CNNMoney suggests Seattle will drop only 2.9%
They claim our peak was in 07Q3 and our bottom will be in 08Q3 with the median price down 2.3%.
(#84 on the list): link
Not nearly the doom and gloom hyperbole that we see on this blog...
(#84 on the list): link
Not nearly the doom and gloom hyperbole that we see on this blog...
Comments
As an example, they predicted San Diego would be +0.90%.
Reality: Case-Shiller Index for San Diego -7.0% YOY as of May 2007.
Here's a summary:
City / CNN 05/2006 Prediction / Case/Shiller May 07 YOY
Phoenix / +6.30% / -5.55%
Los Angeles / +2.30% / -3.28%
San Diego / +0.90% / -6.96%
San Francisco / +6.30% / -3.43%
Denver / +1.10% / -1.44%
Washington / +2.60% / -6.34%
Miami / +0.80% / -3.29%
Tampa / +6.00% / -6.67%
Atlanta / +2.80% / +1.74%
Chicago / +3.40% / -0.56%
Boston / +1.10% / -4.29%
Detroit / +3.10% / -11.06%
Minneapolis / +1.90% / -3.51%
Charlotte / +3.50% / +6.99%
Las Vegas / -3.40% / -4.10%
New York / +2.10% / -2.26%
Cleveland / +4.70% / -2.78%
Portland / +7.50% / +5.71%
Dallas / +4.00% / +1.78%
Seattle / +10.50% / +9.06%
So, of the 20 Case-Shiller-tracked cities, they got the direction of change correct on six.
They did get fairly close for Seattle though, predicting +10.5%, vs. reality of +9.0%. Given how far off they were for many other cities though, it seems like it was more luck than skill.
I'm not saying that it *will* bottom out in 2008 (I doubt it will) but I do think this will be a shorter downturn than has ever been before. I think it will be a steep rate of decline over a shorter time frame.
My reasons are:
1) More awareness of prices and changes due to online databases and the Internet. You can see your neighbors lowering the price the day they do it, and you don't have to wait for the the Sunday paper. Then you can react quicker and lower your price lower than theirs. The race to the bottom can happen much quicker than it has ever happened before.
2) Current situation with financing is so out of control this time that I think this will also help to steepen the decline.
3) The fact that we have this situation when the economy is "doing fine" is amazing and completely counter to the ill-conceived wisdom of not having a housing downtown while unemployment is low. IF we end up in recession in the next year or so (reasonably possible, IMO) then this will also steepen the decline.
Just my 2c.
You can't really compare the CNN forecasts to the Case Shiller data because that's more like comparing apples to oranges. Unfortunately CNN doesn't provide the methodology behind their forecasts so I don't see how you could compare it directly to Case Shiller.
As an aside: when do you expect us to see the full brunt of foreclosures in this area? News out this morning shows that King County continues to have very low rates of foreclosure compared to the rest of the country: http://seattlepi.nwsource.com/local/332 ... ure19.html
Mr"Rational", which CNN forecasts *can* we compare to Case Shiller? Let us know and we'll look back and see how those compare. Or is it the case that we can't compare *anything* to Case Shiller unless it's Case Shiller, which as we know doesn't predict the future as it's a measure of what has happened in the past. So I guess it's always going to be an Apples to Oranges comparison.... Oh well.
LOL on foreclosures. I lived in San Diego a couple of years ago when they were talking about how foreclosures were still low compared to historic figures. Tell me: if something hasn't yet happened, does this mean that it cannot happen? Does a low rate of foreclosure NOW mean that it will always be low?
All I'm saying is that each of these different methods of looking at median prices are going to give you different forecasts and historical readings. Who is right...CNNMoney? Case Shiller? NWMLS? I just brought this up as one particular forecast of where we are headed based on some algorithm that they do not release the specifics of. Comparing it to what has happened with the Case Shiller index as a way of disproving it is not accurate IMO.
Again, I think you're misinterpreting what I'm saying. The article was talking about how we are very low when currently compared to other areas in the country. One theme I see over and over on this blog is that how the alleged "huge" number of people who took on unaffordable loans will cause a large number of foreclosures, increase supply and drive down prices. All I'm asking of Tim is when he feels the brunt of this will come? It's an honest question, I'm not trying to be sarcastic at all. While your anecdotal evidence of San Diego foreclosures is nice, this article was not comparing our current low level of foreclosures to our historical levels.
Obviously having a low rate of foreclosure now doesn't mean that it will always be low but it is also interesting that will all of the problems we're seeing in other markets that our foreclosures are still very low. Maybe you would like to chime in along with The Tim and give me your forecast of when things will get ugly with foreclosures around here...
I think we will probably see 0% YOY price changes according to the Case-Shiller HPI as early as April 2008, or as late as October 2008. However, as I have said every time I make any kind of prediction, there are so many factors that go into it that I recommend taking any predictions made by anybody (myself included) as nothing more than wild guesses, which is what they are.
So basically what you're saying that we can't compare a prediction to the actual real event? You're saying that you can only compare predictions to other predictions? That doesn't make sense.
Like "FinanceGuru" I think you need to change your name to just "Mr".
Of course, I have saved you the trouble of looking. Here's the same comparison for median prices, hopefully accurately transposed...
City / CNN 05/2006 Prediction / REALTOR.Com 06-07 YOY
Phoenix / +6.30% / -2.7%
Los Angeles / +2.30% / +2.8%
San Diego / +0.90% / +0.2%
San Francisco / +6.30% / +7.6%
Denver / +1.10% / 0.0%
Washington / +2.60% / +0.3%
Miami / +0.80% / +2.0%
Tampa / +6.00% / -3.8%
Atlanta / +2.80% / +0.9%
Chicago / +3.40% / +1.7%
Boston / +1.10% / -1.9%
Detroit / +3.10% / -7.1%
Minneapolis / +1.90% / -2.5%
Charlotte / +3.50% / +8.3%
Las Vegas / -3.40% / -3.6%
New York / +2.10% / 6.3%
Cleveland / +4.70% / -7.1%
Portland / +7.50% / +5.2%
Dallas / +4.00% / +1.7%
Seattle / +10.50% / +8.9%
I guess if the goal is getting the sign right, they did ok. I think I could do as well with a dart board.
Predicting seattle is easy. Just shave 1% off the YoY appreciation every month...
Actually, you can get predictions on the Case Shiller indices by looking at the prices of futures contracts on those. These are quite revealing, although the don't trade the futures for Seattle, only for the top-10 markets, and the composite index. The nice thing about these predictions is that they are backed up by money. For example, you can see the futures chain for the national composite index: http://finance.yahoo.com/q/fc?s=CUSG08.CME. The markets think that the national index will drop to 210 over the next three quarters, and that San Diego will drop to 222: http://finance.yahoo.com/q/fc?s=SDGG08.CME.
That's not at all what I'm saying. If we knew how CNN measured the median price then we could compare their historical forecasts to the actual real events (using their definition of median). Don't you notice how everyone has a different interpretation of "median?" NWMLS, CNN, Case Shiller, etc...they all have their own method of computing it. Question is, whose method do you use to test the forecasting ability of CNN? In short, you could only use CNN's interpretation of median to test their forecast and since we don't know those details it's pretty useless to try.
If NWMLS had made forecasts in the past it would be very easy to test their accuracy because we could simply compare them to their interpretation of median which they publish every month. For CNN it's not so easy. Notice how both The Tim and Deejoyah got completely different results when they compared the CNN forecasts to two different interpretations of median?
CNN to actual value vs. Case-Shiller to actual value of a given city and see which method is closer to actual value (ie reality).
But arguments could always be made on the way the actual value is collected or calculated. Someone could argue the right way to get actual median should be this way or that way.
OK. I actually don't think it's going to get too ugly around here. I'm not in the camp that says things are going to hell and we're going to have 80% depreciation, or other figures like that. My best guess is:
-Without a recession, we'll have an increased (not insane) amount of foreclosures and price decreases over two years in the 10-20% nominal range. Most NODs end of next year...
-With a recession, it could get a little more crazy with a lot more foreclosures, but I still don't think it's going to be a disaster. Perhaps 30% off - who knows. Timing depends on the recession in this case.
It's a really tricky formula as people won't foreclose if they can sell, although they might hit default on their loan. People can't sell (for their mortgage amount) if prices go down too much. Prices go down if too many people try to sell at once. Then there's the credit market! It's enough to make your head explode.
Ultimately, I think it's easier to predict the final outcome than the timing. By that I mean that it's likely that mortgages and underwriting will go back to how things were before the bubble, and prices will drop to historic income/value ratios, plus a little bit.
I think you are overcomplicating it.
Median price is median price. It's the number reported by the Local MLS to REALTOR.com. The only uncertainties are 1) whether you have comparable data in terms of geographies and 2) some sites scrape asking prices and report those as medians which can be confusing (housingtracker.net, for instance). The article from cnnmoney pretty clearly states that the forecast is for median prices. Probably a safe assumption that they are using the REALTOR.com numbers
Case-shiller is a totally different approach. They look at paired sales of the same housing stock, and compute an index off of this based on weightings vs. sales volume, price, etc. Very complicated but probably more representative of what is actually happening in the market.
On a side note, I don't know why anyone would take seriously a forecast from economy.com - which is owned by Moodys, which is one of the big ratings firms that screwed the pooch on rating all the CDOs. Seems like the fox in the henhouse to me.
Um, no. There's only one way to compute the median. If they're not computing it that one way then they are lying when they say they're computing the median.
There may be only one way to compute the median but your calculations will be different based on what data you use. Take this simple question: What was the median price change for Seattle in 2005? I'll bet you that CNN, WSJ, NWMLS, Case Shiller, Realtor.com and whoever else will give you different answers. Even though there's only one way to compute it, you will get completely different results depending on the data set.
This was the whole reason why I brought this up to the The Tim when he wanted to compare CNN historical forecasts to actual Case Shiller data. Two completely different data sets will give you very different medians.
nope
Case Shiller doesn't give you a median. it is an index. It was created as an ALTERNATIVE to using median price and is considered more accurate - but is only available for 20 markets
Median price refers only to the MLS data. There are 200+ local MLS entities in the US, all regional monopolies - like NWMLS. NAR/Realtor.com get their data from NWMLS and all the other MLS entities. CNN based their story on Realtor.com/NAR. If you want 290 markets, the realtors have a lock on the data. there is nowhere else to go. WSJ reports on both NAR and C/S.
The whole forclosure business in Seattle wont be quite as bad as the super frothy markets. We went up about 50% less over the past five years than the super overheated markets. Thus the decline will be lower as well.
The Tim is 100% correct that forclosures in any given area wont make a material dent in the housing market until YOY appreciation dips below 0%. However, an economic recession or boom has a direct impact as well.
Last month WA State added over 10,000 jobs when the entire nation lost 4,000 jobs. About 5,000 of WA State jobs were government jobs, yet the Business & Professional (upper income jobs) added about 3,000 in just one month. Most of those are probably in the Seattle area, so the economy is doing just fine right now in this region.
Ironically we have been hiring most people from other high housing cost areas and are happy that prices are lower and love the area.
Okay statements like this add nothing interesting to the discussion. For one, none of us can agree on how much the bubblier places will decline. Secondly, just saying "it will be less than [some place]" will be true for any real value between 0 and whatever percent decline that place sees. For example, the bubbliest places see 50% declines, then your statement is true for any real number from 0 to less than 50 which is a pretty wide range.
Yeah so I guess if what you're trying to say is that since median selling price is inflated due to the change in the mixture of homes that are selling, it will not fall all that much despite the reality being that house prices are falling quickly (as evidenced by Case-Shiller). I can buy that.
And while Case-Schiller indices are somewhat arbitrary numbers, the percent change in the number is supposed to accurately reflect what is happening in the market.
I was not predicting a 50% decline. My estimate is that the rest of the nation will have about a 10% to 20% decline (Seattle probably is in this range) at worst (with super heated areas twice as bad, 20% to 40%).
I read an article in the economist yesterday showing the US RE market has actually had lower appreciation than most world markets at 98% over 5 years (or close to 5), compared to these regions (England, Spain, Netherlands, India, Austraila, and the list goes on). Their main point was that if the US is in trouble, then other countries could get hit much worse (however they mentioned lending standards were not quite as lax in most of the other countries).
I don't disagree that we didn't go up as much and therefore won't come down as much. But the statement is totally obvious and adds nothing to the discussion. You might as well come in here and say "Hay guyz, the earth still revolves around the sun!"
I never said you did. I pulled 50% out of the air to demonstrate why the "we won't go down as much as the areas that saw bigger gains" statement is so obvious and devoid of any interesting information.
Now that you've posited some actual numbers, we have a discussion. Might I ask, why you expect only a maximum 40% decline in areas that "went up by 100% over the past 5 years"?
Just one of the local trends to watch: The number of new apartments in King County is expected to increase over 56 percent in 2008 and by 329 percent in 2009.
Currently: County owner-occupied houses and condos: 425,451
Renter-occupied apartments: 285,465
http://seattlepi.nwsource.com/local/315 ... ate16.html
http://www.city-data.com/county/King_County-WA.html