Fixed Income Derivatives and Model Assumptions

Central Bank Actions and Interest Rates

Central bank actions, specifically the Federal Reserve's adjustments to the Fed funds rate, are often misinterpreted in the financial press. A common misconception is that a $25$ basis point reduction in the overnight Fed funds rate will directly and immediately cause long-term rates (e.g., $10$-$15$ year rates) to decline, thereby stimulating investment. However, the correlation between changes in overnight rates and long-term rates is very small.

While short-term rates can eventually affect long-term rates, the impact is not direct or immediate. Short-term borrowing is primarily used by businesses for immediate funding needs, such as making payroll or bridging temporary cash flow gaps. Financial institutions use overnight rates to manage demand deposits and regulatory capital requirements.

Changes in short-term management costs can indirectly affect the affordability of long-term investments over time, but this chain reaction is slow and depends on many other market factors, economic growth, and investment opportunities. Therefore, the direct causal link pushed by the financial press is often misleading.

Quantitative Easing and Direct Rate Management

A more direct monetary policy tool, often overlooked in public discourse, is a central bank's active buying and selling of longer-term securities (bonds) in the market, known as Quantitative Easing (QE). During the financial crisis, the Federal Reserve extensively purchased Treasury and mortgage-backed securities to inject liquidity and explicitly keep long-term rates low. Central banks, like the Bank of England in a recent example (June/July '24), actively manage their inventory of long-term assets. By reducing the sale of these assets, they aim to keep asset prices up and yields down, thereby directly influencing longer-term interest rates. This direct intervention is crucial for understanding fixed income markets, yet it receives less attention than changes in the Fed funds rate.

Modeling Term Structure and Central Bank Influence

When building models to predict the future evolution of the term structure of interest rates, practitioners typically rely on historical data to estimate parameters. This process implicitly assumes that the underlying factors driving interest rate movements in the past will continue in the future. However, if central bank policies (e.g., QE programs or inventory management strategies) change significantly between the historical period and the projection period, this assumption can be violated. Such changes introduce an additional layer of complexity for risk management and valuation models. It's essential to recognize and potentially adjust models to account for evolving central bank strategies, as these policies, including foreign currency intervention, can significantly impact interest rates.

Fixed Income Derivatives: An Overview

In financial markets, there are four basic derivatives: forwards, futures, calls, and puts. In the interest rate world, these take on specific names and structures:

• Forwards: Known as Forward Rate Agreements (FRAs).

• Swaps: Portfolios of zero-coupon bonds and FRAs.

• Futures: Standardized interest rate futures.

• Call Options: Known as Caps, which are portfolios of Caplets.

• Put Options: Known as Floors, which are portfolios of Floorlets.

Why Portfolios (Caps, Floors, Swaps) Trade More Than Simple Derivatives in Fixed Income

The primary reason that portfolios of simpler derivatives (like caps, which are collections of caplets) trade more actively in fixed income markets, unlike equity or foreign currency markets where single-underlying options are prevalent, is hedging demand. Most derivatives are issued for risk management purposes. The two most common primary debt instruments for corporations and financial institutions are floating-rate debt and fixed-rate debt, both of which represent a sequence of cash flows over time, i.e., a portfolio of cash flows. To manage risk effectively for these multi-period cash flow structures, derivatives that also offer multi-period coverage (like caps and floors) are more efficient and minimize transaction costs compared to bundling individual, single-period options.

Caps (Call Options on Interest Rates)

A Cap is a portfolio of European call options on an interest rate, called caplets. A cap has a notional amount (e.g., $10,000,000$), a cap rate (strike, e.g., $5\%$), and a maturity date (e.g., $5$ years). It consists of $T$ caplets, where $T$ is determined by the number of payment dates until maturity. Each individual caplet (at time $t$) has the following payoff:

$Payoff<em>t = Notional \times \max(R</em>{t-1,t} - K, 0) \times \text{Day Count Fraction}$

Where:

• $Notional$ is the principal amount.

• $R_{t-1,t}$ is the realized spot rate for the period from $t-1$ to $t$, known at time $t-1$.

• $K$ is the strike rate (cap rate).

• $ext{Day Count Fraction}$ (e.g., $0.5$ for six months).

The payoff occurs at time $t$. The critical feature is that the spot rate ($R_{t-1,t}$) determining the payoff is known one period before the payoff time.

Example and Real-World Use of Caps

Consider a cap with a $5\%$ strike rate, a $100,000,000$ notional, and a two-year maturity with payoffs every six months. If the six-month spot rates evolve as 6%, 5%, 4%, 7% over four periods, the caplet payoffs would be:

• Time 1 (Rate at Time 0 = 6%): $100,000,000 \times (0.06 - 0.05) \times 0.5 = 500,000$

• Time 2 (Rate at Time 1 = 5%): $100,000,000 \times \max(0.05 - 0.05, 0) \times 0.5 = 0$

• Time 3 (Rate at Time 2 = 4%): $100,000,000 \times \max(0.04 - 0.05, 0) \times 0.5 = 0$

• Time 4 (Rate at Time 3 = 7%): $100,000,000 \times (0.07 - 0.05) \times 0.5 = 1,000,000$

Caps are vital for managing interest rate risk. For instance, commercial real estate (CRE) companies often borrow at floating rates but have relatively stable, fixed rental income. Lenders may require them to purchase caps to protect against rising interest rates, ensuring they can meet loan payments. When rates rise sharply, caps become expensive (e.g., from $100,000$ to $10,000,000$ for a new cap), potentially leading to financial distress if companies cannot afford to renew their hedging. A treasurer with a floating rate loan (e.g., spot rate + $50$ basis points) can use a cap to cap their maximum interest payment, providing predictability and stability to cash outflows.

Floors (Put Options on Interest Rates)

The structure and notation for a Floor are similar to a cap, with a notional, rate, and maturity. A floor consists of floorlets, which are European put options on an interest rate. The key difference lies in the payoff condition:

$Payoff<em>t = Notional \times \max(K - R</em>{t-1,t}, 0) \times \text{Day Count Fraction}$

A floor pays off when the realized spot rate ($R_{t-1,t}$) falls below the strike rate ($K$). This protects the holder against interest rates dropping too low.

Real-World Use of Floors

In residential mortgages, banks might offer consumers a floating-rate loan with a cap to limit their maximum payment, which costs the bank money. To offset this cost and avoid charging a higher floating rate, the bank might require the consumer to sell them a floor. This means the consumer agrees to pay a slightly higher rate if market rates drop below a certain level. This arrangement can satisfy both parties: the consumer gets protected by the cap (limiting upside risk) and accepts the floor (limiting downside benefit) to secure a floating rate without an explicit premium, while the bank effectively manages its exposure.

By setting the cap and floor strike rates to be equal, a floating rate loan can effectively be transformed into a fixed-rate loan. This concept highlights a deeper principle in options theory, similar to put-call parity.

Swaptions (Options on Swaps)

A Swaption is an option on a swap, meaning it's a derivative on a derivative. A European call swaption, for example, has a maturity (e.g., Time $T$) and a strike price ($K$), which could be zero. Its payoff at maturity $T$ is:

$Payoff<em>T = \max(Value</em>{Swap,T} - K, 0)$

Where $Value_{Swap,T}$ is the market value of the underlying swap at time $T$.

Real-World Use of Swaptions

Swaptions are valuable for managing long-term interest rate risk and maintaining financial flexibility. Consider a treasurer who initially chooses a floating-rate loan (e.g., at $3\%$) instead of a fixed-rate loan (e.g., at $5\%$) at time zero. Five years later, if floating rates rise (e.g., to $6\%$) and fixed rates also increase (e.g., to $7\%$), the treasurer might regret their choice and want to convert to a fixed rate to avoid further increases. However, locking into a swap at the current higher fixed rate of $7\%$ would be disadvantageous.

If, at time zero, the treasurer had purchased a swaption with a five-year maturity and a strike rate of zero, they would have a valuable option. At year five, if fixed rates have risen, the swaption's intrinsic value would be positive, equal to the value of a swap that allows them to pay the original fixed rate (e.g., $5\%$) instead of the current market rate ($7\%$). This means the treasurer can effectively switch their floating loan to a fixed loan at the old $5\%$ rate, thereby insuring against adverse rate movements without having to pay the higher current fixed rate.

The HCM Model: Importance of Assumptions

Beginning with the HCM (Heath, Jarrow, Morton) model, the emphasis shifts to rigorously understanding the underlying assumptions. A mathematical model is defined as the logical equivalence of its assumptions and serves as an approximation of reality. Models generate implications, which are necessary conditions. Rarely are implications both necessary and sufficient to fully represent the model.

Testing Models and Assumptions

Models are typically tested by examining their implications. However, merely confirming implications does not validate a model as true. A model can be rejected by disproving an implication, but it cannot be proven correct solely by confirming its implications, unless those implications are demonstrably necessary and sufficient (a rare occurrence). It is critical to test the model's underlying assumptions themselves, which is seldom done in practice.

For example, two different models (e.g., HCM vs. a risk-neutral model with an incomplete market) might produce the same price for a swaption. While observing that price might confirm one implication of the HCM model, it doesn't validate all its assumptions. If the market is incomplete, certain implications (like effective risk management) may not hold, even if the price is accurate. Getting prices right through