A Pieces Of Data From The Last 10 Years

[Father Fred]

I'm doing some coursework at the minute, based on UK average house prices (land registry figures 1995 to 2005 - I have ignored 2006 as the data is probably not 100% accurate yet due to completions being slowly added over a number of months). Other sources being the government statistics office and the bank of England. In the early stages, but a few things I've seen so far that may or may not be of interest -

In 44 quarters since Q1 1995 average house prices have fallen in real terms on 14 occasions. This is during arguably the biggest housing boom in UK history. ie 32% of the quarters in last 10 years prices have been falling in real terms. These falls are spread pretty evenly across the last 10 years - it's not like there was a dodgy year or two. Interestingly this seems to happen almost every 4th quarter. Sell your house at the end of Quarter 3!

Average house prices are now 109% higher than in 1995. In real terms they are 83% higher.

Wages are 18% higher in real terms.

Interest rates have averaged 5.39%.

[Panda]

Average house prices are now 109% higher than in 1995. In real terms they are 83% higher.

Everywhere i look, and i have looked, i see that since 1995 house prices have doubled, then increased by around 50% once they have doubled. So in real terms a 150% increase.

Wages, well, depends on what you do, where you do it, do you still do it, can you still do it, you are lucky if you can still do it, which way is the wind blowing.

[Father Fred]

Firstly, I thought I might bump this up in case anyone missed it yesterday and was interested.

Secondly, by those figures property is 65% more expensive in real terms than it was in 1995 (is that correct?)

So, how cheap was proeprty in 1995? The reason I ask is if it was "fair value" then maybe we are currently 65% overvalued, whereas if it was 65% undervalued then maybe we are at "fair value" now.

[Rachman]

And factor in the interest rate changes since 1995 and see what difference that makes (they are just over half what they were in the mid to late 1990s....). You can't pick two things in isolation that suit you, there are too many variables to reduce it to that.

[?...!]

Firstly, I thought I might bump this up in case anyone missed it yesterday and was interested.

Secondly, by those figures property is 65% more expensive in real terms than it was in 1995 (is that correct?)

So, how cheap was proeprty in 1995? The reason I ask is if it was "fair value" then maybe we are currently 65% overvalued, whereas if it was 65% undervalued then maybe we are at "fair value" now.

I think it's a bit of a combination of the two.

The market overshot a "fair" value because of the cheap loans available following 9/11 and the fed bottoming out right down at 1%

[kingofnowhere]

In 44 quarters since Q1 1995 average house prices have fallen in real terms on 14 occasions. This is during arguably the biggest housing boom in UK history. ie 32% of the quarters in last 10 years prices have been falling in real terms. These falls are spread pretty evenly across the last 10 years - it's not like there was a dodgy year or two. Interestingly this seems to happen almost every 4th quarter. Sell your house at the end of Quarter 3!

%.

Indeed very true. The best time to buy is December and the best time to sell is June. Have a look at the Haliwide Season adjustements and you will see they adjust up during this period and down in the Spring (Feb to June)

[Magpie]

Secondly, by those figures property is 65% more expensive in real terms than it was in 1995 (is that correct?)

No, not if you're comparing the real rise HPs (83%) to the 18% real rise in wages. In that case it is 55% (because 183 is 55% higher than 118).

What are these figures "real" in relation to I wonder? Might be useful to know...

I certainly think property was undervalued in 1995, though perhaps not by 55%. If about half of the rise was a correction to the equilibrium and the rest was an overshoot we'd now be overvalued by about 20% (in other words 20% real falls would take us to the equilibrium). Whether that's near the truth, or the equilibrium has risen as a result of investment activity I really don't know.

[Father Fred]

And factor in the interest rate changes since 1995 and see what difference that makes (they are just over half what they were in the mid to late 1990s....). You can't pick two things in isolation that suit you, there are too many variables to reduce it to that.

Just looked at Nationwide data and in Q1 1995 real prices were 25% below trend, so by that measure prices are 40% too high in real terms now (other things being equal).

Interest Rates (base rates) were 6.625% in Q1 1995.

If we assume that the result of Bank of England independance is that long term IRs have stabalized at 4.5% (I appreciate my figure here is a bit arbitrary, but it could be the case) then interest rates are only 68% of what they were in 1995, therefore 32% of the 40% "excess" rises may be explained by falling interest rates, therefore we are only around 8% overvalued (ie 2 years of nominal stability and wages catching up with house prices).

If we assume that the result of Bank of England independance is that long term IRs have stabalized at 5.5% then interest rates are only 83% of what they were in 1995, therefore 17% of the 40% "excess" rises may be explained by falling interest rates, therefore we around 23% overvalued (ie 4 or 5 years of nominal stability and wages catching up with house prices).

All these things are clearly massively down to what you assume medium term interest rates will be from now, etc etc.

Indeed very true. The best time to buy is December and the best time to sell is June. Have a look at the Haliwide Season adjustements and you will see they adjust up during this period and down in the Spring (Feb to June)

makes sense, put it ont he market in june for completion at the end of quarter 3.

No, not if you're comparing the real rise HPs (83%) to the 18% real rise in wages. In that case it is 55% (because 183 is 55% higher than 118).

What are these figures "real" in relation to I wonder? Might be useful to know...

I certainly think property was undervalued in 1995, though perhaps not by 55%. If about half of the rise was a correction to the equilibrium and the rest was an overshoot we'd now be overvalued by about 20% (in other words 20% real falls would take us to the equilibrium). Whether that's near the truth, or the equilibrium has risen as a result of investment activity I really don't know.

Magpie

Thanks, I knew I had that wrong but this sort of stuff does my head in.

So 25% compared to 55% we are 30% above real trend, ignoring the affects of interest rates.

If you believed that long term interest rates will be a third lower than they were in the mid nineties (certainly not impossible) then it could be that we are a lot closer to trend now than many people on here think.

(The above does not preclude the possibility that short term rates will rise from their new long term low levels and that we are due a crash that will take us below trend again?!?!?!?)

FF

Real in relation to the Retail Price Index. I believe that most property data is compared to the RPI, whereas a couple of years ago the BoE switched from the RPI to CPI as it's inflation target (CPI tends to be 0.5% lower than RPI)

Edited May 2, 2006 by Father Fred

[Magpie]

Just looked at Nationwide data and in Q1 1995 real prices were 25% below trend, so by that measure prices are 40% too high in real terms now (other things being equal).

Hi Fred,

Sorry to niggle at the %ages again, but that's wrong too. 25% below trend means that the trend would have been say 133.33% of the actual prices. So given 65% "real rises" we'd be 20% over trend(because 165 is 20% more than 133.33). Given 55% real rises (see my last post) we'd be need 14% falls to return to trend, ignoring the interest rate stuff.

So 25% compared to 55% we are 30% above real trend, ignoring the affects of interest rates.

Real in relation to the Retail Price Index. I believe that most property data is compared to the RPI, whereas a couple of years ago the BoE switched from the RPI to CPI as it's inflation target (CPI tends to be 0.5% lower than RPI)

Our posts crossed - see above - 155 should probably be compared to 133.33, giving 16.5% over trend (which means that 14% falls would reach the equilibrium (unless by 25% under trend you mean that the trend figure was 125 instead of 133.33...). Interesting. not to say we mightn't overshoot on the way back down of course...

I bore even myself when I niggle about maths, but I can't help it.

Thanks on the RPI, that helps explain.

Edit - inevitably I've now cocked up (165 is 24% more than 133.33, but 133.33 is 20% less than 165.

Edited May 2, 2006 by Magpie

[kingofnowhere]

Real in relation to the Retail Price Index. I believe that most property data is compared to the RPI, whereas a couple of years ago the BoE switched from the RPI to CPI as it's inflation target (CPI tends to be 0.5% lower than RPI)

Hi Fred

Better to use wage inflation, as earnings are the driver for house prices.

[Father Fred]

I don't mind you niggling, i'm posting here to test my arguments... and maths).

Real house prices Q1 1995 £67596

Trend £88,341.

I make that 25% below trend.

The phrase "this time it's different" or changing long term trends is not something to do very often.

But, by those quick calculations it would appear we are 30% over trend. If we assume that any falls play out over 3 years, then wage growth (if it stays at trend) will account for 10% of the real falls over that time.

If we were to assume base rates staying 20% lower than 6.6% for the long term and trend adjusting for this then we may not be above trend?

It's easy to find reasons for believing the market is grossly overpriced, but it's worth testing the opposite arguments.

(Whether we have a crash or not, prices are hardly going to rush upwards a significant amount so me saying prices may not crash is not the same as saying I advise buying in the current market)

the family cat? warm?

IanBe - it is a heck of a statement to make, and I accept your point. What I am trying to do is clarify my assumptions so that someone who disagrees with those can immediately disagree with the conclusion.

King ON - I am trying to look at wages too... but the starting point of the work is to remove inflation, real house price rises, real wage rises.

Edited May 2, 2006 by Father Fred

[New Darker Law]

Interesting stuff Father Fred, I'm glad you bumped it from yesterday.

I tend to think property may have been under valued in '95. As we have seen time and time again markets overshoot both on the way up and the way down.

Depending on your outlook a 20% reduction may bring us back to equilibirum, but who's to say it won't overshoot on the way down and touch closer to 40%?

Keep up the analysis and post more of your thoughts.

NDL

[Magpie]

Real house prices Q1 1995 £67596

Trend £88,341.

I make that 25% below trend.

Yep, close enough, so ignore the stuff about 133.33 - the comparison should be 155 to 125 - 24% over trend, or 19% falls required to return to trend.

[kingofnowhere]

Yep, close enough, so ignore the stuff about 133.33 - the comparison should be 155 to 125 - 24% over trend, or 19% falls required to return to trend.

Except that because the trend line takes into account the current inputs the trend line moves up. So even if house prices stayed flat and everything stayed flat the trend would eventually rise to the current price.

FWIW Nationwide, when the house price first went above trend 1Q00

Now that point is below trend

[Magpie]

Except that because the trend line takes into account the current inputs the trend line moves up. So even if house prices stayed flat and everything stayed flat the trend would eventually rise to the current price.

Well, yes, but doesn't Fred's original way of doing the sum take this into account in a way? Because it is comparing "real prices" to "real wages" to get the 55%.

I suppose what I should have said is 19% falls required to return to trend if the falls are overnight * - if the falls are spread out over time, obviously the trend line should keep moving upwards so the longer it takes the smaller falls that will be required .

Of course this is all statistical talk rather than real world talk. The falls that are "required" might not be what really happens, and other factors may change - but I think it's an interesting way of looking at the numbers anyway.

I think a few people here tend to get stuck on thinking that since HPs have gone up 150% since 1995, they have to fall by 50% or some similarly huge amount to return to "normal", forgetting that 1995 was not normal. This example makes it clear that while we probably are in an overpriced situation it may not be as extreme as some would assume.

Edit: Or alternatively I could have said 19% "real falls" - to be honest I'm surprised that the increase in real prices is as high as it is - I was tending to assume it was flat or just over RPI.

Edited May 2, 2006 by Magpie

[kingofnowhere]

I suppose what I should have said is 19% falls required to return to trend if the falls are overnight - if the falls are spread out over time, obviously the trend line should keep moving upwards so the longer it takes the smaller falls that will be required .

Thats the bit I was being picky on, but I was only being picky.

FWIW really IMHO you shouldn't use indivdual earnings or RPI, you should you the whole employment earning. After all that is the Total purchasing power of the UK, so as more jobs get created then the prices of housing gets bid up. (This is the real measure for UK house prices) Of course this should be discounted by the increase in the housing stock.

Edited May 2, 2006 by kingofnowhere

[Michael]

No, not if you're comparing the real rise HPs (83%) to the 18% real rise in wages. In that case it is 55% (because 183 is 55% higher than 118).

What are these figures "real" in relation to I wonder? Might be useful to know...

I certainly think property was undervalued in 1995, though perhaps not by 55%. If about half of the rise was a correction to the equilibrium and the rest was an overshoot we'd now be overvalued by about 20% (in other words 20% real falls would take us to the equilibrium). Whether that's near the truth, or the equilibrium has risen as a result of investment activity I really don't know.

Since i gave up on the crash ...i have been saying that prices may be only 20-25% overvalued and unless there's some kinda economic catastrophe such an overvaluation will be corrected by 6 or so years of modest wage inflation against a backdrop of stagnant prices.........

[Magpie]

Thats the bit I was being picky on, but I was only being picky.

FWIW really IMHO you shouldn't use indivdual earnings or RPI, you should you the whole employment earning. After all that is the Total purchasing power of the UK, so as more jobs get created then the prices of housing gets bid up. (This is the real measure for UK house prices)

That's an interesting thought. I'm sure a mega-bear would see it as a VI fiddle, but I can see the argument that this would be a good index of house-purchasing power. On the other hand does more people earning the same amount of money (which would increase this figure) really push prices up? Do you happen to know how the whole employment earning maps against the "real house prices" graph?

[Michael]

Please could you point me to an example in history where nominal prices have held steady for 6 years - or even 2 years come to that?

It just doesn't happen and I can't see why it's suddenly going to happen now considering we're slap bang in the depths of the most enormous debt bubble in history, with rising unemployment, bankruptcies and reposessions to boot.

Nominal prices in my area were more or less stagnant throughout the whole of the 1960s and from 1988-98......

[Magpie]

Please could you point me to an example in history where nominal prices have held steady for 6 years - or even 2 years come to that?

Er, that's pretty much what exactly happened in the two crashes prior to the 1989-95 one. Nominal stagnation, real falls, due to wage inflation catching up. They were in high inflation environments which may or may not be relevant to whether it could happen this time.

Nominal falls are actually quite rare in the UK, because prices tend to be sticky downwards - see the thread "Why Was It Different Last Time?" for more on this.

People here tend to say that after every boom there is a crash, that bubbles always burst. That may be true in real terms, but its questionable in nominal terms.

Edited May 2, 2006 by Magpie

[Michael]

the OECD analysed booms in a dozen countries over 40 years and apart from in a few exceptional cases booms were followed by years of stagnation........although the US ave HP has never fallen in 70 years there have been slumps in local markets.....eg LA -20% in the 80s....

South Eastern England after 88 and Japan after 1990 are other exceptions.....

[Red Baron]

Some excellent posts on this thread.

My own prognosis remains that we will see a combination of small nominal falls of 2-4% p.a. for several years which, coupled with inflation, will conspire to drop the real price of property by around 25%. This would imply that current prices are 33% over-valued.

As a previous poster wisely stated property prices are 'sticky' down and the low level of inflation, unlike in previous crashes, is unable to mask real falls in house prices. In my view this will lead to a protracted correction, rather than an outright crash.

[Father Fred]

Some excellent posts on this thread.

My own prognosis remains that we will see a combination of small nominal falls of 2-4% p.a. for several years which, coupled with inflation, will conspire to drop the real price of property by around 25%. This would imply that current prices are 33% over-valued.

As a previous poster wisely stated property prices are 'sticky' down and the low level of inflation, unlike in previous crashes, is unable to mask real falls in house prices. In my view this will lead to a protracted correction, rather than an outright crash.

I think you could be right.

If what has been posted on this thread is right, AND assuming we overcorrect an equal amount on the way down as we have on the way up, then 10% nominal falls per annum for 3 years is the absolute worst case scenario possible in my opinion (a pretty bad scenario for homeowners, but perhaps not as horrific as some on here would suggest is likely).

[camem']

blah blah blah stagnation for years and years, blah blah, wage inflation, blah blah blah

I love the timescales at work here

*2 years ago* (rising 20% a year) : "Can it really stop rising 20% a year ?"

*now* (stagnation) : "it'll stagnate for years and years"

*2 years time* (falling for 2 years) : "don't buy a house"

(the last one is an extrapolation... )

[Magpie]

I love the timescales at work here

*2 years ago* (rising 20% a year) : "Can it really stop rising 20% a year ?"

*now* (stagnation) : "it'll stagnate for years and years"

*2 years time* (falling for 2 years) : "don't buy a house"

(the last one is an extrapolation... )

Or alternatively:

*2 years ago* (rising 20% a year) : "it's just about to start falling"

*now* (stagnation) : "it's just about to start falling"

*2 years time* (stagnation) : "it's just about to start falling"

(repeat until prices start rising again...)

I wouldn't want to make either mistake, but it's probably best to keep an open mind. If anything this thread has tended to predict real falls - it's just a question of how big and how fast they might be if you look at these figures, and whether they will be nominal falls or just real falls.