For further information, contact:
Paul Denning
Director of Media Relations, MIT
(617) 253-0576 office
[email protected] or
Patricia Favreau
Media Relations Specialist, MIT
617-253-3492
[email protected]
or
Barry List
INFORMS
443-757-3560
[email protected]
Cambridge, Mass, September 8, 2006 – New research headed up by an MIT Sloan School of Management professor found that a majority of homeowners could save money on their homeowner insurance premiums if they increase the deductibles on their policies, due to their reluctance to reportclaims to their insurers.
Co-authored by Michael Braun, PhD, assistant professor of marketing at MIT Sloan, this finding has relevance for both policy holders and insurance companies. While homeowners would avoid paying for excessive coverage, insurance companies could potentially save money by processing fewerclaims along with better identifying those customers more apt to file such claims.
Braun and his co-authors, Peter Fader, Eric Bradlow, and Howard Kunreuther, all of the Wharton School of the University of Pennsylvania, defined each household's selectivity in filing claims as its "pseudodeductible." A household, the study found, will file a claim on an insured loss only if the amount of damage exceeds that household's pseudodeductible, even though it cannot be directly observed.
The study, “Modeling the Pseudodeductible” in Insurance Claims Decisions,” appears in the August issue of the journal Management Science, which is published by the Institute for Operations Research and the Management Sciences (INFORMS). The work has won Michael Braun the INFORMS Decision Analysis Society's Student Paper Award for 2006.
The authors tested this new model using a dataset provided by State Farm, the largest underwriter of personal lines insurance programs in the United States. The company provided deductible and paid claims history information of more than one million records for all homeowners’ policies on single-family homes that were in effect in six Midwestern states for any of the years from 1998 to 2003. The authors did not have access to homeowner names and addresses, and the data was provided solely for use in the study.
Using probability theory and a branch of statistics known as Bayesian inference, Braun and his co-authors identified the policyholders most likely to be selective about filing claims and those most likely to fileclaims.
Braun and his co-authors found that 52 percent of sampled households with at least one claim – and 80 percent of all households in the sample – had thresholds that were higher than the next highest deductible they could have purchased. For example, a customer with a $500 policy deductible and a $1,300 pseudodeductible would not file claims on any losses that are less than $1,300. According to this finding, a customer could increase his deductible to $1,000, save money on the difference in the premium, and still be filing claims on the same set of losses.
Braun and his co-authors further conclude that looking at claims alone can be a misleading measure of a household’s proneness to risk. Since some customers may avoid filing claims because they experience no losses to begin with, other riskier homeowners might be more selective about the claims that they do file. The pseudodeductible model, however, takes this difference into account, offering a critical component in estimating the long-term value of policyholders. The model also allows for segmenting an existing customer base according to their underlying and unobserved loss rates and claim thresholds, rather than depending solely on the directly observed transactions history.
To access a copy of the research paper, please contact the INFORMS Public Relations Department at 1-800-4INFORMs/mailto:[email protected] or the Office of Media Relations at MIT Sloan at [email protected].
For over fifty years, the MIT Sloan School of Management, based in Cambridge, Massachusetts, has been one of the world's leading academic sources of innovation in management theory and practice. With students from more than 60 countries, it develops effective, innovative, and principled leaders who advance the global economy.
The Institute for Operations Research and the Management Sciences (INFORMS®) is an international scientific society with 10,000 members, including Nobel Prize laureates, dedicated to applying scientific methods to help improve decision-making, management, and operations. Members of INFORMS work in business, government, and academia. They are represented in fields as diverse as airlines, health care, law enforcement, the military, financial engineering, and telecommunications. The INFORMS website is www.informs.org. More information about operations research is at www.scienceofbetter.org.
- ### -