Reversion to mean = 37% nominal price drop, maybe 50%

[powayseller]

This chart shows that this year’s hot stock sector is next year’s loser.

Successful investors don’t chase today’s hot stock (or asset class), because they understand reversion to the mean.

Let’s apply this to real estate. Charles Smith at oftwominds.com used Robert Shiller’s finding that housing appreciates 1% nominal per year, and applied it to San Francisco housing prices. His analysis shows that the median price should be $315K, not $715K. That’s a long way to fall.

I did a similar analyis for San Diego homes prices.

I used the most recent data. Per capita income was $35841 in 2003, and adding 3% annual wage increases,it would be $45402 by 2011. Multiply this by 6.8, the last bottom ratio, and you get a median home price of $308,733. That is a 37% drop from today’s median price of $488K. Please correct my math if I made any mistakes.

Now I will take the liberty to make a subjective adjustment. I think this correction will overshoot by a lot, because of the prevalence of ARMs, and the lending made to subprime borrowers.

Historically, 65% of Americans were homeowners, because the rest just didn’t have the financial stability to get a mortgage. That 65% figure was stable for decades, until the late 1990’s. We inched up, and today are at almost 70% homeownership. Those last 5 points of Americans added to the homeownership pool is the subprime borrower, and they are most likely to default. Add to that people with good credit on suicide loans, long time homeowners who refied and HELOCed their equity out, and I think the wave of foreclosures could drive the median house price down 50% to $244,000 by 2011.

That is a stretch, but the bankruptcy waves, bank failures, Fannie Mae failure and government bailout, and the recession caused by contractors and realtors and lenders going bankrupt, will make this bust worse than the others.

Or maybe not? Maybe all bubbles revert to the mean and don’t overshoot.

So the question is: how likely is this bubble to overshoot on its way to reverting to the mean?

I am looking forward to an intellectual discussion. Let’s keep this professional and focus on history and data.

[rankandfile]

It seems that the author(s) of the San Diego Regional report were doing a bit of double-talking. In one sentence they mentioned that there will be a soft landing in the housing bubble due to steady employment, increased incomes, and spending growth; while in the next sentence they express serious concerns over high gas, utility, and housing prices, as well as the resetting ARMs. My question to the author(s) would be where is all the money going to come from to support the 9% annual increases in consumer spending if it is going to cover increasing mortgage payments and rising costs of living?

[equalizer]

PS

I have no qualms about your 37% figure. We may have to lower it a tad to account for the helicopter Ben who is going to cut rates. To counter the ARMS reset we have to examine population density. We now have high rise apt and condos that didnt exist during last downturn, 15+ stories in UTC and off the 15 (Clairemeont area).
If you’ve been to New York, San Francisco, Singapore you know that single family house is a pure luxury reserved for very wealthy. Thats why newer single family houses may go down 20-30%, while condos may fall 30-45%.

[bgates]

First off, the chart you link to does not support your blanket statement. Staying with last year’s hot sector beats switching to last year’s loser 15 of 20 years. If you started with $1000 and invested in the previous year’s best sector every year from 1987-2005, you’d end up with $11770. Switching every year to last year’s worst sector gets you $3081.

The next analysis you link to is a bit suspect since the guy claims 1% increase for 20 years is, yes, 20% (it’s 22 – compounding). That bumps his supported price up to $346k instead of $315. And who says the 1986 price was the ‘right’ one? If you take the 1994 median of ~250k and give 12 years of 3.9% appreciation, you’d predict a price today of $396. If you took the 1997 price of 292k – just at the beginning of the boom/bubble – you’d predict a supported price of $412.

Your San Diego stuff is better, since Rich has collected more convincing (longer term) numbers.

[powayseller]

I love your comments. Have you thought of e-mailing [email protected], to give him your feedback? I’ve done this with him on other topics, and he is very open-minded to feedback, and sometimes changes his position, or explains how he arrived at his conclusions.

Re the stock chart, it shows that asset classes fall in and out of favor. The S&P 500 Growth funds were in the #7 spot in 1993, in the #1 spot from 1995 – 1998, and then moved back down to the #7 spot from 2000-2005. None of the sectors consistently stayed in the top or bottom slots for the period shown.

[Steve Beebo]

Reversion to Mean?

I love it when someone throws out some type of mathematical formula to demonstrate that prices should therefore be dropping by X percent. At some point I think common sense has got to be included in your calculation.

For example, the following metropolitan areas all have median household incomes of a similar amount, all around $45,000 to $47,000 per year:

Indianapolis
Dallas
Kansas City
Austin
Milwaukee
Cedar Rapids
Detroit
San Diego
Santa Barbara

By your reasoning, all of these areas should have median home prices around $300,000. But real estate prices always boil down to supply vs. demand, and the historical trend in San Diego generally has more demand than supply, and eventually we will probably have that again. Who on this forum wants to live in Dallas in the summer, or Detroit in the winter? Right now we do have too much supply in San Diego, so we’re in a down market, but I can guarantee that the median home price in San Diego and Santa Barbara will never drop to $300,000, and that the median home price in the non-California cities above will never rise to $300,000, at least in our lifetimes.

[equalizer]

Steve B

That’s a rash argument. You obviuosly haven’t spent much time here. The comparison here has been with San Diego and its historical figures, not comparing to Autsin or Detroit. (Although Austin is ranked #1 in creativity and mean income is not far lower than San Diego)

Check out Rich’s graphs. If you want more official evidence go to John Burns http://www.realestateconsulting.com, a leading consultant to the home builders. He has in depth data on every city for $400, but occasionally offers free summaries. He is not bearish (or bullish)on SD, but does worry about low paying new jobs and condos.

[powayseller]

Steve, thanks for your reply. You can’t compare the price of San Diego homes to the price of Cedar Rapids homes. (Sunshine tax refers to the higher cost of living here.)

What is the historical price/income ratio for the other cities you mentioned? I would guess that price/income is 2.5 – 3 for Cedar Rapids, 3.5 or 4 for Kansas City. It is usually between 7 and 9 in San Diego. You pay more for the privilege of living in a desireable city.

San Diego’s supply glut could take a decade to consolidate. With the overbuilding, people moving out, and rising foreclosures, it could be many years before the current supply is bought up. Buyer fear, a reverse psychology of the buyer mania of the last 6 years, can make the downturn even worse than any of us expect.

My prediction is San Diego median price is $308,000 by 2011. So it is well above your $300K minimum. What if foreclosures are really high, buyer psychology against buying is stronger than expected, and we get below $300K? Possible? I don’t know.

Have you read the Bubble Primer articles? A lot of people have not read it, but it is well worth the time.

[michael]

Pimco Article

I encourage you to read the above linked article written by Bill Gross. Page 3 references a Fed study dated September 2005.

The yield on the 10 year treasury may have peaked. Friday’s 4.99% yield broke the 5% support level. However, don’t count on lower long term rates to save housing. (Especially in areas such as SoCal that are now, as a result of the recent run up, dependent on the short end of the yield curve.

Notice the chart in the article on the history of 10 year rates back to 1988. It appears that the peak of the 10 year was in 1989 – right before the end of the Fed’s tightening campaign. I don’t think I have to remind anybody as to what happened in 1989 with respect to the housing market in SoCal.

Also notice that rates dropped considerably over the next years. Didn’t help housing one bit!

Reversion to the mean is real. Not even interest rates will help in this case.

Thanks for the comments Powayseller.

[powayseller]

Thanks, Michael. It’s refreshing to hear from people who also research and understand the gravity of this situation, like you, bugs, privatebanker, and others. I get tired of having to explain this stuff to people who just want to argue. I am appreciative and interested when forum members take the time, as I do and you did here, to research and present their findings and analysis. I will read it as my bedtime story. Thanks again.

[sdduuuude]

bgates hit some issues correctly:
1) The stock-related chart is completely unrelated to real estate. Lets just ignore it.
2) You know what I think about that other guy’s article – it isn’t surprising that bgates had the exact same issues with it that I had when you last posted it. Lets ignore that, too.
3) The date from which you start the 1% growth is important, and can affect the analysis significantly. How did you come up with it? 2003 seems random.
4) Your SD analysis isn’t bad, however you make a leap of faith in your assumption that I can’t accept.

Schiller claims that housing increases by 1% per year, nominally, but your analysis is on the ratio of housing prices-to-income. A major, major leap of analysis that has really no merit.

Schiller doesn’t make any claim about the ratio of housing to income. Just housing prices.

[sdduuuude]

Powayseller – I think Steve Beebo has a great comment here:

“Real estate prices always boil down to supply vs. demand”

This is really the crux of the matter. Analyzing charts and trends and ratios of price to income is too far removed from the actual market. Don’t analyze the bottom line, analyze the factors that go into the bottom line and you’ll have a better feeling for what will happen.

May I suggest an analysis that predicts supply and demand by answering the question – “How many houses need to be panic sold in the next 3 to 7 years?” Is is 10? 10,000? 100,000? I don’t know, but I think that answer could help

The demand side may surprise you and others predicing ultimate doom – unlike me, who is predicting your basic doom. There are quite a few people who made lots of money in this boom. That money is sitting around in cash. You sold a house in 2005. You have cash. I sold a house in 2005. I have cash. In fact, if you look at all the sales recorded in 2004 and 2005, you start to realize that for every suicide loan, there is someone with a small truckload of cash – each of whom is a potential buyer in the coming down cycle, even if interest rates are high.

Figure out how many buyers will be out there, and how many houses will be for sale and that will tell you where the price will settle, not the price trend itself.

[rankandfile]

I’d be willing to guess that the number of people who are in trouble outnumbers those who have made cash by selling at the peak. One reason would be those that bought a new home. The developers/contractors would seem to be the ones who make out those instances. But they make out at the expense of many times more people and families.

Another reason is that a percentage of those who sold their homes at the peak likely bought another home at the peak, rather than investing their cash and renting. In these instances the extra cash made in the sale went as a down payment on a new loan (exotic perhaps?) on another over-priced property. Wouldn’t purchasing an over-priced property essentially offset most, if not all, gains made from selling an over-priced property?

[sdduuuude]

Those who sold at the peak may have known it ws the peak, and if everyone who sold bought a new home, we wouldn’t have decreasing sales, now would we?

[powayseller]

suuuude, 3) The date from which you start the 1% growth is important, and can affect the analysis significantly. How did you come up with it? 2003 seems random.
I don’t know where you got 1%. I am extrapolating wage growth at 3% annually, which is higher than it actually was historically. Schiller’s research shows the national housing prices increase 1% real annually, and this was used by Charles Smith, not by me. I used Rich’s prices-to-income ratio.

This thread is about reversion to the mean. The demand argument is all part of it. When bubbles are on their way up, demand is heightened, as buyers are in a frenzy to get in before they are priced out. In a decline, demand works the other way. Buyers start to stay home, so to speak, as they see prices fall and don’t want to buy today, believing that prices could be lower tomorrow. For this reason, although prices are LESS now than last year, the number of buyers KEEPS DROPPING. Why isn’t demand up, since prices have come down so much? So the entire reversion to the mean plays out because psychology shifts on the way down too.

sduuude wrote In fact, if you look at all the sales recorded in 2004 and 2005, you start to realize that for every suicide loan, there is someone with a small truckload of cash – each of whom is a potential buyer in the coming down cycle, even if interest rates are high. I don’t see that at all. How did you get this conclusion?

As to your last comment, many people who sell are leaving San Diego,as I keep writing about. We had 44,000 people moving out in the year ended June 30, 2005. I haven’t seen the data this year, but going by anecdotes from a half dozen friends/acquaintances, it is a lot more. I never heard of people leaving, and these last few months, so many people ar telling me that so-and-so left San Diego for a cheaper city. We have 30% of our listings vacant. So sales are down because the people who sold their homes left San Diego and are not buying anothe home. Some stay and rent, driving up rental demand.

Good questions, keep ’em coming.

[Author]

Posts