Comparative Advantage and Swaps

Here's a puzzle. AAA-rated Northbridge Corp can borrow more cheaply than BBB-rated Del Rio Industries in every single market — fixed-rate bonds, floating-rate loans, all of it. Northbridge has, in economists' language, an absolute advantage in borrowing. And yet Northbridge wants floating-rate debt, Del Rio wants fixed-rate debt, and by the end of this lesson both firms will end up strictly cheaper than if each had simply borrowed directly in the market it wanted. Nobody made an arithmetic error. The interest-rate swap we met last lesson is the mechanism that makes this possible — and the logic behind it is a two-century-old idea from a completely different corner of economics.

Ricardo's idea, borrowed by a trading floor

David Ricardo's theory of comparative advantage in international trade says two countries can both gain from trading even if one is better at producing everything — what matters isn't who's better in absolute terms, but who's relatively better at which good. A swap desk runs precisely this argument on credit spreads instead of crops. Suppose the two companies can borrow $10 million for five years on these terms:

Borrower	Fixed rate	Floating rate
Northbridge (AAA)	4.00%	SOFR + 0.30%
Del Rio (BBB)	5.20%	SOFR + 0.80%
Northbridge's edge	1.20%	0.50%

Northbridge is cheaper in both markets, but not by the same margin: it has a bigger edge (1.20%) in the fixed-rate market than in the floating-rate market (0.50%). Say it this way instead: Northbridge has a comparative advantage in fixed-rate borrowing, and Del Rio — despite being worse at everything — has a comparative advantage in floating-rate borrowing, because its floating penalty (0.50%) is smaller than its fixed penalty (1.20%). That gap between the gaps is called the quality spread differential,

\text{QSD} = (5.20\% - 4.00\%) - (0.80\% - 0.30\%) = 1.20\% - 0.50\% = 0.70\%,

and it is exactly the amount of "free lunch" available to be split between the two firms (and usually a swap-dealer intermediary) if each firm borrows where its comparative advantage lies and then swaps into what it actually wants.

Worked example: running the trade

Northbridge actually wants floating-rate debt; Del Rio actually wants fixed-rate debt. Each borrows where it has the comparative edge, then swaps:

Step 1. Northbridge borrows fixed at 4.00% (its strong market). Del Rio borrows floating at SOFR + 0.80% (its comparatively strong market).
Step 2. The two enter a swap through an intermediary bank. Northbridge pays SOFR to the bank and receives 3.93% fixed; Del Rio pays 4.03% fixed to the bank and receives SOFR.

Now trace each firm's net cost after combining its loan with its swap leg:

	Northbridge	Del Rio
Pays on loan	4.00% fixed	SOFR + 0.80%
Pays on swap	SOFR	4.03% fixed
Receives on swap	3.93% fixed	SOFR
Net cost	SOFR + 0.07%	4.83% fixed
Direct market rate	SOFR + 0.30%	5.20% fixed
Savings	0.23%	0.37%

Northbridge gets its floating-rate debt 0.23% cheaper than borrowing floating directly; Del Rio gets its fixed-rate debt 0.37% cheaper than borrowing fixed directly. Add the two savings: 0.23\% + 0.37\% = 0.60\%, leaving the remaining 0.70\% - 0.60\% = 0.10\% of the QSD as the intermediary bank's fee for structuring and standing in the middle of the deal (bearing the credit exposure of each leg, a topic for a later lesson). Every cent of the 0.70% gap is accounted for. Nobody borrowed anything they weren't already going to borrow — the swap simply let both firms access the market where the other firm has the comparative edge, indirectly.

Northbridge and Del Rio could, in principle, sign a swap directly with each other and split the whole 0.70% QSD between just the two of them. In practice almost no swap works that way. A dealer bank warehouses the swap — trading one leg with Northbridge and an offsetting leg with Del Rio (or, more commonly today, hedging each leg separately in the broader market rather than matching two specific clients at all). This lets each company deal with a single, well-capitalized, professionally rated counterparty instead of doing credit diligence on a company from a completely different industry, and it lets the bank assemble a swap for a client the moment they want one instead of waiting to find an exact opposite-number partner. The bank earns its 0.10% for exactly that service — plus for bearing the risk, covered in a later lesson, that one of the two firms might not pay.

The textbook story makes it sound like 0.70% of value was conjured from nothing. Be suspicious of free lunches in efficient markets — and this one has a real explanation once you look closer. Del Rio's floating-rate penalty (0.50% over Northbridge) is smaller than its fixed-rate penalty (1.20%) for a genuine credit reason, not just market friction: a floating-rate loan reprices every few months, so a lender is only exposed to Del Rio's default risk over one short period at a time before the rate resets to reflect current conditions. A fixed-rate lender, by contrast, is locked into Del Rio's credit risk for the entire five years with no chance to reprice — a genuinely bigger risk, deserving a genuinely bigger spread. The "quality spread differential" is partly compensation for this real difference in risk exposure, not pure friction waiting to be arbitraged away.

It's also not true that the swap makes anyone's underlying credit risk disappear. Del Rio still owes SOFR + 0.80% to its actual lender no matter what the swap does; if Del Rio's swap counterparty defaults, Del Rio loses the offsetting fixed-rate receipts it was counting on and is back to paying a floating rate on debt it thought was fixed. The swap converts which rate Del Rio is exposed to — it does not make Del Rio's underlying credit worse or better, and it introduces a brand-new exposure to the swap counterparty itself. We come back to exactly this counterparty credit risk a few lessons from now.