Introducing the StreetEasy Condo Market Index

Is that suppose to compete with the new Urbandig platform?

Hmmm, SE burying Noah. This is supposed to be Noah's moment Sofia. This couldn't wait until next week? :) Just kidding. I guess all's fair in love, war and the Manhattan transparency market!

SEs response: *News Wire* SE says:
"In no way is the timing of this release a result of the recently release UrbanDigs new platform." "The timing is purely coincidental and we wish Noah Rosenblatt much success in the future"

there u can release this today :)

It's an index, how exactly would that "compete" with Noah?

...besides isn't everyday NoahDay on streeteasy?

headlines bro, headlines, no big deal.

rvargas, u work for SE, don't u.

rvargas, easy bro. Noah seems like a stand up guy. u both can compete and make money.

It appears that Noah gets much of the love on the S.E talk forum because he has been pretty active contributing to the forum with his valuable analysis of the market. We all know he's been working his ass off on his new platform and we were regulary updated on it. Basically, he has done what Street easy hasn't. He directly interacted with his target via the forum, made himself known, loved and respected for his work & contributions.

Ha! We don't issue press releases, but if we did it would say exactly that steveF (-sarcasm).

As everyone knows we are fans and friends of Noah, but this is indeed an index, not a platform.

Besides how could you time anything around UD's release date(s) ;-)

J Miller popped one of his 20 year price charts yesterday too. Big week huh?

http://ny.curbed.com/archives/2010/10/19/three_cents_worth_manhattans_20year_housing_cycle.php

sledge: def. respected not sure about loved. But SE is respected as well. Didn't they get us all together in the first place. One big unhappy family with 2 sides battling it out until that bear buys and he switches sides. Like some of our wonderful politicians out there.

No doubt out that Steve, i was only referring to Noah's direct interaction with its target as the main difference with Street easy approach. Street easy is the most valuable tool ever built to service the industry (& consumers) since the RE broker licence so how can't we love and respect them for that!

sledge..absolutely.

I like it. Case Shiller methodology (give or take) comes to Manhattan. Well done SE

Thank you! Could you do the same for townhouses in Brooklyn?

SE, I am impressed as shit that you undertook this effort. BRAVO!!!

Now, SE bulls and bears can argue with a proper price index.

Maly, not really on Brooklyn townhouses: not enough datapoints in the market. This is about as small a dataset from which one would want to constructs such an index; SE can correct me if I'm wrong...

The result looks plausible: down 14% from the peak, up 5% from the trough, trending vaguely lower.

Front_porch isn't going to like the Upper West Side chart: Peak-to-trough -23%.

Not quite the last word, since it's a condo index and the UWS market is dominated by coops. Still, that drop is a pretty strong indication that the Upper West Side was hit just as hard as Manhattan overall - maybe even harder.

"The result looks plausible: down 14% from the peak, up 5% from the trough, trending vaguely lower."

I thought I read on many threads that the market went down 25-30% and that any evidence of a recovery is to be dismissed an unfounded. This new index is going to be unpopular with SE's resident thought police unless it starts producing some bigger losses pretty damn quick...

inonada, you are correct. There wouldn't be enough datapoints for Brooklyn townhouses to construct an index that was statistically significant. It's the same reason why we were unable to do an index for Upper Manhattan, or to break it down by neighborhoods.

This is only the first phase of the launch. We also ran our model to adjust for inflation to see how the Manhattan condo market performed in real dollars.

Our next stage for analysis will be to address the time lag (from contract to recorded sale) by using our listings database and looking at price changes and time on market. So stay tuned for that.

Let us know if there are other things you'd like to see or if you have any questions on our methodology.

"Besides how could you time anything around UD's release date(s) ;-)"

Now that's the streeteasy spirit!

"Starting in 2000, the StreetEasy CMI shows that the Manhattan condo market’s prices were
increasing at a higher rate than both Case-Shiller metro-New York indices, up until September
2001. Additionally, both Case-Shiller indices show that the metro New York market peaked in
early 2006 (NYXR peaked in June 2006 with an index value of 215.83 and the NYXRC peaked
in February 2006 with an index value of 231.54). They do not reflect Manhattan’s large condo
boom seen in 2007 and 2008 where new development condo transactions made up more than
one-third of all Manhattan residential closings during that two-year period which caused prices
to increase more dramatically. In contrast, the StreetEasy CMI shows that the Manhattan market
peaked in March 2008."

streeteasy, not sure I understand. If the StreetEasy CMI is built on paired sales of like units, how would new development sales impact the index?

New developments were only included when there was a subsequent repeat sale. During the condo boom, we saw a dramatic increase in the number of condo closings as well as condo inventory; new developments were a big reason for this. The subsequent repeat sale (after the initial sale) still produced higher price appreciation rates.

At first we did not include 'flips,' or repeat sales that were less than 6 months apart. But in the final iterations of our model, we decided to include them as we felt that they were really part of the Manhattan market.

Did you try to correct for contract date vs. sale date? Not a big issue in most sales, but a big issue in new devs from a few years back where there could be a lag on the order of a year often. Probably doesn't affect the index circa today much, but shifts the 2006-2008 price changes somewhat, I imagine.

sidelinesitter: I thought the consensus - if there was one - was around 20% from peak to trough overall, with a small bounce since then. That seems to have been pretty accurate. Of course, specific sales reflected a bigger drop. Those sales may have gained outsize attention at the time because they were so spectacular and volume was so low.

I would add 25% peak to trough is the nominal decline. if you factor in inflation the decline is much more pronounced.
For this reason would it be possible to add the inflation corrected manhattan condo line? that would be great (the nytimes used to published beautiful CS index graphs with it , then they stopped).
A very good job from SE anyhow. thank you

w81 - I think you missed the sarcasm in my post, which was more to poke some fun at the it's-down-30%-on-its-way-to-50%-and-don't-bother-me-with-"data"-that-suggest-otherwise / bearish-isn't-a-view-it's-a-way-of-life SE regulars. I'll leave it at that and would rather stop the tangent here lest it distract any more from the useful discussion of the cool new index.

inonada, the long contract period for new dev condos wasn't really an issue since we didn't include the initial sale until there was a repeat sale. Even then, the only time period we took into consideration was the difference between the first sale and the 2nd sale.

nicercatch, we'll be adding the inflation-adjusted index lines shortly.

Guys, this is ALL GREAT for the consumer. SE is a great site, and they are a very smart group! We expect them to come out with innovations like this for the benefit of all. And since we did have 10+ official release dates, no way SE could have timed this against our product release yesterday. I consider SE good friends, and I can only hope they continue to build and innovate so that the consumers get more great tools to track this market.

With that said, we believe our new platform to be quite different than this market index tool. If you subscribe, or a subscriber can confirm this, you will see that. The innovation that is our new site, serves a different kind of end result for what I believe to be a different type of market tracking and measurement. In the end, all these new tools are good!

Trust me, we have plenty more tools coming, just as im sure SE has more coming!! Keep on rocking!

Very nice, StreetEasy.

Will be great to see monthly updates on what's happening with pricing. And, of course, like the variation on Case-Shiller methodology.

Would be nice if you could push back the data to coincide with where Jonathan Miller starts so we could also see what an extended bear market and then sideways market looks like.

Thanks.

"At first we did not include 'flips,' or repeat sales that were less than 6 months apart. But in the final iterations of our model, we decided to include them as we felt that they were really part of the Manhattan market."

I agree that flips were part of the Manhattan market but are they part of the trend? I think you can make an argument on both sides to include it or not, so can you show us both? Would it be possible to see a line that removed flips? That would be really interesting!

Great job streeteasy, really good stuff.

Digs, you too. Nicely done.

I'm astounded by how very different the returns seem to be between the StreetEasy methodology and the Miller Samuel methodology.

On Miller Samuel, they show the yearend 1994 price per square foot of a Manhattan condo to be $325. The figure for 3Q 2010 is $1199. That would suggest that if yearend 1994 were indexed to 1000, the index figure for September, 2010 would be 3,689. (Far different from what StreetEasy is showing!)

The Miller Samuel "median sales price" for yearend 1994 was $272,500. That figure had increased to $1,120,000 for September, 2010. That would suggest an indexed 1000 would have risen to 4110.

I'm very baffled by the huge discrepancy.

Are Miller Samuel numbers a poor representation of what has actually transpired with Manhattan condo prices?

"inonada, the long contract period for new dev condos wasn't really an issue since we didn't include the initial sale until there was a repeat sale. Even then, the only time period we took into consideration was the difference between the first sale and the 2nd sale."

I guess what I'm saying is this. Imagine a 2004 new-dev contract at $1M ends up in a 2006 sale at $1M (it would have been $1.3M with a 2006 contract), and then a 2009 resale $1.1M. This results in a datapoint for +10% return 2006->2009 (over which it was arguably more like -20%), but in reality should have been a +10% return 2004->2009. I imagine the effect goes opposite for 2007 contracts closing in 2008/2009, perhaps the original effect being stronger because fewer of the latter contracts ended up as sales.

Maybe an idea to test my hypothesis. I'm guessing if you remove the new dev, you'll see the peak in your graphs shift to the left, and the ending index value being slightly lower. Not saying you should do that officially, I'm sure you've thought out the implications, but maybe you can give a hint as to what would happen?

Topper,

The StreetEasy index marks January 2000 as the baseline with a value of 1000, has January 1995 at 702, and September 2010 at 1852.

The Miller Samuel data (Manhattan coop + condo average ppsf) has Q1 1995 at $275, Q1 2000 at $483, and Q3 2010 at $1095. Translating to a value of 1000 in Q1 2000 to put it on the same scale as this index, we get 569 for Q1 1995, 1000 for Q1 2000, and 2267 for Q3 2010.

SE, I think you have once again come up with a game-changer. Again with NY RE, you've solved something with an "if someone were going to do this right, how would it be done" attitude. Kudos.

I've always wondered how much the ppsf increase has been due to appreciation vs. quality improvements. Certainly, the average Manhattan apartment of 2010 is much better than that of 1995. I think we can now tease this apart. The JM data has a 3.9x increase in ppsf 1995->2010 (condos only, to keep things on the same scale: $307 1995 vs. $1199 2010) while the SE data has a 2.6x over the same period. This translates to an astounding 48% quality improvement over those 15 years, or 2.6% a year.

Or should we not hedonically adjust, Riversider, and just argue that 15 CPW and the walkups it replaced are the same thing?

Case Schiller, Look Out!!

Thanks, inonada!

SE: question please on this in the methodology?

"In terms of market share, condos made up 50.2% of all apartment
closings in Manhattan, while co-ops made up 49.8%, from January 2006 through August 2010."

Any chance we can get a breakdown annually for this? My thinking is that this time period reflects a once in a generation new development boom where I can imagine a surge in condo sales recorded compared to regular trends? I would think that 2008 and 2009 saw the surge as these condos closed.

What Im interested in is what the more normal breakdown of condo sales are, if the dev boom was excluded? Do condos make up something like 30% of total annual sales in normal times? Higher? Did condos make up 60-70% or so of total sales in 2008 when most closed? Or 2009? I know most of the housing stock is coop, but the condo boom that saw closing ins 2008-2009 likely temporary skewed the general breakdown seen in more regular times?

Thx again for this new tool!

SE: I think the correlation analysis is kinda goofed. I tried to reproduce your results because they seemed pretty wrong given a glance at the data. I found that if you use 1-year returns, the correlation to NYXR is 0.66 while that to S&P 500 is 0.25. If you start looking at, say, quarterly returns you get 0.45 to NYXR and 0.07 to S&P 500.

I tried lots of stuff to see what you might have done, and I can get a correlation of 0.54 if I look at the CMI index value against the S&P500 index value, which matches what you got. I have no idea what to do to reproduce your 0.50 correlation to NYXR.

Can you explain what correlation you looked at exactly? Looking at correlations of prices or levels, if that's what you did, is not what you really want to do for the comparison you are trying to make.

(Feel free to email me as the discussion may get overly technical.)

Inonada, N=186

im spinning

Streeteasy - Thanks for adding more transparency to this market - every little bit helps!

Shouldn't people compare the S&P NYC CONDO index to this, not the general NYC index?

brooklyn condos, please.