS IT
The issuing bank is the borrower. That is the whole re-orientation: the desk is buying a bank's accepted obligation, so the underwriting questions are a bank analyst's — rating and recent direction, size and ownership, history of honouring acceptances, and how much exposure to this same bank the desk already holds across every client's discounted paper. Concentration hides here easily: individually clean transactions stacking onto the same three offshore banks is one decision repeated, not ten decisions made.
Above the bank sits its country. A solvent, willing issuing bank behind newly tightened capital controls is an unpaid obligation with good intentions. Transfer risk, sanctions exposure, and the local banking system's stability are read separately from the bank itself, because they fail separately.
Below the risk sits the paper. LCs run on UCP 600, and under it a discrepancy — a date, a description, a spelling — is lawful grounds to refuse documents. Document examination is therefore not administration; it is the craft that determines whether the client gets paid, done adversarially, anticipating what a motivated checker at the issuing bank could object to. And beneath the paper, the oldest question: is the trade real? An LC verifies documents, never goods. Related parties, off-market prices, and unusual routes are read on every file, because a documentary credit is exactly as sound as the trade it describes.
// 05 — WHERE IT GOES WRONG
Discounting a hope. A client arrives with an LC and an urgent need; the LC has not been accepted. There is nothing yet to discount — only conditions. The product for that moment is pre-shipment finance, priced on the client, and treating the two as interchangeable mis-prices both.
Confirmed and unconfirmed, blurred. A confirmed LC carries a second bank's guarantee alongside the issuing bank's; an unconfirmed LC stands on the issuing bank alone. They are different risk objects with different prices, and the confirmation field of the LC settles which one is actually in hand.
The clean-product illusion. Self-liquidating, short-dated, bank-backed — it is easy to book discounting as nearly risk-free. On the bank's own ledger every discounted bill consumes limit against that issuing bank, and the portfolio question — how much of this one offshore bank do we now own? — is the one most easily forgotten in a product that feels safe one transaction at a time.
The recourse surprise. An exporter who believed the receivable was sold discovers a contingent liability at audit. The with/without distinction — and what it does to price and balance sheet — is first-conversation material, and skipping it costs the relationship more than the basis points ever earned.
// 06 — WORKED ON THIS SITE
The sibling product — trade finance without a bank's acceptance behind it, where the desk underwrites the trade itself — is the trade loan, and reading the two files together shows exactly what an issuing bank's acceptance is worth. Issue 02 works an exporter's facility request where the evidence discipline is the decision. The wider vocabulary lives in the glossary, and the borrower's side of the table in How Banks Read You.
// QUESTIONS PEOPLE ASK
- What is LC discounting?
- An exporter holding a usance letter of credit — one that pays at 30, 60, 90 or 180 days rather than at sight — presents documents, the issuing bank accepts them, and the exporter now owns a bank's promise to pay on a future date. LC discounting converts that promise into cash today: the exporter's bank advances the face value less a discount charge, and collects from the issuing bank at maturity. The exporter gets paid now for work already done; the discounting bank earns the discount for waiting in their place.
- Who actually bears the risk in LC discounting?
- After acceptance, the issuing bank — and this is the product's defining feature. In almost every other working capital product the credit risk sits on the client; here, once the issuing bank has accepted the documents, repayment comes from that bank, not from the exporter's trading. The discounting bank is effectively buying a bank obligation, so what it underwrites is the issuing bank's standing and its country: rating, track record, the regulatory environment it sits in, and whether money can actually leave that jurisdiction at maturity.
- What is the difference between discounting with recourse and without recourse?
- Who eats the loss if the issuing bank fails to pay. With recourse, the discounting bank can claim the advance back from the exporter — cheaper pricing, but the exporter carries a contingent liability until maturity, and it stays on their balance sheet accordingly. Without recourse, the discounting bank absorbs the issuing-bank risk itself — typically priced tens of basis points wider and reserved for stronger issuing banks, but the receivable is genuinely off the exporter's book. The difference belongs in the first conversation, because an exporter who assumes clean sale and discovers a contingent liability at audit time has been mis-sold, whatever the documents say.
- Can a letter of credit be discounted before the issuing bank accepts it?
- No — and confusing the two states is the classic error. Before acceptance, an LC is a set of payment conditions: pay if complying documents are presented. Only when documents have been presented, checked, and accepted by the issuing bank does a debt exist, and only a debt can be discounted. An exporter who needs funding before shipment is asking for a different product — pre-shipment or packing finance, underwritten on the exporter and the trade rather than on a bank's accepted obligation.
- Why would an issuing bank refuse to pay under a letter of credit?
- Almost always on documents. Under UCP 600 — the international rulebook LCs run on — an issuing bank may refuse documents for any discrepancy, and the tolerance is exact: descriptions, dates, quantities and spellings must comply. Some refusals are legitimate protection; some issuing banks also use discrepancies tactically to delay payment. Either way, the defence is the same — document checking is a technical craft, done before presentation, anticipating what a motivated checker on the other side could object to. Beyond documents, the residual refusals are the sovereign ones: capital controls or sanctions between acceptance and maturity, which is country risk, not paperwork.
// SEE THE DISCIPLINE AT WORK
The habit underneath this product — lend against what is evidenced, not what is promised — is the same one that decides Issue 02: a multi-currency request from an exporter who will not name the currencies. Make the call, then read the senior banker's.
Work Issue 02 →