Fair-housing and other civil rights organizations have long used paired testing to provide the most definitive evidence of the presence or absence of unlawful (and in the case of the insurance industry, unfair) discrimination.

The basic concept is fairly simple. Teams of "mystery shoppers" are sent to the same service providers to inquire about a given product or service. Each team consists of individuals or couples who are similar in every respect (for example, income, credit rating, employment status, etc., in the case of housing) except for the trait being tested.

In the case of tests for racial discrimination in housing, white and non-white individuals would visit the same real estate or rental agency and inquire about similar housing. Since these individuals are equally qualified and virtually identical in every relevant way except their race, a reasonable assumption is that they would be treated the same in the marketplace.

If they are not, because all variables except race have been controlled, it can be assumed that racial discrimination has played a role.

Fair housing law enforcement organizations and fair housing groups have used this tool to identify unlawful and unfair discrimination against several major homeowners insurers. The incontrovertible data gathered in this way resulted in significant changes in how these insurers market, underwrite, price and generally serve their communities.

This has resulted in the amelioration of unfair pricing, while bolstering access to products and services to qualified families who previously did not have such access--as well as opening up new, profitable markets for insurers.

Perhaps more importantly, initially confrontational relations between some insurers and some consumer groups have evolved into mutually profitable partnerships.

For example, after settling a fair housing complaint with the U.S. Department of Housing and Urban Development, State Farm and the National Fair Housing Alliance are currently collaborating on a media campaign to promote the benefits of living in diverse, integrated communities.

In addition, following a jury verdict against Nationwide in a fair housing lawsuit filed by NFHA member Housing Opportunities Made Equal, these organizations now work together on a regular basis to expand the availability of insurance products to underserved markets.

The collection and analysis of such data, whether done internally or through contractual arrangements with qualified organizations, should be a requirement for the NAIC's accreditation of any state regulatory program.

As for HMDA-like disclosure, understanding how well an insurer serves a market requires understanding precisely where products and services are being provided, and where they are not.

Just as the Home Mortgage Disclosure Act has long required mortgage lenders to publicly disclose the number and type of loans they make by census tract, property insurers should be required to disclose by census tract the number of policies they write, renew and cancel, along with the number of applications they deny and policies they nonrenew.

(For more details about this, see our Final Say, "More Transparency Needed In Sale of Homeowners Insurance," on page 34 of NU's April 21, 2008 edition.)

Mortgage lenders and regulators attest to the fact that HMDA data have enabled them to serve their markets better and provide more effective regulation. And dozens of effective partnerships involving financial institutions and community groups have emerged in recent years.

The limited ZIP code disclosure requirements for insurers in selected states have had similar outcomes.

Wisconsin's ZIP code data was used as part of the evidence that led to a settlement of a fair housing lawsuit against American Family for discriminatory underwriting practices in Milwaukee. This agreement subsequently led to an effective partnership involving American Family, the local NAACP and other community organizations, resulting in access to insurance for previously underserved communities and new marketing opportunities for the insurer.

Developing the capacity to collect and analyze these data, or contracting with qualified organizations to conduct such work should also be a requirement for the NAIC accreditation of a state regulatory program.

The ability to detect unfair discrimination should be a critical component of any market conduct exam. Having effective procedures in place for the collection and analysis of relevant data is essential for any such evaluation.

Having the capacity to generate and utilize paired testing and HMDA-like data constitute key components of that effort.