Welcome to the World of Real-World Hedging!

In your previous studies, you probably encountered the Black-Scholes-Merton framework, where we assume we can rebalance a portfolio continuously (every micro-second!) without paying a penny in fees. In the "real world" of insurance, specifically when managing Embedded Options in products like Variable Annuities, this is impossible.

In this chapter, we explore what happens when we step out of the "perfect" world and deal with the Costs of Discrete-Time Rebalancing. We will learn how rebalancing at specific intervals (like daily or weekly) and paying transaction fees affects the price and risk of insurance guarantees. Don't worry if this seems a bit math-heavy at first—we’ll break it down into simple steps!

1. The Gap Between Theory and Reality

When an insurance company offers a guarantee (like a Minimum Death Benefit), they are essentially selling a put option to the policyholder. To protect themselves, the company "hedges" by buying and selling underlying assets.

In a perfect world, this hedge is always "Delta-neutral." However, in reality:

  • Discrete Rebalancing: We only trade at set times (e.g., the end of the day). Between those times, the market moves, and our hedge becomes "stale."
  • Transaction Costs: Every time we buy or sell to fix our hedge, a broker takes a cut (commissions or bid-ask spreads).

Did you know? If an insurer rebalanced their portfolio every second, the transaction costs would be so high they would go bankrupt! If they never rebalanced, the risk of a market crash would be too high. Finding the "sweet spot" is the goal of this chapter.

2. The Source of the Cost: Tracking Error

When we rebalance at discrete time intervals (let's call the interval \(\Delta t\)), we experience a Tracking Error. This is the difference between the change in the value of the option we sold and the change in the value of the stocks we bought to hedge it.

The size of this error is closely related to a Greek you’ve met before: Gamma (\(\Gamma\)).

Why Gamma Matters

Think of Delta as the "speed" of your option's price change and Gamma as the "acceleration." If Gamma is high, your Delta changes very quickly when the stock price moves. Because we only rebalance every so often, a high Gamma means our Delta will be "wrong" almost immediately after we set it.

Key Takeaway: Discrete rebalancing creates a variance in our hedging results. The expected cost of this "error" over a small time step \(\Delta t\) is related to the square of the stock price change.

3. Transaction Costs: The Price of Being Right

Every time we adjust our hedge to reach the new Delta (\(\Delta\)), we have to trade. Let \(k\) represent the transaction cost rate (e.g., 0.2% of the trade value).

The cost of a single rebalancing trade at time \(t\) is approximately:
\(Cost \approx k \cdot S_t \cdot |\Delta_t - \Delta_{t-\Delta t}|\)

Where:

  • \(k\): The transaction cost percentage.
  • \(S_t\): The current stock price.
  • \(|\Delta_t - \Delta_{t-\Delta t}|\): The amount of "Delta" we need to buy or sell to get back to a neutral position.

The Trade-off

If you rebalance more frequently (smaller \(\Delta t\)):
1. Your Tracking Error goes down (the hedge is more accurate).
2. Your total Transaction Costs go up (you are paying the broker more often).

4. Total Expected Cost of Rebalancing

In the ALTAM curriculum, we focus on the Expected Total Transaction Cost over the life of the option. Under certain assumptions (like the Leland model), the total expected transaction cost is proportional to the Gamma of the option.

The intuition is simple: When the stock price moves a lot (high volatility \(\sigma\)), you have to trade a lot. When the option is near-the-money (high Gamma), each trade is larger. Therefore, the total cost depends on volatility, the time interval, and the Gamma.

Quick Formula Insight:

The expected cost of rebalancing over a period \([0, T]\) can be approximated by summing up the costs at each interval. A key result often used is that the expected cost is roughly:
\(E[Cost] \approx \sum E[k \cdot S_t \cdot |\Delta_{new} - \Delta_{old}|]\)

Don't worry if this seems tricky! Just remember: High volatility + High Gamma + Frequent trading = Very high costs.

5. Strategy: Choosing the Rebalancing Interval

How do insurers decide when to rebalance? There are two main ways:

  1. Time-based rebalancing: Rebalance every day, week, or month regardless of what the market does.
  2. Move-based (Threshold) rebalancing: Only rebalance if the stock price moves by a certain percentage (e.g., 5%) or if the Delta drifts too far from our target.

Comparison Analogy:
Time-based: Checking your GPS every 5 minutes while driving.
Move-based: Only checking your GPS if you realize you've taken a turn or the scenery looks unfamiliar.

6. Summary and Key Points for the Exam

To master this section for Exam ALTAM, keep these points in your "Quick Review" box:

Quick Review:
1. Discrete Rebalancing leads to Tracking Error because the hedge isn't updated continuously.
2. Transaction Costs are the fees (\(k\)) paid on the absolute value of the change in the number of shares held (\(\Delta\)).
3. The Gamma (\(\Gamma\)) of the option is the primary driver of how much you need to rebalance.
4. There is an Inverse Relationship: Smaller time steps reduce tracking error but increase transaction costs.
5. Leland’s Approach suggests that we can "adjust" the volatility used in the Black-Scholes formula to account for transaction costs, effectively making the option more expensive to cover these fees.

Common Mistakes to Avoid

  • Forgetting Absolute Value: When calculating transaction costs, always use the absolute change in Delta. Buying 10 shares costs money, and selling 10 shares also costs money. You don't "net" them out to zero!
  • Confusing Delta and Gamma: Remember, Delta tells you how much to hold. Gamma tells you how often and how much you will need to change that holding.
  • Ignoring the Stock Price: Transaction costs are usually a percentage of the value traded (\(k \cdot S \cdot \Delta shares\)), not just the number of shares.

You've got this! Understanding the costs of rebalancing is the bridge between theoretical finance and practical actuarial work. Once you grasp that Gamma is the "cost driver," the math starts to fall into place.