The Manhattan Sales Solar System from Miller-Samuels

I don't get it. Do the big planets represent large price declines or small price declines?

Bigger means more number of units sold.

WTF?

ali r.
{downtown broker}

So the Upper West Side flew too close to the Sun (hubris?), and three areas are about to collide, and the Heights are probably feeling pretty chilly, but not too much gravity?

Okeh.

1. ha ha ha ha ha - for the representation of Manhattan being the sun, with all of Manhattan neighborhoods being planets. I intend to place a call to my 3rd grade teacher and let her know that all planets are indeed INSIDE the sun, rather than circling it.

2. HA HA HA (louder) for their data showing that UWS has seen ~5% price increases. REALLY? REALLY? and Tribeca has seen 15% price increases? ROFL!!

They should make the top execs of all remaining large brokerage firms in NYC get together in a room and re-enact it, spinning around and orbiting, while dressed up as each neighborhood.

Whoever gets LIC has to wear a Duane Reade uniform.

alanhart, not rite aid?

not a good representation at all. the chart does not accurately display the data in a clear way. very weak.

Inwood: the new breakout neighborhood.

Mr. Miller.... should we deploy our secret decoder rings?

So if I understand this correctly...prices on the UES and UWS have increased since Q1-08?
I hate data that does not support my wishes!
The sun means that these 'hoods' are HOT!
I hate data that does not support my wishes!
Lets go everybody!
Bring me some bad news...

It says that property prices have declined, it just doesnt show it in a clear way which is the purpose of a chart.

Shouldn't Chelsea have been the 7th planet from the Sun?

PS I agree with Ali.... WTF?

Both Chelsea and FiDi are significantly larger markets than Greenwich Village and Soho and Tribeca combined?

"So if I understand this correctly...prices on the UES and UWS have increased since Q1-08? "

You missed on order of "relativity", it doesn't say they increased, it says that they fared x% better than the other neighborhoods (the old saying "If you are in the water with a shark, you don't have to swim faster than the shark, you just have to swim faster than your friends"). See the note that overall prices went down 2.3%. Assuming the center of the circle is the data point, it seems to say that the UWS did about 6% or 7% better than down 2.3%. What does THAT mean? I'm not sure, because it can mean a lot of different things depending on how you define things, but it certainly *could* mean that $PSF went down 2.14% rather than 2.3% for the UWS and 1.95% down for Soho and Tribeca rather than 2.3% and 2.83% down for Harlem rather than 2.3%, etc.

but of course it's also easy to see how this would make it impossible to graph any neighborhood which actually did increase, so it can't be that. So WTF is it?

OK, think I got this.
UES on the baseline shows a 2.3% change(decline in per sq/ft price). UWS, grester decline, Soho even greater decline. This is Q1-09 vs. Q1-08. The hoods with the biggest run-up in price show the greatest declines in % change. Hoods with tiny run-ups in price show a lesser % change.

I think someone at Millersamuel's saw the Star Trek Movie! Damn Trekies! Get a life!

Where's Uranus in all of this?

i think it's at alpine's jersey house.

Was it the stench?

"OK, think I got this. "

Actually I think you have the principal but the sign is reversed (i.e. you have it backwards. But who knows given the total wackiness of the whole shebang).

but I did just think of another possibility: the chart is actually the straight numbers merely adjusted up 2.3% (i.e. the zero mark is actually -2.3%

"Where's Uranus in all of this?"

that's what i get for being obscure.

30yrs, sorry, a moment of pure malevolent joy.

the stench and the necessary ridicule. perhaps not sophisticted humor, but this was snortful for me.

Well, the label on the Y axis clearly states "% Change in Price per Square Foot", so according to this chart, yes the UWS saw an increase in price of 5% (if we assume the middle of the planet is the data point). This shouldn't surprise anyone who's been paying attention. A lot of the properties that closed in Q1 09 were at 15 CPW which went to contract 12 - 18 months earlier at the peak of the market. I am curious to see the Q2 numbers as they will more accurately reflect the dismal Q4 activity.

I will say, acknowledging that I am more bullish but also more cognizant of macro data trends than many on this board, that those claiming the market is down 30% or more don't know what they are talking about. The percentage up/down varies greatly by neighborhood, product type, price point, along with many other factors. A recently renovated property in a great pre-war building on a great street in a great neighborhood still fetches top dollar - I know because one in my building just sold for 6% below peak '07.

No, its change RELATIVE to the decline in Manhattan prices. So the "best" neighborhoods barely increases, and most decreased....but the higher up, the less they decreased. 0% means the average decline in Manhattan.

Awful.

Jason - I don't think that is correct. I expect Mr. Miller will being along before too long to set us straight. The Y axis says what it says - I know the caption references the decline in PPSF, but the Y axis label would have indicated that it is relative to the overall decline if that were the case.

"30yrs, sorry, a moment of pure malevolent joy.

the stench and the necessary ridicule. perhaps not sophisticted humor, but this was snortful for me."

No. What I meant was that I made the joke first, but was too obscure in my reference.

this thread proves what a poorly designed chart it is

"Jason - I don't think that is correct. I expect Mr. Miller will being along before too long to set us straight. The Y axis says what it says - I know the caption references the decline in PPSF, but the Y axis label would have indicated that it is relative to the overall decline if that were the case."

I think you are wrong there *is* a notation written in white covering the first three columns from about -30% to over -40% on the y axis which explains that the numbers are relative to "all of Manhattan". Or am I missing something?

2.3% decline in Manhattan? What happened to 30%?

I am 100% correct. Read his post on curbed, where he explains its the decrease relative to Manhattan, not the absolute decline. I posted this just because its funny.

Oh, JuiceMan! You are so silly!

The chart is perfectly well designed. It's a bar chart done in circles - the size of each circle represents volume, where it sits on the axis is the relative price change. The sun is 0.

I believe the data is quite consistent with the data J displays in his data tab so there shouldn't be a lot of surprises here.

Manhattan prices peaked in 2Q2008 - not 1Q2008.

Prices are for "condos and coops." A lot of the condo sales simply reflect sales that went into contract a long time ago and thus the current price numbers are biased upward. This is well known.

I believe Jonathan has noted in the past that the better way to look at the data is to look at "existing" apartment sales price changes. A proxy for this is coop sales as there seems to be very little new construction that is coops.

As to the chart format - I think it is a great, sophisticated representation of what is happening in the market. I would prefer a chart, though, which is just coops. And the next quarter's chart which will show prices from the peak will be a lot more interesting. But it will still be biased by the new construction problem. I wouldn't be surprised, though, if the MS data meister were to do a coop-only, or existing apartment chart, in addition to the comprehensive version.

Keep it up, J!

Oops...

Ah hah! So this is a "relative" chart rather than an "absolute" chart.

I'd certainly like to see this same chart in "absolute" space as well.

Regardless, I think my above comments remain relevant. The lag between contract signings and closings creates a big problem for guaging price action in new construction.

Jason - point made. Didn't check out the Curbed article. I for one think it's a good chart, but needs some accompanying text to avoid confusion.

Here is the text from the site - answers most of the questions on this thread:

Because many like to think the Manhattan housing market is in another universe (it’s not), I thought I’d try a different visual approach this week. I wanted to show the relationship of some neighborhoods and regions to the overall Manhattan market. Each of the market areas analyzed were based on their change in price-per-square-foot from the first quarter of 2008 to the first quarter of 2009. The data includes both new development and re-sales because I haven’t parsed out the two types by specific neighborhoods back more than a few quarters.

The inclusion of new development does improve the performance of many markets, but the point I am trying to make here is the relationship a specific location has to the overall trend. Financial District/Soho/Tribeca, which have a high concentration of new development, appear to be the most overstated because the sales closing now reflect the market of 12-18 months ago.

The location of each neighborhood/region (sphere) above and below the 0% line is how much above and below the overall Manhattan market that location performed between the two periods. For example, Midtown East/Turtle Bay fell 19.3% but because Manhattan’s overall price-per-square-foot fell 2.3%, the neighborhood fell behind the overall market by 17%.

The relative size of each sphere is based on their number of sales in 1Q 09. No trending intended for this metric although smaller neighborhoods tend to have more volatility in their results. Larger neighborhoods generally have a more diverse housing stock and tend to have less extreme results.

Manhattan Co-op/Condo % Change Price Per Sq Ft 1Q 09 v. 1Q 08 Relative to Manhattan Overall Change [Miller Samuel]

Previous Three Cents Worth [Curbed]

I like MS stuff, but this is just not a very good graph. Charting how percentages compare to 2.3%? And the Y axis and size being redundant?

He should just stick to numbers and line graphs.