Site size adjustments: a technique to estimate the adjustment magnitude.
Site size adjustment has received relatively little attention in the appraisal literature, When it has been discussed, the major point has been the adjustment technique, which consists of finding two comparable properties of relatively similar size to the subject property and using them to make a linear estimate of the site size adjustment. A graphing concept for the site size adjustment has also been mentioned in the literature, but the discussion of this technique has been solely conceptual. This article presents a practical illustration of the graphing technique to estimate a site size adjustment.
The topic of site size adjustment has two related perspectives into which this article has been divided: the conceptual or theoretical consideration and the practical application of those factors considered in theory.
Site Size Adjustment Concepts
Important information on site size adjustment is discussed by James Boykin in his article and book on the topic. (1) In these works, Boykin illustrates two techniques to estimate a size of site adjustment. Both techniques focus on the analysis of vacant land or vacant sites. The first technique determines the difference in sale price per square foot between two otherwise comparable properties of different site size. (2) The second technique, graphing and curve plotting, receives a relatively short and highly general discussion in Boykin's book on land valuation. (3) The expanded discussion of this technique, which is needed in appraisal literature, is provided in this analysis.
Substantive Comments Regarding Site Size Adjustments
Boykin's article (4) is an excellent primer on the many issues involved in analyzing differences in site size among comparable properties. In that article, Boykin offers the following substantive comments on site size:
Parcels of different sizes generally sell for markedly different unit prices. Thus, their use is limited in measuring the value of appraised parcels of either smaller or larger size. ... there are distinctly different submarkets for various land use activities. In turn, a party interested in a particular land use generally does not consider parcels of distinctly different sizes as viable alternatives, even though they may be in the same vicinity and have the same zoning classification. ... there is a range of sizes within each of the various land submarkets where buyers and sellers would view parcels of slightly different sizes as reasonable substitutes. It is outside of these size limits that parcels are no longer considered comparable by investors, and should be so regarded by real estate appraisers. A parcel may be in close proximity, have the same zoning, similar terrain, and similar access features, but, because of its variance in size, have a different highest and best use and unit value. ... the use of markedly different-size comparable land sales is an unreliable method of estimating the value of an appraised property--unless reasoned size adjustments can be demonstrated. (5)
Boykin further notes that real estate valuation literature up to that time included very few articles "on the hazards of attempting to use markedly different-size land sales to support the estimated value of an appraised parcel or on methods for making size adjustments," and that "The topics of variability of land prices according to size or the difficulty of analyzing comparable land sales that differ in size from an appraised parcel have received cursory treatment in appraisal literature." (6)
Boykin also provides several substantive comments made by other authors on the topic of site size, including the following:
Price also is influenced by the problems that added plot size provokes, namely, higher unit building costs with very large buildings...more speculation on the absorption of the space by the market, higher capital investment requirements, and higher proportionate carrying costs. (7) ... plottage can be a negative factor when the tract under appraisal is larger than the optimum configuration for the prevailing pattern of utilization. (8) Size is generally a less important element of comparison than date and location. Most types of development have an optimal site size; if the site is larger, the value of the excess land tends to decline at an accelerating rate. Because sales of different sizes may have different unit prices, appraisers ordinarily give more weight to comparables that are approximately the same size as the subject property. (9) ... increased size decreases unit price; ... (but) the decrease is not in full proportion; that is, doubling size does not halve unit price. (10) ... there will be a geometric relationship between size and per-acre price. (11)
These comments and statements lead to the following discussion of the prevailing site size adjustments for vacant land or sites.
The Derived Per-Unit Adjustment and Interpolation
Boykin's article provides the comparative illustration displayed in Table 1. The technique simply finds the difference between two sites of different size that are comparable to each other in aspects other than size. They are at least comparable in all of the major aspects, which must include time of sale and location, and should include other aspects, such as zoning, shape of the site, topography, access, utilities, etc. Since only two points are used in the analysis, the analysis is linear in nature. This is a shortcoming that was pointed out by both Boykin and Dilmore.
An inspection of the data in Table 1 reveals several differences between the two sites.
* The difference in acres between the two sites is 14.16 acres (15.27 - 1.11);
* The difference in sale price per acre is $111,949;
* The difference in sale price per square foot is $2.57;
* The difference in sale price per acre divided by the difference in the acreage is $7,906 ($111,949/ 14.16); and
* The difference in sale price per square foot divided by the difference in square footage is $0.0000042 ($2.57/(14.16 x 43,560)).
These figures allow an estimate that puts the subject property of 10.0 acres in a position on a linear relationship between 1.11 acres and 15.27 acres. The estimate can start at either end of the range. Relating the 10.0-acre subject property to the 15.27-acre site, the difference in acreage is 5.27 acres. The per-acre sale price adjustment is $41,665 per acre of difference ($7,906 per acre x 5.27 acres) and $0.956 per square foot of difference ($0.0000042 per square foot x 43,560 square feet per acre x 5.27 acres). The adjusted price per acre is $293,877 ($252,212 + $41,665) and the adjusted price per square foot is $6.75 ($5.79 + $0.96).
An example of the same calculation from the other end of the range also appears in Table 1. It reveals that the size adjustment from 1.11 acres to 10.0 acres is -$1.61 per square foot. This adjustment generates the same $6.75 adjusted price per square foot.
A simple arithmetic proportion provides the same numerical result once the differences are calculated. Consider that the difference between 10.0 acres and 15.27 acres is 5.27 acres, that 5.27 acres is 37.22% of the difference in acreage (5.27/14.16), and that 37.22% of $2.57 is $0.956. Therefore, $5.79 + $0.96 = $6.75.
Using two points makes the analysis linear, but Boykin and Dilmore say the relationship is not linear. Boykin provides a very simple but convenient example of the nonlinear relationship in his book. (12) He shows a best fit line to a set of points on a site size versus price per acre graph. At an unspecified but small acre site, the price per acre is $100. As the acreage increases, the price per acre drops rapidly between one and nine acres, and then slowly between nine and 37 acres. Boykin refers to the shape of the curve as a "hockey stick"; this curve is shown in Figure 1.
[FIGURE 1 OMITTED]
Several observations can be made about this curve. On the positive side, it exhibits the general size of site relationship that is found in the real world. On the negative side, its position and exact shape represent a general relationship that cannot be applied to any specific real estate market. The rate of change can, and more than likely will, differ across spatial and temporal markets. The position of the curve in the coordinate system will also differ across spatial and temporal markets.
On a practical level, the discussion of this curve in the literature in general and in the Boykin publications specifically is not very instructive with regard to what is being plotted and displayed. Hence, a practical application will be addressed.
Relevant Conceptual Considerations Regarding the Best Curve Fit Technique
Several important points need to be raised regarding this best fit curve technique for site size adjustment.
* Boykin's best fit curve or "hockey stick" has a kink at the smaller site size due to site utility. This kink is an important concept to understand. To the left of the kink, smaller sites have less utility; few financially feasible uses exist for a very small site and thus demand is less. To the left of the kink, as site size increases, an upward adjustment is called for if the comparable is so small as to lack utility. To the right of the kink, utility can continue to increase as more financially feasible uses fit on the site but the per-unit price declines as the site size increases. The decline in unit price occurs because the site may contain excess or surplus land not necessary for the use. The downward adjustments on a per-unit basis (acre or square foot) change as a smaller comparable's site size increases. To the right of the kink, the initial stage of decline in per-unit price is rapid, but the rate of change declines and the change becomes smaller within the relevant range of site sizes.
* The buyer might pay a premium to purchase a very small site for assemblage purposes, so an adjustment for conditions of sale or buyer motivation may need to be made. This premium can occur on both sides of the kink.
* From a theoretical standpoint, all adjustments should be made to comparables, except for the size adjustment, and then this adjusted unit sale price is placed on the graph to estimate the site size adjustment. The comparables would be adjusted to the subject for all characteristics, except for size, in an effort to control for all characteristics except for the effect of size on the sale prices. (From a practical perspective, adjusting the comparables for all characteristics may be difficult to accomplish.)
* The adjustment for all characteristics except for size would expand an important concept in appraisal theory. It would be an advanced application of the matched-pair method where two properties are made to be as similar as possible in order to isolate and quantify the effect on value of the unadjusted characteristic. Here, the array of properties is adjusted as a group and made as similar as possible in order to isolate and quantify the effect on value of the unadjusted characteristics.
A Practical Application
The Size of Site and Price per Unit of Comparison Relationship
The conceptual points that need to be made regarding this graphing technique to determine a size of site adjustment are identified and discussed in the illustration shown in Figures 2, 3, and 4. The illustration is taken from an actual vacant land valuation assignment.
[FIGURE 2 OMITTED]
Step 1: Identify the Location of the Subject Property. The subject property is a 4.5-acre site located on a major east-to-west arterial, Oak Road, approximately one-eighth of a mile east of the intersection with a north-to-south interstate highway. The site has 200 feet of frontage on Oak Road and is zoned commercial, which allows for office, retail, hotel, and multifamily uses. The uses can be high-to-moderate density.
Step 2: Identify the "General Neighborhood" for the Subject Property. The general neighborhood of the subject property was expressed as a one-mile radius around the subject property; this is depicted in Figure 2. Within this general neighborhood, recent sales with frontage on major arterial streets and with commercial zoning were obtained. These sales are displayed in an adjustment chart in Figure 3.
Step 3: Identify the "Immediate Neighborhood" of the Subject Property. The immediate neighborhood of the subject property is the set of adjacent or most-proximate properties to the subject property. These properties are the most comparable to the subject with regard to spatial location. The immediate neighborhood is delineated in Figure 2 as properties along, or very near, Oak Road within one-quarter mile of the subject site. Recently sold properties with commercial zoning are the most spatially comparable properties for the analysis.
Comparable properties from this immediate neighborhood should be used to make the site size adjustment. However, in this illustration, as in most situations, the number of comparables in the immediate neighborhood was too few to form the site size adjustment curve. Two points form a line; to form the curve, at least three observations are needed. This necessitated using sales from the general neighborhood. In this case, all recently sold properties in the general neighborhood with acreage in the 0.5-acre to 6.5-acre range were used. This range bracketed the 4.5-acre subject site. Even though sales of recently sold land parcels larger than seven acres existed, they were not used in the analysis because these larger sites differ too markedly from the subject.
Sites of less than three acres were included in the analysis for two reasons. First, they show the rapidly declining price per acre section of the curve between 1.2 acres and 3 acres. Second, site sizes of less than one acre were displayed to show the rapidly rising price per acre section of the curve for small site sizes and to show the kink in the best fit curve.
Therefore, comparable properties from the general neighborhood had to be used and adjusted for differences in the location in that general neighborhood. These properties were located in the following areas in the general neighborhood:
Area 1--Adjacent to the subject property on the west and directly across Oak Road
Area 2--Along Oak Road east of the interstate highway at and near the intersection of Oak and Ash Roads
Area 3--Along Oak Road west of the interstate highway at and near the intersection of Oak and Hickory Roads
Area 4--North of Oak Road on Elm Road
Area 5--South of Oak Road on Elm Road
Area 6--South of Oak Road on Hickory Road
Area 7--North of Oak Road on Hickory Road
All of these recent sales were within the one-mile general neighborhood and two were in the immediate neighborhood--one was west of the subject property at the intersection of Oak and Elm Roads, and the other was across Oak Road from the subject site. The relative rating of these areas as compared to the subject site is shown in Figure 3. Each of these areas is inferior to the subject property area but by differing amounts. This location adjustment could be made by applying the matched pairs technique, but the data did not allow its application in this illustration.
Step 4: Create the Adjustment Grid. Figure 3 displays the 23 properties located in the general neighborhood that sold recently and have commercial zoning (GC, CRC, OI, and RMF) that allows for higher density development. The properties are comparable on property rights conveyed (fee simple), condition of sale (arm's length), shape of the site (irregular), utilities, cash equivalency, and other factors. The properties differ in topography and the stage of site development (raw land versus rough graded). These adjustments are made in the sales comparison adjustment grid. Remember that the focus of this analysis is estimation of the site size adjustment. The properties are arrayed according to size in Figure 3.
Adjustments for time of sale and location in the general neighborhood are made in Figure 3. The period covered for the comparable properties is two to twenty-nine months. This period was judged to be acceptable due to the general stability of the major market factors of that time--steady population growth, stable interest rates, and slowly declining vacancy levels. The magnitude of the adjustment in the general neighborhood over the past three years is judged to be 10% per year, allocated on a monthly basis. The relative location was judged on an inspection of the commercial areas. Areas 2 through 7 were judged to be inferior to Area 1, the location of the subject property, based on the following criteria: frontage on the major arterial versus frontage on less significant streets in the neighborhood, and the volume of traffic to the various retail establishments in each area.
The resulting adjustments for site size are dependent upon the time and location adjustments that are made. The time and location adjustments must be supported by the best available market evidence in order to generate a high degree of reasonableness for the site size adjustment to follow. It may be necessary to view the site size adjustment technique as good general support for the size adjustment. As always, the expert judgment of the appraiser is the final authority.
After the time and location adjustments were made, the resulting adjusted sale prices per square foot were plotted and displayed in the graph in Figure 3. Two important points of interest are apparent in this graph. First, there is a kink in the curve at small lot sizes. Sale prices per square foot for parcels smaller than 1.2 acres rise as the parcel increases in size. This can occur if the parcel is too small to accommodate a variety of uses. For example, a 0.55-acre site is too small to accommodate a high-density office or apartment building. It is also too small to accommodate most freestanding retail establishments. There are fewer potential uses for the small site and thus, a lower per-unit price. At 1.2 acres, the site becomes large enough for most freestanding retail uses, small strip centers, and small office complexes. The demand increases for these sites that can accommodate more potential uses.
Second, the time- and location-adjusted data for site size versus price per unit relationship is not linear and is not a smooth curve. This nonlinear and variable relationship occurs because of variables that have not been handled in the analysis at this point. Only two important adjustments were made to the data: time of sale and location. There is no evidence to suggest that the curve is always variable, but there is a strong basis in logic that it is not linear. Nevertheless, the technique to determine the size adjustment relevant for this general neighborhood still has one more step.
Step 5: Smooth the Site Size versus Price per Unit Relationship in the Adjustment Grid. Figure 3 displays both the nonlinear and erratic pattern of unit sale prices and what Boykin called a "best fit curve" of these points. This best fit curve in Figure 3 is drawn using a purely visual technique. Since the best fit curve is nonlinear according to the existing literature, linear regression is inappropriate. The final task is to mimic the hand-drawn best fit curve by smoothing the cycles in the actual data. This is done in Figure 4 by manually adjusting the sale prices per square foot to form the best fit curve. The manual smoothing of the curve is used in this exposition because it is a visual process that can be inspected as the smoothing occurs and makes the most intuitive sense to a client. (13) The subject site is placed into the site size array at the appropriate place.
The best fit curve reveals the per square foot adjustment between pairs of the properties. For example, the 4.24-acre site would be adjusted by a negative $0.20 per square foot to make it comparable to the 4.5-acre subject property, and the 4.76-acre site would be adjusted by a positive $0.10 per square foot to make it comparable to the subject 4.5-acre property. The magnitude of other adjustments for additional comparable properties can be found in the chart associated with Figure 4.
The best fit curve is a result of the general neighborhood being analyzed and the time period being studied. A study at the nearest commercial area to the immediate area of the subject property in this study will have the same general shape for the best fit curve, but its specific shape will be different. The downward section could start at a larger site size, and the height on the curve in the graph and the slope of the curve could be different.
The best fit curve as shown in Figure 4 is a better representation of reality than the solution that would arise if only two sales were used to obtain the size adjustment. For example, what adjustment would arise using the "derived per unit adjustment and interpolation" as shown in Table 1 if the 1.2-acre and the 6.3-acre sites were used? The calculation is made in Table 2. The adjustment to the 6.3-acre site is +$5.60 per square foot and the adjustment for the 1.2-acre site is -$10.28 per square foot. Using the best fit curve as displayed in Figure 4, the adjustment to the 6.3-acre site is +$0.20 per square foot and the adjustment for the 1.2-acre site is -$14.88 per square fool These are very disparate results, both internal to each technique and between the two techniques. The guiding principle that stems from these results is not to use a wide range around the subject site. The narrower the range, the closer the two estimates for the adjustments should become. The next examples narrow the range to see what happens to the magnitude of the site size adjustments.
If the 3.9-acre site and the 5.1-acre site are selected to bracket the 4.5-acre subject property, the resulting adjustments as calculated in Table 3 are $0.57 per square foot, with the appropriate sign for each comparable. The adjustments for the best fit curve are +$0.20 per square foot for the 5.1-acre site and -$0.70 per square foot for the 3.9-acre site. The disparity is much smaller, both internal to each technique and between the two techniques, as the range narrows.
If the range is narrowed even more to 4.24 acres and 4.76 acres to bracket the 4.5-acre subject site, the resulting adjustments as calculated in Table 4 are $0.29 per square foot, with the appropriate sign for each comparable. The adjustments for the best fit curve are +$0.20 per square foot for the 4.24-acre site and -$0.10 per square foot for the 4.76-acre site. Again, the disparity is much smaller, both internal to each technique and between the two techniques, as the range narrows.
However, remember the nonlinearity and erratic nature of the original data. What would happen if circumstances did not provide data for the 4.24-acre site, but only provided the analyst with data for the 4.167-acre site and the 4.76-acre site? This is a narrow range bracketing the subject property, so on the surface it would make sense to conclude that the disparity should be very small, even negligible. In Table 5, the larger 4.76-acre site has a sale price per acre and per square foot that is greater than the sale price per acre of the smaller 4.167-acre site. This relationship is the opposite of the theoretical relationship. The adjustments in this case are theoretically nonsensical. The adjustment to the larger site of 4.76-acres is a negative $1.66 per square foot when it should be a positive adjustment to make it comparable to the 4.5-acre site. The adjustment to the smaller site of 4.176 acres is a positive $2.12 per square foot when it should be a negative adjustment to make it comparable to the 4.5-acre site. The manual smoothing of the large array of data shown in Figure 4 to generate the best fit curve eliminates this situation from occurring.
In addition, related to Tables 4 and 5 and the two sales examined in the tables, making adjustments for small size differences is not warranted, and such price adjustments do not occur in the typical real estate marketplace. It is only when the site sizes start to differ significantly that buyers and sellers adjust the sale price. The size increase or decrease must be large enough so that significantly fewer or significantly more buyers compete to purchase the property, thereby affecting the demand.
Consideration of the General and the Immediate Neighborhood
Neighborhood analysis is adequately discussed in the appraisal literature. The "general" neighborhood in this article is a term used specifically in this context. It is the spatial area adjacent and in close proximity to the subject property from which sales data for a wide array of reasonably comparable properties can be selected. These reasonably comparable properties will include the best comparables that can be found plus other less comparable properties. It provides the wider array of properties that can be used to generate a size adjustment. The size adjustment is not a value indicator.
The "immediate" neighborhood is the geographic area that is adjacent and in the most immediate proximity to the subject property. It is the geographic area that contains the properties that are most comparable based on location. Ideally, the best fit curve should be drawn from the immediate neighborhood, but realistically there will not be an adequate number of recently sold comparable properties in this geographic area.
Development and Use of the Best Fit Curve Adjustment Technique for Site Size
The application of the best fit curve is illustrated in Figures 5 and 4. The use of the technique requires that the appraiser develop these two spreadsheets. The spreadsheet software creates the charts easily and quickly from the tables. The creation of the tables is a straightforward process once the information for the sites in the general area (the market area) is obtained, and it should take approximately 30 to 60 minutes. Figure 3 starts with the selection of comparables on the basis of a reasonable range of site sizes bracketing the subject site size and the necessary zoning required for the analysis. In the search for comparable properties, information on 23 properties was gathered from a site size of 0.55 acres to 6.3 acres, bracketing the 4.5-acre site. Each of these properties had a form of commercial zoning (office, hotel, and retail) or multifamily zoning. For this analysis, 23 properties was a reasonable number to choose because the case was being prepared for an eminent domain trial. In a typical assignment, a smaller number of properties would be reasonable, say ten but not two.
The zoning, the site size, the sale price, and the date of sale are entered into the spreadsheet. Sale price per acre and per square foot are calculated. Time and location adjustments are made to the sale price per acre and per square foot. The time- and location-adjusted sale price per square foot is selected as the unit of comparison.
Next, a second two-column spreadsheet is created, showing the time- and location-adjusted sale price per square foot and the site size in ascending order. This data is then used to generate the chart. The pattern in the chart is visually analyzed and the best fit curve is drawn. Since the curve is nonlinear, regression analysis is not applicable.
In Figure 4, the time- and location-adjusted sale price per square foot for each comparable property is further adjusted to make the pattern of actual data points conform to the best fit curve. When these best fit adjustments are completed, the site size adjustment can be read from the chart as the difference in best fit adjusted sale price per square foot between the comparable property and the subject property.
A number of issues arise regarding the application of this best fit curve technique. These issues will affect the results of the technique.
Appropriate Data. The technique requires high-quality comparable properties. The appraiser should be comfortable with the comparable properties selected. Time of sale and location were the specific adjustments made for purposes of illustration in this article. In other assignments, a specific adjustment might have to be made for zoning differences or other elements of comparison.
Sufficient Data. The adequate number of comparable properties is difficult to quantify and is left to the appraiser's best judgment. Two comparable properties are too few; three properties can create a curve. Six to ten high-quality comparable properties should be sufficient.
Area of Analysis. The geographic area from which the comparable properties should be drawn needs to reflect the locational or situs environment that affects the subject property. For this analysis, a sufficient number of comparable property sales were available adjacent to the subject property and in very close proximity, constituting the immediate neighborhood. The general neighborhood used in the analysis was a wider geographic area surrounding the subject property but still under the influence of the same general economic and physical environment. The immediate neighborhood was a distance of one-eighth mile to each side of the subject property and the area directly across the street. The general neighborhood was an irregular shape best described as an oval with a radius of approximately one-half mile. As a guideline, the sufficient number of comparable properties should be as physically proximate to the subject property as possible.
The Range of Site Sizes. The site sizes of the comparable properties need to be as close to the subject property size as possible. In an ideal assignment, the comparable properties would be the same size as the subject property. In reality, the range of the comparable properties is the controllable variable and should be as small as possible. The subject property should be the median property in the range; three larger sites and three smaller sites could be sufficient if they are the best of the comparable properties.
This article presents a discussion of the size of the site adjustment. The literature on the subject is reviewed and the technique of bracketing the subject property size with a smaller and a larger comparable property is considered. The analysis illustrates this technique as part of the literature review. Then, a technique illustrated by Boykin for estimating a size of site adjustment is presented. Boykin discussed the issue of the best fit curve as a size adjustment estimator in general terms, but not in specific terms. The literature needs an explicit example of how this best fit curve technique can be used.
A practical illustration of the size of the site adjustment is presented along with an application of the Boykin best fit curve technique using a real world example. The article then compares the results from the best fit curve technique to a series of illustrations using a bracketing procedure around the subject property size and concludes that the best fit curve technique is a better technique for estimating a site size adjustment.
(1.) James H. Boykin, "Impropriety of Using Dissimilar-Size Comparable Land Sales," The Appraisal Journal (July 1996): 310-318; James H. Boykin, Land Valuation: Adjustment Procedures and Assignments (Chicago: Appraisal Institute, 2001), 49-53.
(2.) Boykin, Land Valuation; and Boykin, "Impropriety of Using Dissimilar-Size Comparable Land Sales."
(3.) Boykin, Land Valuation, 52.
(4.) Boykin, "Impropriety of Using Dissimilar-Size Comparable Land Sales."
(5.) Ibid., 310-311, 316.
(6.) Ibid,. 311.
(7.) Ibid., Boykin quoting John R. White and R. Gary Barth, "Land Market Comparison Techniques in High-Density Urban Areas," The Appraisal Journal (October 1974): 504.
(8.) Ibid., Boykin quoting White and Barth, 506.
(9.) Ibid., 312, Boykin quoting Appraisal Institute, The Appraisal of Real Estate, 10th ed. (Chicago: Appraisal Institute, 1992), 303. The Appraisal of Real Estate, 12th ed. states that "Generally, as size increases, unit prices decrease. Conversely, as size decreases, unit prices increase." Appraisal Institute, The Appraisal of Real Estate, 12th ed. (Chicago: Appraisal Institute, 2001), 196.
(10.) Ibid., Boykin quoting Gene Dilmore, "Size Adjustment Tables," The Real Estate Appraiser (May-June, 1976): 23.
(11.) Ibid., Boykin referencing Robert C. Suter, "The Sales Comparison Approach versus the Market Data Approach to Farm Real Estate Values," The Real Estate Appraiser and Analyst (Fall 1983): 28.
(12.) Boykin, Land Valuation, 52.
(13.) Several business function calculators and Microsoft Excel software are able to estimate a best fit curve for such data. The HP Business Consultant is an example of such a calculator. However, the final product from these mathematical and statistical programs will generate estimates that are different from the visual, manual technique used in this article, and will differ among themselves. This occurs because the results depend upon how these programs handle data. Linear regression, LINEST, and SLOPE handle the data as if it were a straight line, which is not the case at small lot sizes. Programs, such as LOGEST, LN, and EXPONENT, handle the data as if it were nonlinear, but they convert the data in a log-linear format. In all of these situations, the results will depend on the "constant" and the "slope" used in the analysis, both of which can be affected by the initial lot size selected. Therefore, appraiser judgment is the important determining factor in the use of a technique.
Joseph S. Rabianski, PhD, CRE, is a professor of Real Estate at Georgia State University in Atlanta. He has taught graduate and undergraduate courses in real estate appraisal and market analysis for 30 years. Rabianski is coauthor of the Appraisal Institute's first Market Analysis course developed in 1982 and coauthor of the book, Shopping Center Appraisal and Analysis, published by the Appraisal Institute in 1993. Many of his articles have appeared in The Appraisal Journal. Contact: Department of Real Estate, Georgia State University, P.O. Box 4020, Atlanta, GA 30302-4020; T 404-651-4609; F 404-651-3396; E-mail: email@example.com
Figure 3 Adjusted Site Size Dollars per Square Foot, Neighborhood Sales Acres $ per Sq. Ft. 0.550 $12.10 0.949 $14.18 1.200 $22.68 1.300 $22.02 1.400 $21.90 1.750 $20.68 1.990 $12.18 2.110 $15.78 2.300 $8.75 2.700 $13.06 3.334 $11.29 3.480 $10.18 3.650 $6.95 3.900 $10.53 4.010 $11.13 4.167 $6.81 4.240 $11.17 4.760 $10.59 5.100 $9.40 5.400 $4.97 5.650 $5.99 5.950 $7.08 6.300 $6.80 Data available from author upon request. Figure 4 Manually Adjusted Site Size Dollars per Square Foot, Neighborhood Sales Acres $ per Sq. Ft. 0.550 $12.10 0.949 $14.18 1.200 $22.68 1.300 $22.02 1.400 $22.00 1.750 $16.00 1.990 $12.18 2.110 $12.00 2.700 $9.80 3.480 $8.80 3.650 $8.50 3.900 $8.40 4.010 $8.30 4.167 $8.10 4.240 $8.00 4.500 * $7.80 4.760 $7.70 5.100 $7.60 5.400 $7.60 5.650 $7.50 5.950 $7.50 6.300 $7.40 * The subject site is placed into the array at the appropriate place in the site size array. Table 1 The Derived Per-Unit Adjustment and Interpolation Sale Price Comparable Sale Price per Square Property Acres Sale Price per Acre Foot A 1.11 $404,219.00 $364,161.60 $8.36 B 15.27 $3,851,283.00 $252,212.40 $5.79 Comp A less Comp B -14.16 $(3,447,064.00) $111,949.20 $2.57 Sale Price Change per Unit of Comparison $7,906.02 $0.0000041666 Subject Property 10.0 Adjustment Adjustment per Acre per Square Foot Comp B less Subject Property 5.27 $41,664.71 $0.956490 Comp A less Subject Property -8.9 $(70,284.49) $(1.61) Sale Price Sale Price per Acre per Square Foot Sale Price per Unit Adjusting Comp B $293,877.11 $6.75 Sale Price per Unit Adjusting Comp A $293,877.11 $6.75 Table 2 The Derived Per-Unit Adjustment and Interpolation: 1.2 Acres to 6.3 Acres Sale Sale Comparable Price per Price per Property Acres Sale Price Acre Square Foot A 1.20 $1,185,529 $987,941 $22.68 B 6.30 $1,866,110 $296,208 $6.80 Comp A less Comp B -5.1 $(680,581) $691,733 $15.88 Sale Price Change $135,634 $0.0000715 per Unit of Comparison Subject Property 4.5 Adjustment per Adjustment per Acre Square Foot Comp B less Subject Property 1.80 $244,141 $5.60 Comp A less Subject Property -3.3 $(447,592) $(10.28) The subject property is $22.68 - $10.28 = $12.40 when compared to Comparable A. The subject property is $6.80 + $5.60 = $12.40 when compared to Comparable B. Table 3 The Derived Per-Unit Adjustment and Interpolation: 3.9 Acres to 5.1 Acres Comparable Sale Price per Sale Price per Property Acres Sale Price Acre Square Foot A 3.9 $1,788,879 $458,687 $10.53 B 5.1 $2,088,266 $409,464 $9.40 Comp A less Comp B -1.2 $(299,388) $49,223 $1.13 Sale Price Change $41,019 $0.0000216 per Unit of Comparison Subject Property 4.5 Adjustment per Adjustment per Acre Square Foot Comp B less Subject 0.60 $24,611 $0.56 Property Comp A less Subject -0.60 $(24,611) $(0.57) Property Table 4 The Derived Per-Unit Adjustment and Interpolation: 4.24 Acres to 4.76 Acres Sale Price per Sale Price per Comparable Acres Sale Price Acre Square Foot Property 4.24 $2,063,036 $486,565 $11.17 A 4.76 $2,195,790 $461,300 $10.59 B -0.52 $(132,753) $25,265 $0.58 Sale Price Change per Unit of Comparison $48,586 $0.0000256 Subject Property 4.50 Adjustment per Adjustment per Acre Square Foot Comp B less Subject Property 0.26 $12,632 $0.29 Comp A less Subject Property -0.26 $(12,632) $(0.29) Table 5 The Derived Per-Unit Adjustment and Interpolation: 4.167 Acres to 4.760 Acres Comparable Sale Price per Sale Price per Property Acres Sale Price Acre Square Foot A 4.167 $1,236,114 $296,644 $6.81 B 4.760 $2,195,790 $461,300 $10.59 Comp A less Comp B -0.593 $(959,676) $(164,657) $(3.78) Sale Price Change per Unit of Comparison $(277,667) $(0.0001463) Subject Property 4.5 Adjustment per Adjustment per Acre Square Foot Comp B less Subject Property 0.26 $(72,194) $(1.66) Comp A less Subject Property 0.33 $92,463 $2.12 Note: In this case, the adjustments are theoretically incorrect. The adjustment to comparable Property B should be positive for a property larger than the subject property, but it calculates as a negative adjustment. This same sign reversal occurs for the adjustment to comparable Property A. It should be negative, but calculates as a positive adjustment.
|Printer friendly Cite/link Email Feedback|
|Author:||Rabianski, Joseph S.|
|Date:||Sep 22, 2005|
|Previous Article:||The effects of a defective construction lawsuit on the value of condominiums: a case study.|
|Next Article:||Quantifying net income attributable to business intangibles for a hotel-operated full-service day spa.|