Geographic Information Systems: Pinpointing Lost Revenues

Article excerpt

The proliferation of information now available creates exciting opportunities as well as challenges for tax auditors. The proverbial "shoe box" of accounting records has been digitized and is recorded in ever-increasing detail. Solutions for efficiently and accurately auditing large amounts of electronic data can be costly and difficult to implement. However, the City of Westminster, Colorado, recently solved one of its most tedious and time consuming audit procedures using a geographic information system (GIS)--an in-house tool employed with minimal training and expense.

GIS is "an information system that is designed to work with data referenced by spatial or geographical coordinates. In other words, a GIS is both a database system with specific capabilities for spatially referenced data, as well as a set of operations for working with data." (1) At its core, a GIS is a database. In addition to the traditional tables of information, a GIS includes specialized tools for managing and analyzing spatially referenced data. GIS also stores points, lines, and polygon shapes representing real objects. These features are precisely located to a specific latitude and longitude, creating complex spatial models.

Behind these features, data tables store detailed information about the graphics. For example, the map may include shapes representing land parcels. When a specific parcel is selected, the GIS can detail the parcel's legal description, owner, taxable value, or associated building permits.

THE IMPORTANCE OF SPATIAL RELATIONSHIPS

GIS, like all databases, allows users to make queries. Specific datasets can be selected from tables based on the values of various attributes. What makes a GIS unique is its ability to select data based on spatial relationships (see Exhibit 1). For example, if the GIS included a polygon representing the area of a special taxing district, a query could be performed to find all of the parcels within that area. Westminster uses spatial queries to determine whether or not tax transactions (represented by point features) occur within city limits.

Colorado, like most other states, has a variety of localities--counties, special districts, cities, and towns--for which the state revenue department collects a retail sales tax in addition to its own. Sales and use taxes are due based upon the point of delivery or point of consumption rather than the point of sale or origin. Colorado is one of four states that also allow home rule municipalities (such as the City of Westminster) to collect and administer their own taxes under independent tax codes. Currently, there are 62 home rule cities in Colorado that collect their own taxes.

While Colorado's home rule system is unique, the problem of accurately determining a tax situs--the legal jurisdiction of a tax transaction--is not. Many types of taxes include a geographic element: Where is income earned? Where are goods and services sold? Where is property located or used? Consequently, validation of tax situsing is a key audit procedure, especially when tax transactions are occurring both within and outside of a jurisdiction.

Exhibit I: Selecting Data Based on Spatial Relationships

Point, line, and polygon features
represent real-world objects, events
and activities on a map. Features can
be selected based upon their spatial
relationship to one another. Selecting
features completely within area Z
would select point A. Area Z itself
could be selected by querying features
that intersect line I. Choosing features
within a distance of area Z might
select point B but not point C.

[ILLUSTRATION OMITTED]

The primary objective of a tax audit is to confirm the correctness of returns. Audit procedures differ depending on variables such as the tax type being audited, the nature of the taxpayer's operations, and size of the taxpayer. Most Westminster audits focus on sales and use taxes, given their prominence as a revenue source. …