INNOVATION AND THE REGULATOR

1
INNOVATION AND THE REGULATOR

• JANET SILVERMAN
• NEW YORK INSURANCE DEPARTMENT
• CASUALTY ACTUARIAL SOCIETY SPRING MEETING
• MAY 7 - 10, 2000

2
RATEMAKING APPROVAL PROCESS

• Innovator
• Insurer
• Regulator

3
CREDIT REPORTS - PROPOSALS

• Used as a rating variable
• Algorithm used to determine rating can have
  black box nature
• Reasons for unfavorable outcome can be kept from
  insured

4
SOME ACTUARIAL REACTIONS

• The case for statistical correlation has been
  made.
• Potential for unfairness is a problem, whether
  actual or perceived.
• Black Box
• Lack of intuitive connection to driving experience

5
SOME ACTUARIAL REACTIONS

• Underwriting factors are not on file at the
  Department perceived unfairness is not so much a
  problem here.
• A 20 billion market attracts a lot of attention .

6
CREDIT REPORTS - APPROVALS

• Discount based on credit characteristics
• No black box approvals
• Reasons for declination must be made known to
  insured.

7
VERY PRELIMINARY REACTION

• What will the proposal look like?
• What kind of support will there be?
• Will there be a transition period?
• What is the effect on the insured population?
• Will our approval be the last word?

8
WHAT WILL THE PROPOSAL LOOK LIKE

• How will data from the black box enter into the
  rate?
• At introduction will the data collected be used
  for the following year?
• Will there be something like a credit based on
  assumed experience to be changed if the results
  are not as expected?
• What will happen to the people who do not want to
  be part of this?

9
WHAT WILL THE PROPOSAL LOOK LIKE

• Will there be a black box aspect to the
  algorithm connecting the technology to the
  premium paid?
• Will the variable be used as a replacement for
  the current system or an addition too it?

10
WHAT KIND OF SUPPORT WILL THERE BE

• Will the supporting data be comparable to that of
  New York?
• Can we be sure that the data cannot be
  manipulated?
• Will we see correlation to loss experience?
• Will correlation to other rating variables be
  recognized?

11
WILL THERE BE A TRANSITION PERIOD

• Will there be a cost to installing the system if
  so who will pay?
• Will there be a problem for people with old cars?
• Will there be a problem for people with cheap
  cars?

12
WHAT IS THE EFFECT ON THE INSURED POPULATION

• Some insureds will benefit some will not what
  is the worst case scenario?
• We must be assured that the rate will not be
  prohibitive for some insureds.
• If there is a voluntary aspect to this, wont
  people who choose this probably be better risks?
• Couldnt its acceptance be used as an
  underwriting variable?

13
WHAT IS THE EFFECT ON THE INSURED POPULATION

• Could this have an unfortunate effect on claims
  handling?
• Will the ultimate market be those who are part of
  the system and an assigned risk population?
• Because of the costs involved will that left-over
  segment contain an abundance of poor people?

14
WHAT IS THE EFFECT ON THE INSURED POPULATION

• PRIVACY ISSUE
• How much do you want your insurer to know about
  you?
• Is this degree of information really necessary to
  underwrite or to rate?
• Can this data be retrieved by law enforcement
  entities?
• Will the information be able to be subpoenaed?

15
WILL THE APPROVAL BE THE LAST WORD

• The other approval process
• Public
• Politicians
• Regulator

16

• The Insurance Department welcomes innovation in
  any form and values companies that put in the
  time and expense to improve the insurance market.
  We take seriously our responsibility -
  effecting positive change with a minimum of
  disruption to the system.