European Swaptions

A pension fund knows it will need to issue a five-year fixed-rate liability-matching swap in one year's time — but not until the plan's actuarial review is finalized. A corporate treasurer has a loan commitment that lets a client draw down floating-rate debt and then swap it to fixed, at the client's option, any time in the next two years. A mortgage servicer's book of prepayable loans behaves like a bond with an embedded option the servicer doesn't control. All three want the same thing: the right, but not the obligation, to enter a swap later, on terms fixed today. That instrument is a swaption.

Payer, receiver — and one option on a whole stream

A European swaption grants the right to enter, on a fixed future date T (the swaption's expiry), an interest-rate swap with a fixed rate K and a fixed remaining life — the tenor — running from T onward.

A payer swaption gives the right to pay fixed K (and receive floating). It's exercised, and is valuable, when rates have risen — the holder locks in the old, now-cheap, fixed rate instead of paying today's higher floating rate.
A receiver swaption gives the right to receive fixed K (and pay floating) — valuable when rates have fallen, locking in the old, now-attractive, fixed income.

Market convention names a swaption by its two time spans: a "1×5 payer swaption" expires in one year, and the underlying swap it delivers runs for five years after that. Compare this with a cap: a cap is a strip of separate options, one per reset — the holder can effectively "exercise" each caplet independently, since each is a wholly separate contract. A swaption is a single option covering the entire multi-period swap at once — on the expiry date you decide once, for the whole remaining stream of payments, all or nothing. That structural difference — many small independent decisions vs. one all-or-nothing decision — is the real distinction between a cap and a swaption, not just "different underlying."

Pricing: Black's model applied to the swap rate

Exactly as a bond option treats the forward bond price as the underlying, a swaption treats the forward swap rate s_0 — the fixed rate that would make the future swap worth zero today — as the underlying, with volatility \sigma and expiry T. The one new ingredient is the annuity factor A = \sum_i \delta\,P(0, t_i), the present value of $1 received on each of the underlying swap's payment dates — it replaces the single discount factor P(0,T) from a bond option, because a swaption's payoff, once exercised, is realized as a whole stream of cash flows rather than one lump sum.

On notional L, with annuity factor A, forward swap rate s_0, strike K, expiry T, and swap-rate volatility \sigma:

Payer swaption: V_{\text{pay}} = L\,A\big[s_0\,\Phi(d_1) - K\,\Phi(d_2)\big];
Receiver swaption: V_{\text{rec}} = L\,A\big[K\,\Phi(-d_2) - s_0\,\Phi(-d_1)\big];
d_1 = \dfrac{\ln(s_0/K) + \tfrac12\sigma^2 T}{\sigma\sqrt{T}}, d_2 = d_1 - \sigma\sqrt{T}.

Worked example: a 1×5 at-the-money payer swaption

Notional L = \$10{,}000{,}000, forward 5-year swap rate s_0 = 5\%, strike K = 5\% (at the money), swap-rate volatility \sigma = 20\%, expiry T = 1 year, and an annuity factor for the five annual payments of A = 4.2.

At the money, \ln(s_0/K) = 0, so exactly as before:

d_1 = \frac{\tfrac12(0.20)^2(1)}{0.20} = 0.10, \qquad d_2 = -0.10, \qquad \Phi(0.10) = 0.5398,\ \ \Phi(-0.10) = 0.4602.

V_{\text{pay}} = 10{,}000{,}000 \times 4.2 \times \big[0.05\times 0.5398 - 0.05\times 0.4602\big] = 42{,}000{,}000 \times 0.05 \times 0.0796 \approx \$167{,}160.

Roughly $167,000 to hold the right, in a year's time, to lock in today's 5% rate on a $10 million, five-year swap — the same at-the-money trick that made the bond-option and caplet examples clean also makes this one clean: at the money, d_1 and d_2 depend only on \sigma\sqrt{T}, split symmetrically around zero.

Why does the prompt call a payer swaption a hidden bond put? Think about what a fixed-rate payer receives: a floating-rate stream. A floating-rate note that resets to the market rate is always worth par (100% of notional) right at a reset date — its coupon is the discount rate, so nothing is gained or lost by holding it. So entering the swap is, cash-flow for cash-flow, the same as being short a fixed-coupon bond (paying its coupons = paying the swap's fixed leg) and long a floating-rate note worth par. At the swaption's expiry, the swap's value to the payer is therefore

\text{par} - B_K(T),

where B_K(T) is the price, at time T, of a hypothetical bond paying coupon K with the swap's maturity schedule. The payer swaption is only worth exercising when this is positive — i.e., when B_K(T) < \text{par}, exactly the exercise condition of a put option on that bond, struck at par. (Rates rising pushes bond prices down, which is exactly when a payer swaption pays off — the same event, described two ways.) A receiver swaption is, by the same logic, a disguised call on that bond. This is why so much of the interest-rate options world — bond options, caps, swaptions — is really one idea wearing different market clothes.

Two mixups that trip up even experienced desks:

Expiry vs. tenor. A "1×5" swaption is not a five-year option — it's a one-year option on a five-year swap. Confusing the two changes both T in the pricing formula and which volatility quote to use (the market's swaption volatility "cube" is indexed by both numbers separately, and they move differently).
Payer/receiver from whose side? "Payer" and "receiver" always describe the swaption holder's position in the underlying swap, not the counterparty's. A payer swaption held by a borrower who wants to lock in a fixed rate is, from the bank that sold it, an obligation to receive fixed if exercised — it is easy to state the trade backwards when writing up a hedge.