What is an automated valuation model (AVM)?
An automated valuation model (AVM) is a term for a service that uses mathematical or statistical models combined with existing property and transaction databases to calculate real estate values. Most automated valuation models compare the values of similar properties at the same time.
Many appraisers, and even Wall Street institutions, use these AVMs to value residential properties. Consumer-ready AVMs also exist on property listing sites like Zillow and Trulia.
- Automated Valuation Models (AVM) are software-based pricing models used in the real estate market to value properties.
- AVMs are more efficient and consistent than a human evaluator, but they are also as accurate as the data behind them, which means they may be out of date or incorrect.
- AVM providers include commercial platforms such as CoreLogic, Freddie Mac, and Equifax, as well as free consumer sites such as Zillow and Trulia.
How do automated valuation models (AVM) work?
Automated Valuation Model (AVM) reports are powered by technology, including proprietary algorithms, and can be retrieved by lenders and agents in seconds. They typically contain a hedonic model (a type of statistical regression analysis) and a repeat sales index, which are weighted and analyzed to generate the price estimate. AVMs generally include the value of the tax assessor, all pertinent information about the property in question, such as its sales history, and an analysis of sales of similar properties.
They are also used to support mortgage and home equity loan underwriting, refinancing decisions, as well as to aid in loss mitigation and credit risk management activities, such as the marketing of real estate holdings in institutional investment portfolios. While AVMs were initially used to value residential real estate, their use has expanded to other types, including commercial real estate.
(AVM) offer their services to a variety of clients, including real estate agents and brokers, mortgage lenders and major financial institutions. AVM’s top vendors include CoreLogic, The Federal Home Loan Mortgage Corporation (Freddie Mac), VeroValue, and Equifax. They can be used by the public through free consumer real estate sites.
Zillow’s popular web-based real estate appraisal tool “Zestimate” is a well-known type of AVM.
Pros and Cons of Automated Valuation Models (AVM) in Real Estate
Despite their widespread use today, AVMs generate some debate, especially about how they stack up against traditional in-person assessments.
Advantages of AVM
The advantages of using AVM over physical assessments are similar to those of any automated system over human effort. Basically, they save time, money and effort. They can do numerous calculations and comparisons in seconds, and do not need to physically drive to view a property or similar properties (“offsets” are a key factor when evaluating and pricing a particular property).
All of this reduces the cost of valuing one property or multiple properties. AVMs are particularly useful for evaluating the value of an entire real estate portfolio. Once configured, AVMs can be operated inexpensively.
In addition to being cheaper and faster, the algorithms are not subject to human error or illegal acts. As objective automata, they remove bias and subjectivity from the equation. Thus, there is less risk of fraud or deliberate pricing, although of course computer programs can be hacked or manipulated.
A 2017 conference paper, “Automated Valuation Models (AVMs): A Happy New World?” by George Andrew Matysiak of the Krakow University of Economics, referenced other studies to address the strengths and shortcomings of these models. “There is little robust unbiased evidence on the accuracy of commercially available AVMs in the public domain,” the document noted. “Despite high levels of average precision, statistically based ratings can be grossly out of place and should be augmented by professional judgment.”
Disadvantages of AVMs
For an AVM to work well, it needs high-quality data in sufficient quantity to be representative. That is where your vulnerability lies.
The most cited drawback against AVMs is that, when determining value, they do not (and cannot) take into account the actual condition of the property. They simply assume an average state, which may or may not be accurate. They cannot notice details or variations in condition.
AVMs are great for making comparisons, but what if there is a dearth of comparable real estate or transactional data on record? For this reason, newly built properties are especially difficult to value and AVMs, being quite literal, tend to lack the imagination to come up with something to serve as compositions. And because an AVM works based on known factors (the historical record), it loses intangible elements that can increase or decrease an estimate.
Finally, an AVM can only work with the data that is provided to it, and there is always the danger that the data will be entered incorrectly. The information you have may also not be up to date, making AVMs unreliable in rapidly changing real estate markets.
The bottom line
The leading providers of automated valuation models tout their accuracy, comprehensive coverage, and time savings. AVMs deal with averages. Therefore, they are especially effective when the real estate stock is very generic. In regions with a wider range of types and styles, they may be less precise and useful.
While their use is growing, AVMs have not supplanted human valuation estimates, especially since most mortgage lenders require a custom appraisal of a property that must be performed by a certified appraiser, in person. Due to concerns about their accuracy, some industry participants suggest viewing multiple AVM results as a way to get a more complete picture and increase confidence in your reports.