Finance-Theory · Unit 7 · Video 3 · Interactive Practice
| Formula | Name | What it says |
|---|---|---|
| Coupon bond = its STRIPS | Identical cash flows carry one price; the video's 3-year 5% bond costs $961.36 | |
| Arbitrage profit today | Short the bond, buy STRIPS of every maturity and more at ; every later date nets 0 (bond at $970.00: $8.64) | |
| When the trade pays | = trading costs, = fee to borrow the side you short; a smaller gap survives, and a ban sets | |
| Owner's trade | A holder sells the bond it owns and buys the STRIPS: no short sale, same future payments ($110 − $100 = $10) |
Key Insight: The identity is held in place by a trade, not by supply equalling demand. Take away the short sale and only holders of the dear side, who must notice the gap and be able to act, push the two prices together: closer, but not to exact equality.
A gap between the bond and its STRIPS is an arbitrage only when it beats trading costs plus the borrowing fee.
💡 The fee levels are illustrative. Lo's argument needs only their ordering: a tripled or quadrupled fee still leaves a finite band, while a ban makes the fee, and the band, infinite. The ban setting, here and in the next panel, is hypothetical for this bond: the September 2008 order covered the shares of financial companies, not Treasuries. The logic is the same for any security.
Under a ban, a gap closes only if someone who already owns the dear side notices it and can trade.
Problem 1 · The Arbitrage Trade
Given: a 3-year bond with $1,000 face value and a $60 annual coupon trades at $1,020.00. STRIPS prices per dollar are , , . Find the net cash flows of the arbitrage.
The package that reproduces the bond is 60 one-year, 60 two-year and 1,060 three-year STRIPS:
The bond at $1,020.00 is the expensive side, so short it and buy the package:
| Position | Date 0 | Date 1 | Date 2 | Date 3 |
|---|---|---|---|---|
| Short coupon bond | +1,020.00 | −60 | −60 | −1,060 |
| Buy STRIPS | −1,012.00 | +60 | +60 | +1,060 |
| Net | +8.00 | 0 | 0 | 0 |
$8.00 today, nothing owed later, none of your own money committed.
Problem 2 · Up to Transaction Costs
Given: the video's bond, whose STRIPS package costs $961.36, trades at $965.00. Carrying out the trade costs $5.00 per bond in all (spreads, fees and the borrowing fee). Is there an arbitrage?
The bond is the dear side, so the candidate trade is short the bond, buy the STRIPS. Before costs it keeps
After the $5.00 of costs it nets
The price $965.00 lies inside the band , from $956.36 to $966.36, so no one can profit from closing it and the wedge survives.
Problem 3 · Price It, Don't Ban It
Given: a bond trades $12.00 above its STRIPS package. Trading costs are $2.50 per bond and the ordinary fee to borrow the bond is $1.50. A regulator quadruples the borrowing fee instead of banning short sales.
What does the short-sale arbitrage now net per bond?
What is now the widest gap that survives without a profitable short-sale trade?
New borrowing fee: . Net of both frictions:
The trade pays only when the gap exceeds :
Before the change the band was and the trade netted $8.00; after it the band is $8.50 and the trade nets $3.50. Under a ban, and no short-sale trade closes any gap.
Problem 4 · The Owner's Trade
Given: short sales of this bond are banned (a hypothetical: the 2008 order did not cover Treasuries). A pension fund holds a 2-year bond with $1,000 face value and a $70 annual coupon, trading at $1,015.00; STRIPS prices are and . What can still narrow the gap, and for how much?
The package that reproduces the bond's $70 and $1,070:
The fund sells the bond it owns for $1,015.00 and buys the package:
Its future receipts are unchanged: $70 at date 1 and $1,070 at date 2, now paid by STRIPS. No borrowing is involved. Whether it happens depends on the fund knowing the gap and being willing and able to trade, which is why a ban leaves prices closer but not pinned.
Solved: 0 / 4