Warning: Cannot modify header information - headers already sent by (output started at /home/theecono/public_html/includes/database.mysqli.inc:134) in /home/theecono/public_html/includes/bootstrap.inc on line 729

Warning: Cannot modify header information - headers already sent by (output started at /home/theecono/public_html/includes/database.mysqli.inc:134) in /home/theecono/public_html/includes/bootstrap.inc on line 730

Warning: Cannot modify header information - headers already sent by (output started at /home/theecono/public_html/includes/database.mysqli.inc:134) in /home/theecono/public_html/includes/bootstrap.inc on line 731

Warning: Cannot modify header information - headers already sent by (output started at /home/theecono/public_html/includes/database.mysqli.inc:134) in /home/theecono/public_html/includes/bootstrap.inc on line 732
The Margin of Safety and Turning Points in House Prices: Great study in the Financial Analysts Journal | The Economic Analyst

Become a member. Sign up here

Become a member

Already a member?

Forgot your password?

warning: Cannot modify header information - headers already sent by (output started at /home/theecono/public_html/includes/database.mysqli.inc:134) in /home/theecono/public_html/includes/common.inc on line 153.

The Margin of Safety and Turning Points in House Prices: Great study in the Financial Analysts Journal

JULY 29, 2011

I have to highlight a great paper published in May/June 2011 edition of the Financial Analyst Journal.  The paper was authored by Mitsuru Mizuno and Isaac Tabner and is titled "The Margin of Safety and Turning Points in House Prices: Observations from Three Developed Markets". 

Hat tip to TC for alerting me to this article. 

In the article, the authors essentially examined how closely house prices have historically tracked measures of fundamental value such as income, GDP, rents, and inflation in the US, UK, and Japan.  Interestingly, these are the very fundamentals we have examined here with respect to the Canadian market.  You can follow the hyperlinks in this article to read all about them.

The authors were also examining how likely it is that when house prices deviate from these fundamentals, that they will revert back to them with time.  The data analysis section is a very heavy read, but I would highlight some of the excellent discussion in the introduction and conclusion:

 

On the importance of housing to the broader economy:

...A procyclical relationship between house prices and bank lending and between house prices and the wider economy has been documented...The reason is that rising house prices increase households’ borrowing capacity, thereby stimulating nonhousing consumption and growth in the economy.... The effect, in turn, stimulates income growth, which enables further house price appreciation. Eventually, it becomes increasingly difficult for production in the nonhousing economy to support the circulation of money in the housing debt cycle. Consequently, expansions in house prices tend to be followed by contractions and deleveraging

This procyclical relationship (ie feedback loop) has been discussed at length on this site.  Consider some of these articles:

The Great Connection

CAAMP puts the cart before the horse

 

Arguably, imbalances between housing and the wider economy arise periodically because housing returns exhibit a high degree of autocorrelation. This characteristic enables econometric models to produce appealingly accurate forecasts in the short to medium term, thus fueling the innate human bias known as “representativeness” or “extrapolation.”

This bias may have resulted in the discounting of long-term macroeconomic imbalances by the application of an “it’s different this time” rationale.

The argument that "it's different this time" is about the only 'rebuttal' you're likely to hear when housing bulls are forced to face the reality of the extreme dislocation between house prices and fundamentals.  Sorry.....it's never different.  The same factors that have worked to drive and limit house price appreciation are still at work today, though they can be muted for a time by low interest rates, an artificial abundance of credit, and a new and temporary willingness on the part of consumers to embrace irrational debt levels.

Neglect of imbalances between housing and the wider economy by key decision makers has contributed to repeated unsustainable expansions followed by rapid contractions and financial crisis.  Specifically, in many circles, widespread denial often exists, even by those who have a responsibility to know better, of the fact that the price of an asset can rise only to the extent that sufficient real wealth is available to pay for it.

Increasing the supply of affordable housing is often stated as an objective in political party manifestos.  Encouraging house prices to fall, however, at least in nominal terms, is widely regarded as political suicide. Various governments have attempted to resolve this paradox either by promoting policies that make mortgage finance widely available or by trying to engineer a faster rise in disposable income than any concurrent rise in house prices. Success in achieving these objectives has been limited. The cause may be partly the complex procyclical relationship between house prices and the wider economy and partly the fact that many members of the political elite have personal as well as political motivations for promoting rising house prices—motivations that potentially override their rhetoric about affordability.

Yup.....This is where CMHC's mission statement is entirely self-defeating

The indispensability of housing services means that prices are unlikely to fall to zero relative to aggregate wealth; similarly, financing constraints impose boundaries on the extent to which price rises can outpace long-term increases in aggregate wealth. Temporary evidence of bubbles and crashes occurs either (1) when home purchasers forget or ignore this rule or (2) when prices react in anticipation of shocks to aggregate wealth. Thus, in the long run, house prices are bounded by the supply of wealth in the economy, so if prices temporarily rise faster than GDP, this imbalance will eventually have to be corrected by an acceleration in the rate of GDP growth, a slowing in the rate of house price growth, or both.

Apart from being an elementary principle of economics, the idea that price expansions are limited by available wealth is evident from an examination of the time-series data presented here. These data, when detrended, reveal long-term mean-reverting relationships between house prices and other macroeconomic variables—in particular, GDP per capita.

And with that, let me remind you exactly what the growth in house prices and GDP per capita has looked like in Canada:

house prices gdp per capita canada

house prices Canada divided by per capita GDP

And for what it looks like on a provincial basis, check out the following posts:

A regional look at Canada's housing bubble:  Part 1

A regional look at Canada's housing bubble:  Part 2

The empirical data continues to suggest that house prices are in fact bound by the fundamentals so often discussed here.  It is not population growth, immigration, lack of land, etc that drive house prices.  And it hasn't been the traditional drivers of house price growth that has led to the unprecedented rise in real estate values over the past decade.  But deviations from these fundamentals can persist only as long as the source of this deviation is similarly at work:  Falling interest rates, abundance of credit, ability/willingness of people to assume massive debt obligations.  These are not sustainable.  It's not different this time.  But don't take my word for it.  Heed the wisdom of the authors:

In fact, despite prolonged cyclical pushes away from the mean, we present evidence indicating a powerful deterministic relationship between
detrended house prices and the macroeconomic variables examined. Hence, when prices deviate from the mean, the question is not if they will revert but when.

Cheers,

Ben

Posted in:

Ben Rabidoux
By Ben Rabidoux

Enjoyed this Post?

Subscribe to our RSS Feed, Follow us on Twitter, Subscribe by email or simply recommend us to friends and colleagues!

18 Comments

  • Jim said:
    • 2 years, 8 months

    Interesting. I have to read the entire article when I get a chance.

    I can see the rebuttal. "Pointy haired academics stuck in their ivory tower", or perhaps "Canada is unique, because of our impeccable banking sector".

    I like the fact that they pointed out the obvious: many policy makers have personal interests in rising housing prices. Anyone who reads a history book on a major metropolitan centre like Las Vegas or New York readily encounters depictions of collusion between politicians and developers. (Not to mention politicians and organized crime figures). Someone really has to do some analytical work on incentive structures in this context.

    Reply
    Post a comment
  • backwardsevolution said:
    • 2 years, 8 months

    Jim, you hit the nail on the head: "personal interests in rising housing prices".

    Reply
    Post a comment
  • Appraiser said:
    • 2 years, 8 months

    In examing your data, a regression line (or line of best fit) would indicate that average Canadian house prices divided by GDP per capita continually rises over the long term. Where is the line on the graph that represents the mean?

    Reply
    Post a comment
  • pv said:
    • 2 years, 8 months

    Appraiser, were you the one saying something along the line of “There hasn't been a correction without a concomitant recession” ?
    Curious what you thought about this morning’s GDP numbers? It looks like we're getting your concomitant recession. Canada's GDP shrank in May and was flat in April. I'll go out on a limb and say Q2 & Q3 will be negative, hence a technical recession.

    http://www.thestar.com/business/article/1032025--canadian-economy-unexpe...

    Reply
    Post a comment
  • rp1 said:
    • 2 years, 8 months

    That's true over the time period because interest rates fell. They can't do that again.

    Reply
    Post a comment
  • Greg said:
    • 2 years, 8 months

    Ben, how long till you admit the financial system is nothing but a ponzi scheme? If it isn't different this time, and history always rhymes, then we know that no fiat currency has ever lasted...ever.

    Reply
    Post a comment
  • Mr.Regression said:
    • 2 years, 8 months

    Hi Appraiser

    For the "Average Canadian house price divided by GDP per capita" graph, a regression line would not have a very good "fit" because of the fluctuations in that graph. The best fitting curve appears to be something like a sinusoidal curve. But there is no denying that over a long period of time, yes - house prices increase over time to a rate similar to the rate of inflation. The fact that a sinusoidal curve is the "best fit" stresses that in the short term, there are good times to buy (and NOT TO BUY). Right now, it's painfully obvious NOT TO BUY.

    Reply
    Post a comment
  • Appraiser said:
    • 2 years, 8 months

    I disagree that the trend line is purely sinusoidal, it definitely expresses an upward trend. Perhaps a more longitudinal study would be beneficial in revealing any real long-term tendancies.

    In regards to the second part of my inquiry - what is the mean? It is clear that the graph fluctuates, but there is no mean to compare it to?

    Reply
    Post a comment
  • Mr.Regression said:
    • 2 years, 8 months

    The upward trend is the whole point of Ben's site - is RE overvalued in Canada?

    Please mistake me if I am wrong, Ben - but ideally shouldn't this be a flat line? If the numerator (average house price) increases at the same rate as the denominator (GDP per capita), then this should be a flat line? From 2000 on, the average house price has increased at a faster rate then GDP per capita, meaning RE in 2010 is overvalued well beyond the mean. Hence, when prices deviate from the mean, the question is not if they will revert but when.

    1980-2010 seems very appropriate. The mean is probably lower for a longitudinal study and would favour Ben's side of the argument. Just for the hell of it, I'd like to know what the mean is for a longer period of time

    Reply
    Post a comment
  • 604x said:
    • 2 years, 8 months

    Using average GDP seems inappropriate for the analysis (although convenient since the data is readily available). I'd guess a better approximation would be:

    House prices to a function of:
    1. Disposal household income ("household GDP"). This grows over time and should lead to long-term appreciation of real estate values.
    2. Mortgage costs (gross debt service carrying capacity for debt)....defines how much people can borrow. This is cyclical - probably biggest driver of the cycle.
    3. Household preference for allocating money to housing (the typical assumption is +/-35% of household income). However in times when people love real estate (like now) they may be willing to allocate more income toward real estate consumption....70% in Vancouver?!

    Combining the above might lead to a flat line over time, with mini cycles driven by mortgage availability (low rates) and the extent people are willing to allocate income to real estate because it is fashionable/hip/cool.

    Ben - you have all the data. Can you get one of your students to crunch the numbers of correlate the factors?

    The big question remains: when will people swing back toward a psychology of not allocating a massive percentage of household income to real estate? The psychological tipping point for this might be increases in unemployment or increase in interest rates.

    Reply
    Post a comment
  • Fiduciary said:
    • 2 years, 8 months

    Just from looking at it, I agree it looks like a sinusoidal curve with a slight upward trend. I'd peg the midpoint of that trend at around, oh, 5.7, which is still much below the 7.0 it's at as of the beginning of this year. That indicated it's still got quite a ways down to go, especially because if it continues the trend it would end up at around 5.0 at the bottom. Which, as Mr. Regression says, makes it not a good time to buy.

    Reply
    Post a comment
  • mfx said:
    • 2 years, 8 months

    I'd look at this graph and draw a upward-trended channel with the mid point currently sitting on 6.2, the ceiling at 6.75 and the floor at 5. I'd also observe where the graph will go. It has to stay above the channel to remain strong market.

    Personally, I'd rather wait for the graph reaching the lower bound of the channel. The price is important. The money management is also very important.

    Reply
    Post a comment
  • mfx said:
    • 2 years, 8 months

    sorry. correction. my lower bound should be at 5.5.

    It doesn't matter how you draw the channel or locate where the buy or the sell is. The important thing is to follow your own signal + good money management.

    Reply
    Post a comment
  • Greg said:
    • 2 years, 8 months

    A small excerpt from a CSNA publication: http://www.statcan.gc.ca/pub/13-605-x/2011003/article/11491-eng.htm

    -The services industries represented 66% of the economy; today they represent 72%

    -Household debt as a percentage of disposable income was 108%; today it is 148%

    -Exports as percent of gross domestic product (GDP) was 39%; today it is 29%

    -Government net debt as a percentage of GDP was 85%; today it is 45%

    So how is the government lowering its debt when exports have fallen from 39% to 29% of GDP? Apparently, the debt is being shifted to household debt (mortgages). But here's the kicker—now that home prices are inflating, new homes valued at 450k+ are being pushed into taxable HST margins, so the government will reap a hefty tax on these new sales.

    This was well planned by the government and is why Mr. Flaherty doesn't care. It's called taxation without representation.

    Reply
    Post a comment
  • backwardsevolution said:
    • 2 years, 8 months

    Greg - I like your posts. You question things. It's refreshing!

    Reply
    Post a comment
  • Greg said:
    • 2 years, 8 months

    It's nothing more then classic over modern economics which supports the argument that the Canadian government did nothing more then find ways to hide inflation and taxation—nearly fooling everyone. This is nothing new and comes straight out of the government and central banking playbook.

    Doesn't anyone question GDP figures when StatsCan just posted (highlighted in blue) that GDP has been revised way back to January 2010! They are fudging numbers to make the economy look better then it is by simply revising (lowering) previous GDP to prop-up current GDP. Same goes for inflation only they revise previous data upwards to lower current rates—it's an absolute joke.

    Reply
    Post a comment
  • bill said:
    • 2 years, 8 months

    Hi,
    Great website. (I found it after reading the article in Canadian Business)
    You have done very thorough analysis of the factors driving house prices in Canada. My question is, why did most of the major metro areas in Canada all take off at the same time - around 2001? and in such a dramatic fashion. No historical precedent. It is a real hockey stick for most cities. You make a very good point on easy credit as the key factor, and I think you are right, but I don't recall credit conditions changing that significantly at that point in time. (i'm sure there is a chart out there somewhere) We basically follow the US up and down. I'm sure that it is a combination of factors (multivariant analysis as the statiticians say) but I think one factor drove the market over the edge.
    To me, Montreal is the most dramatic, because it was a moribund market for the past 50 - 70 years. Renters ruled. then all of a sudden it takes off. makes no sense.

    thanks
    Bill

    Reply
    Post a comment
  • bill said:
    • 2 years, 8 months

    My point is, if we can determine the factor that caused the hockey stick in 2001, then we'll know when that factor goes, the boom will end.
    Seems to be cross canada
    I don't know if other countries experienced the same shift at the same time

    Bill

    Reply
    Post a comment
Post a comment