Market Imperfections and Catastrophe Insurance: Building the Case for Government Intervention

1
Market Imperfections and Catastrophe
InsuranceBuilding the Case for Government
Intervention

• Kenneth A. Froot
• Harvard University

2
Insurance Risk Sharing

• Purpose of insurance is to share risk
• How good is the sharing of pervasive event risks?
• What could explain the lack of large event risk
  sharing?
• What are the implications for intermediation?
• Focus on reinsurance of large catastrophic risks
  from natural perils (e.g., hurricane, earthquake,
  freeze, fire, etc.)
• Objective event probabilities estimated by
  modelers
• Less scope for moral hazard and adverse selection

3
Historically Much Cat Risk is Retained

• Insurers purchasing catastrophe reinsurance
  protect mostly against small events
• Many insurers purchase no reinsurance
• Similar for cat covers of nonfinancial
  corporations
• Not because insurance capital is so large

4
Percentage of US Cat Exposure Reinsured by
Insurance Companies
5
Why So Little Risk Transfer?

• In a perfect world
• elasticity of supply and demand are high

Demand Supply
price
quantity
6
What distortions diminish risk transfer?
Demand

• Two types of explanations
• 1. Supply of Insurance/Reinsurance capacity is
  low

Supply
price
quantity
7
What distortions diminish risk transfer?
Supply

• Two types of explanations
• 2. Demand for Insurance / Reinsurance capacity
  is low

Demand
price
quantity
8
What distortions diminish risk transfer?

• Supply is low (and prices high) because
• 1. Reinsurance capital is not cheap
• 2. Reinsurers have market power
• 3. The private reinsurance production process is
  inefficient
• 4. Moral hazard / adverse selection
• Demand is low because
• 5. Awareness is low (also, mitigation is low)
• 6. Insurers are regulated, and regulation
  suppresses final price
• 7. Ex-post subsidies expected from governments
• 9. Behavioral factors make unlikely events seem
  irrelevant

9
Do these distortions diminish risk transfer?

• Demand-driven distortions cant be solved by
  direct government intervention in bulk risk
  financing
• Can be improved by other forms of intervention
• Increase penetration and mitigation
• Ensure that regulation does not artificially
  suppress price
• Note requires more, not less, reliance on
  market mechanisms
• Create commitment mechanisms making ex-post
  subsidies costly
• Promote information release to overcome behavior
  biases

10
Do these distortions diminish risk transfer?

• Supply-driven distortions are more complex
• If the problem is
• High private capital costs,
• Then intervention may help
• Market power,
• Then regulate dont produce
• High costs of production,
• Then no distortion to counteract
• Adverse selection,
• Then costs of government becoming informed

11
Choosing Among These

• Some facts to help separate out competing
  explanations

12
Reinsurance Price / Quantity Pairs
13
Price Premium / Expected Loss

14
The Impact of Unrelated Perils on Prices
15
Capital Supply Changes Help Explain Facts
Demand
Supply
price
quantity
16
Capital Supply Changes Help Explain Facts

• Specifically, three corporate finance distortions
  interact
• 1. Internal capital valuable because capital
  market imperfections make external finance
  costly.
• 2. Internal capital valuable because product
  market imperfections make insureds overpay for
  safer coverage
• 3. Internal capital costly because agency costs
  and corporate taxation add a corporate cost of
  carry
• 1 and 2 push companies toward holding excess
  capital, 3 pushes toward reduced capital

17
In the absence of these distortions, corporate
value increases 1-for-1 with capital
Value with Perfect markets
Value
0
Quantity of capital deployed (given risk
underwritten)
18
With distortion 3, value increases less than
1-for-1 with capital
Value
Value of internal capital Value of capital
deployed less taxes and agency
0
Quantity of capital deployed (given risk
underwritten)
19
Adding distortions 1 and 2, value increases
strongly at low amounts of capital
Value
Value with agency, taxes, capital
product market imperfections
0
Quantity of capital deployed (given risk
underwritten)
20
Some Implications of these Distortions

• 1. Privately deployed insurance capital is
  always scarcer than under perfect markets

21
Some Implications of these Distortions

• 2. There is an optimal amount of external
  capital to deploy it minimizes distortions for
  given risks.

22
Some Implications of these Distortions

• 3. An additional dollar has higher marginal
  value as internal capital rather than as external
  capital. Firms should treat internal capital as
  scarce. A marginal event loss has a greater
  impact on undercapitalized companies.

23
Some Implications of these Distortions

• 4. Uncertainty in underwriting performance
  reduces insurer value, and does so by relatively
  more for less-well-capitalized firms.

24
Some evidence to support this

• Event losses from September 11, 2001 had a more
  negative impact on the market value of
  more-poorly-rated insurance firms

From Cummins and Lewis, 2002, Catastrophic
events, parameter uncertainty, and the breakdown
of implicit long-term contracting in the
insurance market
25
Some evidence to support this

• Excess, risk-adjusted returns 1995-2001 on a
  panel of insurers are decreasing in the realized
  volatility of company earnings

Controlling for company rating makes these
results stronger
26
But Markets are Resilient Marginal Reinsurance
Coverage 8 Years After Hurricane Andrew

27
Can Government Intervention Solve?

• 3. Internal capital costly because agency costs
  and corporate taxation add a corporate cost of
  carry
• Tax policy can reduce corporate capital carry
  cost, but not agency-based carry costs.
• Government financing is pay-as-you-go,
  eliminating capital carry.
• 2. Internal capital valuable because product
  market imperfections make insureds overpay for
  safer coverage
• Little that government should do to interfere
  with private tastes.
• 1. Internal capital valuable because capital
  market imperfections make external finance
  costly.
• Some governments have lower credit risk than the
  best firms, potentially reducing deadweight costs
  of external finance.
• Short-run costs of raising external capital are
  greater than long-run costs.

28
Intervention creates distortions as well

• Dynamic inefficiencies
• Market responses and adjustments in pricing and
  cover.
• Experiment in US with terrorist protection.

29
Conclusions

• There is both theory and evidence to suggest that
  financial and product market imperfections
  prevent the first-best provision of catastrophic
  protection.
• Policies aimed at removing the distortions may be
  counterproductive (e.g., reducing tax-based carry
  costs) but should be explored.
• Policies aimed at partially supplanting market
  finance (for the bulk of exposures) should focus
  primarily on temporary distortions