Changes to USPAP 2018-2019 | Part I: Understanding the data
In his Origin of Species, Charles Darwin states, “It is not the most intellectual of the species that survives; it is not the strongest that survives; but the species that survives is the one that is able to adapt to and to adjust best to the changing environment in which it finds itself.” Every two years, an appraiser’s ability to adapt and adjust is tested when another version of USPAP is minted. 2018 will be no different.
For most appraisers, one of the most important aspects of USPAP 2018-2019 will be Advisory Opinion 37 (AO-37), which specifically addresses the multitude of analytics tools appraisers now have at their disposal.
First, a little background: in 1997, the Appraisal Standard Board (ASB) ratified Advisory Opinion 18, which addressed the use of tools known as Automated Valuation Models (AVMs). Back then, some appraisers regarded AVMs as a dystopian danger to the industry, while other appraisers embraced AVMs as a useful arrow in their quiver of valuation tools. A careful reading of AO-18 demarcated how AVMs could be used in the valuation process. Specifically, it cautioned that AVMs could be used in the context of an appraisal, but the AVM output, standing alone, could not be considered an appraisal.
Now, fast forward twenty years, and appraisers find themselves viewing a new valuation landscape. New regulations have not replaced AO-18 because it still offers specific, useful guidance about AVMs, but a new world of valuation tools has created a need for additional regulatory guidance moving forward. This change reflects the movement from an artisanal to a scientific approach in the appraisal industry. The newest regulation, AO-37, will stipulate which requirements and cautions appraisers must heed when using the next generation of appraisal analytic tools.
The main thrust of AO-37 is that computer assisted valuation tools can be wielded poorly or properly. If wielded poorly, the appraisal report will lack credibility and the appraiser’s license will be in peril. Conversely, if the next generation of appraisal analytics tools are wielded properly, the appraisal report will have credibility and statistical support that were nearly impossible only a decade ago. To use analytics tools properly, it is the responsibility of the appraiser to understand the assignment conditions and appropriately make use of the necessary tools. Once the assignment conditions are understood, the data must be correctly selected and reviewed (or scrubbed).
An example of scrubbing: a few years ago, I found that a regression model was not returning appropriate values for the land or the improvements. The problem was not my computer or the model, but the data. I had failed to scrub the data appropriately — a Realtor had incorrectly put in a lot size of 192,000 acres instead of 1.92 acres. One problem with the data and the whole process was sabotaged! AO-37, speaking to this issue, states, “Regardless of the tool chosen, the appraiser is responsible for the entire analysis including the controlling input, the calculations, and the resulting output.” Here, controlling the input is necessary to understanding the economics of the market area (a model). Without looking for erroneous sales data (scrubbing), the resulting model, analysis, and appraisal will be misleading.
Some computer assisted valuation tools opaquely output numbers without allowing the appraiser to see sales data.. HouseCanary Appraiser provides a downloadable spreadsheet of the MLS data chosen by the appraiser in the form of an Excel CSV (comma separated values). Conveniently, this not only meets the USPAP requirement to control the input, but also adds transparency and assists with work file record keeping requirements.
We hope this has been helpful in learning about the latest data scrubbing requirements. Interested in learning more about regulatory changes? Check out the second of our three-part series, a guide on choosing the right data for your appraisals.