After a wonderfully fun — and occasionally frustrating — evening of hacking around, I’m please to present this mashup of Bay Area housing price trends. The graphs are from our friends at Altos Research, and the Google Maps were created using Zeemaps from Zeesource.

For a larger map, click here.

[Read more →]

Hot on the heels of Redfin’s controversial news that analysis of the MLS data conclusively shows that Redfin agents are better negotiators, 3 Oceans is pleased to announce the results of a study showing, again conclusively, that agents who take lots of property pictures are better negotiators.

After downloading all transactions in 2006 in the REIL MLS, the results are striking: properties in which the listing agent only took 1-3 pictures sold on average for just under $900K, while properties with 4-6 pictures sold on average for nearly $930K. But the really trigger happy agents did the best: their properties with 7-9 pictures sold for a whopping $1,077,933 on average.

Here’s my only question: What happens when a 9-picture-taking listing agent goes head to head with a Redfin negotiating agent?

P.S. List with me. I’ll take 9 pictures of your property. It’ll sell for more.

[Read more →]

Though temporarily put out of commission by a mild form of the Martian Death Flu that recently afflicted Kris Berg — and perhaps an overdose of Thanksgiving Tryptophan, not to mention a stronger-than-expected Q4 — I came to life upon seeing Mike Simonsen’s insightful post on the relationship between Palo Alto home prices and the NASDAQ.

I thought I’d take myself up my own challenge (posted on Mike’s blog) of seeing how well that pattern fits going further back. The result? Broad-stroke parallels useful for thinking about large trends, but (alas) no neat regression-based formula that would help you to predict, say, next month’s median prices. (Of course, if I had found something like that, trust me, I would not be posting it on my blog! I’d run as fast as my legs could carry me over to Sand Hill Road and hit up some VC’s for a few billion dollars to buy and sell Palo Alto real estate!)

Here’s the chart comparing Palo Alto median home prices with the NASDAQ; both numbers are indexed to 100 = February 1997.

The key observation seems to be that, very roughly speaking, NASDAQ’s movements are followed several months later by corresponding movements of about half the magnitude in Palo Alto prices. In other words, when the NASDAQ starts trending upwards, Palo Alto prices start doing the same a few months later. If the NASDAQ’s climb ends 50% higher, Palo Alto’s ends about 25% higher. Observe: (Numbers refer to chart below.)

• 1 to 2 The NASDAQ , as we’ll all recall, had a glorious run, peaking in February 2000 at a level more than triple its February 1997 level; Palo Alto prices also had a glorious run during that time, though the peak took place a few months later and was “only” 2.5X higher than in February 1997.
• 2 to 3 The NASDAQ’s initial plunge from February 2000 to May 2000 shaved roughly 30% off its value; in response, Palo Alto prices dipped 13% from June to September 2000.
• 3 to 4 The NASDAQ’s brief rally from May to August 2000 was followed by a brief home price rally from September to December of the same year.
• 4 to 5 Following its brief rally, the NASDAQ went into free-fall from August 2000 through September 2001, during which it lost over 60%. Palo Alto’s corresponding price drop started a good 6 months later and ended in December 2001 30% lower.
• 5 to 6 The NASDAQ had another brief rally through November 2001; Palo Alto waited till December and climbed for a good half year.
• 6 to 7 Unable to sustain itself, the NASDAQ dropped 40% from January to September of 2002; Palo Alto prices fell 18% from July 2002 through January 2003.
• 7 to 8 With a few brief interruptions, the NASDAQ has been marching slowly upward from September 2002 through now, slightly more than doubling during that time. From February 2003 through now, Palo Alto prices have climbed by about 50%.

[Read more →]

A previous post of mine generated an interesting series of email exchanges between myself and reader Greg. Part statistics, part epistomology, and part real estate, I thought the discussion — a tribute to liberal arts education — worth posting on its own. The end state, alas, was inconclusive.

Readers who cringe at memories of college stats classes are excused from reading this entry.

Greg:

Kevin, I am dying to know how the Palo Alto medians you report compare to sales volume during that period. While the median is decent summary statistic, I think it would be even more interesting and helpful to look and the distribution and number of sales on the graph directly. Could you plot all Palo Alto sales on a day by day basis, so we could see what the cloud of points actually looks like? (BTW, you may need to toss a couple of outliers out so they don’t compress the scale on the other points).

Kevin:

I want to make sure I understand what cut of the data you’re intrigued by. Is it something like the attached [following] PA scatter diagram? If so, I’m happy to post it, along with any insightful comments that either you or I come up with. At first glance, this data doesn’t tell me a compelling story one way or the other; I’m assuming by your comments that one would expect a declining volume of sales during the time that prices were declining, and an increasing volume of sales during the time that prices were increasing?

Greg:

Thanks for the attachment. I’ve attached it [following] back with one new sheet which shows trailing 2-week sales volume. What I’m trying to get at is this: what distribution of sales prices are responsible for each of your median calculations?

My line of thought is this: there is a real population of home prices/values out there, but we’re limited to looking at a small sample, i.e. the prices of homes that transacted. As the number of sales goes down, the confidence interval we have around our population mean estimate should go up. I am trying to figure out how to calculate this confidence interval, and interpret what it means for home buyers.

I’m over in China right now on a project, so I don’t know if I’ll be able to figure this out to my satisfaction in the immediate short term.. but I do look forward to thinking about it more and hearing your thoughts on it as well.

Kevin:

First, your diagram: I’m not sure that I’m pulling any real insights from it, apart from it partially reflecting the natural cycle of real estate sales, which tend to peak in the spring.

Secondly, I’m not sure I agree with your paragraph on the “real population of home prices.” This may be more of an epistomological argument than a real estate one, but the way I look at it is that value is conferred only through an arms-length transaction between two willing parties. Homes that haven’t sold obviously also have a value, but it’s an unknowable value (though estimatable) until they go on the market and sell as well. Thus the universe of homes that have sold is not, in fact, a sample set from which we can figure out a confidence interval about the “real population”; instead, the universe of homes that have sold is, for all practical purposes, the real population.

[Read more →]

You gotta love it when you win a contest you didn’t even know you’d entered — sort of like winning a lottery when you weren’t even aware of having a bought a ticket!

Well, that’s what happened this last week with my post on real estate numbers that I entered into the Carnival of Real Estate. Alas, I didn’t win — the prize went to a better post from a better writer, Jim Cronin at the Real Estate Tomato — but what I did win was The Best Post of the Week, Anywhere at PoliticalCalculations (permalink most likely this, but it’s not live yet.)

Political Calculations apparently tracks money and business-related carnivals every week, and I’m honored they came across my post and selected it.

Another numbers-related post that’s also mentioned is on “Interpreting Median House Prices.” Sounds like a yawner, right? Well, actually, no, not at all. The author, James Hamilton, notes an interesting paradox: according to Census Bureau numbers, median national home prices fell 1.7% from 2005 Q3 to 2006 Q3…even though regional median home prices are up in every single region! Confused? Intrigued? Head over and read what James has to say.

[Read more →]

A friend of mine named Eilam, also a numbers sort of guy, made an interesting comment about my previous post in which I dissected the year-to-date real estate numbers for Palo Alto, CA. He challenged me to look still deeper into the numbers and try to figure out whether I had the cause-and-effect in the right order.

I essentially asserted that, “Homebuyers were buying increasingly smaller homes from January to July, and increasingly larger homes thereafter, and that’s what led to prices going down between January to July, and upwards thereafter.”

Eilam asks if instead it might be this way around: “From January to July, homebuyers decided they wanted to buy less expensive homes, so they simply bought smaller ones. Thereafter, they loosened the purse strings, and therefore bought more.”

His exact quote:

Thanks. Good analysis.

However - it does beg a follow-on question. What can we learn from the down-tick and later up-tick in sizes of houses being sold ? Is there rhyme-and-reason underling the observed trend (which you very accurately analyzed) ?

I think ’sizes=total price’ (i.e. it isn’t that people are all of a sudden interested in larger or smaller houses at certain times of the year).

What is interesting to ask is: is this a pattern change in demand or in supply (i.e. is the reason smaller houses were selling for a while because larger ones were not on the market, or because they were on the market but people were not buying ?).

Also - what would be interesting is to overlay it with a possible ‘culprit’: interest rates.
Is the swing in ‘house sizes’ (again in my mind possibly an ‘alter-ego’ of total-purchase-price) correlated to changes in interest rates ?

Eilam

Great question — I would have expected nothing less from Eilam — and my answer for now is, “I’m not really sure.”
Here’s what I (think I) know and don’t know so far:

1. Since Altos Research’s data only goes back a year, I’ll have to go to another data source to find out if this is a seasonal trend — ie for whatever reason, people buy increasingly smaller homes in the first half of the year, and increasingly larger homes later on. I strongly, strongly doubt that’s the case. I have the data that will answer that question locked up in a 250MB Access database, but I haven’t had time to release and analyze it.
2. Interest rates may indeed be the reason. I don’t have a quick-and-dirty way of putting interest rates and Altos Research data on the same graph, so for now this will have to do:
   
   Palo Alto median home prices:
   Interest rates (from Freddi Mac)
   

   30-year fixed mortgage rates climbed steadily from 6.15% in January to 6.76% in July, and then crept down to 6.40% in September. As mortgage rates were climbing, home prices were going down, and when mortgage rates started going back down, home prices started going back up. Eilam’s theory may indeed be correct in that prospective homebuyers ratcheted down their size requirements to stay within a budget.

3. If this theory is correct, we should see a similar pattern in other towns. The results below don’t show a similar pattern; I’ve included towns both more and less expensive than Palo Alto.Menlo Park:
   
   
   Los Altos:
   Woodside:
   
   
   

   Redwood City:
   

4. Conclusion? I’m still not sure! Possibilities:
   • It’s unlikely that Palo Alto home buyers are more interest-rate sensitive than their neighbors in Menlo Park, Redwood City, Los Altos, and Woodside.
   • It’s possible that people who buy homes during the school year tend to have fewer or no kids and thus need less space, and those who buy during the summer tend to have more kids and need more space. Sounds like a good theory, but wouldn’t it be the same for other towns as well?
   • It may just be something completely random, an artifact of the characteristics of the homebuyers that happened to be on the market in Palo Alto this year.

[Read more →]

Late Friday afternoon, I heard back from our local MLS about realtor.com’s displaying of sold data on their web site. Their interpretation is that Realtor.com is not breaking any rules after all because…their sold data comes from a 3rd party aggregator, not the MLS.

So let’s follow a home’s journey on realtor.com from when it’s first listed to after it’s sold. The following simple diagram will help:

1. The listing agent uploads the property into the local MLS, which in turn feeds the property into Realtor.com.
2. The property sells. Poof! The listing disappears (temporarily) from Realtor.com because it’s not allowed to display sold listings if the source is the MLS.
3. When the property closes escrow, the escrow officer submits the paperwork to the county.
4. A county employee enters the data into its system.
5. Onboard, a real estate services provider, sucks the data in from the county records.
6. Realtor.com sucks the data in from Onboard. Voila! The property appears again.

Hmm…let’s look at the rules again. Our local MLS has the following regulation [sic]:

[I assume the first line is supposed to say, "No one may display the following...]

So if the data on a sold listing comes from some place other than the MLS, then it’s no longer considered a sold listing? Seems kind of shady to me…

Again, don’t get me wrong. I think the prohibition on displaying sold data is a silly anachronism, and I certainly don’t begrudge Realtor.com a little bit of enterpreneurialism. I just think it should be a level playing field for everyone. Based on my past experience, if I were to bend the rules like that, I suspect I’d be called to task pretty quickly.

[Read more →]

I rarely have a need to go to Realtor.com, but when I did so last night I had quite the surprise — there were not just active listings, but recently solds as well!  Their source for that data?  Not the MLS, but a 3rd party data aggregator called OnBoard.  I’m getting a headache.

It’s taken as a given in this industry that by withholding information from consumers, you encourage them to call a Realtor, which has been the rationale behind the archaic prohibitions on showing data on sold properties.  Alas, this thing called the Internet came around and set a whole bunch of data free, much of which is gleefully basking in the sun at Zillow.com.  So now consumers who want sold data can simply ignore Realtors completely.  Bad move on our part.

As a competitive response, Realtor.com, the “official site of the National Association of Realtors,” now displays sold data…in violation of the recommended rules of…you guessed it, the National Association of Realtors.

From the NAR site:

Our local MLS, which provides the data feed for this area to hundreds of web sites, including Realtor.com, is pretty explicit on this issue:  [I assume the first line is supposed to say, "No one may publish the following kinds of information..."]

So let me see if I have this straight…REIL, our local MLS, operating under guidelines from NAR, explicitly prohibits the display of sold listings.  REIL licenses its data for this area to Realtor.com, NAR’s official site, which then breaks REIL’s own rules by displaying sold data?

Perhaps they’re able to get around the prohibition simply by using sold data from a 3rd party aggregator, instead of from the MLS.  Ingenious, and disingenuous, at the same time.

Real time update…I just called REIL’s compliance department and spoke to a very friendly and helpful person.  She promised to look into it and get back to me.  I explained to her that while I think it’s a silly rule, if other people have found a way around it, I’d like to be able to do the same.

[Read more →]

This just in … thanks to a contribution from Marlow Harris, we now have the skinny on the accuracy of the Seattle real estate data from which Zillow spins its magic, and the results are pretty much the same as here in Menlo Park: the mid 80%’s. If you have access to the MLS in your area and would like to contribute 10 data points, please email me at and I’ll send you an invitation to my iRows spreadsheet.

[Read more →]

[Read more →]