My comment: I love the “20 cats” comment. I appraised a house with at least that many cats. They were on all the tops of dressers, cabinets, etc., Very Creepy! I still have nightmares about it sometimes. A volunteer owned it for a local animal shelter. She had a large outdoor enclosure. Did not smell, but…
By Jonathan Miller
Zillow Offers As A Proxy For ‘Big Data’ Shows The Lack Of Qualitative Analysis
Yes, big data usually infers ‘quantitative’ analysis, as in “relying on numbers.” The Zestimate legacy of profound inaccuracy finally reached a devastating conclusion with the collapse of Zillow Offers this week and the loss of hundreds of millions in shareholder…
Yes, big data usually infers ‘quantitative’ analysis, as in “relying on numbers.” The Zestimate legacy of profound inaccuracy finally reached a devastating conclusion with the collapse of Zillow Offers this week and the loss of hundreds of millions in shareholder equity. Zillow never figured out the qualitative part that enables the actual precision in the pricing of a home sale.
This Just In: The ‘A’ in ‘Zillow’ Stands for ‘Accuracy’
Excerpt: Yet it would seem unlikely that Zillow Offers used something completely separate and conceptually very different from their ‘Zestimate’ because it would be quite expensive and extremely difficult to keep a radical new valuation concept a complete secret. All we know at this point is whatever valuation methodology they used was a complete fail. And to go a step further their Zestimate valuation methodology has long been a complete failure in the accuracy department. But it hasn’t been a complete failure in the consumer credibility department at all.
Excerpt: Situated in the small town of Talkeetna, Alaska, the fantastical house is what most of us would describe as something we’ve seen in a Dr. Seuss book. Unlike anything Dr. Seuss wrote about, however, this “house is real, it’s not a fantasy” says Phillip Weidner, creator, builder, and owner of the towering home, which stands 185 feet high.
Weidner did originally only plan on building a two-story log cabin, then he let his mind (and engineering degree) take over. He decided to add another floor. Then another, and another, and it just kept going up. Depending on how you look at it, and which ladders and staircases you climb to get to the top, there are somewhere between 14 and 17 floors.
My comment: Too Tall for Me and Too Many Stairs and Ladders!! My last scary inspection was a room over the tall garage. I went up the steps and the whole structure bumped in all directions. I disclosed it in my appraisal. Fortunately date of death appraisal and not for a lender.
For more info and photos, google Dr. Seuss House Alaska.
Excerpts: An interesting twist on one of the TV story interviews is the homeowners’ incorrect assumption that just because they SPENT 1.3 Million Dollars building their ‘dream’ house, it would APPRAISE for that amount. A comment was (paraphrasing) “we spent all those dollars buying nails to build this place, so we don’t understand why it didn’t appraise for what we paid.”
That, of course, is a common misconception. It’s that way in large part because, “we” in this industry, have done a poor job explaining to the public what an appraisal is, how one is done, and the factors that are considered in stating an OPINION of value for the property. There’s a disjointed disconnect between public perception and our reality.
Frankly, before I became an appraiser, I had no clue about anything appraisers do.
Somewhat helpful, though, in the story covered by the TV station, there is a video showing an appraiser explaining to the film crew how on-site observations for an appraisal are done (see video below). It’s the first time I’ve seen this presented this way in a story related to bias. Kudos to him for doing that!
To read more and watch the video news segment, click here
My comment: Great that an appraiser was interviewed to include comments.
2. Appraisal Today blog: https://appraisaltoday.com/blog (posted by noon Friday) To get notices when newsletters are posted, sign up in the upper right of the blog. No ads in the blog post. You will receive an email to confirm.
(posted early Friday) The link is above and to the right of the big yellow email sign-up form. Newsletters start with “Newz.” Contains all recent emails sent.
4. Link to 10 most recent posts and newsletters (no ads) at www.appraisaltoday.com. Scroll down past the big yellow signup block.
NOTE: When you have set up one of the choices above, and it is working, you can unsubscribe from this email list, which includes getting ads, by clicking the unsubscribe button at the bottom of every email newsletter.
ASC ‘Roundtable’ 2 on Bias & Appraisal Inequities November 9, 2021
I attended this virtual presentation. Very good. There were two breakout sessions with lots of appraiser questions and input. One was about diversity and new appraisers.
I attended the other breakout about Freddie Mac’s recent report on differences between White, Black, and Latinx census tracts regarding appraisal under contract price vs. appraisal data. The speakers were Danny Wiley, Senior Director of Single-Family Property Valuation, and Vivian Li, Senior Manager of Fair Lending Analytics, Modeling, Econometrics, Data Science & Analytics. Danny is an appraiser and spoke about appraiser issues. Vivian explained the data and how it was analyzed.
No appraiser bashing, similar to the first Roundtable. It was set up to get appraiser input in the breakout sessions.
The big takeaway that Danny said repeatedly is that they have not looked at causation and are planning to do more data analysis.
Lots of good questions from appraisers. One person asked how Freddie compares with the recent AEI analysis of the bias issue. Freddie has data directly from millions of appraisals. AEI used secondary data. They compared appraisers working in and out of the affected census tracts. About 40% of appraisers had not worked in those tracts, which may be why so many appraisers don’t understand why they are affected.
Freddie uses about 12,000,000 appraisals. There were “raw” differences among appraisers, which need to be analyzed for causation factors. Appraisals done by appraisers who worked in White and non-White census tracts were studied. I worked for over 20 years in Oakland, before 2006, appraising in all neighborhoods, so I am interested in future studies of the data.
To download the 9-21 Freddie PDF report: Racial and Ethnic Valuation Gaps
in Home Purchase Appraisals click here. Scroll down and look for the tables you want with explanations.
My comments: I wrote about the first Roundtable in the September 24 issue of this newsletter. It did not include appraiser-focused breakout sessions.
Video Comments: Racial and Ethnic Valuation Gaps – 19 min.
Excerpt: Speakers: Scott Reuter, Chief Appraiser and Director of Valuation of Freddie Mac, and Vivian Li, Senior Manager of Quantitative Analytics of Freddie Mac. We’ll be getting the inside scoop while discussing the studies and data surrounding this hot button issue.
How was the research conducted for this Freddie Mac study regarding this topic? In what ways are we applying what Freddie Mac has shared? How is this information beneficial to the future of appraising? These questions and much more are answered as Jim, Scott, and Vivian discuss this subject.
Excerpts: “The Mataja Residence posed two particular challenges from the onset: the clients’ desire for a retreat from an urban lifestyle while simultaneously maintaining a passion for the technological car culture, and the constraints of building in mountainous undeveloped land with no roads, water, or power source,” states a description of the project on the designers’ website.
To overcome those constraints, the solutions they devised were both clever and elegant. They wrapped the structure around an existing granite boulder outcropping, using a California courtyard house design, allowing the boulders to form and complete the courtyard.
Five-bedroom, 7,500-square-foot house, car gallery, a five-plus car garage with auto lifts, and a 20-car motor court.
To read more and see lots of interesting photos, click here
HOW TO USE THE NUMBERS BELOW. Appraisals are ordered after the loan application. These numbers tell you the future for the next few weeks. For more information on how they are compiled, go to www.mbaa.org
Note: I publish a graph of this data every month in my paid monthly newsletter, Appraisal Today. I have been following this data since 1993. For more information or get a FREE sample issue go to https://www.appraisaltoday.com/products.htm or send an email to email@example.com . Or call 800-839-0227, MTW 7 AM to noon, Pacific time.
Mortgage applications increased 5.5 percent from one week earlier
WASHINGTON, D.C. (November 10, 2021) – Mortgage applications increased 5.5 percent from one week earlier, according to data from the Mortgage Bankers Association’s (MBA) Weekly Mortgage Applications Survey for the week ending November 5, 2021.
The Market Composite Index, a measure of mortgage loan application volume, increased 5.5 percent on a seasonally adjusted basis from one week earlier. On an unadjusted basis, the Index increased 5 percent compared with the previous week. The Refinance Index increased 7 percent from the previous week and was 28 percent lower than the same week one year ago. The seasonally adjusted Purchase Index increased 3 percent from one week earlier. The unadjusted Purchase Index increased 0.1 percent compared with the previous week and was 4 percent lower than the same week one year ago.
“Mortgage rates moved lower for the second week in a row for all loan types. The 30-year fixed rate decreased to 3.16 percent and has declined 14 basis points over the past two weeks. Although overall activity remains close to January 2020 lows, homeowners acted on the decrease in rates. Refinance activity was up 7 percent overall, with gains in both conventional and government refinances. Additionally, the average loan balance for a refinance application was the highest in a month,” said Joel Kan, MBA’s Associate Vice President of Economic and Industry Forecasting. “Purchase applications were also strong last week, increasing just under 3 percent and down only 4 percent from last year’s pace. The dip in rates might have helped to bring some buyers back into the market, but housing inventory is still extremely low and price growth remains elevated.”
The refinance share of mortgage activity increased to 63.5 percent of total applications from 61.9 percent the previous week. The adjustable-rate mortgage (ARM) share of activity decreased to 3.1 percent of total applications.
The FHA share of total applications decreased to 8.8 percent from 9.2 percent the week prior. The VA share of total applications increased to 10.2 percent from 9.9 percent the week prior. The USDA share of total applications remained unchanged from 0.5 percent the week prior.
The average contract interest rate for 30-year fixed-rate mortgages with conforming loan balances ($548,250 or less) decreased to 3.16 percent from 3.24 percent, with points remaining unchanged at 0.34 (including the origination fee) for 80 percent loan-to-value ratio (LTV) loans. The effective rate decreased from last week.
The average contract interest rate for 30-year fixed-rate mortgages with jumbo loan balances (greater than $548,250) decreased to 3.26 percent from 3.29 percent, with points increasing to 0.32 from 0.27 (including the origination fee) for 80 percent LTV loans. The effective rate decreased from last week.
The average contract interest rate for 30-year fixed-rate mortgages backed by the FHA decreased to 3.18 percent from 3.29 percent, with points decreasing to 0.31 from 0.38 (including the origination fee) for 80 percent LTV loans. The effective rate decreased from last week.
The average contract interest rate for 15-year fixed-rate mortgages decreased to 2.52 percent from 2.58 percent, with points increasing to 0.33 from 0.29 (including the origination fee) for 80 percent LTV loans. The effective rate decreased from last week.
The average contract interest rate for 5/1 ARMs decreased to 2.82 percent from 2.88 percent, with points increasing to 0.25 from 0.11 (including the origination fee) for 80 percent LTV loans. The effective rate decreased from last week.
The survey covers over 75 percent of all U.S. retail residential mortgage applications, and has been conducted weekly since 1990. Respondents include mortgage bankers, commercial banks and thrifts. Base period and value for all indexes is March 16, 1990=100.