Can this bubble correct at a new higher plateau?

User Forum Topic
Submitted by powayseller on July 18, 2006 - 5:48pm

edna_mode made an interesting comment about the piggington graph of per capita income/median home price, which historically reverted to a multiple of 7.

" Sometimes fundamental circumstances really do change the mean to a new metastable state. For example, just because the mean for SD price/wages has been 7 is not any guarantee that a disequilibrium has not occurred recently (hidden during the huge runup) that may not cause the new mean to be a different number."

Studies of past bubbles by Charles Kindleberger and the International Monetary Fund show that all asset bubbles correct to their pre-bubble value.

Kindleberger labeled 35 major bubbles in 400 years, including in coins, cotton, land, coffee, ships, canals, wheat, housing, silver, railroads, eurodollars, oil, banks, and British exports to Brazil. The details differ, but the results are the same.

The IMF study of housing bubbles in industrialized countries from 1970 to 2000, found 20 crashes in 14 countries. The average housing market downturn was 30%, and took an average of 4 years, at a cost to GDP averaging 8%.

Let me quote from Bubble Man by Peter Hartcher, which shows that exceptions to bubbles popping to their pre-bubble state have not been found:

"In a ...sampling of bubbles, Jeremy Grantham, co-founder and chief strategist of the Boston-based funds management firm Grantham Mayo Van Oterloo, researched 28 bubbles of the last century or so. He cast his research net beyond stocks and land to incldue currencies and commodities including oil, gold, nickel and cocoa.

He came to an interesting conclusion about the fate of the prices of all of thee items once the bubble had burst: "We found that in every case, in every market, the market returns to where it was before the bubble began. It happened in commodity markets, it happened in currencies, it happened in stockmarkets."

Grantham presented this finding to audiences totalling 2400 professional investors and analysts and wanted to know if any of them nkew of any exceptions to this rule. "I asked them to find me an exception, just one exception. Put up your hand, send me an email, maybe I missed Kuwait in the '70s or something. Not one of these guys can find an example.

If you believe in a new era, you carry a terrible burden of proof. History is full of bubbles where everyone believed there were in a new era, but they all turned out to be just bubbles."

So, the burden of proof is on you edna_mode. Please bring us even one example of a bubble, defined by several deviations from the mean and mass frenzy, that did not correct to its pre-bubble value. Just one. And then you can e-mail your results to Jeremy Grantham, too.

Submitted by SD Realtor on July 18, 2006 - 6:03pm.

PS if this analysis is correct then this market will take a pretty sever beating to get back to a multiple of 7. At least that was what the other two peaks dropped back to wasn't it?

Aye caramba....

Submitted by Tone on July 18, 2006 - 6:11pm.

Again if this is true, just what is the "pre-bubble" level for this current RE bubble? I noticed SD Realtor mentions something about "a multiple of 7" in the previous post. How can I figure out a rough estimate of what price my home might drop down to according to this study?

Sorry if this might be obvious to some of you. I'm trying to wrap my mind around this. I'm just a caveman.

Submitted by edna_mode on July 18, 2006 - 11:53pm.

Madam, I did not mean to throw down a gauntlet down before someone I hold in such high esteem...however, since we find ourselves in a position of misunderstanding, please allow me to elaborate: I speak not of bubbles per se, but examples of overreliance of the reversion of the mean for future predictions, specifically with predictions of how long the system will take to revert to the mean, and especially on how to measure the mean (the chosen time frame, statistical methods and ways of grouping the data can materially affect one's conclusions, for example). There is no objective way to do this process; honing one's judgement in how to treat the data meaningfully is no less of an art than pre-meds spending much of their residencies examining perfectly healthy patients -- they do this to calibrate their sensibility on what "the average healthy person" *is*, and is likely to be *in the future*.

I do not take issue broadly with any of your examples, or the analysis Rich or you have done to demonstrate that the recent housing market is vastly overpriced. I am however concerned that in the strong conclusions that I am hearing, there may be a lack of appreciation in how significant the difference of the sample mean might be from the population mean (ie how representative is your sample set in terms of describing the behaviour of the *whole* system, past and present), and also the difference between descriptive statistics (dissecting data from a specific sample set) and inferential statistics (projecting the analysis of your sample set onto a population, which is intimately related to my first point).

Since I started with _Against the Gods_ (mine is from 1996), I will continue from there...searching the index, "Regression to the Mean, overreliance, illustrations of"...ah, pages 182-183.

I'll refer to the second of Mr Bernstein's examples below (I hope my paraphrasing falls under "fair use" -- but just to be sure, I encourage everyone to rush out to the public library, read through this book once for free and determine if it is worthy of a permanent place in your personal library):

Quote (p183-185, all words in [] are mine):

"Up to the late 1950s, investors had received a higher income from owning stocks than from owning bonds. Every time yields got close, the dividend yield on common stocks moved back up over the bond yield. Stock prices fell, so that a dollar invested in stocks brought more income than it had brought previously. That seemed as it should be. After all, stocks are riskier than bonds. Bonds are contracts...if borrowers default on a contract, they end up in bankruptcy, their credit ruined and their assets under the control of the creditors. With stocks, however, the shareholders' claim on the company's assets has no substance until after the company's creditors have been satisfied. Stocks are perpetuities: they have no terminal date on which the assets of the companies must be distributed to the owners...the company has no obligation ever to pay dividends to the stockholders. Total dividends paid by publicly held companies were cut on nineteen occasions between 1871 and 1929; they were slashed >50% from 1929-1933 and ~40% in 1938. So it is no wonder that investors bought stocks only when they yielded a higher income than bonds. And no wonder that stock prices fell every time the income from stocks came close to the income from bonds.

Until 1959, that is. At that point, stock prices were soaring and bond prices were falling. This meant that the ratio of bond interest to bond prices was shooting up and the ratio of stock dividends to stock prices was declining. The old relationship between bonds and stocks vanished, opening up a gap so huge that ultimately bonds were yielding more than stocks by an even greater margin than when stocks had yielded more than bonds. [Discussion of how inflation underwent a serious inflection point around 1940: cost-of-living had risen an average of only 0.2% a year from 1800-1940 (140 years!!)] Under such conditions, owning assets valued at a fixed number of dollars was a delight; owning assets with no fixed dollar value was highly risky.

But from 1941-1959, inflation averaged 4.0% a year...The relentlessly rising price level transformed bonds from a financial instrument that had appreared inviolate into an extremely risky investment. By 1959, the price of the 2.5% bonds the Treasury had issued in 1945 had fallen from $1,000 to $820 [which] bought only half as much as in 1949!

Meanwhile, stock dividends took off on a rapid climb, tripling between 1945 and 1959...No longer did investors perceive stocks as a risky asset whose price and income moved unpredictably. The price paid for today's dividend appeared increasingly irrelevant. What mattered was the rising stream of dividends that the future would bring. Over time, those dividends could be expected to exceed the interest payments from bonds, with a commensurate rise in the capital value of stocks. The smart move [at that point] was to buy stocks at a premium because of the opportunities for growth and inflation hedging they provided, and to pass up bonds with their fixed dollar yield.

Although the contours of this new world were visible well before 1959, the old relationships in the capital markets tended to persiste as long as people with memories of the old days continued to be the main investors. For example, [Bernstein's] partners, veterans of the Great Crash, kept assuring me that the seeming trend was nothing but an abberation. They promised me that matters would revert to normal in just a few months, that stock prices would fall and bond prices would rally. [As of 1996], I am still waiting..."


So from this example we see how the noise of WWII hid the disequilibrium shift in inflation, which became the main factor in answering the question “should I buy stocks or bonds?” If we were to judge our mean based on 1800-1940, we would conclude one way; if we were to judge after WWII, we would conclude the opposite on a much smaller data set. The winds really had shifted for the foreseeable future.

I am not making any sort of prediction, with regards to housing, the economy or anything else. What I am doing, is pointing out that something as hugely distorting as the recent run-up in prices, largely due to irrational exuberance, may very well have also hidden more subtle yet persistent factors that could very well change the dynamics of the outcome. If sentiment is the kinetics of the system, then factors like monetary policy, inflation targets, the strategic importance of San Diego (militarily or economically to the country, affecting what kinds of incentives are created by private groups or the government for people to live here) that cannot be captured with a headline or a soundbite are the thermodynamics. And I am interested in characterizing each of these appropriately.

But your posts are highly educational, “I enjoy our visits” ;)

Submitted by lamoneyguy on July 19, 2006 - 7:41am.

Very interesting. I enjoyed Bernstein's book several years ago. Highly recommended. So, the takeaway is that fundamental shifts in price relationships are possible. Extraordinarily rare, and only due to a massive shift in underlying economics. In this case the average inflation rate increasing twenty-fold! As was written, the "noise" of WWII hid the disequilibrium shift. So, two questions:

1) Is the war on terror, or any other current events loud enough to hide any such current disequilibrium shift.

2) If such a disequilibrium shift is taking place, what COULD it be?

You mention the possibility that the strategic importance of living in San Diego may have changed, but we clearly see bubble like price increases in many major cities around the world. What sort of disequilibrium shift could have fundamentally changed the valuation of real estate globally? I can think of none. But I would thoroughly enjoy a discussion on this idea.

Submitted by powayseller on July 19, 2006 - 8:34am.

What would be the factor causing a new shift to higher prices? To me, wages would be an obvious factor. If we had inflation, and wages went up 30%, then housing prices would be cheap, and people would keep buying. Problem solved.

If I had to consider the effect of wages on future housing prices, I see a greater likelihood of flat wages for decades, supporting the case for housing prices to fall. Global trade and competition from low-wage countries, and illegal immigration have kept our real wages flat for 7 years.

Housing prices are dependent on wages and employment. So if there were a sudden shift to more high paying jobs, housing prices would increase. But this argument I provide does not answer your well-made point about a permanently higher plateau. Rising wages could allow housing prices to stay high, because the multiple of 7 is maintained.

What has increased the multiple today? Loose lending practices. If loose lending became permanent, then the multiple would stay higher.

I'm sorry I didn't read the book yet, but could you tell me in layman's terms what caused stock market valuation to shift to a higher level, and if that shift is indeed permanent.

What do you mean with the difference between the sample mean and population mean? The sample mean is the houses that were sold. Do you think that the types of homes sold over the last 5 years is skewed? Perhaps, to bigger homes. New homes today are much bigger, skewing the median upward. We sell 30K - 40K homes a year in a city of 1.1 million homes, so those annual home sales are setting the price for the entire city. But nobody selected which homes could be sold, so nobody influenced the sample mean. I don't see the point. We are not selecting a sample (as the pre-med students), but measuring transacations over time. So there is no influencing of the sample.

To date, has any bubble ever corrected to a new mean? This is the Grantham challenge I noted above.

A wonderful disccusion, but I think it is a disguised version of "This time is different, because the mean was not properly determined, or the mean will change because this time is different." Burden of proof is on you, edna.

Submitted by lamoneyguy on July 19, 2006 - 11:15am.

I'm sorry I didn't read the book yet, but could you tell me in layman's terms what caused stock market valuation to shift to a higher level, and if that shift is indeed permanent.

My understanding is that stock market valuation made a permanent shift relative to bonds when comparing bond yields and stock dividends. In other words stock price appreciation became valued above dividends. Not unlike the way that RE investors came to value potential cap gains above rents.

However, that does not preclude price declines. As we have seen, despite this permanent shift in comparative yields, stocks still experience sharp declines in price when they get irrationally overvalued.

Submitted by edna_mode on July 19, 2006 - 11:23am.

I am not interested in debating whether or not this particular bubble will revert to mean. Regression to mean is a very powerful tool and one ignores it at one's peril; I never indicated that I was in disagreement with you of this phenomenon. Nor do I appreciate being categorically lumped in with the mouth-breathing Visigoths who are unable to appreciate history as an educational tool. However being short of time, I will offer the following analogy:

The seasons turn in order from spring, to summer, to fall, to winter. There is a great deal of historical evidence to agree with the prediction that this cycle is a powerful one and is likely to continue into the foreseeable future. Any fool who watches the season turn from spring to summer and then declares a permanently high plateau of summertime temperatures would rightly be regarded as a fool. As would the person who asks what the mean yearly temperature is in Wisconsin; the answer would be, it depends on the season.

However, evidence that there has *always* been a cycle of seasons tells us nothing about the effects, if any, of global warming on the extremities of temperatures on each season. Human activities may be causing an changes in the mean temperatures in any given season from this point forward. Another example is that the climate of the Sahara has undergone enormous variation between wet and dry over the last few hundred thousand years, give an ice age or two and discussion the mean temperature of the Sahara over that time frame is meaningless. This evidence is not mutually contradictory with the force of yearly seasonal changes -- rather, it is merely a matter of difference of scope (kinetics vs. thermodynamics). My suggestion of examining larger effects is neither a contradiction or a challenge to your analysis of kinetic effects, which I affirm in the highest -- I agree with your identification of the kinetic main effects. I just would feel remiss in not adding to your analysis with this addition of scope.

Using evidence that fall and winter will surely follow the summer (which no reasonable person would dispute) cannot say anything about what the future long-term trends will be on any given season based purely on the information that the seasons change and they always follow that cycle. Perhaps an ice age is coming. Perhaps not. Of course Rich's graph shows a situation so extreme that even returning to just the historical mean would be extremely painful, let alone overshooting. If we put a pencil with one end at the earliest time point on his graph so that it sits in the centre of the cycles, we can see that whatever points are furthest out in time (the future) will affect the angle of the pencil the most (points far out on the y axis have the greatest leverage on the linear regression). But it is such a huge distortion of prices over the past few years that maybe a better characterization of the system would be to break your pencil into line segments where each segment covers a full cycle. Then each mean for each cycle would be somewhat different. And your judgement about where to break your pencil (the edges of what you consider a cycle) could radically affect your conclusions.

The way to study longer term questions would be to examine the edges and the extremes of the seasons and try to identify what changes are due to causes that had not existed previously, whether or not those causes are likely to persist (credit liquidity, the game theory involved with international economic factors/trading incentives, saber rattling, etc), and even then there would be a great deal of uncertainty in the prediction.

And as in any chaotic system quantitative, specific predictions are hard to make with any reliability. Picture the gazelle that has been gravely wounded by the lion and then somehow managed to get free (eg the housing market wounded by change in sentiment), only to find itself now faced with a circle of jackals (inflation, monetary policy determined by a new Fed, credit liquidity, changes in the Bretton-Woods agreement (our currency is no longer gold backed and other countries may decide to stop pegging their currencies to ours)) -- it may very well be impossible to judge which particular jackal leaps for the throat of the gazelle to make the final killing blow. So while it might be interesting to line up potential "jackals", no one of them might be obviously strong enough to deal the killing blow -- but the combination of them might be. Or there might be few enough jackals at the moment and the gazelle might manage to escape one more time, for now -- and everyone who might benefit from eating gazelle tonight goes hungry for another day.

Submitted by VCJIM on July 19, 2006 - 11:53am.

I find Enda's thoughts intriguing; it is wonderful to have an idea proposed that makes one rethink those things he is definite about. She is clearly not saying that she thinks things will not revert; only proposing other possibilities based on sound technical principles. Being technical, I can appreciate not only her arguments, but also her writing style.

Those with even the slightest brush with science are aware that, not too long ago, the sun, moon and stars revolved around Earch and it was sacrilegious to think otherwise. Only through the strength and wisdom of those that consider other possiblities are we now able to communicate ideas through this site, agreeable or disparate as they may be.

Submitted by bob007 on July 19, 2006 - 1:13pm.

it is possible that san jose real estate market stays high given the number of high income couples who are living in apartments. lot of home owners have mortgage payments of less than $1000 on their primary residence. they have amassed plenty of money to scoop up apartments and condos should the real estate market go soft.

Submitted by lamoneyguy on July 19, 2006 - 3:58pm.

Why are the high income couples living in apartments right now? Are there a large number of newly created high income jobs within the last 3-5 years? Or have those jobs been there all along, and perhaps even more high income jobs 5 years ago versus now?

As for the owners with low mortgage payments, not eveyone wants to be a real estate investor.

According to the San Jose Dept of Housing, average monthly rent is $1,269, and the median home price is $765k. Why on Earth would you ever buy with this disparity? No wonder there are high income households renting. Prices would have to come down at least 50% for many of those renters to consider buying.


Submitted by realtycheck on July 19, 2006 - 5:38pm.

Consistent with some of you arguments, Rich's graph of prices to income covers only 30 years. That is a very short period of time. While the 2 most recent cycles have reverted to the mean or slightly below, that does not mean that that current cycle will do the same.

Submitted by masayako on July 19, 2006 - 6:11pm.

I'm considered high income (6 figures) and I simply refuse to buy and pay for other's retirement. My wife and I will rent as long as it takes. If the market does not drop by 30%, we won't buy. We are just as happy to rent and use the money to invest.

It's stupid to buy. Period.

Submitted by rankandfile on July 19, 2006 - 6:43pm.

Then both you and her need help. I don't mind some challenging reading from time to time, but comon. A single thought drawn out over a seemingly endless tirade of words, and over-peppered with analogies? I'd rather drag my boys across broken glass.

Nothing personal to either of you, but maybe provide a link to an executive summary next time for us breathing Visigoths.

Submitted by bob007 on July 19, 2006 - 7:10pm.

Some folks like me started working in Silicon Valley after 2000. They did not have the down payment to bid for homes in 2002. They got priced out 2003 and later.

High income jobs have been around. I do not think there is a net increase in high income jobs. There is no assurance jobs will stay around five years from now.

Submitted by powayseller on July 19, 2006 - 8:24pm.

enda_mode, I understood the point you were trying to make in your first post.

I don't think you can compare evolution and changing environmental conditions to asset bubbles. So once again I ask you the question I posed above:

To date, has any bubble ever corrected to a new mean? This is the Grantham challenge I noted above.

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.