Appraisals and ValuationsOpinion

What AVMs can and can’t do in the appraisal process

While not as reliable as complete appraisal reports, AVMs can be helpful in supporting them.

Fannie Mae recently announced that it will begin to “include a desktop appraisal option in Desktop Underwriter beginning March 19, 2022. According to Fannie Mae, “Desktop appraisals are disclosed as “Appraisal” as the property value is obtained through an appraisal that was completed by a licensed or certified appraiser.

Further, Fannie notes, “Leveraging the wealth of big data, the power of advanced analytics, and the speed of technology, appraisers can perform desktop appraisals without physically inspecting the property.” Amazing, right?

What about AVMs? Automated Valuation Models (AVMs) are commonly used today in the mortgage industry to help support the appraiser’s concluded value. The mortgage industry has begun advancing initiatives to allow technology and big data to take the next steps in the development of things like more accurate AVMs.

But, as noted by Clear Capital on one of their AVM reports, “This analysis has not been performed in accordance with the Uniform Standards of Professional Appraisal Practice which requires valuers to act as unbiased, disinterested third parties with impartiality, objectivity and independence and without accommodation of personal interest. It is not to be construed as an appraisal and may not be used as such for any purpose.”

Though AVMs can sometimes be helpful, there are things that AVMs do not currently do. Here are six misconceptions surrounding what AVMs currently do for appraisal valuation:

1. AVMs don’t account for important property details.

Appraisals often include details such as a pool, solar panels, landscaping, patio, casita, barn, detached garage with an office above, and other things that may contribute to real estate value. AVMs do not consider such things nor can they provide an estimated value to adjust any sales prices. 

Appraisals outline the housing trends for particular markets. These housing trends can’t be identified by an AVM nor can appropriate adjustments be derived to account, for example, for increasing value trends. Perhaps AVMs could link to Housing Index data, but even then, I’m not sure if it is currently being done.

3. AVMs don’t account for adverse site conditions.

Sometimes there are adverse site conditions for properties, such as busy streets, flood zones, commercial influence, etc. This information is not known by AVMs, nor do they calculate possible value adjustments. They do not know how much value to give either.

4. AVMs do not account for upgrades or renovations.

Sometimes properties have increased in value due to recent upgrades or renovations. AVMs don’t know that a subject property is a recently flipped house or that the owners just installed new floors, upgraded the bathrooms and kitchen, or completed a landscaping project in the backyard. Once again, they can’t calculate possible dollar adjustments to the unadjusted sales prices of the closed comparables.

5. AVMs don’t account for commercial influence.

Sometimes properties have heavy commercial influence, such as a home located in commercial zoning surrounded by businesses and retailers that may be non-conforming and result in across-the-board adjustments to account for its marketability and value. AVMs do not have this information nor do they calculate the possible adjustments.

6. AVMs don’t account for geography.

Geography plays a large role in real estate valuation. For example, a home located in Utah, California, or Colorado that has mountain views may be more valuable than homes in those states that don’t have scenic views. Notable lakes, rivers, valleys, and even amenities such as golf courses may contribute to value. AVMs do not currently have this information nor do they calculate possible adjustments to be made for these details.

So, what can AVMs do? While they are not as reliable as complete appraisal reports, they are helpful in supporting them. AVMs are mathematical models combined with data that are built and designed by human beings. Further, AVMs don’t know what a gazebo or dock is worth unless the data is given to it. Mortgage lenders wouldn’t want to conclude value based solely on AVMs. 

How do we produce more accurate AVMs? At least one thing is to include the data listed above. Perhaps future appraisals will be completely automated using the ‘leverage of big data,’ but that doesn’t necessarily mean that appraisers won’t have a job, their job will just change as the industry does.

James Dawkins is a certified residential underwriter and collateral underwriter at Cardinal Financial Company.

This column does not necessarily reflect the opinion of HousingWire’s editorial department and its owners.

To contact the author of this story:
James Dawkins at [email protected]

To contact the editor responsible for this story:
Sarah Wheeler at [email protected]

Leave a Reply

Your email address will not be published. Required fields are marked *

Most Popular Articles

3d rendering of a row of luxury townhouses along a street

Log In

Forgot Password?

Don't have an account? Please