Karl Guntermann at Arizona State University’s W.P. Carey School of Business; Center for Real Estate Theory and Practice puts together an index of home prices in the Phoenix area based on repeat home sales.
Guntermann takes the analytical methodology used by Case-Shiller and applies it to the Valley of the Sun. Case-Shiller only publish information for the metropolitan Phoenix area as a whole while Guntermann gets more “granular” and runs an index for several cities in the metro area as well as a composite index for Phoenix as a whole.
In this graph below you can see that the home price index dropped below trend in July. (The latest published data is for July. It takes the economists a while to run all the math on the home sales numbers to get the index.)
You, of course, might draw a different trend line. And home prices might still be above your trend line.
Nevertheless, you get the point.
Prices were way down and near trend… in July.
And prices are lower now than in July.
{ 17 comments… read them below or add one }
Brian 11.15.08 at 7:17 am
And the scary thing, as I have suggested before here, is that big corrections like this one in the housing market, normally overshoot the trend, before eventually coming back to the long term trend. This seems to be a physical law and it can be observed in many different kinds of systems.
My hope was that the “regulators”, with their depth of wisdom and experience, could anticipate and manage the correction so that it settled back on the trend line. It still could happen, but something has to change quickly for it to do so.
A very promising idea recently making the rounds is that the leftover TARP money, $450B, should be used to guarantee a much lower Fannie/Freddie 30 year fixed mortgage rate of 4% or so. Fannie/Freddie would finance mortgages at the market rate, and the Feds would subsidize the difference. This would be a giant “buy down” program. (I heard this first from Ed Yardeni on CNBC).
Yes, it continues the socialization of the housing market. But as Ed pointed out, it has been so since the 30s. So let’s not kid ourselves we have had deductible mortgage interest against income tax, subsidized mortgage agencies, etc., for decades. Home ownership is the centerpiece of the American Dream and right or wrong, it has become a birthright, not just a privilege.
The main point is that since it was the housing market collapse that got us into the economic mess we are in, that is what has to be fixed first to get us out. That means stopping the declines of home prices by creating home buying demand. Artificially cheap mortgage rates would really help that along.
Sam 11.16.08 at 9:16 am
John,
One of these days you may have to define the term “bottom”. To me it means that the value of a home would not go down further. Some people keep saying that prices have reached 2004 levels, but they fail to see the contrast between 2004 and now. The major ones are
1. The inventory in Nov 2004 was 10K while it is 56K today!!!
2. You didnt need to have an average credit to buy a home. All you needed was the ability to sign.
3. 30-yr fixed mortgage rate was at a historic low, while it is at 6.25% today.
4. 3-yr ARM rate was 4.25%, while today it is 5.75%
In the last year, inventory level managed to drop only by 4K even though sales have been good since June. So I am not really sure why we are nearing a “bottom” given the high supply of homes.
Peter Fork 11.16.08 at 10:34 am
Some reasons why the median price should be below the trend:
1- The median salary didn’t follow the historical trend since 2000 (it didn’t increase in real terms).
2- Phoenix is in recession, so prices should be lower than usual.
3- Immigrants (legals and illegals) stopped coming to Phoenix. Even illegal immigrants were buying houses a few years ago.
4- Credit is harder to get than in recent years (it may have inflated the trend since the 90s).
5- There is no such thing as a trend in the short term.
John Wake 11.16.08 at 12:28 pm
@Brian. Overshoot is a very real possibility.
That buy down idea is new to me and would no doubt soften the landing and help minimize the overshoot. My first impression is good.
Another problem is PMI with today means buyers have to put down 10%. Once prices bottom out, that might go back to the traditional 5% which would greatly increase demand. So as we hit bottom, demand should increase significantly.
But we can’t hit bottom until we hit bottom. Delaying the bottom delays the recovery.
@Sam. To me the bottom means the median price stops declining in that zip code.
That 10K in November 2004 was as much an outlier as today’s inventory. I think the “normal” inventory would be around 35K to 40K. The market could get there by June. Of course, if you read this blog you’ll see that is just one man’s opinion and nothing more. I didn’t think the median home price for the Valley would get to $175K (the current number) and I certainly had no clue that it would go to $265K in 2006.
@Peter. You, Brian and I agree that an overshoot is possible. How much of an overshoot do you guess? $10K? $30K?
Peter Fork 11.16.08 at 8:25 pm
@John
I think it will overshoot the “trend” on your graph, but I can’t say by how much. It will depend on the region’s economy and immigration rate. I don’t think credit is much of an issue, we’re just back to responsible times.
I am watching a few zip codes using your very helpful pages. It is very hard to call a bottom, because of the diversity of the market.
In some zip codes, the median price per square foot is below the 2003 level (for example Queen Creek 85242). But I suspect a lot of homes in this zip code are empty, so prices can still go down. A lot of new neighborhoods are in this situation.
In some others, like Scottsdale 85255/85260, the median price per square foot is still 50% above the 2003 level. I suspect those richer neighborhoods are still in denial, so prices can go down a lot over there. I may be wrong on that one, and I don’t know if the upscale markets affects the median for the whole Phoenix area anyways.
The job market is the most important aspect to consider. For example, companies like Intel, Motorola, Freescale and Microchip have about 20000 employees in Chandler. If they lay off 10% of them (a number I heard this week), it would affect over 2000 families in this city, which won’t help house prices there.
These are a lot of bearish prospects for those already invested in the Phoenix market. But Phoenix has a lot of real estate potential on the upside in the long term, because of the affordable quality of life and good climate, if we can only fix the economy and bring in more migrants.
John Wake 11.17.08 at 7:44 am
@Peter,
In places like Queen Creek where there is a lot of new home construction, the ultimate bottom may be at the cost of new home construction. I don’t know what that cost is but have been using in my head 2002/2003 prices as a proxy for the cost of construction in those areas.
ks 11.17.08 at 1:12 pm
The trend line is generated from part of the data. The part of the data that the trend line is using is 1989-2003. Since the bubble in AZ did not start until late 2004, this should be a good way of determining the non-bubble long term trend. However, it would be nice to know what type of function is being used.
I assume that the trend line is a simple exponential: function = A*exp(r*time), where A is just the initial price in 1989, r is the infintesimal rate increase with respect to time, and time is measured from 1989. This can be more easily visualized when expanding the exponential. The expansion leads to function= A + A*r*t + A*r*t^2/2! + A*r*t^3/3! + … In small time intervals it can be aproximated as funtion ~ A*(1+r*t). The parameter r can now be easily seen as the inflation rate for houses over the period 1989-2003.
If we knew the actual functional form (here I assume an exponential), as well as, the value of the floating parameter (here I assume only two free parameters: A, & r) then we could see how reasonable the trend line is.
By eye, the trend line appears to roughly double over the period 1990-2004. This would imply 2~exp(14*r). A bit of algebra yields r ~ (1/14)*ln(2) = 0.0495. This would mean that houses were appreciating at about 5% over the period 1990-2004.
The question then becomes: Has wages been increasing at a 5% rate?
This was a very rough analysis. It would be nice to have the actual fit function and fit parameters. However, the numbers don’t seem too far from reality. Perhaps we are getting near the bottom.
John Wake 11.17.08 at 2:31 pm
ks,
5% has been my rule of thumb, although I once saw a study that had long term Phoenix appreciation at 4.6% or so. Anyway, most years will be between 4 and 6 percent nominal appreciation.
Brian 11.17.08 at 3:01 pm
The Schiller studies that go back over 100 years (see “Irrational Exuberance 2″) show real inflation in housing prices of less than 1%. Add to that 1% value the average inflation in the dollar of around 4% and you get 5%, so that probably is a good number over a very long time period. I didn’t check those numbers from my memory of reading the book, but you can by going to his website and downloading the economic data going back to the 1880s. http://www.irrationalexuberance.com/
Some might be surprised that housing prices appreciate at something less than 1%, making housing a questionable investment. But it does make intuitive sense. The utility of a home to humans does not change over time. It provides shelter and that is it. Any appreciation over the time period is probably due to added value in the way of comfort and amenities like indoor plumbing, electricity, floor coverings, nicer finishing and added square feet. In some highly desirable areas with limited available real estate (Manhattan), supply and demand may also play a role, but not for most of America where more land is always available.
Brian McMorris 11.17.08 at 4:37 pm
“show real inflation in housing prices” from above. That should read: show real appreciation in housing prices.
John…what is the prospect of allowing us to edit our posts since I always seem to post with grammatical errors?
person 11.17.08 at 6:46 pm
i cant see a thing on the graphs
Brian McMorris 11.17.08 at 8:12 pm
I ran the formulas in Excel for @RATE (which calculates compounded rate over a period of time). Here is what I found: from 1890 to 2007 the American CPI increased by an average of 2.84% per year. The average home price increased by an average of 3.44% per year. So, the “Real Return” on a home over those 117 years is the difference which is 0.60% (0.58% according to the Real Home Price figures in Column B of Figure 2.1).
This is definitive for the total American market. So, if not for the leverage of mortgages (which proved to be high risk the past couple years for the perceived high reward), homes are a pretty poor long term investment.
ks 11.17.08 at 11:13 pm
The couple of things that CAN make housing a decent investment:
1) As a primary residence, not only does a person have to live somewhere, an apartment yields zero return on investment.
2) As a rental property, the landlord has someone else pay the leverage on investment, and if done properly the property will provide positive cash flow. (Note: being a landlord does not sound like an easy way to make money, but some people seem to like doing this sort of thing)
The only truly safe investment is treasuries and FDIC backed CDs. Everything else requires an element of risk. Managing the risk is the trick.
Here is the funny part: Two years ago, when everyone should have been scared to death of investing in housing, people bought houses like crazy. Now that the prices are getting in line with historical trends, people are scared. The one thing that you can be sure of is that buying a house today is much better than buying in 2006 (when the price of houses were far from historical trends). The questions become: Is it better to wait for a year or two? How close to the bottom do you need to be? Is there any way to spot the bottom before it is over?
John Wake 11.18.08 at 7:14 am
I love these definitions.
A boom is when people buy more when prices go higher.
A bust is when people buy less when prices go lower.
A basic assumption of economics works the opposite way.
John Wake 11.18.08 at 7:28 am
Brian, I’ve seen blogs where authors can edit their posts for about 15 minutes after posting. Would that work?
Brian 11.18.08 at 7:49 pm
John, That would be great! Let me fix all those typos. Is this a tool available in Wordpress?
Also, I want to be clear that real estate is an okay investment. Many people have become wealthy owning real estate, especially with good timing (buying in 1992 and selling in 2006 for instance). But RE depends on leverage for its success. With 20% down, the leverage is 5X. So, a 0.5% real return (after inflation) becomes 2.5%, which is better than you can get with TIPS (Inflation Protected Treasuries). However the downside is possibly being forced to sell during a downturn like this one, because of that leverage and having the loan called. If one could ride out the downturn and have a 20 year investment time frame, then RE is a fine investment (albeit a lot more hassle than TIPS to manage).
Another alternative: I have run the numbers on owning REITs (or REIT fund) on leverage. If one were to use 5X leverage (borrow 80% of the investment if that were possible) and buy REITs, assuming one could more than cover the interest payments for the loan with the dividends from the REITs, then one could obtain a similar return as individually owned RE with a lot less management hassle.
Assume a 5% annual REIT dividend which is near the long term average, multiply times 5 for an 80% loan to value, for a 25% annual return on the original investment, and then subtract 8% for the loan interest, considering a 10% rate on the 80% of the investment financed which would generate a net 17% annual return on the original investment, which is probably as good as it gets over time for owning RE outright, using leverage and accounting for management and tax costs.
This is a very simple analysis without tax considerations.
sophia jason 12.03.08 at 12:41 am
The collapse of housing market has created a great mess for homebuilders. Some drastic measures have to be taken to stop the decline of home prices any further. By lowering home prices one may increase home buying demands. Phoenix area has lots of real estate potential with the pleasant nature of life in the suburbs. Builders just need to exploit that area to raise the sale.