USPAP (Uniform Standards of Professional Appraisal Practice) ethics rule specifically states that “An appraiser must not perform an assignment with bias.”  It defines Bias as “a preference or inclination that precludes an appraiser’s impartiality, independence, or objectivity in an assignment.”

As we noted in a prior Analogue Blog, there are two types of bias:  personal bias, and analytical bias.  It appears that the USPAP focus is on the appraiser, rather than the analysis.  However, the next reference to bias is in the “Scope of Work” rule, where it prohibits the client’s use or objectives to bias the “assignment results.”

So it appears that bias is prohibited for both the appraiser’s behavior, and in the results of the appraisal.

In reading the first (relevant) 24 pages of the actual standards for real property appraisal, I find no other mention or instructions on how an appraiser is to avoid conscious or unconscious personal bias.  So the rest of our inquiry must focus on whether USPAP condones, enables, or helps prevent analytic bias.

An area of concern is regarding the “acceptability” of the scope of work.

The key requirement is that the research and analyses are credible (worthy of belief).  The believability is judged by two things:  1) what users expect, and; 2) what appraisers’ peers do.  Herein lies problem #1.  If clients expect bias, and other appraisers respect this bias, then this standard permits, or even requires “commonly accepted” levels of bias in the work scope.

Standard 1-1(a) states the appraiser must employ “recognized methods and techniques.”  But what if these methods enable or foster bias?  It appears that if recognized methods are expected by users, and are used by peers, then the bias is OK, or even required!

[It appears that this methods and techniques requirement may contradict the requirement that the result be unbiased.]

Standards Rule 1-4 states the appraiser must “collect, verify, and analyze all information necessary.”  However, the goal of the rule is to be credible (worthy of belief), not unbiasedness.  At times it is easier to use information which favors believability, rather than unbiasedness!  In fact, being worthy of belief is more important than avoiding bias.

But we have good news!

Standard 1-4(a), about the comparison approach, requires the use of “such comparable sales data as are available.”  [Similar requirements are there for the cost and income approaches.]

Note that what is available may be different from sales which are believable!  Hmmm.

The issue is that traditional appraisal education and practices require the appraiser to pick comps before doing any analysis of the market.  Now — this opens the door for picking sales which may bias the analysis and the result.  The bias may be conscious (if the appraiser is aware of ‘hit the number’ psychology), or unconscious (if the appraiser ‘anchors’ on the sale price, for example).

The problem here is multi-fold.  First, we have no analytic definition of ‘what is a comparable.’  It is not to be found in the standards, nor in appraiser education.  Second, sales “as are available” is a vague and inadequate guide for the right selection of comparables.  Finally, the ‘market’ – of actual competitive sales – may comprise two or three sales, ten sales, or dozens of sales.

Three comps on a grid, or six comps in a spreadsheet is arbitrary.  Understandable to the human brain, but not optimal to best accuracy and reliability.  We need something better.