(Reposted from the last thread, since this is where it belongs...)
May's number is not nearly as good as it looks, IMO.
I've got a spreadsheet going that tracks C-S using a 12-month rolling average rather than the 3-month average that S&P reports, which to my layman's mind makes a hell of a lot more sense (since real estate sales are, you know, seasonal).
And in fact, using this method the year-over-year rate of change went from -8.9% in April to -10.1% in May ... so the downward move in prices has not slacked off but, to the contrary, is picking up speed.
I suspect you won't hear CNBC lead with that conclusion today, though.
That is how it looks to me as well, I was puzzled by CR's focus on the month over month numbers...unless I misunderstood his "the rate of price declines has slowed a little."
As of early 2007 a number of CS cities were more than twice their average prices in Jan 2000: Phoenix, Los Angeles, San Francisco, San Diego, Washington DC, Miami, Tampa, Las Vegas and New York.
How many of them are still twice as expensive? As of May, none.
Hopefully CR will update and repost his excellent graphs of LA (C/S) in the early '90s boom/bust and this last decade. That illustrates that this decline in nominal numbers isn't even halfway over.
I was puzzled by CR's focus on the month over month numbers ... unless I misunderstood his "the rate of price declines has slowed a little."
In his defense, even the year-over-year percentage declines appear to have "slowed a little" ... if you take the figures provided by S&P at face value.
But right there on the C-S Index home page, you see: "The S&P/Case-Shiller Home Price Indices are calculated monthly using a three-month moving average and published with a two month lag."
For an asset as illiquid and as seasonal as real estate, constructing trend lines by going back only 3 months makes no sense to me at all. And if you apply a 12-month moving average to these same data points, the slope of the line tells you a very different story.
But then again, I'm just a layman, not a highly-paid S&P economist, so what do I know?
Extreme makeover family is MER in miniature, except, they were used to living precatiously and can be excused for thinking short term. Merril and pals supposedly knew better and used other people's money--both will end up on the dole.
It does seem like the rate of decline is slowing a little. But I can't help but think it is a head fake, a little eye-of-the-hurricane calm while we finish up the sub-prime wave and await the option-arms.
some investor guy writes:
As of early 2007 a number of CS cities were more than twice their average prices in Jan 2000: Phoenix, Los Angeles, San Francisco, San Diego, Washington DC, Miami, Tampa, Las Vegas and New York.
How many of them are still twice as expensive? As of May, none.
How much is that same dollar worth compared to Jan 2000?
All of the price models are badly flawed, CS included.
There is no data base that I am aware of that can incorporate all of the factors of an apples to apples price discovery.
We may judge that prices are declining but at what rate and from what base. If people buy cheaper houses, the total housing price base is thought to have declined.
Incredible that people would use flawed and mis-leading data. Admitedly there are distressed areas of the country and distressed individual properties. Some will bottom tomorrow and some in 10 years.
I suppose it is still possible to drown in a river that 'averages' 2 feet deep.
Normally UK auctioneers will not accept properties unless the seller is seriously prepared to sell. The results speak for themselves.
This has been the pattern of the last month or so of UK auctions. Most of this stuff is not so desirable but even so. Presumably by the Autumn stuff will be priced to sell no matter what price is offered.
Britain has a massive crisis as does Spain. The dominos are falling one by one.
Does anyone else find these drops in oil oddly timed to situations where the banks are poised to drop? Gee, it is almost as if it is being manipulated. But I figure at $3-4 dollars per shot, they've only got 5-6 bullets left.
I hate to sound like I'm channeling Anonymouse here, but there's literally nothing but bad news on the business page of Google News today, yet the market is up 1%-plus.
If it can finish this strongly in the face of this much pessimism (granted, a big "if"), that would IMO be a big flashing indicator of a sharp (if possibly short-lived) bear-market rally in the making.
Ross: "If people buy cheaper houses, the total housing price base is thought to have declined."
Case-Shiller uses PAIRED sales to take care of this. In other words, they look at what the same house sold for at different times.
They may have other flaws (like why use a 3-month average), but the use of paired sales fixes the common problem with the NAR's worthless "median" figure.
Median and average prices are, of course, affected by market mix. CS is matched sales of the same house. Can't get much more "apples to apples" than that.
"It does seem like the rate of decline is slowing a little. But I can't help but think it is a head fake, a little eye-of-the-hurricane calm while we finish up the sub-prime wave and await the option-arms."
-kis
I was talking with a neighbor last weekend who is a professor at the local state college and told her about the coming option-ARM mess and she hadn't heard anything about it. People were pissed about "sub-prime", wait till this hits the fan. When I told her about negative amortization and cap triggered resets, she just goes: "WTF!, You're shitting me right!"
Any 1 month change in and of itself should be considered statistical noise. I would expect plenty of noise given the drastic change in sales mix to REO & short sales. I wonder if auctions (black box transactions) are factored. A lot of shadow activity with vulture groups etc as the banks get serious about unloading NPAs.
Mook:
My data shows not only downward price acceleration in May and June but also in July. CS charts are no doubt accurate and attractive, but in a market falling this fast, price trends in May are getting old.
The startling price drops in my local zip codes are being set by bank REO sales occurring 10-20% below comparable listed asking price averages. Daughter's gated tract continues to show comparable asking prices averaging $220/sf, even with 2 recent REO sales at $150 and $167/sf, and 6 more coming. That's the front edge of this wave for me, so I'm tracking preforeclosure and bank REO inventories, both of which are accelerating faster than REO sales.
The rate of price decline won't moderate until preforeclosures and bank inventory numbers start down, and I'll see all that here way before it shows up in 2 month old Case charts, nice as they are.
I know you are just north of me. My body shop guy just dropped a repair off. He mentioned he just signed a lease on a new building in Port Charlotte. Old building lease...Just shy of 12/sq ft. Brand new lease...4/sq ft. 5 years no increase. Next 5 tiny increase. Option to get out at 10 or sign for 10 more with a tiny bumps. He mentioned even after 20 years he will barely be at his old payment.
Instant 100k of freed up cash per year...Deflation central around here.
If you take out the "problem markets" of S. Florida, S. California, Nevada and Arizona and average the remaining cities, prices were actually UP in the rest of the country in May!
"If you take out the "problem markets" of S. Florida, S. California, Nevada and Arizona and average the remaining cities, prices were actually UP in the rest of the country in May!"
If you ignore the heart attack the victim would still be alive.
"a professor at the local state college and told her about the coming option-ARM mess and she hadn't heard anything about it."
I am continually amazed by people I know who are far more extensively educated than I am PhD.s, professors, professionals, etc who are completely unaware of most economic and political issues that are discussed here routinely. I've come to the conclusion that while they may be well-read experts in their particular field, when it comes to matters outside their specialty they rely on TV and whatever the pathetic excuse for a local paper they subscribe to.
some investor guy writes:
As of early 2007 a number of CS cities were more than twice their average prices in Jan 2000: Phoenix, Los Angeles, San Francisco, San Diego, Washington DC, Miami, Tampa, Las Vegas and New York.
NY real estate isn't anywhere close to a bottom. The northeast generally is on a 6 month lag to the southern cone of US. Focusing ion price to the exclusion of affordability and or rent ratios is useless
ostkptrnr writes:
If you take out the "problem markets" of S. Florida, S. California, Nevada and Arizona and average the remaining cities, prices were actually UP in the rest of the country in May!
ostkptrnr | 07.29.08 - 10:29 am | #
Looking at the data, I see 7 areas with higher prices month over month: Den, Atl, Bos, Minn, Charlotte, Portland, Dal.
I graphed the month-over-month changes for the last 10 years. You can see some seasonality. I believe that you will find the rate of decrease has declined remarkably in the last few months even after you take out seasonality.
I actually average the seasonal changes and subtracted them off the trend for the last 12 months. Things are turning better very quickly.
How much of the CS index is self-selecting? Sales volumes around me are off 50%, mainly because sellers are not reducing price to clear inventory. Does that leave a higher proportion of sales at or near asking price? Or do 'May' sales actually represent March and April contracts, so we are looking back at unfounded optimism then?
This really does look like a tale of two markets. On the one hand, you have So Cal, Florida, Arizona, Las Vegas, where prices are 30% off peak and freefalling.
Then you have Denver, Charlotte, Dallas, Portland, etc., where prices seem to be down 5-10% from peak and rising again.
.........Riverside........San Diego
7/12....16,400...........10,000
7/23....17,900...........11,300
7/26....18,300...........11,600
7/28....18,500...........11,700
Bank REO inventories in 6 SoCAL counties on 7/28: 64,082.
June REO sales, same 6 counties: 7100 est.
Months of bank REO inventory at June sales rate: 9 (Nine).
You all in the rest of the U.S. go ahead and start your real estate recovery. We'll catch up with you later.
Nothing happens 'very quickly' in housing markets...they move like a fully loaded VLCC (super tanker), LOTS of momentum in whichever direction they are moving.
Mook made a point in your first post. Case Shiller was created for trading purposes, S&P makes $$ every time its published.
THere's little visibility into the index but from what I've read, its weighted towards more volitile and expensive markets.
Also, if their looking for pairs, when the high end isn't selling, they're only used the low end for tracking, pairing up prior overpriced sales of tract homes with their foreclosure prices.
Yes there is some huge pain out there.
Even though your dealing with highly respected PHDs, these are the tabloids of our day - shocking data selling stuff.
I understand CS uses matched sales. The problem is that more 'tails' are being matched. A very large percentage of the housing stock never turns over. How can you match what isn't bought to what was never sold.
Exceptions: Condos in Miami, McMansions in SD. So what part of the total housing stock turned over these last 5 years. Some houses 4 or 5 times. Others-never.
CS studies are the best we have. My problem is that they are no where good enough to rely on when making investment decisions.
White noise or is it possible to glean something out of these numbers? I dunno.
But here's a thought--the Lucky Seven (ATL, DEN, DAL, MINN, etc.) were the areas that participated least gleefully in the recently ended boom. The max pain seems to remain focused in the areas that bubbled the most.
I have this enduring question which is, are we all pigs sliding through this python together, or will be impact of the boom and bust be felt differentially throughout the process? In other words, will the worst of the hurt be concentrated where there was most of the gain, or will everything eventually be dragged down about the same, and it's just a matter of timing, when it hits where?
I think these numbers support the proposition that the worst of the hurt will be concentrated in the markets that boomed most, but I'm not sure. And I certainly don't rule out the possibility of contagion driven by a long term souring of the economy, tainting everything in the end.
So how come there are no (noticeable) price drops in the heart of Silicon Valley, Manhattan etc? Do these places just lag?
My take (living very, very close to Silicon Valley) is that a good fraction of people there have access to a little more money than average, as well as access to a little more self-deception than average about what is going on.
They must not read "The Ant and the Grasshopper" in California schools.
"I am continually amazed by people I know who are far more extensively educated than I am PhD.s, professors, professionals, etc who are completely unaware of most economic and political issues that are discussed here routinely."
I work at a university and am acquainted with a fair number of tenured academics: some of these guys barely wipe themselves, much less write a coherent sentence.
I used to work in the registrar's office, and I remember one prof who wouldn't take a lecture hall with a stage, because she was obsessed with the idea of students looking up her dress. There was another guy, a geo lecturer, who bitched and whined because he'd have to carry his rock samples 100 yards from his car to his classroom. Heard of dollies, homeboy? He was the same guy who turned in a catalogue course description that began "How are earthquakes happen?" Native-speaker, too.
Rant off, but jeez; most of these guys are mediocre minds at best, but they've got arrogance enough for a genius.
... The problem is that more 'tails' are being matched. ... My problem is that they are no where good enough to rely on when making investment decisions. ...
My understanding is that real estate prices are set on the margin, so the Case-Shiller methodology captures the right mix.
To clarify, during the run up in prices, the fact that 90% of the comparable housing stock had been bought for a quarter of the current asking price never deterred buyers or sellers from completing transactions. The same principle holds on the way down modulo the stickiness of real estate which just slows the drop.
Feel free to correct my understanding. Constructive criticism is appreciated.
Case Shiller averages three months. So May data is average of March, April, May.
The slight inflection in the curve is what would normally be the "spring bounce" as home prices are normally a bit higher in the spring to early summer.
In this case, there is no bounce, just a slight upward inflection in the general trend of rapidly falling home prices.
"Confused writes:
So how come there are no (noticeable) price drops in the heart of Silicon Valley, Manhattan etc? Do these places just lag?"
Case Shiller's parent company suggests not only do they lag, but they just do better over the long run. All things being equal, closer to the core is the best place to be.
While I often watch Housingtraker.net for more updated information, it has list prices. It's a nice data series that is current each monday.
What is changing is there are larger and larger discounts to original listing price, and even the final list price before selling. As an example, in Malibu actual selling prices are typically around 20% off of initial listing price Real Estate Blog - Have We Bottomed Out Yet? . At the peak, sales prices were slightly above list prices.
"Then you have Denver, Charlotte, Dallas, Portland, etc., where prices seem to be down 5-10% from peak and rising again."
Those same cities had a similar price increase April to May in 2007. Is this the buying season effect? Maybe people are more likely to pay a premium during peak season?
Compare CS to PMI's Summer US Market Risk Index. Then take a look at the historical appreciation rates for those with the lowest downward risk, like Pittsburgh and Indianapolis, and you get a bit clearer picture.
What the CS numbers don't reflect in more expensive bubble markets like San Diego, is that while what sold in 2005 for $500k is now $300k, the higher end is only off about 20% over the same time frame, but with the cost of jumbo money going up, the high end will start to see the same dramatic drop.
The CS index for Phoenix was 100 in Jan 2000. At 5% annual appreciation for 8 1/2 years, the index would be at 151. Today's number is 157. I would say valuations in Phoenix are pretty close to where they should be. You can even get SFH's in good neighborhoods that cash-flow.
Of course, the inventory is still high, and prices will probably over-correct.
Mike in AZ writes:
The CS index for Phoenix was 100 in Jan 2000. At 5% annual appreciation for 8 1/2 years, the index would be at 151.
I just checked the Phoenix CPI minus shelter from Dec 2001 (admittedly a little late) to second half 2007 from here. It's 16%, or about 2.75% annually.
So, at 2.75% annual appreciation, the index "target" would be at 126, suggesting there might be 20% more downside to regress, excluding effects from overbuilding as well as pricing and demand destruction from economic effects. (and the 2.75% increase per annum vs the time the remaining regression takes).
It is worth looking at the historical C/S data from the last downturn to note there is a relative maxima almost every August (summer bounce) before the general downward trend continues. We'll know things are really bad if we can't even muster that bit of "happiness".
Ross said: "I understand CS uses matched sales. The problem is that more 'tails' are being matched. A very large percentage of the housing stock never turns over. How can you match what isn't bought to what was never sold."
I've always thought the CS numbers, being "capitalization-weighted" (heaviest weight given to the housing markets that are most expensive), were exaggerating the problem.
Your point about the CS data being "self-selecting", along with your more-recent post, indicate even more exaggeration that hadn't even occurred to me.
The peak in sales was a couple of years back, which was also the peak of prices. So what are CS's paired sales skewed towards? The people who bought near the top of the market and are now being forced to sell. IOW, CS data is skewed towards the experience of homeowners who are in absolutely the worst position.
It's no different than assuming that all the stock market investors who bought at the recent market top in October, 2007 and are now looking at the heaviest losses are representative of every stock investor, including the ones who bought before the peak, after the peak or bought better-quality, better-valued stocks that haven't suffered nearly as much as the broader indices.
El primero.
On topic, this is what happens when you give a house away on tv, the people just yank out all the equity they can and then let the foreclosure begin:
Yahoo! 404 - Page Not Found
Sadly not surprising at all. Glad we get to bailout, I mean modify the mortgage of these classy down on their luck people.
What, first Merrill is down, and now, housing is down too! Whats next?
(Reposted from the last thread, since this is where it belongs...)
May's number is not nearly as good as it looks, IMO.
I've got a spreadsheet going that tracks C-S using a 12-month rolling average rather than the 3-month average that S&P reports, which to my layman's mind makes a hell of a lot more sense (since real estate sales are, you know, seasonal).
And in fact, using this method the year-over-year rate of change went from -8.9% in April to -10.1% in May ... so the downward move in prices has not slacked off but, to the contrary, is picking up speed.
I suspect you won't hear CNBC lead with that conclusion today, though.
Futures up, Merrill Lynch writedown looked at as a healthy purge......
Yehawwwwwwwwww.........
Also worth noting that in May, for the first time, a metro market in the C-S blew through the "30% off its peak" figure.
In fact, three of them managed the feat, all in the same month. Take a bow, Phoenix, Vegas, and Miami!
Mook,
That is how it looks to me as well, I was puzzled by CR's focus on the month over month numbers...unless I misunderstood his "the rate of price declines has slowed a little."
As of early 2007 a number of CS cities were more than twice their average prices in Jan 2000: Phoenix, Los Angeles, San Francisco, San Diego, Washington DC, Miami, Tampa, Las Vegas and New York.
How many of them are still twice as expensive? As of May, none.
Hopefully CR will update and repost his excellent graphs of LA (C/S) in the early '90s boom/bust and this last decade. That illustrates that this decline in nominal numbers isn't even halfway over.
I was puzzled by CR's focus on the month over month numbers ... unless I misunderstood his "the rate of price declines has slowed a little."
In his defense, even the year-over-year percentage declines appear to have "slowed a little" ... if you take the figures provided by S&P at face value.
But right there on the C-S Index home page, you see: "The S&P/Case-Shiller Home Price Indices are calculated monthly using a three-month moving average and published with a two month lag."
For an asset as illiquid and as seasonal as real estate, constructing trend lines by going back only 3 months makes no sense to me at all. And if you apply a 12-month moving average to these same data points, the slope of the line tells you a very different story.
But then again, I'm just a layman, not a highly-paid S&P economist, so what do I know?
Yahoo has a story about the 2005 Extreme home makeover home this morning:
Now being foreclosed.
Lucky recipient family used it as an ATM machine, took out 450k on it.. at a wild guess, spent it on consumer crap.
damn sorry beaten by miles to this story didn't read before posting
Are we on the other side of the inflection point?? It seems so...
OT Newsflash : Merrill Lynch Prices 380 Million Share Offering at $22.50 a share. (Story to Come)
Actually, the Extreme Makeover family apparently sank the loot into a failed construction business.
That's just beautiful.
AP's lead conflicts with CR's conclusion. I know who I trust, but this is what everyone will hear today:
"A closely watched housing index shows home prices fell by the steepest rate ever in May, as the housing slump continued to deepen nationwide."
Extreme makeover family is MER in miniature, except, they were used to living precatiously and can be excused for thinking short term. Merril and pals supposedly knew better and used other people's money--both will end up on the dole.
It does seem like the rate of decline is slowing a little. But I can't help but think it is a head fake, a little eye-of-the-hurricane calm while we finish up the sub-prime wave and await the option-arms.
some investor guy writes:
As of early 2007 a number of CS cities were more than twice their average prices in Jan 2000: Phoenix, Los Angeles, San Francisco, San Diego, Washington DC, Miami, Tampa, Las Vegas and New York.
How many of them are still twice as expensive? As of May, none.
How much is that same dollar worth compared to Jan 2000?
I just heard a "Dow Jones Report" on the car radio - they mentioned "upbeat" home sales report as the reason for the market moving up. Weird.
All of the price models are badly flawed, CS included.
There is no data base that I am aware of that can incorporate all of the factors of an apples to apples price discovery.
We may judge that prices are declining but at what rate and from what base. If people buy cheaper houses, the total housing price base is thought to have declined.
Incredible that people would use flawed and mis-leading data. Admitedly there are distressed areas of the country and distressed individual properties. Some will bottom tomorrow and some in 10 years.
I suppose it is still possible to drown in a river that 'averages' 2 feet deep.
some results of a UK mortagee auction being held today by
Property Auctions UK London Nationwide - Barnard Marcus Auctions
41 unsold 93.5
42 unsold 118
43 unsold 79
44 unsold 95
46 sold 250
47 unsold 75
48 sold 150
49 unsold 134.5
50 unsold 285
51 unsold 183
57 unsold 477
57a unsold 12.5
58 unsold 172.5
59 sold 140
60 unsold 102.5
62 unsold 116.5
63 unsold 195
64 unsold 155
65 unsold 125
67 sold 115.5
70 unsold 135
72 sold 61.5
79 unsold 197
80 unsold 71.5
80A unsold 122
82 unsold 54
83 unsold 95
85 unsold 144
86 sold 83
88 unsold 36.5
90 unsold 71
91 unsold 53
92 unsold 315
93 unsold 152
94 unsold 415
95 unsold 149
96 unsold 222
97 unsold 193
98 sold 13
Normally UK auctioneers will not accept properties unless the seller is seriously prepared to sell. The results speak for themselves.
This has been the pattern of the last month or so of UK auctions. Most of this stuff is not so desirable but even so. Presumably by the Autumn stuff will be priced to sell no matter what price is offered.
Britain has a massive crisis as does Spain. The dominos are falling one by one.
Does anyone else find these drops in oil oddly timed to situations where the banks are poised to drop? Gee, it is almost as if it is being manipulated. But I figure at $3-4 dollars per shot, they've only got 5-6 bullets left.
And one offhand comment to add:
I hate to sound like I'm channeling Anonymouse here, but there's literally nothing but bad news on the business page of Google News today, yet the market is up 1%-plus.
If it can finish this strongly in the face of this much pessimism (granted, a big "if"), that would IMO be a big flashing indicator of a sharp (if possibly short-lived) bear-market rally in the making.
Ross: "If people buy cheaper houses, the total housing price base is thought to have declined."
Case-Shiller uses PAIRED sales to take care of this. In other words, they look at what the same house sold for at different times.
They may have other flaws (like why use a 3-month average), but the use of paired sales fixes the common problem with the NAR's worthless "median" figure.
Ross-
Median and average prices are, of course, affected by market mix. CS is matched sales of the same house. Can't get much more "apples to apples" than that.
"It does seem like the rate of decline is slowing a little. But I can't help but think it is a head fake, a little eye-of-the-hurricane calm while we finish up the sub-prime wave and await the option-arms."
-kis
I was talking with a neighbor last weekend who is a professor at the local state college and told her about the coming option-ARM mess and she hadn't heard anything about it. People were pissed about "sub-prime", wait till this hits the fan. When I told her about negative amortization and cap triggered resets, she just goes: "WTF!, You're shitting me right!"
Any 1 month change in and of itself should be considered statistical noise. I would expect plenty of noise given the drastic change in sales mix to REO & short sales. I wonder if auctions (black box transactions) are factored. A lot of shadow activity with vulture groups etc as the banks get serious about unloading NPAs.
Mook:
My data shows not only downward price acceleration in May and June but also in July. CS charts are no doubt accurate and attractive, but in a market falling this fast, price trends in May are getting old.
The startling price drops in my local zip codes are being set by bank REO sales occurring 10-20% below comparable listed asking price averages. Daughter's gated tract continues to show comparable asking prices averaging $220/sf, even with 2 recent REO sales at $150 and $167/sf, and 6 more coming. That's the front edge of this wave for me, so I'm tracking preforeclosure and bank REO inventories, both of which are accelerating faster than REO sales.
The rate of price decline won't moderate until preforeclosures and bank inventory numbers start down, and I'll see all that here way before it shows up in 2 month old Case charts, nice as they are.
Anonymouse | Homepage | 07.29.08 -
I know you are just north of me. My body shop guy just dropped a repair off. He mentioned he just signed a lease on a new building in Port Charlotte. Old building lease...Just shy of 12/sq ft. Brand new lease...4/sq ft. 5 years no increase. Next 5 tiny increase. Option to get out at 10 or sign for 10 more with a tiny bumps. He mentioned even after 20 years he will barely be at his old payment.
Instant 100k of freed up cash per year...Deflation central around here.
Chris
If you take out the "problem markets" of S. Florida, S. California, Nevada and Arizona and average the remaining cities, prices were actually UP in the rest of the country in May!
"If you take out the "problem markets" of S. Florida, S. California, Nevada and Arizona and average the remaining cities, prices were actually UP in the rest of the country in May!"
If you ignore the heart attack the victim would still be alive.
sheesh. Those parkets ARE center of gravity.
"a professor at the local state college and told her about the coming option-ARM mess and she hadn't heard anything about it."
I am continually amazed by people I know who are far more extensively educated than I am PhD.s, professors, professionals, etc who are completely unaware of most economic and political issues that are discussed here routinely. I've come to the conclusion that while they may be well-read experts in their particular field, when it comes to matters outside their specialty they rely on TV and whatever the pathetic excuse for a local paper they subscribe to.
some investor guy writes:
As of early 2007 a number of CS cities were more than twice their average prices in Jan 2000: Phoenix, Los Angeles, San Francisco, San Diego, Washington DC, Miami, Tampa, Las Vegas and New York.
NY real estate isn't anywhere close to a bottom. The northeast generally is on a 6 month lag to the southern cone of US. Focusing ion price to the exclusion of affordability and or rent ratios is useless
Wow, we've been screaming "wheeeeeeeeeee" on the downside of this roller coaster ride for quite awhile now.
How many times have we paused to catch our breath?
ostkptrnr writes:
If you take out the "problem markets" of S. Florida, S. California, Nevada and Arizona and average the remaining cities, prices were actually UP in the rest of the country in May!
ostkptrnr | 07.29.08 - 10:29 am | #
Looking at the data, I see 7 areas with higher prices month over month: Den, Atl, Bos, Minn, Charlotte, Portland, Dal.
I graphed the month-over-month changes for the last 10 years. You can see some seasonality. I believe that you will find the rate of decrease has declined remarkably in the last few months even after you take out seasonality.
I actually average the seasonal changes and subtracted them off the trend for the last 12 months. Things are turning better very quickly.
Effe
How much of the CS index is self-selecting? Sales volumes around me are off 50%, mainly because sellers are not reducing price to clear inventory. Does that leave a higher proportion of sales at or near asking price? Or do 'May' sales actually represent March and April contracts, so we are looking back at unfounded optimism then?
This really does look like a tale of two markets. On the one hand, you have So Cal, Florida, Arizona, Las Vegas, where prices are 30% off peak and freefalling.
Then you have Denver, Charlotte, Dallas, Portland, etc., where prices seem to be down 5-10% from peak and rising again.
Glad I live in Denver.
In June, sales activity was way off. May numbers are a bit out of date, IMHO.
Bank REO inventories in SoCAL, weekly:
.........Riverside........San Diego
7/12....16,400...........10,000
7/23....17,900...........11,300
7/26....18,300...........11,600
7/28....18,500...........11,700
Bank REO inventories in 6 SoCAL counties on 7/28: 64,082.
June REO sales, same 6 counties: 7100 est.
Months of bank REO inventory at June sales rate: 9 (Nine).
You all in the rest of the U.S. go ahead and start your real estate recovery. We'll catch up with you later.
It is interesting to superimpose Presidencies on the charts and see how they fared. The Bushes just aren't lucky with the economy!
"So Cal, Florida, Arizona, Las Vegas, where prices are 30% off peak and freefalling."
Spanish curse?
Its not over in Denver yet. Outside the bubble areas its the job market and the broader economy which will be the major catalyst for declining prices.
So how come there are no (noticeable) price drops in the heart of Silicon Valley, Manhattan etc? Do these places just lag?
Nothing happens 'very quickly' in housing markets...they move like a fully loaded VLCC (super tanker), LOTS of momentum in whichever direction they are moving.
Mook made a point in your first post. Case Shiller was created for trading purposes, S&P makes $$ every time its published.
THere's little visibility into the index but from what I've read, its weighted towards more volitile and expensive markets.
Also, if their looking for pairs, when the high end isn't selling, they're only used the low end for tracking, pairing up prior overpriced sales of tract homes with their foreclosure prices.
Yes there is some huge pain out there.
Even though your dealing with highly respected PHDs, these are the tabloids of our day - shocking data selling stuff.
I understand CS uses matched sales. The problem is that more 'tails' are being matched. A very large percentage of the housing stock never turns over. How can you match what isn't bought to what was never sold.
Exceptions: Condos in Miami, McMansions in SD. So what part of the total housing stock turned over these last 5 years. Some houses 4 or 5 times. Others-never.
CS studies are the best we have. My problem is that they are no where good enough to rely on when making investment decisions.
Stepping off soapbox.
White noise or is it possible to glean something out of these numbers? I dunno.
But here's a thought--the Lucky Seven (ATL, DEN, DAL, MINN, etc.) were the areas that participated least gleefully in the recently ended boom. The max pain seems to remain focused in the areas that bubbled the most.
I have this enduring question which is, are we all pigs sliding through this python together, or will be impact of the boom and bust be felt differentially throughout the process? In other words, will the worst of the hurt be concentrated where there was most of the gain, or will everything eventually be dragged down about the same, and it's just a matter of timing, when it hits where?
I think these numbers support the proposition that the worst of the hurt will be concentrated in the markets that boomed most, but I'm not sure. And I certainly don't rule out the possibility of contagion driven by a long term souring of the economy, tainting everything in the end.
So how come there are no (noticeable) price drops in the heart of Silicon Valley, Manhattan etc? Do these places just lag?
My take (living very, very close to Silicon Valley) is that a good fraction of people there have access to a little more money than average, as well as access to a little more self-deception than average about what is going on.
They must not read "The Ant and the Grasshopper" in California schools.
"I am continually amazed by people I know who are far more extensively educated than I am PhD.s, professors, professionals, etc who are completely unaware of most economic and political issues that are discussed here routinely."
I work at a university and am acquainted with a fair number of tenured academics: some of these guys barely wipe themselves, much less write a coherent sentence.
I used to work in the registrar's office, and I remember one prof who wouldn't take a lecture hall with a stage, because she was obsessed with the idea of students looking up her dress. There was another guy, a geo lecturer, who bitched and whined because he'd have to carry his rock samples 100 yards from his car to his classroom. Heard of dollies, homeboy? He was the same guy who turned in a catalogue course description that began "How are earthquakes happen?" Native-speaker, too.
Rant off, but jeez; most of these guys are mediocre minds at best, but they've got arrogance enough for a genius.
Ross said:
... The problem is that more 'tails' are being matched. ... My problem is that they are no where good enough to rely on when making investment decisions. ...
My understanding is that real estate prices are set on the margin, so the Case-Shiller methodology captures the right mix.
To clarify, during the run up in prices, the fact that 90% of the comparable housing stock had been bought for a quarter of the current asking price never deterred buyers or sellers from completing transactions. The same principle holds on the way down modulo the stickiness of real estate which just slows the drop.
Feel free to correct my understanding. Constructive criticism is appreciated.
Case Shiller averages three months. So May data is average of March, April, May.
The slight inflection in the curve is what would normally be the "spring bounce" as home prices are normally a bit higher in the spring to early summer.
In this case, there is no bounce, just a slight upward inflection in the general trend of rapidly falling home prices.
"Confused writes:
So how come there are no (noticeable) price drops in the heart of Silicon Valley, Manhattan etc? Do these places just lag?"
Case Shiller's parent company suggests not only do they lag, but they just do better over the long run. All things being equal, closer to the core is the best place to be.
http://www2.standardandpoors.com/spf/pdf/index/052708_Housing_bubbles_collapse.pdf
While I often watch Housingtraker.net for more updated information, it has list prices. It's a nice data series that is current each monday.
What is changing is there are larger and larger discounts to original listing price, and even the final list price before selling. As an example, in Malibu actual selling prices are typically around 20% off of initial listing price Real Estate Blog - Have We Bottomed Out Yet? . At the peak, sales prices were slightly above list prices.
I hope for San Diego this is an overshoot. Otherwise we're headed back to 2000 levels and the whole region is going to be underwater!
"Then you have Denver, Charlotte, Dallas, Portland, etc., where prices seem to be down 5-10% from peak and rising again."
Those same cities had a similar price increase April to May in 2007. Is this the buying season effect? Maybe people are more likely to pay a premium during peak season?
I remember one prof who wouldn't take a lecture hall with a stage, because she was obsessed with the idea of students looking up her dress.
I had a similar obsession, but from the opposite side of that equation.
In any case, did you suggest that she wear pants?
I call the bottom when the prices drop to Jan 2002 levels. What'ya think? Too optomistic?
Compare CS to PMI's Summer US Market Risk Index. Then take a look at the historical appreciation rates for those with the lowest downward risk, like Pittsburgh and Indianapolis, and you get a bit clearer picture.
What the CS numbers don't reflect in more expensive bubble markets like San Diego, is that while what sold in 2005 for $500k is now $300k, the higher end is only off about 20% over the same time frame, but with the cost of jumbo money going up, the high end will start to see the same dramatic drop.
" MiTurn writes:
I call the bottom when the prices drop to Jan 2002 levels. What'ya think? Too optomistic?
MiTurn | 07.29.08 - 12:23 pm | # "
In real terms or nominal?
I predict 2004 levels in nominal dollars for the DC/Balt region.
The CS index for Phoenix was 100 in Jan 2000. At 5% annual appreciation for 8 1/2 years, the index would be at 151. Today's number is 157. I would say valuations in Phoenix are pretty close to where they should be. You can even get SFH's in good neighborhoods that cash-flow.
Of course, the inventory is still high, and prices will probably over-correct.
Spring sales bump is all dat be. Ain't no prices risin' for least 7-10 years, my pappy was right 'bout dat.
Sheee-eeeet, home skillet, Effen got more wrong wit his head dan dat baby I be drop in 1959. Never did talk raht after dat.
Shiller da man. Don't be doubtin'.
Mike in AZ writes:
The CS index for Phoenix was 100 in Jan 2000. At 5% annual appreciation for 8 1/2 years, the index would be at 151.
I just checked the Phoenix CPI minus shelter from Dec 2001 (admittedly a little late) to second half 2007 from here. It's 16%, or about 2.75% annually.
So, at 2.75% annual appreciation, the index "target" would be at 126, suggesting there might be 20% more downside to regress, excluding effects from overbuilding as well as pricing and demand destruction from economic effects. (and the 2.75% increase per annum vs the time the remaining regression takes).
That my baby daddy up there. Ignore that. He work at Merrill.
Great point, callous. Thanks for the link.
It is worth looking at the historical C/S data from the last downturn to note there is a relative maxima almost every August (summer bounce) before the general downward trend continues. We'll know things are really bad if we can't even muster that bit of "happiness".
Ross said: "I understand CS uses matched sales. The problem is that more 'tails' are being matched. A very large percentage of the housing stock never turns over. How can you match what isn't bought to what was never sold."
I've always thought the CS numbers, being "capitalization-weighted" (heaviest weight given to the housing markets that are most expensive), were exaggerating the problem.
Your point about the CS data being "self-selecting", along with your more-recent post, indicate even more exaggeration that hadn't even occurred to me.
The peak in sales was a couple of years back, which was also the peak of prices. So what are CS's paired sales skewed towards? The people who bought near the top of the market and are now being forced to sell. IOW, CS data is skewed towards the experience of homeowners who are in absolutely the worst position.
It's no different than assuming that all the stock market investors who bought at the recent market top in October, 2007 and are now looking at the heaviest losses are representative of every stock investor, including the ones who bought before the peak, after the peak or bought better-quality, better-valued stocks that haven't suffered nearly as much as the broader indices.
Sebastia
Eliminate seasonality with Year-Over-Year (YOY) change (7/07 v. 7/08)
Multi-month rolling averages lessen volatility/noise (smooth the curve)--a purpose distinct from annual seasonality.
The more months in the rolling average, the more you are measuring the past.
A 120-month (10yr) rolling average probably would show "current" house prices still rising.
Ah yes Sebastian.
But the current market prices are reflective of the current market.
After all those folks who got $10 for their BSC still only got $10.
Someday this war's gonna end...