"What is the property worth?" is probably the most frequently asked question in the real estate industry. The question is fundamental regardless of whether an individual, household or organization is thinking about buying, selling, renting, mortgaging, brokering, managing, taxing (or whatever) a property. Accordingly, if one were to pick just one facet of the real estate industry for supplying or using geographic software, hardware, data or services, property valuation would be a logical choice.

Comprehensive, well-defined valuation professions have of course evolved in both the private sector ("appraisers") and public sector ("assessors"). Traditionally, properties have been valued using three methods: comparable sales, income-based, and replacement cost. In recent years a fourth method has been gaining acceptance: statistical extrapolation, frequently called computer-aided mass appraisal (CAMA). At the risk of just "skimming the surface of the tip of the iceberg, " in this column I will delineate GIS tools and data relevant to each of the four methods. (Also at the risk of misapplying professional designations, I will use the term "appraiser" to encompass anyone valuing real estate, whether in the private or public sector, and whether certified or not.)

Comparable sales is the most widely used method, especially for residential properties. The appraiser identifies recently sold properties (the "comparables" or simply "comps") similar in type and location to the property to be valued (the "subject"). The appraiser then adjusts the selling price of each comp upward or downward to reflect differences between the comp's and the subject's size, age, location, quality of construction, vacancy rate (if a commercial property), and on and on. The appraiser then averages the adjusted sales prices of the comps to obtain an estimate of the subject's value.

GIS software and data can be very useful in identifying appropriate comps. An appraiser can display a map showing all recent sales. The appraiser can then "click on" each sold property to show its attributes, to determine whether the property is sufficiently similar -- as well as spatially close enough to -- the subject to be a reliable comp. The appraiser might select different buildings and calculate new valuation scenarios. Once the best comps have been chosen and the valuation completed, the appraiser can readily produce a hardcopy map showing the location of the comps and subject, attributes of the properties, neighborhood characteristics (such as census data), etc., to append to the final appraisal document.

To create the requisite comps data base, the appraiser can input proprietary data from past appraisals and/or purchase third-party data from "comps services" operating in most communities; shown below is an example of relevant comps retrieved from a vendor's data base and automatically mapped on the appraiser's monitor. [Insert DataQuick graphic.]

Income-based and replacement cost valuations can be conducted in a similar manner. In the income-based approach the appraiser first forecasts a commercial property's pre-tax earnings ("net operating income" or NOI) from assumptions on rents, vacancy rates, miscellaneous income, operating expenses, etc., for one year, five years, or even further into the future. The actual valuation then consists of applying one or more coefficients (e.g., gross rent multiplier, capitalization rate) to the first year NOI, and/or calculating the present value of several years' NOI plus the present value of an assumed sale of the property in the final forecast year. In the replacement cost approach, the appraiser determines the cost of reconstructing the current residential or commercial building(s) at the given location; the logic is that no purchaser would pay more for the land plus existing improvements than the purchaser would have to pay for the land plus brand new buildings.

As with the comparable sales method, GIS can again be used to identify properties similar to the subject. The average rents, operating expenses, construction costs, and so on of those buildings can then be entered into the subject's income-based and replacement cost pro formas. Unlike the comparables sales method, though, the appraiser need not have the information available for individual buildings. Data on average rents, operating expenses, etc. for the subject's submarket (e.g., the central business district), entire city, or even metropolitan area might be sufficient. Also in contrast to the comparables sales method, the income-based method and (to a lesser extent) the replacement cost method are dependent on geographic analyses of population and employment growth rates, income levels, zoning restrictions, construction permits, and similar demand-side and supply-side "data layers" essential to forecasting the pro forma inputs. Consequently, in the final documents, income-based and cost replacement appraisals are likely to contain more computer-generated thematic maps than comparable sales appraisals.

CAMA-type valuations are useful when many appraisals must be conducted quickly and inexpensively. Prerequisites include a statistically sufficient number of comps, sophisticated software, and highly trained personnel -- all of which increase in proportion to the heterogeneity of the properties to be valued. Essentially, the appraiser runs statistical routines such as regression analysis to determine the average portion of selling price (dependent variable) that can be explained by size, age, construction quality, etc. (independent variables) in a given sample of properties; the appraiser then applies those regression coefficients to the characteristics of a subject property to estimate its probable value. Property tax assessors use this valuation method the most. The method is receiving increased attention, however, from the holders of large real estate portfolios. A notable example are banks, which need to monitor the collateral values underlying their mortgage portfolios, to predict which credits might become problematic and to ensure sufficient capital reserves.

Geographic analyses bolster traditional statistical routines in at least three ways. First, by mapping the locations of the properties in a proposed statistical sample and also by displaying other data layers (e.g., census data, recognized neighborhood boundaries), the appraiser can select the best statistical sample. Second, the appraiser can introduce geographic variables into the regression equation; examples include distances to key transportation facilities, shopping malls, schools, and earthquake faults. Third, the appraiser can simultaneously display the properties with known values and the properties with estimated values, to visually check for anomalies.

A final comment on the utility of GIS in property valuation: Regardless of whether an appraiser is using one, several, or all of the above methods, significant dollars are always at stake. Accordingly, the cost of implementing and maintaining a GIS can easily be recaptured in even one transaction -- for example, an institutional investor discovering that an office building can be purchased for far less than the initial asking price, or a bank officer identifying a problematic loan in time to work with the borrower to avoid (expensive) foreclosure proceedings.