<p>ARLINGTON, Va. – Part of the percolating debate about whether, and how, credit unions should report their service to lower-income or underserved people (see related stories page 1 and 3) includes how credit unions would be able to provide the necessary empirical data. Greg Gemerer, Research Officer with the National Federation of Community Development Credit Unions (NFCDCU), believes geocoding, the practice of plotting membership, loan and other data on easy to read maps, might provide part of the solution to that problem. Before a small session of the NFDCU's recent annual meeting Gemerer made a presentation that, quickly and clearly, showed where the members of two community development credit union members lived and where the credit union made their loans. In six simple overlaid graphics, (three per credit union) Gemerer was able to demonstrate that the majority of each credit union's members lived in low-income areas and that the bulk of each credit union's lending went there as well. "Geocoding can provide credit unions with an extremely illustrative and useful tool for examining their data," Gemerer told the meeting. Geocoding is the practice of taking addresses and assigning latitude and longitude values to them. The addition of the latitude and longitude data allows the addresses to be plotted on maps against data from other sources, like the U.S. Census, which show where low-income people live or where unbanked people likely live. Further, adding the data about latitude and longitude allows a financial institution to geographically map a host of other data such as deposits, loan balances, rates of loan rejection and use of the institution's services. Bruce Moore, director of business intelligence for the $1.1 billion Credit Union of Texas, based in Dallas, reported that, like most credit unions that use the service, the credit union uses geocoding primarily for planning and marketing. "We use it to plan branches," Moore said, "and estimate driving time." At this level, Moore reported, data from credit union members can be analyzed from many different angles and often with a correspondingly high cost that might put the practice out of the reach of many credit unions. But geocoding offers simpler reports too, according to Christina Speh, director of strategic planning for PCi Corporation, reports that don't have all the marketing bells and whistles and also lack their higher costs. PCi is the leading provider of compliance services that both banks and credit use, according to the company. Both banks and credit unions, but mostly banks, use PCi's "CRA Wiz" program to present data on their mortgage lending, but other packages that can look at other lending data are available as well, Speh said. PCi prices its services for these reports in terms of how much data the financial institutions want to map and how much of a breakdown of the data needs to be done, according to Todd Cooper, product manager for CRA Wiz. A map that shows the data from 500,000 credit union members is going to cost more than one that maps the data from 50,000. Likewise, a map that shows data from only members with loans taken out before or after a certain data, for example, is going to cost more than a map which only shows members who have active loans. Depending on what a financial institution wants PCi to do, the price per record can run from a penny to a dollar, Cooper said. Over a dozen credit unions use PCi's services, according to Speh, and most of those do so for marketing maps, although some want a more basic product too, which does not need to cost so much. If a credit union only wanted to show, for example, where its members lived in relation to low income areas and where it made its loans, Speh estimated a credit union's cost would run between $1,700 and $6,000. But Bruce Moore pointed out that, to some degree, geocoding can provide an inaccurate picture of a credit union's activities. "You don't necessarily know whether a dot in a low income area represents someone who actually has low income," he pointed out, and a tighter data survey can cost more money. But Speh countered that, for this purpose, credit unions would not want the data for marketing which needs the additional specificity, but for a more general demonstrative purpose where the institution would to highlight the trends in a given area. In addition, Speh reported that if a credit union is capable of assembling its information in a least one electronic format then PCi will be able to help them geocode and map their information. CRA Wiz has a "very flexible data intake port," said Cooper. He reported PCi can take data in 8500 or 9000 different formats. "Banks can have thirty of forty different data sets," Cooper said by way of explanation of the high number. There can also be a problem making geocoding work for rural areas. A mapping system that relies on each data point having an address can stumble in rural areas where many people may only live on rural routes without addresses. But Robert Thorbus, director of consulting for Liberty Marketing Services counters that in rural areas, where the distances can result in very large maps, might be better served by mapping not to the individual point but to zip codes. Darker colors in a given zip code can indicate more members or more loans in the zip code versus another. Liberty Marketing Systems is a branch of Liberty, the credit union service company based in Mounds View, Minnesota. Liberty Marketing has about 230 clients that use its geocoding analysis primarily for marketing purposes, Thorbus reported. [email protected]</p>