Trading the Yield Curve: Terminology, Short End and Slope Strategies
This is a retelling of Chapter 6 (Part 1) from “Trading the Fixed Income, Inflation and Credit Markets: A Relative Value Guide” by Neil C. Schofield and Troy Bowler (Wiley, 2011, ISBN: 978-0-470-74229-7).
Chapter 6 is all about trading the yield curve. That means taking positions across different maturities to make money from how the curve moves. Could be parallel shifts, steepening, flattening, or changes in curvature. This first half covers the basics: terminology that trips people up, trades at the short end, and slope strategies.
The Terminology Problem
Fixed income jargon is a mess. Let’s start with “long” and “short.”
In most markets, long means you bought something and short means you sold it. Simple. But in bonds, things get confusing fast. When someone says they’re “long the curve,” they mean they benefit when yields go down. That’s because lower yields mean higher bond prices. So being long the curve is really about being long price.
Being “short the curve” means you profit when yields rise and prices fall.
Same deal with “rallying.” In bonds, a rally means prices went up and yields went down. A “sell-off” is the opposite: prices fall, yields rise.
Then there’s “long duration” and “short duration.” This is about choosing how sensitive your portfolio is to rate changes. If you expect rates to fall across the board, you’d want to be long duration. That means buying longer-dated bonds, buying bond futures, or receiving fixed on swaps. All of these increase your sensitivity to falling rates.
Carry and Roll Down
These two concepts are the bread and butter of yield curve trading. If you only remember two things from this chapter, make it these.
Carry is what you earn (or pay) just for holding a position. It’s your income minus your costs. For a bond funded in the repo market, carry equals coupon income minus repo expense. If carry is positive, you’re making money every day just by existing.
Roll down is the profit or loss from a bond aging along the yield curve. If the curve slopes upward and nothing changes, your bond gets revalued at lower yields as it gets closer to maturity. That means higher prices. Free money, assuming the curve stays put.
Here’s a quick example. Say 10-year bonds yield 5% and 9-year bonds yield 4%. You buy the 10-year bond with a 12-month horizon. Assuming the curve doesn’t move, your bond rolls down by about 8.3 basis points per month. That’s the 100 basis point difference spread over 12 months.
The forward breakeven price ties these together. It’s the price your bond needs to hit on the forward date for you to break even after accounting for carry. If carry is positive, the breakeven price is below the current spot price. The curve can move against you a little and you still don’t lose money. That’s your cushion.
Steepening Trades and How Carry Works
A steepening trade bets that the gap between short-term and long-term rates will widen. You buy the short-dated bond and sell the longer-dated one, sized to be DV01 neutral. That way parallel shifts don’t hurt you.
The relationship between spot spreads and forward spreads tells you about carry. Consider a steepener between 5-year and 10-year swaps. If the spot spread is 108 basis points but the forward spread is 92 basis points, the forward spread is lower. That means the position carries positively. The curve can actually flatten by up to 16 basis points before you start losing money.
The rule is straightforward:
- Steepener with forward spread less than spot spread: Positive carry. The curve can flatten a bit before losses.
- Steepener with forward spread greater than spot spread: Negative carry. The curve needs to steepen beyond the carry cost.
- Flattener: Reverse the logic above.
Trading the Short End
The short end of the curve has its own ecosystem of trades.
Money market loans and deposits are the simplest play. If you think rates will rise, borrow at a fixed rate and then re-lend when rates move up. If the curve slopes upward, you could borrow overnight on a rolling basis and lend at a longer maturity like one month.
Interest rate futures are cleaner because they don’t need upfront cash. Eurodollar futures (referenced to 3-month LIBOR) let you lock in a forward rate. The contract is quoted as 100 minus the rate, so a 2.5% rate shows up as 97.50. Each basis point is worth $25.
Spread trades are popular here. If you think the short end will steepen, you buy the near-dated contract and sell the far-dated one. The beauty is that even if both prices move in the same direction, you still profit if the spread between them widens.
You can also play cross-market spreads. Say you think the differential between Eurodollar and EURIBOR rates is too narrow. You’d buy Eurodollar futures and sell EURIBOR futures, adjusted for the exchange rate. Since a Eurodollar basis point is worth $25 and a EURIBOR basis point is worth 25 euros, you need to weight the trade by the EUR/USD exchange rate.
Basis trading got interesting after 2007. The 3-month vs 6-month LIBOR basis used to be stable and boring. Then the financial crisis hit and it got volatile. Banks were suddenly paying very different rates to borrow for 3 months versus 6 months. You could trade this through two offsetting swaps or a single-currency basis swap (paying 3-month, receiving 6-month LIBOR).
The FRA vs EONIA trade expresses a view on the banking sector’s creditworthiness relative to the government sector. If you think that gap will widen, you receive LIBOR on an FRA and pay on an OIS of the same maturity.
Trading the Slope
Now we get to full curve slope trades.
STIRs vs bond futures spans the entire curve. You sell a short-term interest rate future and buy a bond future (like the Bund). Both have the same expiry, and the trade is DV01 neutral. If the curve flattens as expected, the short-dated future loses more value (in rate terms) than the long-dated future.
For a practical example: selling 377 EURIBOR futures against buying 100 Bund futures. The ratio comes from the DV01 difference. A week later with rates having flattened, the EURIBOR leg profits more than the Bund leg costs, netting a tidy gain.
Bonds and swaps can also express slope views. The classic 2s10s steepener: buy 2-year bonds, sell 10-year bonds, DV01 neutral. The 2-year has a lower DV01, so you need a bigger nominal position in the short end.
The math is pretty clean. Both legs have roughly the same DV01 exposure, so parallel shifts cancel out. If the curve steepens, you make about the DV01 amount per basis point of steepening. In the book’s example, that’s roughly 1,963 euros per basis point.
Conditional Curve Trades
When you use swaptions instead of plain swaps to trade the slope, you get “conditional” curve trades. They’re conditional because the payoff depends on whether rates actually move enough for the options to be in the money.
The trade matrix has four quadrants:
- Bullish steepener: Buy receiver on short-dated swap, sell receiver on long-dated swap.
- Bearish steepener: Sell payer on short-dated, buy payer on long-dated.
- Bullish flattener: Sell receiver on short-dated, buy receiver on long-dated.
- Bearish flattener: Buy payer on short-dated, sell payer on long-dated.
These are typically structured to be zero cost and DV01 neutral. The strikes are set so that no premium changes hands, and the notionals are chosen so that if both options get exercised, you’re insulated from parallel moves.
Swaptions can also help you identify slope trading opportunities. By comparing the ratio of implied volatilities across different tenors with regression-based historical ratios, you can spot when the market is implying something unusual about how different parts of the curve will behave.
What About Volatility and Rates?
There’s generally an inverse relationship between percentage (lognormal) volatility and the level of interest rates. When rates are low, lognormal vol tends to be high. When rates are high, vol tends to be low.
This makes sense when you think about normalized (basis point) volatility. Normalized vol is just the rate times lognormal vol. If this relationship didn’t hold, you’d get absurd conclusions. Low rates with low vol would imply the market expects almost no rate movement. High rates with high vol would suggest wild swings. But empirically, rates move by roughly similar amounts in basis point terms regardless of the level.
When curves are steep (usually a low-rate environment), there’s more demand for receiver swaptions, pushing implied vol higher. The steep front end means forward rates are high, making receivers attractive from a carry perspective.
On the other side, structured product investors sell volatility to get yield in a low-rate world. The net effect on vol depends on whether option demand outweighs this supply.
That covers the first half of Chapter 6. We’ve built up from basic terminology to full slope trading strategies. Next up: curvature trades, butterfly spreads, and structured products.