Housing Market Distortions, Like Breakfast, Aren't Always Good For You

Housing Market Distortions, Like Breakfast, Aren’t Always Good For You

Do you want some really good real estate advice? How about: Buy Low, Sell High? But before you do that, have a good breakfast (this article was emotionally devastating because there is something magical about it). But don’t confuse the issue by serving spaghetti in pie form because facts aside, some things are just wrong. Like trying to time the housing market.

Timing is not Everything

Every day I hear the droning sound of market participants trying to time the market. i.e. “Oooh, mortgage rates ticked down .03%, lets buy now.” I really enjoyed an article in Brick Underground this week on this subject: Why trying to time the market is a waste of time (and can actually backfire). I shared my views from the standpoint of buyers intent on owner occupancy:

Not to mention the not-small consideration of your own life and comfort. “I think there’s a certain amount of shaming that goes on, like ‘You sold too early!'” says Miller Samuel appraiser Jonathan Miller. “But what if you sold because you wanted a smaller place, or to move out of state?”

In other words, hanging onto an apartment that no longer suits your needs—say, a cramped one-bedroom after you’ve just had a second child and need more space—in hopes of maxing out the market isn’t generally going to serve you well.

I’ve always looked at this thought process as an attempt by buyers to rationalize their purchase, rather than engaging in some sort of investment strategy. We’re human after all, and to error is, well, human. During the insanity of the now decade ago housing bubble where home buying was equated with a stock purchase, I maintained that home buying should be a way to serve your need for a change space and/or location needs and a bit of working to improve your quality of life hopefully balanced with affordability and common sense. And if you end up hedging inflation and making some money in the process, all the better.

There is nothing wrong with Zestimates, unless you want to know what your home is worth

The people at Zillow are smart and built a strong ground breaking brand, but that doesn’t always mean they are making the right decisions. Little did I know, when I met one of the founders at a party the day before they launched a decade ago, how much disruption they would cause. I innocently asked the question, “So, what do you do?” And in the response I heard things like “Expedia” and “Rhymes with Pillow.” Their intro to the public began with the “Zestimate” which unleashed a property narcissism within us as we have checked the value of our homes and compared those values to the houses of friends, colleagues, neighbors, celebrities, etc. That search tool was later de-emphasized as they focused on listings and building a nationwide property database.


Over the years I’ve spoken with a number of Zillow senior execs, middle management and one of the founders about the issue of lack of accuracy of the Zestimate. I’ve never come away with any answers to the accuracy problem. I believe the lack of accuracy is a constant drag on their branding efforts. I am an appraiser and automated valuation models (AVMs) like the Zestimate product are marketed as a reliable tool in the property search process.

This was evidenced in a recent news item that discussed the sale of Zillow CEO Spencer Rascoff’s Seattle home with a great analysis by Inman News.

The day after the home sold for $1,050,000, the Zestimate showed a value of $1,750,405. This indicates that their CEO took a 40% haircut on the value of his home which was exposed to the market for a reasonable time and sold for 19% below its list price. But of course he didn’t dump the property. It couldn’t have been worth anything close to the Zestimate since the property was exposed to the market for a reasonable period of time and sold well below the list price which was well below the Zestimate.


For the Seattle market, Zillow says that 63.8% of all Zestimates are within 10% of their sales price. While I appreciate the transparency, it confirms how inappropriate a Zestimate or any AVM product for that matter are as a tool to estimate the specific value of a home. In the real world, an appraiser that was accurate to within 10% of the sales price on a mere 2/3 of their assignments would not have any clients. In the relocation business, where appraisal variations from selling price are measured, 3% is on the high side and that would be for 100% of the assignments. Aside: Some of the best residential appraisers in the U.S. are members of RAC.

I have no problem with the Zestimate product itself since they are doing the best with what realities they have to deal with. My point of contention has always been about the precision that is implied in the branding of the tool and its misapplication as a reliable way to value a specific property. Then again, if that were the way this tool was marketed, no one would have ever heard of Zillow.

The Zestimate is an AVM, and AVMs are used by lenders for risk assessment. For example, take a bank who has a loan applicant, a homeowner with a $500,000 mortgage, applying for a $50,000 line of credit. The bank would run an AVM to understand the probability of the value being $1 million at that location in order to issue the loan.

And one last point. In the Zillow CEO sale example above, the Zestimate was $1,750,405. The precision of the analysis was to the nearest $5. I contend that is not possible with the data they have because that precision doesn’t exist. So one one hand they are being transparent by sharing their accuracy, but at the same time, they are presenting their valuation results to the nearest dollar.

I get questions like this in my appraisal practice from homeowners:

“If I changed my closet doorknobs to chrome from brass, how much will effect the value of my condo?”

Imagine if I responded with “we would Jestimate™ (my proposed valuation tool) a contributory value to your home of $242.53.” Most consumers who probably doesn’t know what an AVM is or what I would do to calculate such a number, would probably take this information as fact.

Over the coming decades, the quality of data and these analytics will continue to improve, and maybe we can have that conversation at that time. Until then, the concept of what precision is currently available and what precision actually exists will remain distant from each other.

A 25% Drop versus a more than 50% Drop in Super Luxury Contracts

One of the unfolding stories in 2016 has been the price correction that has already occurred in the super luxury condo market despite many players hunkered down in the fetal position remaining in full denial. I was talking about the significant oversupply back in the fall of 2014 after I grew concerned over what I saw that summer. No need to watch the video, I just needed to prove my timing. 😉

A few days ago there was a Real Deal article featuring the results of the Olshan Realty’s contract report of sales above $4 million: The hype is real – Manhattan luxury sales are way down in 2016. Donna Olshan is a friend of mine and has data on this market going back a number of years. She’s even let me play around with her raw data to make some charts a while back. Here is one of TRD’s charts using the Olshan report.


The headline of this article seemed to spread like wildfire through the luxury real estate broker community. At first glance one might assume that brokers were upset at how negative the drop was. Down 25%!!! Yet I spoke to a number of brokers, received a number of emails or was looped into email chains about how the report undercounted the decline. Huh?

There is has long been this false stupid narrative that real estate brokers are focused on wanting prices to be up so they make more money – that happy housing news keep consumers in the market. This is patently false and dumb. Real estate brokers are transactional and if a property is over priced, especially in a flat to declining market, the listing will never sell. What their reaction showed me was that there are plenty of buyers in the current Manhattan super luxury market. Its just that they are not willing to pay 2014 prices which I believe are generally 10% to 20% above current market conditions. By not getting “real” about the market, sellers/developers hold out, thinking the days of market yore are coming soon and won’t negotiate price. So sales remain at a trickle.

The overwhelming message from the luxury brokers I known was that the decline in contracts in the super luxury market (I define as above $5 million and Olshan defines as above $4 million) is separate and apart from a fairly active new development market below the $5 million threshold in Manhattan. The luxury brokerage community felt that the super luxury sales had fallen more sharply since last year, at least double the 25% number in the Olshan report. I agree and have said so since last fall when I chronicled the year over year decline in new development listing inventory to be down about 50% year over year in each of the past 3 quarters. I explained and illustrated this phenomenon back in the January 22, 2016 Housing Note.

The reason the 25% drop was likely under represented current conditions was because last year’s numbers were too low. In early 2015, new development contracts were being reported later than when they actually occurred. In other words, if a development project sold 5 units during June 2015, they might report 1 listing under contract in each of May, June, July, August and September to show steady activity. They could feel the market stalling and thought it would be better to push out the activity as a steady stream. This isn’t a new concept in the Manhattan market. In a number of super luxury projects we have looked into, there hasn’t been a sale since last summer. When market trend distortion is caused by a biased source like real estate developers, it’s always necessary to elaborate on the possibility of such a distortion when reporting results seem off. The report remains the best resource we have for contract volume at the high end, just keeping in mind the potential flaws in reporting the results. This is also a reminder on how contract trends are not the bees knees (did I really just say that?).

More Distortion Chronicled in this week’s Curbed Column on Inflation

I wanted to get in touch with our inflated sense of Manhattan real estate. Admittedly, it has been a confusing year with a slowdown in a super luxury condo sales with lots of new product slated to enter the market. New development prices are skewing everything higher because many of the units closing now sold one or two years ago. The resale market is seeing more inventory come to the market but it remains choppy, with some markets (like high-end Park Avenue) nearly stalling.

Inflation is something not consistently applied to housing results in many forms. Ironic, because it is touted as a good hedge against inflation, although Robert Shiller would disagree. What is missing from Shiller’s discussion is the use of leverage—that is, how you can hedge against inflation.

In our market reports, I’ve always preferred providing a raw benchmark and then the user can do whatever they want to the results. We could also adjust for seasonality, but I prefer relying on year over year comparisons for that. In the appraisal business, experience teaches you that the less adjustments made to a comparable sale, the more, well, comparable it is. But I digress.…

This week, I thought I would compare Manhattan inflation-adjusted (real) results to reported results at the time (nominal) to compare current conditions to the pre-Lehman median sales price high in 2Q08. [read more]


40% of Manhattan Buildings Could Not Be Built Today

There was an epic NYT research piece and visualizations of Manhattan buildings that could not be built today because of an overly aggressive zoning code that turns 100 this year. I really enjoyed reading the comments on this piece.

“Look at the beautiful New York City neighborhoods we could never build again,” Mr. Smith said. “It’s ridiculous that we have these hundred-year-old buildings that everyone loves, and none of them ‘should’ be the way they are.”

“Look at the beautiful New York City neighborhoods we could never build again,” Mr. Smith said. “It’s ridiculous that we have these hundred-year-old buildings that everyone loves, and none of them ‘should’ be the way they are.”


Zoning can be a very good thing but at the same time it can and will be abused.


Pending Home Sales Highest in a Decade

Despite the recurring theme that housing remains “soft at the top”, overall contract volume just reached it’s highest level in a decade and the non-seasonally adjusted April contracts were up 3% m-o-m and 2.9% y-o-y. Here’s my version of the NAR PHSI results in stereo.


Its all sounds pretty good, but we know PHSI is very volatile. So are Census’ new home sales although I wouldn’t titles my explanation of this: Don’t Put on Your Party Hats, Homies.

You can see the volatility here:

Source: Bloomberg Gadfly

Distortion of Homeownership can be Insulting

While I tend to focus on the for sale market than the for rent market, distorting the advantages of home ownership can be tone deaf, even in a brutal rental market like San Francisco.

Photo by Brock Keeling / Curbed SF

If you need something rock solid in your life (particularly on Friday afternoons), sign up for my Housing Note here. And be sure to share with a friend or colleague. They’ll feel good about breakfast, you’ll time the market and I’ll be listening to one of my favorite bands, Social Distortion.

See you next week.

Jonathan Miller, CRP, CRE
Miller Samuel Inc.
Real Estate Appraisers & Consultants

Reads, Listens and Viewings I Enjoyed

My New Content, Research and Mentions