Housing: Where Is the Bottom?

Thursday morning (February 26) a new U.S. Department of Commerce report found new home sales in January at a lower level than any time since they started keeping records in 1963 (New-Home Sales Tumble To Record Low; Prices Fall). The day before, the National Association of Realtors announced a 12-year low in sales of existing homes for January (January Existing Home Sales).

This is a good time to update and extend a comprehensive review of the national housing market posted nearly two months ago (link cited below).

Single Family Home Prices

The graph below contains two more months of data (November and December) available since the previous article on this subject, Because there has been a continuation of the rapid drop in prices from the earlier months, the new projection of a slightly more gradual price decline gets to the target box in late 2009 or early 2010. This target is somewhere around $120,000 to $125,000, or about 20% below the December average price. The target price is near the same value as two months ago ($120,000 to $130,000 then), but the time to reach the target box is now projected to be 3 – 12 months less. If the rate of house price decline slows significantly, then the time to reach target would become much longer and the target price would rise slightly ($1,000 to $2,000 per year).

Data just released this week (February 25) revealed a 5.3% drop in existing home sales for January to a 4.49 Million annual rate (January Existing Home Sales ). In January, 45% of all sales nationally were distressed sales (foreclosure and pre-foreclosure sales).

Housing Starts – Single Family

The National Association of Home Builders maintains a history of housing starts and makes projections into the future. The graph below was obtained on their web site on February 24. The graph shows a bottom on housing starts projected for December, 2008 at 622,000 units (annualized). The actual data is in the graph through year-end 2007. On the same web site one can obtain the actual monthly through January, 2009, most recently updated on February 18. The projection in the graph for December, 2008 was around 600,000 units; actual was 395,000. The projection for January, 2009 was for an increase from the December level; the actual number was a drop from December of 12% to 347,000 units. I have added a (red dotted) line representing the actual data since the graph was produced.

[Source]

The January number of housing starts (annualized) seems quite a low number, but single family housing starts will have to fall another 20% to reach my January projection of 275,000 units annualized at the bottom.

Using the 2005 average as a reference (1716 units), the January 2009 number represents a drop of 80% in single family housing starts.

Note: More recent projections than the one I quoted here (graph above) can be obtained by subscription at http://www.HousingEconomics.com

Housing Starts – Multi-Family

The NAHB (links above) also publishes the housing starts data for multi-family housing. This is summarized in the following table. Multi-family housing starts have declined by 66% from 2005, significantly less than the 80% drop in single family housing starts.

If some of the single family housing sales are to investors (and, therefore, increase the number of rental properties), this may further depress the multi-family housing construction business.

Note: The 2005 multi-family total is one (thousand) units greater than the two component categories, and the 2005 U.S. total is one (thousand) less than the sum of one unit and multi-family. This is presumed to be due to round-off errors.

Regional Differences

It is all well and good to talk about national markets, but real estate can vary significantly between different local markets. The National Association of Realtors (http://www.realtor.org/research/research/ehsdata) gives separate data for the four regions of the country. The following two tables show some of that data for prices and sales volumes. The data base does not show any 2005 data or the highs in price and volumes for 2006, just the total for the latter year. This data shows the region (West) with the largest price drop also has the most positive sales volume numbers.

The West has the smallest drop in volume from 2006 and actually has an increase in volume for 2008. Mark Perry (California Real Estate Market Is Up) recently reported an increase in sales volume in California on drastically lower prices. Since the West has had a larger loss in house price thus far, it is not unreasonable that region would have the most favorable sales volumes.

Summary

Some additional conclusions from this update:

1. The continued rapid drop in average home price has shortened my time estimate for reaching the “target equilibrium price” around $120,000 to $125,000. My time estimate is now late 2009 to early 2010. This is as much as 12 months earlier than my estimates just two months ago. If the rate of price decline slows in the next 2-3 months, the time estimate will be increased again.

2. In the previous article I estimated the drop from 4th quarter 2008 to a bottom in housing starts of 275,000 (annual rate) to be about 33%. With the unexpectedly lower housing starts rate of 347,000, the drop to the projected bottom is now 20%.

3. Multi-family housing starts have fallen much less (66%) from the cycle peak than have single family housing starts (80%).

4. There are significant regional differences in the housing market outlook. The West has seen prices fall much further than the other three regions. Not surprisingly, sales volume (existing homes) in the West has increased during 2008 while the other three regions have continued to see volume declines.

5. Sales of distressed homes (foreclosure and pre-foreclosure) constituted 45% of sales in January.

The increase in distressed sales is undoubtedly contributing to the continued rapid decline in sales prices. As shown by the West region, sales volumes will probably increase as prices drop. As prices approach historic price to income ratios, inventories should start to decline. If housing starts do not continue their decline, however, the entire process could be delayed. Also, if government action attempts to support prices above the “equilibrium” level, this would be another cause for delayed bottoming.

Comments

[D. McHattie:]

John, your methodology is sound and very scientific. What is your opinion of a couple additional factors which could significantly increase the extent of declines?

1) The tendency of price distortions to overcorrect

2) The resetting and recasting of alt-a and option arm mortgages over the next 2 years

Personally, I am very reluctant to rush back into real estate or to lacerate my net worth on the falling knife.

Feb 27 07:45 AM

[Ergo:]

Interesting article. Thanks for your views.

Feb 27 08:01 AM

[markg:]

I would again ask, do the sales numbers for existing houses reflect that when houses are taken back by the bank these are counted as a "sale".

I have brought this up several times, and have yet to see a SA writer acknowledge this.

Feb 27 08:50 AM

[alivewweb:]

Elsewhere it was mentioned that the average Bear market last about 30 months and we are now 16 months into this one. For this bear market to end means we will be heading up by that time (hopefully). It has also been suggested home prices can't stabilize until the credit market become unfrozen which is what the treasury is attempting to do right now ala the "dreaded" partial nationalization of some money center banks. No doubt the stock market will react more quickly to first light at the end of the tunnel than housing but things are cheap now and expectations are low.

It appears to me that we are at the beginning of the end of this crisis and its way too late to be afraid but a perfect time to be thinking about getting rich.

Again. Just my opinion.

Feb 27 09:02 AM

[CLH:]

Two things that you didnt mention. Markets always overshoot in both directions. Expect the bottom to be below the true value. More important is demographics which tells up that boomers had few children and less housing is needed in the future..

Feb 27 09:14 AM

[kelm:]

Everyone has flagged it - the market is likely to overshot. However, you analysis serves to flag when we are in a historic equilibrium position. If we continue through that to lower levels this provides an excellent measure of whether we are dealing with an overshoot or f, over time, we see that something fundamentally has changed in the US market.

Feb 27 09:21 AM

[Urbane Gori...:]

markg I can't answer that. I can tell you that there is a backlog of bank-owned units that are not making it to the MLPs. This has been variously explained as (1) Flood of repos too big and they can't handle the volume, (2) Banks don;t want to flood the market and drive prices lower and (3) Taking a wait and see approach pending government intervention.

My gut feeling is that many repos are sold directly by the banks to investors and may not make it into these figures. In particular I suspect that unfinished tract areas may be moved in blocks to speculative buyers. I'm not convinced that these sales are included as I presume the figures that we are presented with are MLP indexed. In my mind that means the cheapest deals are worked behind closed doors and are not counted in these reports.

jegan

On Feb 27 08:50 AM markg wrote:

> I would again ask, do the sales numbers for existing houses reflect
> that when houses are taken back by the bank these are counted as
> a "sale".
>
> I have brought this up several times, and have yet to see a SA writer
> acknowledge this.

Feb 27 09:32 AM

[nym:]

Thanks for your insight Urbane, but I think the banks are trying to limit their losses, not wait for a better market: they've been right in that for at least a year.

Feb 27 09:53 AM

[nym:]

John, I think you need to re-examine that average family income extrapolation.

Feb 27 09:56 AM

[John Lounsbury:]

D. McHattie - - -

The upcoming Alt-A and Option Arm defaults are indeed a problem. These were discussed in my previous post on the housing market (link provided in the article).

markg and Urbane Gorrilla - - -

I also can not answer with certainty if repo's by banks are included in the sales figures, but I doubt that they are until they are sold out of foreclosure to a home owner or an investor.

In my own neighborhood, there have been seven foreclosed properties. Four of these have been sold privately by the lender without ever being listed by a realtor, a fifth is still for sale "by owner" (the lender) and two have been on MLP (realtor listed). One MLP property has been sold and the other is still on the market. These are anecdotal observations, but interesting, none-the-less and relate to Gorrilla's comment.

CLH - - -

You are very correct that changing demographics will have a significant impact on the housing market. This was discussed in detail in my early January article (link provided in this article).

To all readers - - -

Overshoot is indeed a possibility. There have been widely varying estimates of what (and when) the bottom in housing will be. These were reviewed in my first article in early January (link in this article).

I have tried not to repeat topics covered in the early January article unless projections have changed. I wanted to provide only updated news here. Please use the link to the first article if you want more depth and older news.

Feb 27 10:20 AM

[John Lounsbury:]

nym - - -

You are very correct in questioning my average family income projection. If we stay in a recession for another year, it is hard to see how this will not drop significantly. If we could not make any significant gain in average family income since 2000, how can that fail to drop in a two-year (or longer) recession?

If average family income drops, then the "equilibrium" ratio of house price to income will require home prices to come down accordingly.

Another factor not discussed in the article is the depressing effect that increased availability of foreclosed homes as rental properties would have on rents. If rents are depressed by increased supply, then sales prices will come under further pressure (buy/rent ratio).

Feb 27 10:28 AM

[ROLEXROLEX:]

If houses where bought at least with 10-20% downpayment, all this crash wouldn't happen at all.
When you buy real assets for free, everything can happen, bottom is 6 years away.

Feb 27 10:38 AM

[mrresponsible:]

John: I appreciate your analysis. If you don't know about Freddy Hutter's charting of existing and new home prices @ trendlines.ca I would recommend it as a resource.

Feb 27 11:13 AM

[John Lounsbury:]

mrresponsible - - -

Thanks for the link. I was not aware of Hutter's charting.

Feb 27 11:17 AM

[phaf:]

ROLEXROLEX- even with substantial downpayments, many folks find themselves underwater currently. From personal experience, it's painful to sell on the downside of the curve. Plus, in many of the overheated markets (Miami/LA-CA/Vegas/Pho... making that 10-20% downpayment would have been prohibitive for virtually everyone.

But you're right that if prudence were involved by all players leading back to the prebubble days, the crisis would have been averted. Easy and plentiful credit plus buyer psychology on the rising tide is a toxic combination (unless you timing is very good...)

Feb 27 11:31 AM

[Ricard:]

John Lounsbury,

Great follow-up to one of the best articles I've read on SA. The discussion has also been enlightening.

I do wonder where you got the average family income extrapolation. That seems to be a bit higher than what I've read to be the mean, but then I realized I was working with a per capita figure. Seems the census leans towards a higher household income, but I couldn't find a national average there.

Feb 27 01:25 PM

[OldLimey:]

Mr. Lounsbury: thank you for an impressive analysis which has stimulated informative comments.

Feb 27 01:37 PM

[Pilm:]

Very good article, especially the affordability chart showing relation of avg home price to avg family income. The first thought I have is that avg family income may become flat, or even decrease somewhat in the next 1-2 years, which would extend the time to reach a suitable buyer/seller price. There is also the potential impact of the gov't providing very cheap money (once again), which might tend to shorten the time to recovery, as well as the possible impact that soaring inflation will make more expensive homes appear cheaper. Just a lot of factors to consider.

Feb 27 02:06 PM

[John Lounsbury:]

Ricard - - -

The future estimated average family income is my SWAG. As pointed out in my comment above (responding to nym), a protracted recession could bring future incomes below my projection, which is roughly in line with the long-term historic trend. Since average family income is actually lower for 2007 (in 2007 $) than in 2000, my projection may well be too optimistic.

The historic data was obtained from the U.S. Census Bureau historic data tables. The last entry there is for 2007. My 2008 value is simply a repeat of 2007. Link for data table is

www.census.gov/hhes/ww...

Feb 27 02:27 PM

[John Lounsbury:]

Ricard - - -

Sorry. The link above does not work. Try this:

www.census.gov/hhes/ww...

Feb 27 02:29 PM

[John Lounsbury:]

Ricard - - -

Link is bad again. Just go to www.census.gov
and try to follow links to find Historic Income Tables - Families (Table F-7).

Feb 27 02:31 PM

[Andrew E Bu...:]

Nice piece John, I agree basically with everything you say.

A few small points:

1: On your first chart you have the S&P Case Shiller in 2000 looking suspiciously like $100,000, but that was when it was indexed to 1000, I'm not sure that in 2000 exactly average house prices in USA were exactly $100,000. for coincidentally every market covered by the index?

2: Looks like you are going for about 40% peak to trough, that's seems to be about consensus, it's about where I ended up in September (on Market Oracle), after accounting for the slump, Nouriel Roubini said 35% to 44% in December.

3: One comment I saw about overshoot is correct, in my view your line is about the right angle but in my view the wrong place, basically the overshoot happened in about September, housing in USA is lower than the long-term average valuation, just it;s overshooting, like markets always do (see Bob Farrell).

4: I think you will find you get better models if you look at family income per house, that is dimensionally correct.

5: Interest rates are also a factor - this is explained in my article on Market Oracle in September.

Feb 27 02:47 PM

[John Lounsbury:]

Andrew E. Butter - - -

Thanks for the comments.

I obtained my Case-Shiller data from www.macromarkets.com/c...
The CSXR (Composite-10) number is exactly $100,000 for 2000 in those tables. It seems to be a coincidence because I can find no evidence that the data was normalized to that year.

I am going to look up your Market Oracle article and put it in my reference file.

Feb 27 03:43 PM

[billddrummer:]

John, excellent analysis and further confirmation of the long-awaited 'bottom.'

Some local data regarding my region (northern NV, including Reno, Sparks, Lake Tahoe basin):

Nearly 80% of the homes sold last month were either short sales or REOs;

High value homes purchased with low-rate option ARMs are now resetting at their cap percentage, with 30-40% increases in mortgage payments;

Sales activity is brisk below the $200,000 price point. But for homes above $1 million, there's a 16 year supply of listings, based on current sales volume.

Feb 27 04:02 PM

[Lilguy:]

I've been looking at the existing home sales market through a slightly different prism: 12-mo. rolling inventories. What I've seen is I think the peak in these inventories in November 2008 at 10.5 mos.--about double the "normal" inventory for sale. Otherwise, between August and January, the inventory has been at 10.4 mos., to which it creeped up over the last two years.

Still, I think it will take 2-4 years for that inventory and price stability to return. Why?
1. I'm not certain (although reasonably confident) that inventory has peaked. Sellers who have stayed out of the market because prices were too low (!) or they couldn't sell their homes may re-enter the market in the spring and drive the inventory even higher. (I am encouraged by what's happened in California--double the house sales at half the price).
2. It took us four years to get here and I can't think of a good reason for the reverse trend to take less time--even on falling prices.
3. We still haven't reached the peak (2011-2012) in those less-than-prime mortgages and their many resets. We're talking hundreds of billions of dollars worth per year. The resets, defaults, etc., will keep downward pressure on prices through the next 3 years at least.

Anyway, I think we have a long way to go before we get to market stability in the home real estate market. Nonetheless, I greatly appreciate your thorough and thoughtful analysis of our situation. Please keep up the good work.

Feb 27 04:24 PM

[John Lounsbury:]

To all Readers - - -

The Market Oracle article Andrew Butter refers to is
www.marketoracle.co.uk...

This is a very good analysis of overshoot and undershoot in housing prices and the effects of interest rates on housing value. While at this article, click on the "Archives" link at the end of the article to get a listing of all Andrew's posts. He has several other good posts on housing. Warning: Andrew's analysis is very technical - fine for people like me but maybe not for the less nerdish.

Feb 27 04:26 PM

[Steven Hansen:]

john, another great presentation.

because of the dynamics of the economy, it will be anyone's guess which factors will be effecting housing. the 4Q 2008 gdp drop is larger than my wildest dreams. bernanke just said we will leave 2010 with an 8% to 8.5% unemployment.

it is everyones advantage for the housing market to overshoot the bottom to demonstrate the bottoming process is complete. the problem remains the lack of buyers at the current housing price level.

there is a lot of uncertainty now.

one other point, the assumption of title by the mortgage companies is not counted as a sale in either nevada or california - and neither is conversion of properties to trusts.

Feb 27 07:16 PM

[jstratt:]

Great article and responses!

Regarding average family income I suspect it will drop significantly do several factors.

1) Huge drop in banking, financial and real estate earnings
2) Layoffs
3) Small business earnings drop off
4) Huge Hedge Fund earnings by a few would affect a mean average greatly

I add this but I am not necessarily arguing for a longer or deeper housing fall than the authors excellent analysis.

Feb 27 07:17 PM

[pacific alpha:]

Your statistic that new single home commencements has fallen 80% YOY to January 2009 itself shows what is in store for the economy. Hang on for a (thankfully) once in a life time experience.

Feb 27 11:36 PM

[Norcal1:]

Really good stuff... My wife and I have felt this recession and housing bubble coming for years now and have planned our life accordingly to put ourselves in a position to capitalize on this downturn. Your chart has helped to clarify a few things and give us some more critical information as we decide our future investment startegies. This is a painful but necessary correction in our economy. I completely agree with your collection of quotes on "The Benefits of a Recession" article.

Feb 28 12:36 AM

[John Lounsbury:]

The hand - - -

Thanks for the comment. I appreciate the information about mortgage company title assumption in Nevada and California. I believe the same is true here in North Carolina, but have not been able to document it.

Re- Bernanke's 8 - 8.5% unemployment estimate for the end of 2010. That sounds reasonable, but probably optimistic, if we start a recovery before the middle of 2010. The real question is what will happen in the rest of 2009. Will we go to 10% or more? If so, 8.5% may be too optimistic. Employment will come back very slowly in this business cycle.

pacific alpha - - -

The 80% drop in new single family house starts is not Y over Y, but from the top of the building boom in 2005. It has taken 3 years to drop to 20% of the peak.

Feb 28 01:27 AM

[CRS:]

The housing boom was quite simply an epidemic. The Fed and our policy makers created and supported the environment, private unregulated mortgage companies and brokers were the original carriers and hard working few, and the contagion itself was a highly distorted version of the American dream - the "own your own ATM" version - which was made available by the rapid maturation (ok, failures) of the securitization market. The pain that is currently being inflicted upon those who participated directly, and the innocent by-standers, will impair the willingness side of the demand equation for years to come. Furthermore, no tax break or subsidized loan is going to make me forget any faster, and its certainly not going to recreate the exuberance that got us here. Oh yeah, the ability side of the equation...that's just starting to fall apart.

Nothing but more downside surprises in this story for years to come.


On Feb 27 10:20 AM John Lounsbury wrote:

> D. McHattie - - -
>
> The upcoming Alt-A and Option Arm defaults are indeed a problem.
> These were discussed in my previous post on the housing market (link
> provided in the article).
>
> markg and Urbane Gorrilla - - -
>
> I also can not answer with certainty if repo's by banks are included
> in the sales figures, but I doubt that they are until they are sold
> out of foreclosure to a home owner or an investor.
>
> In my own neighborhood, there have been seven foreclosed properties.
> Four of these have been sold privately by the lender without ever
> being listed by a realtor, a fifth is still for sale "by owner" (the
> lender) and two have been on MLP (realtor listed). One MLP property
> has been sold and the other is still on the market. These are anecdotal
> observations, but interesting, none-the-less and relate to Gorrilla's
> comment.
>
> CLH - - -
>
> You are very correct that changing demographics will have a significant
> impact on the housing market. This was discussed in detail in my
> early January article (link provided in this article).
>
> To all readers - - -
>
> Overshoot is indeed a possibility. There have been widely varying
> estimates of what (and when) the bottom in housing will be. These
> were reviewed in my first article in early January (link in this
> article).
>
> I have tried not to repeat topics covered in the early January article
> unless projections have changed. I wanted to provide only updated
> news here. Please use the link to the first article if you want more
> depth and older news.

Feb 28 01:50 AM

[Stephen Kanitz:]

One thing missing. Mortgage Interest Tax Deductions. If you factor in that 100% of NOMINAL interest is tax deductible and americans know about this tax deduction, you should:
1. Either increase family incomes to include this tax deduction
2. Reduce the present value of the house, my calculations comes to nearly 30% of the home.
This means families lose tax benefits if they wait and see for too long, and your target date may be sooner by 3 to 4 months,

Feb 28 11:58 AM

[socphd71:]

A small point: The median is preferred to the mean as the average of choice for skewed distributions such as income and house prices. Both are bounded on the left (one can't easily have a lower income or house price than zero) and unbounded on the right (the sky seems to be the limit for both) In both cases, the median figure will be lower than the mean figure.

A larger point: Housing is incredibly local. House prices in some neighborhoods may have already overshot. In other, possibly nearby, neighborhoods, they may yet have a long way to fall.

My stepson is looking for a house in Oakland. All the houses he has looked at are bank foreclosures. Most have had several (more than three) offers within a few (less than three days) of being on the market. This may be a symptom of overshoot.

Feb 28 04:05 PM

[CVGal:]

Please explain how properties are sold *privately*. Every realtor I've asked about this says it isn't done. But they may not want to discuss it.

On Feb 27 10:20 AM John Lounsbury wrote:

>
> In my own neighborhood, there have been seven foreclosed properties.
> Four of these have been sold privately by the lender without ever
> being listed by a realtor,>

Mar 02 02:47 PM

[mr freddo:]

Thanks for the interesting article John. Very well researched and reported. Since the housing market started this swirling vortex of financial destruction, many believe that we cannot see a bottom until the housing market bottoms.

My concern about your current one year projection to bottom is the issue of declining income. You allude to this in one of your comments. That and the overshoot factor should stretch your bottoming time frame to more like two years.

Unless lenders throw in the towel and speed up the bottoming process by taking any reasonable first offer on foreclosures. I am going to test this theory tomorrow by putting in a low ball offer on some foreclosed real estate I know of.

Good luck out there. Remember to breath...

Mar 02 09:28 PM

[John Lounsbury:]

User 6825 - - -

In the area I live a lender has repossessed several lots, one partially constructed house and one finished house that was never lived in. He is having the partially finished house finished and landscaped. He has listed that property with a realtor. The unfinished house and the other five properties have all been "repossessed" by title transfer from the builder without completing a foreclosure process. The finished house was sold with only a "For Sale by Owner" sign in the yard. The sale price was 70% of the original listing price when the builder had it with a realtor. The house had competing bids within 30 days. One came from a buyer represented by a realtor (buyer's agent) and the bid accepted was from a private party with no realtor representation.

Three of the four vacant lots were also sold to private individuals with no realtor representation and were sold directly by the lender-owner. The fourth lot still has a "For Sale by Owner" sign.

I have talked with the employee of the lender who has worked with these properties. He says that most of their properties taken from builders facing bankruptcy have been transferred by private agreements after the first step in bankruptcy (a bankruptcy hearing) and before actual foreclosure. He also said that they have sold less than 20% of such properties where a realtor was involved on either side (seller's agent or buyer's agent) and more than 80% have been sold "by owner".

I have been told by several realtors that it is illegal to sell any real property without a licensed realtor. What the realtors don't say when they tell you that is that it is illegal to be the agent of sale for someone else's property. You can always sell your own property without a realtor. Several realtors have admitted that to me when I pinned them down.

mr freddo - - -

Be prepared to buy. You may find a motivated seller. There is one property near me that was sold after foreclosure. It was in contract in 2006 for $719,000, but the deal fell through on failure to get mortgage. Before it was sold, the last listing was $369,000. I was told by a neighbor of the property that it went for $320,000, but the title transfer has not been posted in the county records office yet, so I can't be sure. This is in an area that did not have a huge housing bubble. (Prices were up 25-30% from 2000 to the peak in 2006.) Those houses that are selling now are generally between 10-20% below peak prices.

So, you may find a motivated seller for one of your foreclosed property offers. Good luck.

Mar 03 12:40 AM