05_Mildenhall_pt2 Flashcards

1
Q

Three Types of Insurance Demand

A
  1. Risk transfer – the insured purchases insurance to shift their risk to insurers. They do this by pooling their risks with other insureds
  2. Satisfying demand – the insured purchases insurance to satisfy statutory, regulatory, or contractual requirements
  3. Risk financing – the insured purchases insurance to finance uncertain future contingencies in an efficient manner
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

Identify five forms of insurance services an insurer supplies.

A
  1. Sales
  2. Marketing
  3. Risk surveys
  4. Loss control
  5. Risk bearing
  6. Pricing
  7. Underwriting
  8. Customer billing and support
  9. Investment Management services
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

Briefly describe two critical functions when managing a risk pool.

A
  1. Controlling access to the pool through UW and pricing
  2. Ensuring the pool is solvent by funding risk-bearing assets through the sales of liabilities
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

Briefly describe three ways in insurers bundle UW, pricing, risk-bearing, claims handling, and customer service when managing a risk pool.

A
  1. Some companies offer all of these services in-house
  2. Customer-facing functions are included with a direct writer, but farmed out to independent agents or brokers for other types of writers
  3. Managing general agents (MGAs) might only provide UW and pricing services
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

Parametric Insurance Contracts

A
  • Policies pay based on an objective event outcome, such as hurricane intensity and landfall
  • These policies are easy to underwrite since they do not depend on the characteristics of the insured
  • These policies are difficult to design in such a way that they eliminate the potential for an insured to profit from a claim
  • Under these policies, the insured is exposed to basis risk, which is the mismatch between the insured’s subject loss and the insurance recovery
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

Dual-Trigger Insurance Contracts

A
  • Policies pay based on an objective event outcome AND the event causes an economic loss to the insured
  • The indemnity payment is a function of the underlying subject loss suffered by the insured
  • These policies apply an indemnity function 𝑓() to the subject loss 𝐿 that combines limits, deductibles, etc
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

Explain why franchise deductibles fail to meet the indemnity function requirements underlying a dual-trigger contract.

A

The indemnity function 𝑓() has the following conditions:
* 0 ≤ 𝑓(𝐿) ≤ 𝐿
* 𝑓(𝐿) is a monotonic function of 𝐿
* 𝐿 − 𝑓(𝐿) is a monotonic function of 𝐿
* 𝑓 is continuous
Under a franchise deductible 𝑑, if 𝐿 < 𝑑, then 𝑓(𝐿) = 0. But, if 𝐿 > 𝑑, then 𝑓(𝐿) = 𝐿 instead of 𝐿 − 𝑑. Thus, 𝑓 is not continuous (jump at 𝑑) and 𝐿 − 𝑓(𝐿) is not monotone (increases until it hits 𝑑, then drops to zero).

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

Explain why excess of loss covers meet the indemnity function requirements underlying a dual-trigger contract.

A

The excess of loss indemnity function is 𝑓(𝐿) = (𝐿 − 𝑎)! for some threshold 𝑎.
Clearly, 0 ≤ 𝑓(𝐿) ≤ 𝐿. Also, 𝑓 is continuous and monotonic since it is zero before 𝑎, and then increases from zero after 𝑎. Finally, 𝐿 − 𝑓(𝐿) is monotone since it increases to 𝑎 before 𝑎 and says there after 𝑎.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

Describe the general payment order upon liquidation of an insurer.

A

From highest (i.e., senior) to lowest priority:
1. Receiver expenses
2. Insurance claims from policyholders
3. Debtholders
4. Shareholders

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

Capital vs. Equity

A

Capital (i.e., policyholder surplus) equals assets net of liabilities owed to policyholders.
this is the portion of money set aside on top of the loss reserves to fund the losses.

Equity equals assets net of liabilities owed to all parties EXCEPT owners.
Equity is also defined as the owner’s residual value or value of the owner’s interest in the company

Equity is usually lower than capital as capital would include the impact of any debt (which is excluded from equity)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

Four Types of Insurer Capital

A
  1. Common equity – comprised of capital stock (par value of shares issued), additional paid-in capital (excess of amounts paid-in over par value), and retained earnings. Note that capital stock and additional paid-in capital only apply to stock companies
  2. Reinsurance capital – a reinsurance transaction can be thought of as the sale of the residual value for a specific portion of the business
  3. Debt capital – when an insurer cannot issue shares (ex. mutual), it can use surplus notes to raise capital
  4. Preferred equity – blends characteristics of debt and equity
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

Six Reasons Why Equity Capital is Expensive

A
  1. Principal-agent problem – investors and management do not have perfectly aligned incentives
  2. No independent validation of insurance pricing
  3. Equity requires a long-term commitment to a cyclical business
  4. Returns are left-skewed (investors prefer right-skewed returns)
  5. Regulatory minimum capital standards can force an insurer into supervision before it is technically insolvent (could lead to dividend restrictions)
  6. Double taxation – insurers’ corporate profits and dividend distributions are taxed (essentially taxes investors twice)
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

Four Ways in Which CAT Bonds Address the High Cost of Equity Capital

A
  1. CAT bonds are not equity and do not have market risk
  2. CAT bonds are a diversifying, zero-beta (i.e., independent of the financial market) asset class
  3. Catastrophe pricing can be validated
  4. There are no principal-agent problems since CAT bond cash flows are contractually defined
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

Provide a major disadvantage of CAT bonds.

A

CAT bonds are illiquid and trade in a thin market.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

Weighted Average Cost of Capital (WACC)

A

The WACC combines the cost of all forms of capital into a single figure. It is the weighted average of the various forms of capital.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

Briefly describe two theories of capital structures.

A
  1. Trade-off theory – the debt and equity mix trades off the costs and benefits of each capital type. It compares the expense of equity and the right of debt holders to force bankruptcy
  2. Pecking order theory – informational asymmetries between management and owners makes equity more expensive, in which case retained earnings are the favorable form of financing

Notes: The Trade-Off Theory suggests that a company balances the tax benefits of debt (since interest is tax-deductible) against the costs of potential financial distress (such as bankruptcy costs). This theory proposes that there’s an optimal capital structure where the marginal benefit of debt equals the marginal cost of financial distress.

Notes: The Pecking Order Theory is a financial theory that explains how companies prioritize their sources of financing when they need capital. It suggests that firms prefer to use internal funds (like retained earnings) first, and only turn to external sources of capital if necessary.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
17
Q

Regulatory Regimes Around the World

A
  1. Solvency II – used in the EU; assets are based on market value and liabilities consist of a best estimate plus a risk margin
  2. NAIC SAP – used in the US; focuses on the balance sheet; assets that cannot be readily converted to cash are considered non-admitted on the balance sheet
  3. GAAP – used in multiple jurisdictions; loss reserves are booked at the undiscounted best estimate; allows deferral of acquisition expenses
  4. IFRS – focuses on market value; loss reserves are valued on a discounted basis and include a risk adjustment
  5. Rating agency evaluations – rating agencies typically use adjusted statutory or GAAP financials in their internal capital models
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
18
Q

Top-Down Pricing

A

This textbook follows top-down pricing. This means that it begins by pricing an insurer’s total portfolio such that aggregate premium satisfies a stability criterion. A stability criterion essentially establishes a constraint.

For example, we may want to price the total portfolio such that the probability of ruin is at most 5%. A bottom-up pricing approach would focus on pricing individual risks rather than the portfolio in total.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
19
Q

Timing Risk and Amount Risk

A

Insurance loss payments have both timing risk and amount risk, which may be dependent risks since larger claims often take longer to settle.

Timing risk refers to uncertainty in how and when a claim will be paid out.

Amount risk refers to uncertainty in the amount of the claim.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
20
Q

Assumptions Underlying CAPM and Black-Scholes Financial Models

A
  • Competitive – there are many small sellers and buyers and undifferentiated products
  • Perfect – there are no information or transaction costs, and no bid-ask spread. In addition, everyone can borrow or lend at the same risk-free rate, there are no restrictions on short sales, and there are no taxes
  • Complete – there are enough securities to replicate any set of future period cash flows by securities trading
  • Arbitrage-free – there is no potential for a risk-free gain on an initial investment of zero
  • General Equilibrium – supply equals demand, no trader has an incentive to trade to improve their position, and everyone agrees all products are fairly priced. There is no arbitrage in when prices are in general equilibrium
21
Q

Myers-Cohn Fair Premium Condition

A

A premium is fair if “whenever a policy is issued, the resulting equity value equals the equity invested in support of that policy.”

If premium increases equity, it is unfair to the policyholder because it transfers wealth from them to investors.

If premium decreases equity, it is unfair to investors because it transfers wealth from them to the policyholder.

Fair rates mean that prices are in equilibrium. Otherwise, investors would have an incentive to write more or less insurance.

22
Q

Discuss an argument for using the risk-free rate for investment income in the DCF model.
Discuss an argument for using the anticipated market return for investment income in the DCF model.

A

Proponents of the risk-free rate argue that premium profit margins should not depend on actual investment portfolios, but only on the risk-free rate and systematic underwriting risk. Since policyholders do not share in the investment risks of the insurer, they should pay the same premium regardless of the firm’s investment strategy.

Proponents of the anticipated market return argue that adding investment risk increases the probability of default for policyholders and changes their expected loss recoveries. Hence, policyholder premiums should reflect the anticipated market return.

23
Q

Describe the IRR model.

Provide one advantage and one disadvantage of the IRR model.

A

The IRR model is from the investor’s viewpoint. It estimates all cash flows to or from the investor (called “equity flows”) and then computes the discount rate (i.e., the IRR) required to produce a present value of zero. If the IRR ≥ the investor’s hurdle rate, then the project should be undertaken.

An advantage of the IRR model is that it doesn’t require us to determine discount rates for each cash flow.
A disadvantage of the IRR model is that it requires a hurdle rate (i.e., required rate of return) to make the final decision.

24
Q

Describe the DCF Model.

A

The DCF model discounts each cash flow in a project at an appropriate risk-adjusted interest rate. Some cash flows are positive (ex. recurring revenue) while others are negative (ex. initial investment). If the total DCF is positive, then the project should be undertaken.

25
Cummins Conclusions on the DCF and IRR Models
1. Ratemaking is prospective, and prospective business should stand alone from legacy business and reserves 2. Investment returns should estimate yields expected during the policy term. In other words, we should use the anticipated return for the firm rather than the risk-free rate 3. Regulation and accounting are relevant only if they affect cash flows 4. It is difficult to determine the appropriate cash flows associated with writing a new policy 5. Both models depend on a liability discount rate, or equivalently, a cost of capital
26
Discuss transaction costs in terms of an insurance market imperfection.
In a perfect market, insureds would purchase insurance directly from investors. However, this doesn't happen due to the existence of underwriting expenses. Underwriting expenses are a form of transaction costs. In the event that an insured tried to purchase insurance directly from an investor, it's unlikely that the investor would have the underwriting expertise to properly evaluate the insured. As a result, insurance intermediaries and/or insurers are used to perform this task (which obviously has a cost associated with it).
27
Frictional costs of capital are another insurance market imperfection. Briefly describe three types of frictional costs of capital.
1. Agency and informational costs – agency costs refer to the principal-agent problem in which manager incentives do not align with owners. Informational costs refer to an insurer's failure to control adverse selection and moral hazards of insureds due to information asymmetry 2. Double taxation – insurers’ corporate profits and dividends are taxed 3. Regulation – regulation can allow insurer assets to be seized or temporarily controlled by a regulator if the insurer fails minimum capital standards. Regulation can also restrict investment opportunities and payments to investors
28
Discuss bid-ask spreads in terms of insurance market imperfections.
In addition to transaction costs and frictional costs of capital, there are market frictions such as bid-ask spreads. These spreads are driven by information asymmetry. For example, underwriters may price in positive loadings even when there is no risk due to uncertainty around adverse selection.
29
Suppose Unit A and Unit B are the same size. Briefly discuss the difference in gross premiums and gross loss ratios between the units if Unit B is riskier than Unit A.
Since the units are the same size, the riskier unit (Unit B) will have a higher gross premium and lower gross loss ratio than Unit A.
30
Suppose Unit A and Unit B represent non-cat and cat business, respectively. Briefly describe the expected difference in implied loss ratios between the two units.
Since cat business is riskier, we would expect lower implied loss ratios for Unit B.
31
An actuary calculates stand-alone and total capital figures for a portfolio comprised of two thin-tailed lines. The actuary also calculates stand-alone and total capital figures for a portfolio comprised of one thin-tailed line and one thick-tailed line. Briefly describe the difference in diversification between the two portfolios.
Thinner tailed distributions diversify more effectively. Hence, we expect a much stronger diversification benefit for the two thin-tailed lines (i.e., the reduction in capital between “the sum of the stand-alone capital figures” and the “total capital figure” is expected to be greater for the first portfolio).
32
Explain why the diversification benefit for premium is expected to be less than the diversification benefit for capital.
Since premiums include expected losses, the diversification benefit is expected to be smaller than the diversification benefit for capital.
33
Bernoulli Layer
* A Bernoulli layer pays $1 when 𝑋 > 𝑎 and 0 otherwise, with no partial losses * Its pricing is entirely described by its probability of loss
34
Total Pricing Problem
Deciding how to split asset funding between policyholder premium and investor capital.
35
Describe the “shared liability.”
Insureds and investors have a shared liability to fund assets in excess of expected losses (𝑎 − 𝑆̅(𝑎)), with the funding coming from: * Margin (𝑀) from policyholder premium * Capital (𝑄) from investors
36
Layer Funding Constraint
We can think of 𝑃(𝑎) as the fair premium for the Bernoulli layer 1_X>a. The layer funding constraint that says that 1 = 𝑃(𝑎) + 𝑄(𝑎). We can further break the premium into loss and margin. This tells us that each Bernoulli layer can be thought of as 1 = 𝑆(𝑎) + 𝑀(𝑎) + 𝑄(𝑎).
37
Formulas for the Layer Premium and Capital Densities Using a Distortion Function
𝑃(𝑎) = 𝑔(𝑆(𝑎)) 𝑄(𝑎) = ℎ(𝐹(𝑎)) = 1 − 𝑔(𝑆(𝑎)) Where 𝑔() is a distortion function, ℎ() is the dual of the distortion function, 𝑔(𝑠) is the premium to write an insurance policy with Bernoulli payout having probability 𝑠 of loss (i.e., layer premium density), and ℎ(𝑝) = ℎ(1 − 𝑠) is the value of a bond with a Bernoulli payout having probability 𝑝 of full payment and 𝑠 = 1 − 𝑝 of defaulting (i.e., layer capital density).
38
Desirable Properties of 𝑔 and ℎ
* Insurance for impossible events (events with zero probability) is free. Thus, 𝑔(0) = 0 * A bond certain to default is worthless. Thus, ℎ(0) = 0 * Insurance for a certain loss has no risk and no markup. Thus, 𝑔(1) = 1 * Since higher layers respond in a subset of the events triggering lower layers (events that breach lower layers may or may not breach higher layers), 𝑔 must be increasing * When investors are risk averse, we expect they discount uncertain assets. Thus, ℎ(𝑝) ≤ 𝑝 and 𝑔(𝑠) ≥ 𝑠
39
Distortion Function Mathematical Requirements
Formally, a distortion function is a function 𝑔 that the maps an input on [0,1] to an output on [0,1] and satisfies the following: * 𝑔(0) = 0 and 𝑔(1) = 1 * 𝑔 is increasing * 𝑔(𝑠) ≥ 𝑠 for all 𝑠
40
Fully describe why using a distortion function always create a positive risk load in the premium.
Since g(s) ≥ s, the function 𝑔 thickens the tail of the survival function. This increases the expected value of any excess layer versus the objective probabilities defined by S(x). The distorted survival function 𝑔(𝑆(𝑥)), creates risk-adjusted probabilities that increase the probability of extreme events. Since the distorted survival function increases the expected value of any excess layer, it means that all layers have a positive risk load.
41
Properties of an SRM
* ρ_g is law invariant (LI) since it is a function of the survival function of 𝑋 * ρ_g is positive homogeneous (PH) for λ > 0 since ρ_g(λX) = λρ_g(X) * ρ_g is comonotonic additive (COMON) since ρ_g(X + Y) = ρ_g(X) + ρ_g(Y) * ρ_g is translation invariant (TI) if and only if g(0) = 0 and g(1) = 1. In other words, if 𝑔 is a distortion function, then ρ_g is TI * ρ_g is monotone (MON) * ρ_g can be expressed as a weighted average of TVaRs if and only if 𝑔 is increasing and concave * ρ_g is coherent (COH) since it is MON, TI, PH, and SA
42
Four Representations of an SRM
The four statements below are equivalent: 1. 𝜌 is coherent (COH, comonotonic (COMON) and law invariant (LI) 2. 𝜌 equals an SRM ρ! for a concave distortion function 𝑔 3. 𝜌 equals a weighted average of VaRs, where the weights ϕ(p) are positive and increasing 4. 𝜌 equals a weighted average of TVaRs, where the weights are positive
43
Simulating Losses from a Distorted Distribution
* Simulate a random uniform variable u_i = g(s_i) * Transform the uniform random variable to obtain g! = g^-1(u_i), where 𝑔 is from the distribution 𝐺 * Invert 𝑔 to simulate the distorted loss S^-1(g_i)
44
Compare and contrast the premium for body risk and tail risk between the PH, dual moment, TVaR, CCoC, and Wang transforms.
* The CCoC distortion function is the most expensive (highest premium) for tail risk, but cheaper for body risk (i.e., the non-tail portion of the distribution) * The PH and Wang transforms are also expensive for tail risk and cheaper for body risk * The TVaR and Dual Moment distortions are the most affordable for tail risk, but are expensive for body risk
45
Identify six things that should be considered when selecting a distortion function.
1. Simplicity 2. Transparency 3. Fairness 4. Objectivity 5. Data-based 6. Best practice
46
Briefly describe how an actuary should calibrate a one-parameter SRM.
In terms of calibrating a one-parameter SRM, actuaries can use financial metrics such as the weighted average cost of capital, required return, loss ratio, or margin. The actuary should set the parameter of the SRM such that the final results align with the expected cost of capital, required return, loss ratio, etc. The actuary can also use market information such as cat bond data to calibrate prices.
47
2 reasons for using the risk free rate when calculating investment income in ratemaking.
1. Premium profit margins should not depend on actual investment portfolios, but only on the risk-free rate and systematic underwriting risk. 2. Since policyholders do not share in the investment risks of the insurer, they should pay the same premium regardless of the firm's investment strategy.
48
one reason for using the anticipated market return when calculating investment income in ratemaking.
adding investment risk increases the probability of default for policyholders and changes their expected loss recoveries.
49
Basis risk
Mismatch between an insured’s subject loss and the insurance recovery An insured should never profit from a claim