Automated Valuation Model AVM

These days house values are all the rage, especially as they’ve surged to all-time highs in recent years in the United States. The value of a piece of property can be the basis for a sale, purchase, or financing, so it’s important to ensure house values are calculated accurately. After all, if no one takes the time to actually look at the subject property, someone could end up with a money pit.

One form of property valuation is the use of an AVM, or “Automated Valuation Model.” This is a computer program that performs a data analysis using an automated process, calling on the database to provide a series of data sets to perform a comparable approach.

Let the Robot Decide

The engine relies on linear and multiple regression to compile statistics and form the analysis. Linear regression adjusts for one variable, such as the age of a home. For each change in age, the same change in price is applied to the value of the home. Multiple regression is a similar process, though it uses a series of variables simultaneously.

After the data is compiled, the AVM produces a report with the value range and a proposed value of the property. This can form the basis for an appraisal which can then be reviewed by an appraiser, or the appraisal process can be deemed complete.

This is where things can get a little gray. Over the last few years, many banks and lenders, including the likes of Fannie Mae, Freddie Mac, Countrywide and Washington Mutual have relied upon AVMs to take the place of real-live appraisers.

The decision was based on the fact that automated valuation models would be sufficient to determine property value and price range, and that these valuation engines could effectively eliminate the need for an enhanced review.

Automation Fail

The problem with AVMs is that simply put, they are automated. They don’t take into account the human element, the present condition of a home, its actual size, views, location, improvements, and more.

There’s a good chance AVMs will not have up-to-date information, including new sales comparables, and they’ll likely compare two homes with dissimilar views, such as oceanfront versus non. In addition, recent improvements may be noted on one comparable, but not the subject, and the present condition of a neighborhood may also be missed. The result can be deceptive and cost banks, lenders, and homeowners thousands.

Unfortunately, AVMs haven’t necessarily swung in one direction value wise. There have been a number of homes undervalued and overvalued, with the lack of precision being the problem. Of course this hadn’t been much of an issue until recently when home prices stopped their miraculous run about a year ago.

Now, as home prices have stagnated or dropped, appraisals have become increasingly important again. As a result, many banks and lenders are requiring the use of field reviews and broker price opinions (BPOs) as Wall Street investors demand more extensive inspections before buying mortgage notes in bulk. The result should be less reliance on automated appraisal reports until the real estate market gets back on its toes.


Leave A Response