Inventory - Question for UD

my new data shows 1,258 listings removed from market over last 30 days...I have OFF MARKET inventory around 10,877 but we are not yet done with research to figure out exactly how we should fine tune how we measure that statistic...I dont want to reveal our formulas here, but Ill give an example of what I mean.

Every listing is handled by an agent and updated. We get all this data. Now, a listing goes from ACTIVE to either TOM or POM...so it gets counted as OFF MARKET inventory. Now, what main update timing rule do you put in - along with many other rules to filter our noise? For example, what do you consider off market? A listing that was changed from ACTIVE to TOM/POM up to 8 weeks ago, 12 weeks ago, 52 weeks ago, 2 years ago, etc..? Clearly you need to really do your research and figure out the proper way to count these things for every metric. Its not as simple as just figuring out how many listings' status's are set to TOM/POM, then you would have 60,000 shadow inventory that clearly is not correct. Hence the difficulties and time its taking to build this platform so far. But as I said, we have temp rules in place that put shadown off market inventory around 10,877 right now. This is down from about 13,455 or so in early 2009 and UP from about 5,200 or so in mid-2007..the surge higher occurred around AUG 2008 right before Lehman, as if sellers took listings off when the market started its plunge down. They seem to put them back on when markets got active again..very interesting

thanks. still pretty high. and pretty complicated.

UD. Please pardon my ignorance. Does your shadow inventory include apts in new developments that have never been listed? For example apts at Tempo that have never made it into a listing but clearly is part of NYC inventory. I also noticed that there are apts now listed at The Harrison that were not listed when I asked the broker for a listing of every remaining available apt for consideration this fall. It seems that the sponsor held back some trophy apts and they are now on the market. If we assume that other developers do the same, is that included in shadow inventory.

I guess what I'm asking is this: When you look at a new development like the Rushmore eg, do you count it as 289 units minus the closed sales and contracts, or do you count only the apts that the developer has officially listed.

let me ask my programmers..give me time to get a response...I think the answer is NO, as if sponsor holds back release of these units, then its NOT in the REBNY broker sharing systems at all until they enter them in as ACTIVE...therefore, this is more of a existing OFF-MARKET tracker and less of a SHADOW tracker...one reason why I didnt use the term SHADOW INVENTORY..

Nice work, Noah!

oops I did say shadow in first response, but probably shouldn't have...thanks topper! trying my best!

one more clarification please. does your new listings number include properties that have been relisted? i'd think not, given what i see on SE, your new listings numbers and your inventory numbers.

or to put it another way, does today's inventory = yesterday's inventory + new listings + relistings - delistings. is that correct?

Well, UD, if you don't mind -- do you have any answer to this: what do you think the new dev pipeline (complete or nearly complete by say year end) is, and do you have any metric by against which you would measure this (say total housing stock, or units closed on avg each year) to gauge how significant this inventory is?...in short, how "big" is the issue

first off, my new systems are totally different than what you've seen on UD since late 2007, powered by SE..my data source is new, and my team and I built everything from ground up to serve certain purposes of the new platform for users.

NEW LISTINGS on my new systems includes brand NEW ACTIVES + BOM's listings with rules to filter out noise and flaws.

ACTIVE INVENTORY has tons of rules in it because broker status updates can be flawed in 100s of ways that would otherwise skew the correct inventory number. But yes it does add new listings, BOM's, and it removes delistings, contracts signed, and POM/TOM...however, the broker last update record plays a big role (again, as I described above - and this is something many will have problems understanding) in how we calculate ACTIVE inventory and we have that preset so it will not be as cut & dry...if an ACTIVE listing was not updated by the broker for X days and it hits our preset rule, it will be removed from the count. So imagine again, lets say a broker does NOT update a listing for 14 days, 28, days, 40 days, 60 days, 100 days, etc...yet status is still set to ACTIVE? Is it really ACTIVE if broker didnt update for 100 days? The answer is no. So rules must be placed and research done to figure out the proper measurements for all these...dont want to get into this any further but is one of may issues we were dealt with facing.

jimSE - hmm, really dont know and hadn't looked into it yet. But the only thing I will say with confidence is that whatever the # of shadow inventory for new devs/conversions is (that is, soon to be ACTIVE inventory that is not yet released by sponsor for marketing purposes), its a heck of a lot LESS than it was in 2005, 2006, and probably 2007.

jim, just consider the sheffield, manhattan house, mirival and 30 west 63rd. just those four properties. and it's close to impossible to tell how many units there are because there are, i believe, still RS tenants in some of the buildings. but the sheffield alone has over 800 units and has sold 250ish, etc., etc.

bjw suggested we try to put together a list by neighborhood, but it would be close to impossible. quite a few of the new developments don't show up on the new development page. and in some developments you'd have to spend hours going through sale by sale to figure out what was a resale and what was a sponsor sale. SE has started adding that info to the listings, but it is very incomplete. i'd really like search parameters that would allow me to find sponsor sales vs. resales. and then you have buildings that may go rental. and buildings that are stalled permanently and those that somehow will miraculously and bizarrely become unstalled. some developers are sitting on some very cheaply acquired land.

Thanks. Since you parse all this data so carefully it surprises me you aren't more focused on this issue.

Is that because your gut feel is that is just isn't a significant issue size wise relative to this market , i.e., that it doesn't represent at a minimum a dampening factor on price movement, and potentially very destabilizing factor if pretend/extend strategy ever cracks and buildings lower prices to actually clear inventory.

Lots of people on this thread tend to think this is a very big issue....and if it is, some rough estimate of the number would obviously be important.

Noah, I know I am a chronic nag to you on this issue, but closings/sold figures are so important.
There is so much emphasis on in contracts, and certainly it is crucial metric. But closings/sold is just as important.Especially today with tightened lending. It is a tough metric as though the bulk average closing from in contract period is 30 to 90 days, it serves as an audit for in contracts. The lagging nature of it is difficult enough, but when nobody anywhere posts any numbers on it until a month after a full quarter, it becomes an worse lagging metric.
It would be great to have even January's sold/closing figure today. We heard how robust the close of the year (Nov/Dec)was for in contracts. We should see a sign of confirmation of it in Jan/Feb.
For example, let's say(making up numbers) Nov. in contracts was 700 and Dec. in contracts was 750, but Jan closings was 550 and Feb closings was 500. That would be some form of a heads up that that market is on a different path. If March proved just as bad, that would indicate too big a difference. it would explain a spike in inventory possibly (units back into inventory) or growth in shadow inventory (listing staying off).
Is it difficult to calculate from acris?
Right now, you'd be the only one posting these numbers.....
One more log on the fire, how about having active in contracts (0 to 120 days) and inactive in contracts (120 days or more).

AR, I see that it's hard to do.

But with all due respect to everyone, all this endless toing and froing about "inventory" without more focus on the new dev numbers seems odd to me, unless the implicit view is that it just isn't such a big deal.

By way of contrast, articles on Miami crash often mention the new dev pipeline as a huge issue.

And Noah, don't think us ungrateful. We know there is a lot of work put into this.
We have become a generation of information junkies, and I think we'll always want more.

truthskr10 - i hear you! i want it too. Fact is, there is a 30-day lag free xls file of rolling sales available on nyc.gov property site, but its incomplete and lagging. The only way to get complete data is through ACRIS Electronic Data Systems...getting 6 years of data, that is just sitting there and public record, as a full dump will cost just under $40,000...arghhhhh!

Getting real time complete data on public record as they are captured, is about $420/mth for Manhattan only. It adds up. I already put tons of money into this, and now the thought of all these costs for just one man like me, is scary. Now, I dont necessarily need 6 years of closed data, I could get 3 for likely half the cost, and the monthly nut for real time ACRIS is not that bad too, but I need to see that these costs are warranted.

Once I launch, I will leave it up to users of my new platform to decide for me. Ill give data away for free, but there will also be a small subscription premium upgrade. I hope people use this, if anything to get a great tool that you will soon see and to show their good intentions to support what I am doing. If I see subscribers sign up, Ill dole out the cash to expand the suite of tools offered to you and make the system better and better. Its up to the support of the users. If nobody uses it, the people will have spoken and I cant justify pouring more money in to build what I want to build

I will tell you I already know what I want to do with that data real time and historical. Just waiting to see if I can justify the costs! I hope I can.

UD -- I too appreciate all your efforts, and your objective market assessments.

But I want to be clear on this new dev question....sorry...again...if the purpose of tracking inventory so carefully is to gauge the market condition, isn't the stock of new dev not yet listed an issue of equal or potentially greater importance (unless it isn't in your view)?

Fair enough.
Now I understand fully the obstacle for these stats.
That's a lot of money.
Looking forward to 2.0.

jimster

Problem is, where do you cut off what's shadow or not. A building that's complete but units just not listed, that's easy. But what about unfinished or stalled? You can't confirm a completion date.That building may stay incomplete for 2 years, transfered to someone or something else, like full floor work lofts, a rental building, or just torn down and into parking lot waiting for the next development boom.

of course not truthsker...these are all great questions and I will always be here to answer and help if my thoughts are asked for.

your new dev question is clear and it is an issue, but if the source of data does not have that information because the sellers/sponsors/developers dont release it, it might be quite an effort to use alternate means to get this and account for that shadow market...I personally dont see it as anything more than interesting..i dont see it as directly impacting buyer confidence or other market dynamics that are affected by outside forces. I see lending rates and stock market movements as more directly impacting buyer and seller confidence than the trend in new dev shadow unreleased inventory

Tryskr. I see those nuances.

Nevertheless, if new dev is a major issue for this market, then the scale of the issue is important. Frequently, articles on re disaster markets discuss the new construction volumes and the unsold completed units as a major factor.

I'm sorry but I don't think it makes sense to spend hours and threads discussing inventory without also really trying to estimate unlisted new dev pipeline, which is afterall the MOST important kind of inventory in terms of market impact because it is has to be sold (or rented) at some point. However, if the number isn't so great then this wouldn't be a big deal but many on here cite new dev pipeline as a significant negative issue.

truthskr, exactly.

jim, people will start talking about it more as things like riverhouse and one madison park occur with greater frequency.

and it's fairly disingenuous to say nobody talks about. miller has been talking about it for ages. i think it's readily apparent that at this rate of absorption plenty of buildings are at risk. when and how many die i'm sure some very sophisticated analytical companies are busy figuring out. but likely they work for people who don't want you to know.

"but it would be close to impossible. quite a few of the new developments don't show up on the new development page. and in some developments you'd have to spend hours going through sale by sale to figure out what was a resale and what was a sponsor sale. SE has started adding that info to the listings, but it is very incomplete. i'd really like search parameters that would allow me to find sponsor sales vs. resales. and then you have buildings that may go rental. and buildings that are stalled permanently and those that somehow will miraculously and bizarrely become unstalled. some developers are sitting on some very cheaply acquired land."

aboutready, you're probably right. Doing the tally I did for Williamsburg took a bit of work, but it wasn't so bad because a) I live there, walk by these sites all the time, and generally harass anyone working there for info, and b) a good amount of information is out there, on a smattering of sites, it just takes some digging. I don't think the latter is true of all neighborhoods, unfortunately. It would be great to get a clearer understanding of all this mess though. I think Noah's work will help a lot, but there will inevitably still be questions and doubts. Any transparency is helpful though.

UD -- thanks again. I see your point. It seems to me the implicit premise in your answer that it is only interesting is that the scale of the new dev pipeline is not so big in your view that it would "wag" the market. (eg, if you were in brickell avenue miami i would guess you would say it was more significant)

yes if I were there I would say it was significant..But in my world in Manhattan, new dev shadow inventory was highest around 2006, the year before the boom and before the bulk of contracts were signed and when building boom went from planning/approval stages in 2004-2005, to pre-marketing stages in 2006, the year before the peak when most units were so called 'shadow'.

When I look at the market behavior at that time, the HUGE shadow inventory did not stop buyers from rushing in to get a deal before the first wave of units sold, and a new amendment issues by sponsor with a price hike for next wave of units...this continued for a while..so to me, the fear was greater than the reality when it these shadow unsold and unmarketed units were way higher than I see out there today.

AR -- do you agree with UD that " I personally dont see it as anything more than interesting..i dont see it as directly impacting buyer confidence or other market dynamics that are affected by outside forces. I see lending rates and stock market movements as more directly impacting buyer and seller confidence than the trend in new dev shadow unreleased inventory"

Onemadison....what a great example.
You have some units sold (so it affected sales figures of the recent past), you have unfinished upper floors (who and how will they be finished), you have a second tower that was supposed to be built (is that really shadow?) you have a slew of units that are listed "already in contract" (which are not likely to close and in my eyes very much active inventory but technically off the grid as in contract).

jim: I am with you on this one. I was looking to buy in the UWS and as I came to realize how much inventory was coming onto the market in that area -- both rentals and new dev shadow inventory -- I backed off. I saw it all in Miami (I am an investor there) and didn't want to get burned. And, after I decided to rent for another two years in NYC and see where the market goes, I was shocked when I started to look at the rental stock in that area. I would say that 60 - 70% of what I was seeing was owned by foreign investors. The rents had plummeted and they were sitting on the market for months (albeit this was off peak months of Jan/Feb). How many months of negative carry do these investors take before they put the apts on the market? And when those 500 new luxury rental units come on the market beginning next month, how much pressure will that put on all the investors in the area? Of course you can't track that shadow inventory but tracking the # of nev dev units would help.

I do think new dev inventory is key. And, truthskr: there has to be a way to measure. For completed buildings, list number of units - sales = one number. Units - sales +contracts= another number. For buildings not complete, you take the date of close from the operating plan and for that month you have units-contracts. Next month you have units - sales= one number, etc. For stalled buildings, you do same (start from start of proposed operating budget) under the assumption that those are units that some vulture will take over. If it becomes a parking lot or other resolution, you remove the number. But you would get some idea of how many units are out there. And, more importantly, what is coming down the road.

we have all that (total units, sold, avail, in contract, off market, etc.) but now you have to TAG each building the proper way to so that new rules dont overextend...and what if the sponsor left out some units, then your new algo will do the math right, but in reality its wrong...lots of potential problems. its never easy.

Thanks apt23. You're looking at the kinds of issues I think are key, and it does surprise me there isn't more discussion along those lines. I've said before, I'm sure that developers and lenders are very focused on the pipeline numbers as they decide strategy going forward.

However, it does seem interesting, make me wonder what to think, when I see UD's take on this as not so critical because he follows this market very closely obviously.

I think some better estimate numbers would make it clear just how big a deal this is or isn't.

well the only thing I would say is, the dev boom was 2004-2005-2006, with much of the shadow inventory hitting around 2005-2006 or so; meaning unreported units held back that were going to come to market...there was much more shadow stuff then there is now...so, how do we explain the markets rise into late 2007 even in the face of all these shadow units? Clearly other forces were playing a bigger role? I dont deny there is an impact from this, but I dont think its over weighted as some think, thats all.

It sounds like a daunting task.
And I see a terrible problem already existing in completed buildings. The ones that have a bunch of units listed "already in contract."
Now when you have an agency doing that, it's a double whammy affect on metrics.
If you had a building with say 60 units. That's 60 proper inventory units.
But now you have a cute broker listing 30 of those to "in contract." When only say 5 are.
Not only do you have 25 units left off the true inventory count, you have 25 units applied to "in contract" numbers, skewing perception of market conditions.
These 25 turn into a disparity of 50. (-25 to inventory, + 25 to sales)

UD: Can't you get the number of units from filing plans? How could sponsor hold back units if you know how many units there are. Rushmore is 289 units. If dev holds back or only lists 223 units, it should still be reflected as 289 units, no? Of course, more problematic buildings exist like Apthorp, but they have to file too and it seems all numbers should begin with the filings, not what developer says.

I take your argument about huge inventory that didn't frighten buyers in 2007 but that was a different time -- and anyone who has had to resell has been proven wrong. But it did affect the developers -- they knew it was too much to absorb and they pulled back -- except those that chose to continue to soak banks as they did in Miami. Richard LeFrak started proposing giving green cards with apt purchase in 2008. Now that a form of reason has hit the market, accurate info on inventory is key. I applaud your efforts.

I guess the issue is what's the new dev overhang that, now that the music has slowed, has to be absorbed, and is it significant enough as a percentage of the market that it will dampen prices, or more if banks stop pretending/extending.....

truthskr 10 These 25 turn into a disparity of 50. (-25 to inventory, + 25 to sales)

Yes but if we had a rolling monthly number of in contract numbers that never close, those disparities would be outed and should be an alarm bell for buyers. your point on in contract is a good one. you are right, if 20% say of "in contract" never close, it is something important to look at -- tighter lending, growing #s of unqualified buyers, distressed developers trying to hoodwink banks, etc.

except for downtown, RSB, a couple of midtown west buildings, and maybe chelsea for some reason, most of our new development didn't start closing until 2006 at the earliest. and there was huge demand, we had had almost no new condo developments in the '90s.

some of the buildings downtown have been closing for ages, and look where they are.

at some point the absorption rate becomes just too unappealling to the banks.

at some point the absorption rate becomes just too unappealling to the banks.

And the owners. Can't tell you how many people recently have told me they are ready to sell their Manhattan apts, they are just waiting for the market to move back up. Meaning that they just don't want to face the losses quite yet. Of course, I travel in a lot of middle aged, newly, empty-nested circles. They will eventually take the hit to their bottom line just like the took it to their stock portfolios. Their mindsets are that there is one or two million of net worth built into those apts. The thought that they may still come up with a couple hundred thou in profit cannot assuage the thought of the loss from the peak or the loss of esteem they sit with when they contemplate their all important net worth. My guess is that when taxes go up and the market hasn't, there will be a middle aged exodus.

apt23

yeah it's a point Ive been hammering(apparantly in poor or unclear fashion until this thread) on various inventory threads for three months.
The way I put it in my last post is more clear.
However, Noah has illustrated the challenges to making this a reality.

Have a look at UD's chart on pending and closing sales.
Particular attention to January '08.
The discrepency between closings and pendings Jan/Feb is the real birth point of impending doom.
However with 4/5 month lagging data on real sales (and limited to quarterly figures pushing you really to end of Q2),who could really now how bad it was coming until June of '08?
http://www.urbandigs.com/2010/02/pending_sales_90day_ma_closing.html

BTW Noah, could you explain what the left and right column numbers represent?
The 90,80,70,60, etc. and the 8000,7000,6000,5000, etc.

UD: It seems that an accurate number for unit inventory - including shadow, contracts, closings would be so important for investors and govt agencies that I wonder if you have tried to use the Neilson model. That is, get the Reits, govt agencies, vulture funds etc to pay for your costs to get the info in advance. If this really isn't being done with any accuracy it would seem that REITS should not only back you, they should pressure the govt to give you access to this public information for free --or at least much smaller administrative fee.

I would think a big sales argument would be that the only guy with a handle on the actual inventory info, (Miller from prudential) went into biz with a hedge fund and is now in competition with the REITS. Clearly if all the REITS had access to real, accurate info through subscription, it would stave off competition from the hedgies.

apt23
That's a good point for both shadow and closings.

aboutready is a toilet whore

the 2000-8000 right y axis is pending sales...the 40-90 right y axis is the daily 90 day moving average of closed sales from ACRIS...Im about to make a big decision to get more data that will cost me like 30k, and 420/mth for real time closing data..but will be the final cog in our quest of the best analytics...basically Im making a bet that users will not hesitate to sign up for the premium services at about $15-$20/mth or so rate.

Large surge in listings today -

Real estate for sale
in Manhattan
We found 377 listings updated after March 9, 2010
Median price: $1,550,000 Median size: 1,410 ft² Median price per ft²: $1,131
Information on Manhattan