The post In Search of “Perfect” Adjustment Factors appeared first on Appraisal Buzz.

]]>I think we would all agree that the accuracy of an appraisal is primarily dependent upon the selection of the best comparables. The Sales Comparison Approach uses each of the selected comparables as a starting point and then adjusts the actual sales price for differences in date of sale, sales concessions, site size, location, view, condition, quality, condition, gross living area (GLA), bath count, parking, porch/patio/deck configuration, fireplaces, swimming pools and other features impacting value.

The more similar the comparables are to the subject, the fewer adjustments will be required. And the adjustments will be smaller as well. Adjustments are calculated by multiplying an adjustment factor times the quantity difference between the subject and comparable. For example, if the GLA for the subject is 2200 sq ft and for a comparable, 2000 sq ft, the difference, 200 sq ft would be multiplied by the adjustment factor. If the GLA adjustment factor was determined to be $60/sq ft, the adjustment would be $60/sq ft x 200 sq ft = $12,000 added to the comparables sales price. Obviously, the smaller the difference between GLAs, the smaller the adjustment, regardless of the magnitude of the adjustment factor. So, the greater the similarity between the subject and comparables, the smaller the impact of a too high or too low adjustment factor.

In selecting the comparables to be used in the sales comparison grid, it is also important to “bracket” the high impact variables like GLA, condition and site area. By selecting comparables with variables that are higher and lower than the subject, placing the subject towards the middle, the positive and negative adjustments will tend to cancel each other out. This also reduces the impact of using adjustment factors that are too high or too low.

My point is, that while it is important to use reasonable adjustment factors, If I were to follow the current push toward comprehensive detailed support for every adjustment factor used for every appraisal, it would increase the amount of time spent to the point that my clients simply wouldn’t pay for it.

In addition to good comparable selection, what’s the solution? Put time and effort into high impact parameters and use low impact adjustment factors that are judged to be reasonable, resulting from previous analysis; matched pair or regression.

But what happens if you use a “wrong” adjustment factor? The impact will probably be minimal as long as the comparables are reasonably similar to the Subject and bracket the Subject’s high impact parameters.

The following “Sales Comparison Sensitivity Illustration” spreadsheet illustrates the point. The hypothetical sales comparison grid contains five comparables along with the subject, adjustment factors and adjustments. The comparable sales prices have been adjusted so the comparable adjusted values are all equal at $320,000. That makes the adjustment factors “perfect”.

The grid titled “Results w/ Various Adjustment Factors Applied” shows what happens if too high or too low “imperfect” adjustment factors (in red) are used. Case #1 uses the “perfect” adjustment factors for every parameter but GLA. In this case, the GLA factor used is 30% of the average sales price per square foot or $44.24/sq ft compared to the perfect adjustment factor of $73.74/sq ft (50%). When this is done, the weighted adjusted value of the five comparables is $319,000 which is 99.7% of the correct $320,000. Not so bad for such a big “mistake” with a high impact parameter!

Case #2 substitutes nine adjustment factors and leaves the GLA perfect adjustment factor. The result is a weighted adjusted value of $324,000 or 101.3% of $320,000. Cases #3, #4 and #5 similarly show the impacts of various mixes of perfect and imperfect adjustment factors on the weighted adjusted value. The range of the five cases is $319,000 to $324,000 or approximately 99.7% to 101.3% of $320,000.

The conclusion to be drawn from this example is that the use of imperfect adjustment factors does not necessarily compromise the validity of the results. This is not to minimize the importance of using reasonable adjustment factors. It is to point out the fact that good comparable selection – including bracketing – can go a long way to mitigate the adverse impact of the use of imperfect adjustment factors. This should be kept in mind when prioritizing limited resources.

The post In Search of “Perfect” Adjustment Factors appeared first on Appraisal Buzz.

]]>The post A Spreadsheet Solution for Land Value Extraction appeared first on Appraisal Buzz.

]]>But, what is a “reasonable” number of comparable sales? I normally look for at least six comparable sales to consider by analyzing lot sale price per square foot vs lot size, then use the power function to calculate the relationship between the two. The resulting formula is then used to calculate the subject’s estimated land value. As an additional benefit, the formula can be used to calculate lot size adjustments to the comparable improved property sales included in the sales comparison grid.

Unfortunately, there are not always enough comparable sales of vacant land to permit the use of the sales comparison approach. When this happens, a possible solution is to use the extraction method.

Using this approach, improved properties with lots comparable to the subject’s lot are identified. In order to complete this step, the value of the improvements is calculated using the cost approach to estimate the replacement cost new. Using effective age and remaining economic life estimates to calculate the depreciated value of the improvements, then subtracting the depreciated value of the improvements from the sales price will produce a viable land value. This is done for several properties to produce the relationship between lot value per square foot and lot size as described above for the purpose of calculating the subject’s land value and lot size adjustments used in the sales comparison grid.

There is a lot of “estimating” involved with this approach, so it works best when the value of the improvements are a low percentage of the sales price so that inaccuracies with the estimates have minimal impact on the lot value estimate. This strategy includes selecting properties with small, older homes that have not been remodeled and with minimal site improvements. Of course, the sales should be recent and in the subject’s neighborhood and the lots should be as similar (views, topography etc.) to the subject’s lot as possible.

The first step in the process is to search MLS using the developed criteria to produce a list of potential improved property sales that contains information such as bed and bath counts, year built, garage spaces, GLA, sales price and sale date. Using this information, the next step is to select the “best” sales to use for the statistical analysis. At least six should be selected. The selected sales can, then be imported into an Excel spreadsheet similar to that shown for calculating their lot value and lot value per square foot.

The MLS sheets and their descriptions and photographs should be reviewed to determine the condition of the improvements. Based upon the condition rating, age and total economic life of the improvements, the spreadsheet can be set to calculate the effective age, remaining economic life and percent depreciation. The cost new is calculated based upon GLA, garage spaces, and unit prices. The cost new is reduced by the calculated amount of depreciation to produce the depreciated improvement value.

The improved sales prices are adjusted for date of sale, based upon the effective date of the appraisal, the estimated monthly price trend percentage for existing homes in the subjects market area, and the comparables’ dates of sale. For each of the selected property sales, the estimated lot value (Lot Val) is calculated as the difference between the adjusted improved sales price (Adj S$) and the depreciated value (Dep Val) of the improvements: Lot Val = Adj S$ – Dep Val

Finally, this information is used to estimate the value of the subject lot and the adjustments to the lot sizes for the comparable properties on the sales comparison grid. In the attached example, the Lot Value/SQFT and the Lot Acres are plotted and analyzed using regression analysis with the Power trend line. The equation is shown on the graph along with the square of the correlation coefficient. In this example, the equation is y = 7.476x^{-.650} where y is the lot value/sqft and x is the lot size in acres. ^{ }The Subject’s lot size is .175 acres: y = 7.476 x .175^-.650 = $23.215/sqft. $23.215/sqft x .175 acres x 43560 sqft/acre = $176,966.

In my experience, the exponent normally falls between -.4 and -.9 with a minimum R2 of about 0.3. If there are “outliers” impacting these values, they should be removed. This process, if used sparingly, should produce reasonable coefficients and results. In summary, if it is necessary to use the Extraction method, using the process outlined above will help to produce credible and defensible results.

The post A Spreadsheet Solution for Land Value Extraction appeared first on Appraisal Buzz.

]]>