Welcome to Insurance Payments!
Hey there! If you’ve ever looked at a car insurance or health insurance policy, you’ve probably seen words like "deductible" or "out-of-pocket limit." In this chapter, we take those real-world insurance rules and apply Probability to them.
Our goal is to figure out how much an insurance company expects to pay out based on a random loss amount. Don't worry if this seems a bit "wordy" at first—once you see the patterns, it’s just a bit of calculus and logic. Let's dive in!
1. The Basics: Loss vs. Payment
In Exam P, we usually deal with two different random variables:
1. \(X\) (The Ground-Up Loss): This is the total amount of damage that actually happened. It’s what the loss would be if there were no insurance at all.
2. \(Y\) (The Insurance Payment): This is the actual amount the insurance company writes on the check after applying all the policy rules.
Think of it like this: If you drop your phone and it costs \$500 to fix, the loss is \$500. If your insurance company only gives you \$300 because of your policy rules, the payment is \$300.
2. Deductibles: Your "Entry Fee"
A deductible (usually denoted as \(d\)) is the amount the policyholder must pay before the insurance company pays a single cent. It "deducts" from the total loss.
The Rule:
If the loss \(X\) is less than the deductible \(d\), the insurance company pays nothing.
If the loss \(X\) is more than the deductible \(d\), the company pays \(X - d\).
Mathematically, we write the payment \(Y\) as:
\( Y = \max(0, X - d) \)
Calculating Expected Payment with a Deductible
To find the Expected Value \(E[Y]\), you integrate the part of the loss that "survives" the deductible:
\( E[Y] = \int_{d}^{\infty} (x - d) f(x) dx \)
Quick Trick: The Survival Function Method
For any non-negative random variable, there is a much faster way to calculate this! If you know the survival function \(S(x) = P(X > x)\), then:
\( E[Y] = \int_{d}^{\infty} S(x) dx \)
This is a huge time-saver on the exam!
Key Distinction: Per Loss vs. Per Payment
This is where many students get tripped up! Read the question carefully:
1. Expected Payment per Loss (\(E[Y_L]\)): The average payment over all losses, including the ones where the company pays \$0.
\n2. Expected Payment per Payment (\(E[Y_P]\)): The average payment given that a payment actually occurred (i.e., given \(X > d\)).
The Relationship:
\n\( E[Y_P] = \frac{E[Y_L]}{P(X > d)} \)
Key Takeaway: Deductibles shift the loss distribution. They eliminate small claims and reduce the amount of large claims.
\n\n\n\n
3. Benefit Limits: The "Ceiling"
\nA benefit limit (or cap, usually denoted as \(u\)) is the maximum amount an insurance company will pay. They aren't going to pay for your \$10 million mansion if your policy limit is \$500,000!
\nThe Rule:
\nThe company pays the full loss \(X\), but never more than \(u\).
\n\( Y = \min(X, u) \)
Calculating Expected Payment with a Limit
\nThe integral looks like this:
\n\( E[Y] = \int_{0}^{u} x f(x) dx + \int_{u}^{\infty} u f(x) dx \)
Survival Function Shortcut:
\nAgain, the survival function makes this much easier:
\n\( E[Y] = \int_{0}^{u} S(x) dx \)
Did you know? Using the survival function \(\int S(x)dx\) is essentially calculating the area under the curve. For a limit \(u\), you just stop calculating the area at point \(u\).
\n\nKey Takeaway: Benefit limits protect the insurance company from massive, catastrophic losses.
\n\n\n\n
4. Coinsurance: Splitting the Bill
\nCoinsurance (denoted as \(\alpha\)) is a fixed percentage that the insurance company pays. If you have 80% coinsurance, the company pays 80% of the covered loss, and you pay 20%.
\nThe Rule:
\n\( Y = \alpha \cdot X \)
This one is the easiest for calculations! Because of the linearity of expectation:
\n\( E[Y] = \alpha \cdot E[X] \)
\n\( Var(Y) = \alpha^2 \cdot Var(X) \)
Key Takeaway: Coinsurance scales the entire loss distribution down by a factor of \(\alpha\).
\n\n\n\n
5. Inflation: When Everything Gets More Expensive
\nInflation affects the Ground-Up Loss \(X\). If inflation is \(r\), every loss \(X\) becomes \((1+r)X\).
\nCommon Trap Alert!
\nDeductibles and Limits usually stay the same even when inflation occurs. If your deductible is \$500 and inflation hits, the loss gets bigger, but you still pay the first \$500. This means the insurance company ends up paying more than just the inflation rate!
Step-by-Step for Inflation Problems:
1. Let the new loss be \(X' = (1+r)X\).
2. Apply the policy modifications (deductible, limit) to \(X'\).
3. The new payment is \(Y_{new} = \max(0, \min((1+r)X, \text{limit}) - \text{deductible})\).
6. Combining Everything: The Mega-Formula
Most Exam P questions will combine these. A typical policy might have a deductible \(d\), a limit \(u\), and coinsurance \(\alpha\).
The standard order of operations (unless the problem states otherwise) is:
1. Apply the Limit (on the loss).
2. Apply the Deductible.
3. Apply Coinsurance.
The Payment Formula:
\( Y = \alpha \cdot [\min(X, u) - d]_+ \)
(Note: The \([]_+\) notation just means "take the maximum of 0 and this value")
Summary Table for Quick Review
Deductible (\(d\)): Insurer pays \( (X-d) \) if \( X>d \). Shortcut: \( \int_d^\infty S(x) dx \).
Limit (\(u\)): Insurer pays \( X \) up to \( u \). Shortcut: \( \int_0^u S(x) dx \).
Coinsurance (\(\alpha\)): Insurer pays a %. \( E[Y] = \alpha E[X] \).
Per Payment: Divide the "Per Loss" expectation by \( P(X > d) \).
Final Words of Encouragement
Don't be intimidated by the integrals! Many Exam P questions use Uniform or Exponential distributions for these topics. For Uniform distributions, you can often solve these problems by simply calculating the area of a trapezoid or triangle on a graph rather than doing full calculus. Keep practicing, and these "insurance modifiers" will become second nature!