Long-time readers may recall Deejayoh’s inaugural Seattle Bubble article from June 2007: Why Inventory Matters. In it, he postulated that the Seattle-area Case-Shiller Home Price Index could be relatively accurately predicted fourteen months in advance by looking at year-over-year house inventory on the MLS.
Let’s extend Deejayoh’s analysis fourteen months into the future to see if it will predict a bottom.
Here are our basic assumptions for the Inventory-Based forecast (see Why Inventory Matters for more detail).
- Inventory changes forecast home price changes by 14 months.
- Equation: HomePricesYOY = (-0.386 * InventoryYOY) + 0.092
- I’ll be using King County SFH inventory only, as that provides the best fit through 2007 and 2008.
Given these assumptions, here’s a rough picture of what Seattle’s Case-Shiller Home Price Index would look like through late 2009:
Using the Inventory-Based forecast model, Seattle-area home prices (as measured by the Case-Shiller HPI) will hit a bottom early this year, and proceed with a sharp “V-shaped recovery,” ending the year well above their current levels.
As Deejayoh demonstrated in 2007, the Inventory-Based forecast model has a pretty good record from 2001 through 2007. However, the forecast for the coming year does not pass the “sniff test.” No other signs currently point to such a sharp recovery this year, despite what this model is predicting.
Unfortunately, it would appear that the Inventory-Based forecast model has reached the end of its useful life.
Method 1: Inventory-Based Forecast (Summary)
Bottom Month: April 2009
Bottom Value: 20.1% off peak
Likelihood*: 10%
* Likelihood is a totally subjective value assigned according to The Tim’s gut feeling. Treat it accordingly.
Bottom-Calling Week on Seattle Bubble
- Introduction: Bottom-Calling Week on Seattle Bubble
- Method 0: Blind Optimism
- Method 1: Inventory-Based Forecast
- Method 2: Dollars per Square Foot Linear Forecast
- Method 3: Simple Mirror Forecast
- Method 4: Affordability Index Forecast
- Method 5: San Diego Lag Forecast
- Conclusion: So Where’s the Bottom?
The Tim » Feb 16, 2009 at 12:11 pm
For anyone that is interested, here’s what the forecast looks like if you use combined inventory for King, Pierce, and Snohomish:
The month of the bottom doesn’t change, nor does the “v-shaped recovery,” but the value at the bottom does. I find this even more unreliable, as we are quite likely already beyond 14.1% off-peak.
AMS » Feb 16, 2009 at 12:44 pm
The Tim-
Really this is all about trend analysis versus the efficient market hypothesis. I don’t care how many different models you publish, and there are literally hundreds of these models.
Also the divide between the trend people and the EMH people will probably always be there.
Those on the far end of the trend analysis go to extents such as “solar mortality theory” and “Elliott Wave Theory” and so on. The Elliott Wave Theory has so long of a life cycle that we will all long be dead before it is proved or disproved.
This reminds me, in some way, about how there are always those who will predict the end of the world, and such predictions spike in years that precede the XX00 years.
How does all of this reconcile with possible Fourier Series and Laplace Transforms might be a far more interesting question to answer. In some way it’s only a matter of getting the model right, if we are given an infinite supply of sine and cosine waves. Yet the EMH suggests that no matter if we are given an infinite supply from the past, you cannot predict the unknown event of the future. So can we use behavioral patterns to develop a sufficient model to predict the future?
DrShort » Feb 16, 2009 at 1:05 pm
From 2000 – 2007, low inventory was a result of houses being quickly snatched up by eager buyers.
Low inventory in 2008 – 2009 is the result of potential sellers holding off on trying to sell.
Scotsman » Feb 16, 2009 at 1:15 pm
The Titanic was modeled to be unsinkable, engineered to perfection. As far as I know, it’s still on the bottom. No “V” shaped recovery there either.
Things are different this time, from all the other times gone past. No, really, they are.
80% off peak.
demo_kid » Feb 16, 2009 at 2:11 pm
The problem with arbitrary fitting like this is that it’s… well… arbitrary. There is no guarantee that the trend will continue.
The other problem is that the statistical model that Deejayoh created indicates that the natural appreciation rate with no net inventory change is around 9%. That sounds just fine for the 2001-2009 period, but could you really make the case that an inventory-price relationship during a bubble would apply during a burst?
I mean, I think that the relationship is pretty strong if noise is taken into consideration, but I’m guessing that a number of variables might track well with market price changes…
eric » Feb 16, 2009 at 2:12 pm
While I agre with potential sellers not selling. I’d say another thing that will keep inventory low is New Housing Starts, as well as, lack of new permits.
deejayoh » Feb 16, 2009 at 2:42 pm
I think that you’re right about the limited applicability of the model as it applies to the “bounce back” – but I still think that it may be useful for predicting when we see the end of price declines.
If you push out 14 months from today, that means prices stop dropping next spring. Personally, I can believe that will be what happens. Another 15-20% off and we are back in historical ranges on price:income and price:rent
Where it goes after that is anybody’s guess. Probably not bouncing back to 8-10% appreciation
However – FWIW I ran a view of average appreciation rates of home prices back to ~1971 using OFHEO data. I don’t think I found a single 5 year period with less than a 5% average. That is pretty interesting perspective
Kary L. Krismer » Feb 16, 2009 at 2:50 pm
The problem with things like this is that they work until they don’t work. I forget which way it’s going, but either the NFC or AFC has won the Superbowl coin toss for something like 13 times in a row. Or more to the point, there are things that periodically appear in the press where a certain team winning determines winner of the Presidential race, or maybe who wins a certain county.
There’s a huge difference though between the period for the chart above and today. The period for the chart above doesn’t include any time where we were in as dire of economic straights as what we’re in today. I think it’s Sniglet predicting deflation–others are predicting severe inflation. I don’t know which is more likely, but each are far more likely than at any time covered by the chart above.
So basically, even assuming the chart above could actually have told you something a year or two ago, I don’t think it means much of anything today. Too many circumstances have changed.
deejayoh » Feb 16, 2009 at 3:13 pm
By demo_kid @ 5:
I thought the same thing so I CAGR for seattle going back to 1971 is 8.4% – so that is not such an unbelievable stat for the “natural” rate.
Jillayne » Feb 16, 2009 at 3:17 pm
I’m definitely in the “inventory” camp. I know I’m just one person but I must interact with literally thousands of Realtors every year all across the state but mostly in King, Sno, and Pierce. Here are some trends I’m seeing this winter.
1) There is a growing percentage of inventory that could be catagorized as “distressed.” This means short sales, pre-foreclosures, and already foreclosed REOs. As this percentage of inventory increases in a given market area, the home values will decrease due to the lower price on the distressed homes, and the pressure on appraisers to not ignore the REO comps.
2) There is a growing number of Realtors who are reporting that homeowners with homes currently listed are considering walking away and letting the home foreclose because the homeowner HAS assets to make up the short-fall (on a short sale) and does not want to part with those liquid assets. A foreclosure appears to be the better option for some homeowners. These are mostly people who own more than one home. They’re going to just let that rental or second home “go” rather than part with their cash. This will mean more distressed, REO inventory in the latter part of 2009.
3) There are many homeowners who would LIKE to sell and maybe even move up but they don’t want to lower the price on their first home any more because they need that equity to buy something nicer. So when no offers are forthcoming, they take the home off the market. We could call this “shadow inventory” that will come back to haunt us when some Realtors think we might have hit bottom.
4) Home builders continue to build new homes and condos.
5) Lending guidelines are continuing to tighten, leading to fewer homebuyers able to qualify.
Massive government intervention, to be announced on Wed will likely slow new foreclosures…for a while. As we found out with Indymac, a high percentage of loan modifications fail, which will push distressed inventory further on down the road.
Foreclosure moratoriums will also give banks time to catch up on all their existing cases and then we’ll see a flood of new foreclosures into 2010.
jon » Feb 16, 2009 at 3:25 pm
The period from 1971 to present includes a few years of high inflation, the emergence of Microsoft, and then the bubble. Going forward I expect we will see inflation once again as the federal debt is monetized, but a new Microsoft occurring in Seattle is statistically unlikely.
However, the bailout money should be a pretty good stimulus for Seattle, as new project offices being set up will generate orders for MSFT products and services, and business air travel will be needed as projects are organized.
Jeff » Feb 16, 2009 at 3:34 pm
Inventory is a big factor, but job market, # of foreclosures/short sales shouldn’t be ignored.
There is no way RE market starts appreciating in April.. Boeing and Microsoft is starting to lay off people this month.. Sure it may bottom out in April, but I don’t see it appreciating for awhile until job market in Seattle improves.
EconE » Feb 16, 2009 at 3:44 pm
RE: deejayoh @ 9 –
What would the figures be if you ran it back to 1951 instead of 1971? The reason I ask is that my parents bought in 1972 for 2x the cost of the house in 1925. Why would 8% appreciation be a “natural” rate? Wouldn’t we possibly be in a similar mess (not as bad however) further out into the future?
RE: Jillayne @ 10 -
RE #2: If someone is walking away from a second home (and the mortgage)..isn’t that a recourse loan? Can’t the lender come after other assets?
RE #3: In a way…that is backwards thinking. If a person is “upsizing” and depending on equity to get them there…does that really work?
For example.
Person buys house for $200k. (let’s assume zero down for simplicity)
House was worth $400k at peak and the desired home to upsize to was $600k.
Let’s say they decided to upsize at the peak.
Person sells house for $400k, takes $200k profit and a $400k mortgage for the $600k house.
Can they handle a $400k mortgage when the last one was $200k? Perhaps their income doubled…so I’ll give them the benefit of the doubt.
Person can now sell house for only $300k….but the $600k house is now $450k
Person puts down $100K and needs $350K loan.
So….in reality, they may have received less for their home, but the “upgrade cost” in the end was only $350k instead of $400.
And if they could afford a $400k mortgage and only have a $350k one, they can spend money on all those other things that help keep people employed.
Overly simplified of course…just food for thought.
Jillayne » Feb 16, 2009 at 3:58 pm
Hi EconE,
Hope all has been well in your world the last few months. Regarding number 2, lenders seem to be so overwhelmed with foreclosures and seem to be waiting for the government bailout that I’m sure many banks just simply won’t go after the homeowner.
Regarding the move up buyer, when they purchased the 600K home, maybe they did so with an option arm and have been accruing neg equity so their 100K to put down on the new place isn’t 100K. Or maybe they took out some of that equity to purchase whatever or payoff debts.
Consider the tighter underwriting guidelines for qualifying. And when they see the $350K home for the upgrad, what they really want is the 850K home, priced at 700K today. ;)
DrShort » Feb 16, 2009 at 3:59 pm
RE: EconE @ 13 –
It’s hard to “upsize” these days because of lack of bridge loans. Everyone I knew who upgraded from 2001 – 2007 owned two homes for a period of time while they tried to sell the old house. I think that’s much harder to do from both a lending standpoint and a time standpoint (takes much longer to sell).
Moving up is much more complicated now.
EconE » Feb 16, 2009 at 4:20 pm
Thanks Jillayne & Dr. Short. :^)
Jillayne…I can understand the banks hope to collect $’s lost through bailout funds, however, I believe that the funds received will be far less than the losses. A little help is better than nothing. I just have a feeling that when all is said and done, the banks will go after the “low hanging fruit” (people with lots of assets and future income) while letting some others off the hook. I doubt they’ll let a surgeon or some other high payed professional off the hook if that professional was flipping condos and just decided to walk from their “investments”.
You can’t squeeze blood from a turnip (shoeshine boy flipper with a liars loan), but I’ll bet that you could squeeze quite a bit from a physician.
TheHulk » Feb 16, 2009 at 4:22 pm
Unless you purchased your first house before Y2K, it would take some serious cojones to “move up” these days.
george » Feb 16, 2009 at 4:50 pm
First I would like to see the model that seriously answers a much more simple question: Why did the housing bubble form in the first place?
Jillayne » Feb 16, 2009 at 5:06 pm
EconE,
Just heard one a few weeks ago. Guy owns five homes in the Seattle area. All five are listed as short sales. He HAS the money
to make up the shortfall on all five but doesn’t want to pay it. Since he’s not a U.S. citizen his plan is to default on all five mortgages and move back to his country of origin. Realtor will be taking them all off the market soon and they’ll all foreclose and end up as REOs later on this year.
One loss mit counselor I talked with recently had 1600 files on her desk. I’ve been trying to imagine what that looks like….
TheHulk » Feb 16, 2009 at 5:22 pm
By Jillayne @ 19:
I would love to find out which irresponsible lender gave this person the money to purchase the 3rd, 4th and 5th houses. And no doubt, this person got greedy as well. He kept doubling down his bets. Sadly enough, we will have to pick up the pieces when the lender goes hat in hand to the govt for the bailout.
I think the fact that this person is not a citizen is not relevant here since we are a non recourse state. I figure he will still lose whatever tiny percentage he put down as down payment (I am guessing not more than 3% of all the asset value at purchase). I can only hope he did not get any HELOCs and get more money from the idiotic lenders.
Kary L. Krismer » Feb 16, 2009 at 5:31 pm
By Jillayne @ 10:
The last time I looked, approximately 1/2 the listings in King county were “vacant” listings. That’s a pretty good indicator that the owner really wants to sell, and if it’s been on the market for some time, that they need to sell.
Kary L. Krismer » Feb 16, 2009 at 5:34 pm
By DrShort @ 15:
I think Washington Federal still does bridge loans, and try Googling “Washington Loan” and Windermere. Maybe that web page is just stale, but as to the former, I’ve talked with a loan officer of Washington Federal within the last 60 days about them.
Mason Boswell » Feb 16, 2009 at 5:36 pm
Tim, for someone who’s analysis is usually so well founded in data, I found your conclusion a bit hasty. Why not try to figure out how the model might be right rather than just discarding it. It could be that the current inventory numbers signal a sharp enough drop in prices that buyers will find the market very attractive and produce the upward spike. Interest rates are low, if prices drop 20%, I know I will think about putting some money in real estate, because the return in other areas (stocks, even CDs) is certainly not attractive right now.
This market has also shown a kind of bandwagon mentality, where once people think things are back to “normal” and prices start climbing there will be those that worry that they will be priced out. We, of course, know the fallacy of that, but that doesn’t stop the herd.
patient » Feb 16, 2009 at 5:50 pm
Mason B said,
“This market has also shown a kind of bandwagon mentality, where once people think things are back to “normal†and prices start climbing there will be those that worry that they will be priced out”
If you are talking psychology I think this saying is much more likely to be observed this time: “Burnt child dreads the fire.”
David Losh » Feb 16, 2009 at 6:21 pm
It’s bottom week on the bubble.
I’ll bet my bottom dollar that the chart is a red herring.
Pegasus » Feb 16, 2009 at 7:26 pm
Senator Patty Murray gets a well deserved public spanking. Next up Maria Can’tdowell.
http://michellemalkin.com/2009/02/15/sen-patty-murrays-worthless-stimulus-earmark/#comments
Kary L. Krismer » Feb 16, 2009 at 7:57 pm
Cantwell was just appointed chair of an energy subcommittee. She’s an idiot when it comes to energy issues. Expect long lines and blackouts in the future if she has her way. And coincidentally, she’ll probably cause the power lines that Murray is seeking to never be built!
Jillayne » Feb 16, 2009 at 9:13 pm
Pegasus, Patty receiving a spanking from Michelle on a post called “bottoms.” thanks for putting that image in my mind.
Maybe it was the movie I just watched: Vicky Christina Barcelona.
EconE » Feb 16, 2009 at 9:41 pm
RE: Jillayne @ 19 –
Wow. Just…wow.
AMS » Feb 16, 2009 at 10:42 pm
The Tim-
In your end conclusion, will you please address how no matter how many models you present, there may be a better one. In fact, if you actually had a perfect model, you probably would not tell us about it.
In other words, to show that some models are bad does not suggest that all models are bad, and furthermore, to show that we cannot use trend analysis effectively, we need to show that all models are bad.
This is a very difficult task. I don’t think too many people would suggest that all models are good, so it is already clear that there are plenty of bad trend models.
If a good one exists, how can it be identified?
DrShort » Feb 16, 2009 at 11:01 pm
RE: AMS @ 30 –
Market models generally fall short in that they don’t measure the psychological factors. There are rent/price and income/price ratios are that relatively predictive when looking at long time periods, but fail miserably in the short term. Over optimism can cause a bubble. Pessimism and fear can cause dramatic crashes. The problem is that market psychology is hard to measure, impossible to predict, and can shift very quickly. It plays a huge part in short term market dynamics.
There’s also a “post bubble” mentality where investors shun the asset class for decades after the crash. It took 20 years for gold to return from it’s early 1980s bubble levels (not adjusted for inflation). The NASDAQ is still 70% below its highs in 2000. Oil has crashed, but it’s still too early to see how that turns out. Housing will probably be better since it’s a necessity, but it just makes it harder to predict.
There’s no modern era housing data to guide us through what’s happening right now, so be cautious of any predictions.
Thomas B. » Feb 17, 2009 at 12:22 am
Okay… since we are talking about bubbles and bottoms. Does anyone have anything to say about the economic concepts of “race to the bottom” and “race to the top”? I bring this up because during the bubble era, it seemed like buyers were racing to the top to beat out other buyers. Now, is it possible that there could be a race to the bottom (i.e. sellers racing to the bottom to unload their house and beat other sellers)?
Kary L. Krismer » Feb 17, 2009 at 7:30 am
RE: Thomas B. @ 32 – I call that the herd mentality. People feel more comfortable doing what others are doing.
Also, I think it’s driven by the media. On the way up it’s buy while you can and other optimism. On the way down it’s whatever they can come up with to scare people, and other pessimism.
Cris » Feb 17, 2009 at 9:11 am
MS is trying to reach 5000 lay offs and will keep laying off quarterly both FTEs and contractors. We heard a big chunk of contractors are being released in July. Many contractor positions have already been shifted and merged with other departments.
Boeing is crying out with order delays. Probably they will keep laying off this spring.
Many car dealers at 0% APR with huge discounts just to empty their inventory. Eventually many of them may go out of business before summer as well.
T-Mobile cancelled all market trips until June in parallel to hiring freeze. They kept sending back contractors this month as well.
I don’t see any good news that may cause a V shape increase in such near future.
Cris » Feb 17, 2009 at 10:34 am
Inventory numbers showing a flat or negative trend. While it is interesting to look at the trend in “months of supply†data, there is to much manipulation to trust the data. Developers are intentionally de-listing properties to keep the supply numbers low. Banks are intentionally holding back REO properties so as not to flood the market. Delinquent home “owners†are not being foreclosed on in a timely fashion.
http://www.geldpress.com/2009/02/requirements-for-buying-a-house-dont-lose-money/
The Tim » Feb 17, 2009 at 10:57 am
RE: AMS @ 30 – Good points. I definitely plan to get into those topics in the conclusion.
Kary L. Krismer » Feb 17, 2009 at 1:59 pm
By Cris @ 35:
I don’t think developers ever list all their properties in a development. It would be very annoying to agents. They just list a representative for each major type.
Bottom-Calling: So Where’s the Bottom? | Seattle Bubble — News & discussion about real estate & the housing bubble in the Seattle area. » Jul 30, 2009 at 8:03 pm
[...] Method 1: Inventory-Based Forecast [...]