Demand for homeownership remains at an all-time high from coast to coast, while home inventory is at a record low. Moreover, Bank of America predicts that U.S. home prices will increase 10% in 2022—nearly double the average annual home price growth (4.6%) posted in the three previous decades. According to a Realtor.com survey, 53% of sellers expect to get their asking price and 24% expect to get more than that.
Home sellers most definitely have the advantage. That being said, do home value estimates on the larger real estate sites (e.g., Zillow, Redfin, Realtor.com) influence their expectations? Real estate coach Tim Harris who, along with his wife, Julie, is an industry trainer, believes so. “It goes to a broader question: How do you price a house in a market that’s expected to appreciate 10% in a single year? Appraisals are assessing value on what was whereas sellers assess value on what will be. So a seller in a hot market is justified to ask for what the price will be in a month or two, or three.”
Harris explained that, although it might seem that way, the rapid increase in home prices didn’t happen overnight. It began back in 2009, he said, when the federal government started buying mortgage-backed securities (MBS), which are bonds secured by home and other real estate loans. “I cannot see why the housing market would slow down unless some unexpected recession occurs, or interest rates rise to a crazy number. There is pent-up demand for homes right now from millennials and people who are upsizing or downsizing. Plus, this country is about to witness the greatest wealth transference in history when $22 trillion will pass from baby boomers to their heirs. That massive wealth transfer will include a lot of homes coming [on the market]. So, even with the war in Eastern Europe and inflation that’s expected to reach double digits by the end of this year, you’re still going to have a strong housing market.”
How do you price a home?
Are these online valuations setting expectations? Online valuations typically provide an approximation of what a home is worth based on features like the address, square footage, number of bedrooms and bathrooms, lot size, and age of the property. “Every market is different,” said Michael Murgatroy, a Boca Raton-based mortgage loan officer licensed in all 50 states. “Properties can be in the same ZIP code and the algorithms can be [different] due to certain variables.” There may be inaccuracies in property taxes paid or tax assessments, for example, and the valuations may not include any upgrades or improvements made by homeowners.
Online valuations use aggregated Automated Valuation Models (AVMs) that essentially take average comps of what’s for sale. “Zillow waved the white flag a while ago and basically admitted that Zestimates were just for entertainment purposes,” said Harris. “They ran some contest [offering] a $1 million bounty if somebody could come up with an algorithm for a systematized approach. They were unable to do so because every property is different. Consumers can use an AVM as a [starting] data point.”
At the end of the day it comes down to the fact that automated CMAs and online home valuations don’t buy houses. “Those are just spreadsheets,” Harris said. “People buy houses and everything else based on emotions. So if your emotional response to something is not commensurate with the asking price and you like it more you’re going to overpay. It’s all dependent on how you feel.”
Both sellers and buyers put a lot of stock in online valuations, said Sacramento-based appraiser Ryan Lundquist of Lundquist Appraisal Co. “People almost look at Zillow as the standard and I frequently get asked, ‘Why is your valuation different?’” Zillow has said that its algorithmically generated home valuations (or Zestimates) are independent and never factor in list prices, but that isn’t always the case.
Lundquist previously tweeted about a single-family home in Carmichael, Calif., which was listed for $380,000—with a Zestimate of $380,414—but after it went under contract for $335,000, the Zestimate dropped right away to $346,364. “Zillow’s algorithm tends to match the list price,” he said, adding that an agent sent him a screenshot of a recent three-bedroom, two-bath listing in Galt, California, that had a Zestimate of $353,300 before it sold for $435,000. “About one week later, the Zestimate rose to $429,488. This happens all the time.”
Murgatroy noted that online valuations can also be spot on. “I recently received an appraisal in Merritt Island, Florida at $515,000 and cross-checked Zillow and [the Zestimate] was $516,000. That’s the second one I’ve seen in a week, in totally different areas.”
Why are some valuations higher than others?
We spoke to iBuyer.com, which tracks cash offers from their top performing iBuyers against the big consumer-facing real estate directories. “We like to understand all offers we receive before we deliver them to our home sellers,” said co-founder and president Mark Weibel. “So we look at home valuations across many sources including our own. Over the last month, we have been seeing a good portion of our offers exceeding the larger portal valuations.”
Weibel said iBuyer.com is definitely seeing patterns, and some are consistently higher than others. “A home is really worth what someone will pay for it. If the amount for the home is in the ballpark of the valuations we leverage, then we have additional confidence that the home seller has a strong offer plus the convenience of a no-fee, quick transaction. Knowing what a cash offer will be should always be the starting point before a prospective seller determines if they want an agent to list their home on the MLS, or if they want to take a nontraditional approach. This is a real valuation because it’s what someone is going to pay, not what some algorithm reveals. Then, if the homeowner believes it’s worth more than that, they can test the waters.”
“Where my wife, Julie, and I live in Puerto Rico all eight villas have basically the same square footage,” Harris said, adding that the measurement of living space is where the sameness ends. “The villas face different directions. Some get the ocean breeze and others do not. One is closer to the road and in other units you cannot hear the road noise. Those nuances [are not present] in empirical data.”
“This is where the divide of data versus human subjectivity takes place and presents a valuation challenge,” Weibel added. “Emotions are subjective, though, and not everyone looks at things the same way. The data, however, is strong enough that companies are buying literally billions of dollars of real estate based on these valuations. That’s significant.”
Leslie C. Stone is a freelance journalist based in Vero Beach, Florida.
This column does not necessarily reflect the opinion of RealTrends’ editorial department and its owners.