Academic journal article Risk Management and Insurance Review

Unfair Discrimination and Homeowners Insurance Availability: An Empirical Analysis

Academic journal article Risk Management and Insurance Review

Unfair Discrimination and Homeowners Insurance Availability: An Empirical Analysis

Article excerpt


There has been an active public policy debate about the availability and cost of homeowners insurance in urban markets, and insurers have been charged with intentional discrimination, or redlining, against minority buyers and neighborhoods with a large proportion of minority population. Studies of relative price, loss costs, agency location, and product quality have attempted to determine whether insurers unfairly discriminate against minority homeowners in urban areas. This study focuses on insurance availability, and examines whether there is a systematic difference in the breadth of insurance coverage in the market that corresponds to the proportion of minority residents in a state. The market share of dwelling fire insurance polices across states for years 2000 and 2003 is analyzed, and evidence is presented that shows a positive correlation between the proportion of minority homeowners in a state and the share of more restrictive dwelling fire policies. However, this difference is not statistically significant once other risk-related and economic factors are included in the analysis.


Allegations of insurance redlining have been made against the property casualty industry since the urban riots in the 1960s revealed that a large number of damaged properties were un- or underinsured. This again became a pressing public policy concern after the Los Angeles riots in 1992. Since that time, a number of investigations by state and federal regulators have attempted to determine whether insurance is available to all homeowners on a basis that is not unfairly discriminatory.1 To that end, eight states collect information on underwriting and claims costs at the zip code level, and three of these make that data available to the public so that compliance with antiredlining statutes can be independently analyzed (Squires et al., 2001). For example, California requires insurers to submit annual community service statements to the department of insurance. These statements include zip code level data on homeowners exposures and premiums, and also must report the race, national origin, and gender of each applicant, and the number of applications accepted and declined.2 Although redlining is illegal in every state, several of the largest homeowner insurers have settled lawsuits arising out of this charge.3

Prior research that has examined unfair insurance discrimination takes one of two tacks. One method used to detect unfair discrimination is through the use of paired-testing. Several studies based on paired-testing have found evidence of unfair racial discrimination (see, for example, ACORN, 1993; Smith and Cloud, 1997). A more recent study using the same methodology, however, found no evidence of systematic discrimination in the provision of replacement cost coverage on dwellings in Phoenix or New York (Wissoker et al., 1998).

A more rigorous approach to examining the issue relies on statistical analysis of insurance price and/or availability. Statistical studies have attempted to measure the extent of discrimination by looking at consumer complaint data as a proxy for product quality differences (Chan, 1999), agency location in minority neighborhoods (Schultz, 1995; Squires et al., 1991), and the relation between minority population and insurance price and availability (for example, Minnesota Department of Insurance, 1994; Klein, 1995). Each of these studies finds some support for the idea that there are differences between urban and outlying areas in the provision of homeowners insurance. However, these studies do not control for other factors that might influence insurance prices or availability. Klein and Grace (2001) use zip code level data from the Texas insurance department to perform careful analysis of homeowners loss costs and coverage availability for urban versus outlying areas in Texas. Although they do find a positive correlation between minority population, loss costs, prices, and more restrictive insurance coverage, the correlation is not significant when other economic and demographic factors are controlled for. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.