Welcome to the World of Assumptions!

In the previous chapters, you’ve learned about identifying risks and building models. But a model is just an empty shell without assumptions. Think of a model like a high-tech GPS: it’s useless unless you tell it where you are starting from and how fast you expect to travel. In Actuarial Practice (CP1), setting assumptions is one of the most critical steps in "Developing the Solution."

Don't worry if this seems a bit abstract at first. By the end of these notes, you’ll understand that setting assumptions isn't just about picking numbers—it's about making informed judgments based on the past, the future, and the specific goal you’re trying to achieve.

1. What are Assumptions and Why do They Matter?

An assumption is a value or parameter used in a model to represent a future uncertain event. Since we can't predict the future with 100% certainty, we use assumptions as "best guesses" or "safety-first" values.

Quick Analogy: Imagine you are planning a road trip. You assume your car will get 40 miles per gallon, gas will cost $4.00, and you’ll drive at 60 mph. If your assumptions are wrong, you might run out of money or arrive late. Actuarial models work exactly the same way!

2. Types of Assumptions

Before we dive into how to set them, we need to know what they look like. We generally group them into two buckets:

A. Economic Assumptions

These relate to the "outside world" and financial markets. They are usually the same for all players in a market.

  • Investment Returns: What will the stock market or bond yields do?
  • Inflation: How much will the cost of claims or expenses rise?
  • Discount Rates: The interest rate used to bring future money back to "today's value."

B. Non-Economic (Demographic/Statistical) Assumptions

These relate specifically to the people or events being insured/modeled.

  • Mortality/Morbidity: When will people die or get sick?
  • Withdrawals/Lapses: When will people cancel their policies?
  • Expenses: How much will it cost to run the office and pay staff?
  • Claim Frequency/Severity: How often will accidents happen and how much will they cost?

Key Takeaway: Economic assumptions are about money; non-economic assumptions are about events and behavior.

3. Key Considerations in Setting Assumptions

When you are asked in an exam "What should you consider when setting assumptions?", use this checklist to structure your thoughts.

I. The Purpose of the Model

The number you pick depends entirely on why you are picking it.
Example: If you are calculating Statutory Reserves (money required by law), you use prudent (cautious) assumptions. If you are Pricing a product in a competitive market, you might use Best Estimate assumptions to avoid being too expensive.

II. Data Availability and Quality

You can't set a good assumption without data. You must consider:
- Relevance: Is old data still useful? (e.g., medical data from 1950 is useless for 2024).
- Reliability: Is the data accurate and complete?
- Credibility: Do you have enough data to be confident in the results?

III. Past Experience vs. Future Trends

The past is a guide, but it’s not a map of the future.
- Past Experience: Analyze what happened previously (Experience Analysis).
- Future Trends: Adjust for changes. For example, if medical technology is improving, we should assume people will live longer than they did in the past. This is known as secular trends.

IV. Materiality

Materiality is a fancy way of asking: "Does this actually matter?"
Don't spend three weeks perfecting an assumption that only changes the final result by 0.01%. Focus your energy on the "heavy hitters" (like interest rates or major claim types).

V. Legislation and Professional Guidance

Sometimes, the choice is taken out of your hands.
- Regulation: The government might mandate a specific discount rate.
- Accounting Standards: (e.g., IFRS 17) may dictate how to value liabilities.
- Actuarial Standards: Professional bodies (like the IFoA) provide TASs (Technical Actuarial Standards) that you must follow.

Quick Review Box:
Ask yourself: P-D-M-E-R
Purpose, Data, Materiality, Experience, Regulation!

4. Best Estimate vs. Margins for Adverse Deviation

This is a core CP1 concept. You have two main "styles" of assumptions:

1. Best Estimate: This is the "mean" or "expected" outcome. There is a 50/50 chance the actual result will be higher or lower. It is used for internal planning and realistic pricing.

2. Prudent (with Margins): This adds a "buffer" to the best estimate.
- In General Insurance, a prudent assumption for claims would be higher than the best estimate.
- In Life Insurance, a prudent assumption for investment returns would be lower than the best estimate.
\( Assumption = Best Estimate + Margin \)

Why add a margin? To ensure that even if things go slightly worse than expected, the company can still pay its claims.

5. Consistency Between Assumptions

Assumptions do not live in isolation. They are friends that hang out together! If you change one, you might need to change another.

Examples of Consistency:
- Inflation and Interest Rates: Usually, if you assume high inflation, you should also assume high nominal interest rates.
- Lapses and Interest Rates: If interest rates in the market go up, people might "lapse" (cancel) their old, low-interest savings policies to move to better ones.

Common Mistake to Avoid: Setting assumptions independently. If your model assumes 10% inflation but only 2% investment returns, your model will likely show the company going bust very quickly! Ensure they are internally consistent.

6. The Step-by-Step Process for Setting Assumptions

If you are tasked with setting an assumption for a new project, follow these steps:

  1. Identify the need: Which parameters does the model require?
  2. Gather Data: Collect internal data (the company's own history) and external data (industry statistics).
  3. Analyze Experience: Look for patterns in the data.
  4. Adjust for the Future: Consider trends, planned changes in business strategy, or changes in the law.
  5. Incorporate Expert Judgment: Talk to underwriters, claims managers, or economists.
  6. Apply Margins: If the purpose requires prudence, add your buffers.
  7. Review and Document: Record why you chose that number. This is vital for professional accountability.

Did you know? Actuaries often use Delphi Technique for assumptions where data is scarce. This involves asking a panel of experts for their opinions and refining them until they reach a consensus.

7. Summary Checklist

Before you move on to the next chapter, make sure you can answer "Yes" to these:

  • Can I distinguish between Economic and Non-Economic assumptions?
  • Do I understand that Purpose dictates the level of Prudence?
  • Can I explain why Consistency between assumptions is vital?
  • Do I know that Data Quality and Materiality limit how precise an assumption can be?

Final Encouragement: Setting assumptions is as much an art as it is a science. It requires a balance of mathematical data and common-sense judgment. Keep practicing with past exam questions, and you'll soon start to see the patterns!