- Miller Samuel Real Estate Appraisers & Consultants - https://www.millersamuel.com -

April 6, 2018

Housing’s ‘Dilly Dilly’ Moment

I’m no longer interested in playing golf anymore, but I found it odd that the 2018 Masters [1] banned crowds from cheering “Dilly Dilly” because, well, they can. Between you and I, do you really think they will throw out hundreds of spectator [2]s for cheering a great drive by chanting that phrase?

And then there’s this display of Dilly Dilly [3] entertainment…

Carrying the torch for Tiger’s old Nike Golf commercials – here’s the original and a few other versions.


But I digress…

Market Report Gauntlet Q1 2018 Week 1: Manhattan

As most Housing Note readers know, I’ve been writing the Elliman Report series [6] for Douglas Elliman Real Estate [7], the (now) third largest real estate brokerage firm [8] in the U.S. since 1994. The market report focused on the drop in sales, the largest year over year decline in 9 years and the lowest total in 6 years.

In other words…Dilly Dilly.

The Bloomberg piece that covered the report was the number 1 most read article on Tuesday all day.

[9]

I don’t see this decline in sales phenomenon as the onset of a continued linear trend but more as a reset event. And in many ways, this event helps get the issue out in the open and could actually help buyers and sellers come to terms with new conditions.

[9]

I spoke with Bloomberg Radio [10] on this and had a fun discussion about this on Bloomberg TV [11]:


Here are a couple of Manhattan market charts [12]:


[13]

This Week in Aspirational Pricing

The 432 Park Avenue new development in Manhattan continues to dominate the price record books with the lion’s share of top level sales: Wife of Hedge-Fund Titan Pays $60 Million for Manhattan Apartment [14] [WSJ]

New in the Real Estate Lexicon

“Correction” – Speaking to a group of real estate agents and brokers the other day, it was clear to me that there was some nuance lost with the word “correction” as in “the market experienced a price correction this quarter.” The word “correction” conveys that the prior pricing was “wrong” rather than economic conditions that created value had changed. Values change over time as economic conditions and tastes change.

Perhaps “reset” is a better term, since it doesn’t convey right or wrong pricing, that values returned to levels that reflect prior economic conditions. What’s more nuanced than simply saying “Dilly Dilly” when the situation calls for it? Just a nuanced thought.

Housing Reality Shows Convert Dollars to Cupcakes

Waco Texas, home of Joanna and Chip Gaines’s hit HGTV show Fixer Upper [15] has helped Waco become a tourist destination and as presented in this Curbed read, reduce its association with cult leader David Karesh [16]. I remember seeing t-shirts in the early 1990s that said: “My parents went to Waco and all I got was this lousy AK-47. [17]” Thanksfully times have changed.

With all the bad and confusing news in the world these days, it’s refreshing to see a housing reality show have such a positive impact on a city.

[18]

Appraiserville

It’s Time to Change the Narrative

My friend and data management savant Mark Stockton penned a great article on statistics being tossed around by the major players in the valuation industry is misleading. Mark, Phil Crawford and I got to talking about this topic after reading a recent Appraisal Buzz article: Evolution in the Industry [19]. Well intentioned I’m sure, but presents a tired and misleading message that I interpret as “embrace technology for the sake of technology” as if the default position on technology is that it always is “better” than the alternative.” The reality in valuation is that most of it is crap, but the big dogs in the space have a vested interest in making this 100% about automation.

Here’s Mark’s article:


It’s Time to Change the Narrative

According to Fannie Mae: “Market value is the most probable price that a property should bring in a competitive and open market under all conditions requisite to a fair sale, the buyer and seller, each acting prudently, knowledgeably and assuming the price is not affected by undue stimulus.”

This is a very nice, neat definition to describe something that is extremely elusive.

In his introduction to “The Adverse Effects of Single Approach Appraisals and Single Point Valuations” (Recognizing and measuring the market impact of volatility in residential real estate purchase prices and valuations), written in 2015 by Bill King President, Chief Valuation Officer at Real Info, Inc., he states:

“Real estate markets are imperfect markets; we should expect a certain lack of precision when measuring both sale prices and appraised values. Even in the most homogeneous neighborhoods, no two properties are truly identical, and no two buyers operate with the same information or motivations. Value, like fairness, is in the eye of the beholder. Values concluded in residential appraisal reports are said to be “in error” when those values disagree with actual sale prices. Closed sale prices are believed to be the best evidence of market value as defined in the federal register. While there are many studies of the error rates in appraised values, there are few studies of the error rates in sale prices.”

Mr. King performs several analyzes in order to draw conclusions about error rates in sale prices. Among them, he looks at sales for a subset of homes that, “Aside from minor differences in maintenance and upkeep…are nearly as substitutable as stock certificates.” He observes price differentials reaching nearly 15% in a single month.

Among his conclusions:
“a purchase price variance of up to 10% should be expected when all other factors, including buyer ability to pay and overall motivation to buy are otherwise the same.”
“Price variances must be expected because ultimately, there really is no one true market value number to the exclusion of all other numbers.”

(Bill King has more than 35 years of housing industry experience, including work as a senior executive, qualified forensic expert, speaker, trainer, author, appraiser and real estate broker.)

Fannie Mae’s definition of market value does not state that sale price equals market value. As Mr. King’s analysis makes abundantly clear, real estate markets by nature are imprecise, and the correlation between sale price and true market value can be difficult to reconcile. However, sale price is the benchmark most often used to test the accuracy of real estate valuation models.

Every national valuation company that utilizes automated valuation technology represents their value estimates are accurate, within certain parameters. These accuracy statistics are generated in blind testing where value estimates are compared to actual sale prices (sometimes contract prices and appraised values, as well), and differences are reduced to a percentage of error.

By way of illustration, “Zillow’s accuracy has a median error rate of 5.6%”, according to their website on June 15, 2017. The site elaborates on this statement: “This means half of the home values in the area are closer than the error percentage. For example, in Seattle, Zestimate values for half of the homes are within 5.4% of the selling price, and half are off by more than 5.4%.”

Those of us who deal with real estate analytics on a day-to-day basis recognize that statements like these are entirely misleading. When the market itself is less than exact, how can one expect value estimates benchmarked against sale prices to be predictably accurate?

We need to change the narrative from a discussion of accuracy to a discussion of “defensibility”.

A valuation provides an estimate or opinion of value. The accuracy of that estimate cannot be conclusively determined from a mere representation of precision, unaccompanied by documentation that adequately supports that statement. The results of a test which compares value estimates with sale prices might seem reasonable at first blush, but we know without reservation that sale prices are not precise measures of “market value”. If our target or benchmark value is inexact, how can we make definitive statements about the accuracy of estimates that are compared to those values?

In addition, accuracy statistics published by valuation companies relate to tests performed on large sets of data; they are not granular enough to be meaningful at the property level. It is doubtful that someone living in Houston, TX cares about the statistical accuracy of values computed for all of Harris County. They likely are concerned with the accuracy of the value estimate for their home – and they won’t get that from the valuation company.

On the other hand, if each valuation is accompanied by thorough documentation that allows the recipient to understand with complete transparency how the value was derived, the recipient can then form an educated opinion as to the reasonableness of the value conclusion. If they can form an opinion about the value conclusion – whether they agree with it or not – the estimate becomes a useful tool that can be used in a decision-making process. Otherwise, it is useless.

So, while the results of tests performed against large volumes of sale prices are useful to the analysts who strive to improve their valuation models, they have little or no application for those who are concerned about the accuracy of individual home values, or groups of values as might exist in a portfolio.

Defensibility is key to understanding whether a value is “accurate”, i.e. reasonable in light of supporting documentation. We need to move away from the absurd reliance on accuracy rates we know to be false, and move toward reliance on definitive, factual documentation that allows us to form an honest opinion about the reasonableness of value estimates.


The second to the last paragraph is powerful so I’ll repeat it:

So, while the results of tests performed against large volumes of sale prices are useful to the analysts who strive to improve their valuation models, they have little or no application for those who are concerned about the accuracy of individual home values, or groups of values as might exist in a portfolio.

FHFA: AMCs Provide No Contributory Value to the Mortgage Appraisal Process

In reference to the damning FHFA white paper [20] published last month, Phil Crawford breaks it down on his Voice of Appraisal Podcast: E196 The AMC Report Card…OUCH! [21]

Brilliant Idea #1

If you need something rock solid in your life (particularly on Friday afternoons) and someone forwarded this to you, or you think you already subscribed, sign up here for these weekly Housing Notes [22]. And be sure to share with a friend or colleague if you enjoy them because:

Brilliant Idea #2

You’re obviously full of insights and ideas as a reader of Housing Notes. I appreciate every email I receive and it helps me craft the next week’s Housing Note.

See you next week.

Jonathan Miller, CRP, CRE
President/CEO
Miller Samuel Inc. [23]
Real Estate Appraisers & Consultants
Matrix Blog [24] @jonathanmiller

Reads, Listens and Visuals I Enjoyed

My New Content, Research and Mentions

Real Estate Blockchain Reads

Appraisal Related Reads

Extra Curricular Reads