How housing slumps end

Agustín Bénétrix, Barry Eichengreen and I have a piece over at Vox looking at the end of house price collapses. Historical patterns don’t suggest that Irish residential prices will stop falling any time soon; the best way to ensure that they do is to let them adjust downwards as speedily as possible.

Update: in light of a recent article in the Sunday Tribune, I should clarify that nowhere in the Vox piece do we present estimates of the extent to which house prices will decline in Ireland. When asked by the journalist in question how far they would have to fall, I replied that I agreed with Morgan Kelly’s analysis, or words to that effect. When pressed as to what that meant, I gave the figure mentioned in the Sunday Tribune. I am always happy to cite and give credit to Morgan’s work in this area, but I am not happy to be presented as an independent source of analysis on the subject, much less to be described as a “leading housing researcher”.

There. That feels better.

38 thoughts on “How housing slumps end”

  1. Really interesting piece Kevin–just today I was reading Ed Leamer’s Housing *is* the Business Cycle (http://bit.ly/9KrAKr). Leamer points to the Volume cycle on page 25 of his report as being of significance for GDP statistics. Leamer writes:

    “…housing has a volume cycle, not a price cycle. Home prices are very sticky downward, and faced with a decline in demand, it is the volume of sales that adjusts not the prices. With the decline in sales volume comes a like decline in jobs in construction, finance and real estate brokerages.

    “Don’t forget that the P in GDP refers to production, and the appreciation of your home as you sip your chardonnay in your hammock in your back yard is not production and doesn’t count as part of GDP. But jobs in construction, finance and real estate do count. Thus volumes matter for real GDP accounting, not prices.”

    Do you feel the conclusions you draw in your work be substantially changed if you were looking at volumes of residential and commercial construction across business cycles, and not prices?

  2. I’m I right in my interpretation of the conclusions, that there’s a

    – 0.3 % chance that house prices have stop falling here (based on Q1 of this year).
    – If the the falls are drawn out over a total or more than 30 quarters that it’ll have a further negative impact on the probability of a recovery?

    Evidence that the current ‘fire sales’ are really just ‘trying to get out before we loose more than our shirt’ sales.

    My guess is we wont see the recommended fast falls, as our lot, NAMA/banks/professional-fee gougers, will continue to drip feed or recommend to drip feed property to the market.

    Irelands lost decade and a half. Japan eat your heart out.

  3. @ KO’R,

    I’m in the process of reading your article in bits and pieces.

    A bit off topic: What I do find quite interesting myself, is the legislative context in which housing development happens in North America say, versus in Europe. I thought I knew my way fairly well around European energy conservation policy, having reading through the 12 articles of the EU directive, which I printed on a few sheets of A4. Recently, I had to review several case studies for cutting edge home design/build projects in North America. I kept on running into different accreditation schemes, standards and metrics used to rate performances of one home versus another. That was in addition to thinking in BTU’s and square feet, instead of KWh’s and meters. I decided to take a quick look at the federal government of US, department of energy website. The distinct impression I have now, is that the government bureaucracy in places such as North America is trully vast, complex and un-manageable. It is a wonder it operates at all. Given the way they operate the system in North America, I am left in no doubt, it would be hard to regulate production of housing units along any dimension. I never realised the federal system over in North America is so different to our own, here in the EU. BOH.

    http://www1.eere.energy.gov/femp/financing/superespcs_fda.html

  4. Stephen Kinsella says:

    “…housing has a volume cycle, not a price cycle. Home prices are very sticky downward, and faced with a decline in demand, it is the volume of sales that adjusts not the prices. With the decline in sales volume comes a like decline in jobs in construction, finance and real estate brokerages.

    The point to bear in mind Stephen, is that house construction exists within a massively complex regulatory and environmental planning policy mega-framework. I hate to use that many terms to describe it in English, but that is exactly what it is. The mega-policy framework operates within its own cycles too. By the end of 2006, we had really stretched the then-existing framework beyond its elastic limits. In other words, we were still ‘building’ a type of world, which was designed to fit around a society, economic and political reality that existed back in the late 1980s and early 1990s. We managed to stretched above 15 years+ from that policy framework cycle, within which the operation of house construction has to happen. We did quite well to even get that far I think. The rush now is to create a sustainable policy framework, which can serve us for the next phase of rolling out housing stock. The trouble being, these mega-policy frameworks involve so many parties and require so much time to incubate, they are almost fizzled out, by the time the light turns ‘green’ again, and housing production can resume. We probably have the bones of a good policy framework at this stage, but we still haven’t got a clue, how to go about financing and integrating it, with feasible projects. I guess, this is why about 1.0 billion people on the planet inhabit un-authorised shanty town styles of development, where they don’t even have legal title to the four walls they occupy. My main point being, that housing doesn’t exist so much in the form of bricks and mortar, but moreso in the form of legislative framework. The sub-prime debacle was a legislative framework, which responded to the fact that Kennedy rolls out a massive house building program in the early 1960s, for whites only, in the United States. The subprime lenders were then able to take advantage of the occupants of red-lined areas, which had been avoided by president Kennedy’s building program and legislative policy. BOH.

  5. BOH @ “Recently, I had to review several case studies for cutting edge home design/build projects in North America. I kept on running into different accreditation schemes, standards and metrics used to rate performances of one home versus another.”

    Many or most of these regs are set by the state (not federal/national) and counties/municipalities can also have their own special schemes.

    The USA functions like 50 different countries in many ways. It was designed that way on purpose. There is no single national election administration system, for example. With few exceptions the detailed elections rules are set by the state and the laws are completely different from one state to another–and most Americans don’t even realize this. It’s no surprise that building & energy regs are also completely different from one state to another.

    National trade associations can be good points of contact to get an overview and understand the state & local peculiarities in relation to any particular industry. They might help simplify the complexity, or help you identify states or regions that will be easier to deal with.

  6. @Stephen, I think that is an important topic and the subject for another paper!

    @Celtic Phoenix: yes, you are right. We found that the probability that the Irish market has bottomed out is essentialy zero. We also found some evidence that after 30 quarters or so price declines become self-sustaining. If that is not something people want then we should get the required price adjustment over with as quickly as possible.

  7. @ OAC,

    Many or most of these regs are set by the state (not federal/national) and counties/municipalities can also have their own special schemes.

    Not to mention the plethora of private utility companies that people do business with in North America, from one side of the continent to the other. Many of those utilities (from my limited understanding), would rather not have to produce over capacity of power, and have to handle all of the associated overheads of power distribution lines etc. That is, they would be more than happy to grant you money to install some local, on-site generation capacity through PV panels, hydro or whatever.

    That is an excellent point about the US electoral system btw. I hadn’t thought of it that way, ever before. Regards, BOH.

  8. I would have liked a ‘plain text’ version – the stats can be quite off-putting. Only one point of contention. The US housing market is in an ‘iffy’ situation. It is improbable that there will be a ‘recovery’ for some time (debt levels are too high, and credit uptake is actually declining). SK’s comment about ‘volume’ is very interesting. Makes sense.

    B Peter.

  9. @ BoH:
    “That is, they [private utility companies that people do business with in North America, from one side of the continent to the other] would be more than happy to grant you money to install some local, on-site generation capacity through PV panels, hydro or whatever. ”

    Not necessarily. Different states in USA have completely different policies and regs regarding feed-in tariffs and distributed generation–amount of grants, whether or not there is a feed-in tariff, whether or not they have schemes to help (or require or prevent) you hooking up to the grid if you install renewables.

    There are grid issues there, like there are here (old, failing grid not capable of handling some aspects of distributed generation).

    Consider that virtually *each* US state is much bigger than Ireland, and *each one* functions like an independent country in ways Europeans often don’t realize. Ditto for educational policy, sales tax, etc.–it’s set at state and local levels.

  10. To come back on topic, the regs about lending & mortgages (e.g. recourse vs. non-recourse) are set at the state level, not the national level.

    While it’s useful to analyze the USA as a whole (e.g. its position on the table in the article referenced here by KOR), the situation varies widely from one state to another, in part because state laws regulating mortgages differ.

    OTOH the US bankruptcy law is a federal law.

  11. @KOR
    The way this is getting read by the media and online forums, from what I can see, is that Ireland has two options: (a) a big sudden adjustment, or (b) a prolonged slow adjustment.
    I think many parts of the country are heading for option (b), the way they’ve adjusted the past three years.
    However, in at least one part of the country, Dublin City Centre, we’ve already seen a significant adjustment, down 48% from the peak. Suppose you’d put that into the model, how would that change the probability?
    (Quarter by quarter data available if needed!)

  12. @BOH:
    “That is an excellent point about the US electoral system btw. I hadn’t thought of it that way, ever before”.

    There is *no such thing* as a “US electoral system”, and there never has been.

    Each state has its own state constitution, and sets its own way of choosing or electing the electors who in turn elect the President.

  13. @ OAC,

    The points you are making in relation to the US market, also have a lot of relevance when discussing other aspects of the economy. Types of industries, centres of innovation and what not. Often in the Irish scene, you hear discussion about R&D, venture capital and so forth, as if that existed all across the north American continent. When the reality is, there are only isolated pockets. I watched an episode of the Frontline on PBS a while ago, which featured a capital of the credit card industry in the United States. There was something purely created out of state level policy directives. Ireland became a centre of some kind of financial services activity over the past decade also. That certainly did impact on the housing market here. Stewart Brand makes some very interesting comments about rural de-population. There are schemes to offer people 100 acres of free land in the middle of the US, he says, to encourage people to move there. More than likely, a state level incentive scheme. I can appreciate in that situation, how the competitive landscape for utility companies must vary. BOH.

  14. @ Kevin

    Thanks for the reply. The work makes interesting reading.

    I hope Irelands potential first time buyers take note. The more people fooled on the way down, the slower the fall.

    Also, a lot of what makes us uncompetitve simply boils down to rent prices which are somewhat perportional to house prices.

    Hopefully we’ll see a copper fastening of credit rules. i.e. nail lending for principle residents to 3.5 times a persons salary.

    Wouldn’t it be great if we take this oppopportunity to end the boom bust housing rubbish. Looking back on the last decade, all one can say is ‘What a bloody waste waste of time (and resources)’, and unfortunately not just, a waste of time, no one can say we’re in a better place now than we were in 2000.

    End of rant.

  15. House prices are inversely proportional to the monies in your pockets and the beats of the feets on the streets.

    Competitiveness is proportional to poverty.

  16. @Celtic Phoenix – “Hopefully we’ll see a copper fastening of credit rules. i.e. nail lending for principle residents to 3.5 times a persons salary.”

    The CB is on the case:

    http://www.irishtimes.com/newspaper/ireland/2010/0722/1224275196494.html

    Does anyone know where I can find a link to this CB report?

    This sort of noise is hardly likely to do anything other than continue the decline. I wonder where the bottom is? Just looking in the window of an estate agent a couple of days ago, there is still a lot of property around priced at ‘hopeful’ (i.e. hopefully and idiot comes along) levels.

    It all looks like it’s going to continue to be kind of painful. For banks and home-owners.

  17. Don’t all the historical analyses distinguish themselves from Ireland today in that all of those countries had their own floating currencies ( though was it the case that the Nordic countries had some pegging which went out the window when the crashes came?)

    Isn’t it the case that a property crash is accompanied by an economic contraction? With a floating currency, wouldn’t your currency tend to lose value and inflation tend to rise with imports increasing in price? No chance in Ireland – France and Germany trump any disturbance on the periphery. So we’re going to have low inflation for the forseeable future (and indeed prices have increased by 1% since the peak in Jan 2007 to June 2010).

    The UK in 1989 – 1996 is a case in point – prices dropped 10% from peak to trough in nominal terms (as you say 30% in real terms over 6 or so years) so in nominal terms the crash was less serious for those with mortgage debt and negative equity for example did not become a barrier to recovery.

    Japan’s lost decade(s) would be the exception which was distinguished by its low inflation (deflation) – is that the most likely scenario here? Hope not because in 2010 there are still property prices in Japan that have not recovered to their 1990 values.

  18. @Stephen Kinsella

    “Don’t forget that the P in GDP refers to production, and the appreciation of your home as you sip your chardonnay in your hammock in your back yard is not production and doesn’t count as part of GDP. But jobs in construction, finance and real estate do count. Thus volumes matter for real GDP accounting, not prices.”

    This is exactly the point. Perfectly expressed.

    When talking of housing booms/slumps, we need to distinguish two things:

    (a) booms/slumps in product (ie the number of new houses built and houses traded), which is included in GDP

    (b) booms/slumps in house prices, which is not included in GDP

    All those economists/commentators who say that the economic growth up to 2007 was based on high house prices are talking through their hats. The high house prices (or low house prices now) are not and never were included in real GDP.

    It is generally agreed that an excessive boom in house prices is a bad thing, with very few compensating benefits. A boom in housing product is generally considered a good thing, except for environmental wackos, bringing increases in employment and contribution to GDP.

    It is possible to have one of the above booms/slumps and not the other, or to have both or neither. For example, Ireland had a boom in both housing product and house prices. The UK had a boom in house prices but not in housing product.

    In the years up to 2007, while house prices rose sharply in the UK, at more or less the same rate as in Ireland, the number of new houses built slumped and was much lower than it was 40 years ago, while in Ireland the number of new houses built rose sharply. That’s why homelessness, overcrowding and bad housing conditions have deteriorated in the UK in recent years, while improving in Ireland. Ireland now has the most modern housing stock in the EU, while the UK has the oldest, with consequent social and health effects. One only has to drive through Ireland and the UK to see this.

    Booms/slumps in housing product are primarily driven by demographics. Over the long run, bank lending practices have little effect, while they are the principal effect on booms/slumps in house prices. It could even be argued that a boom in housing product is less likely to lead to a boom in house prices, as supply will be greater. And, it could also be argued that a slump in house prices is more likely to lead to a boom in housing product, as affordability will be greater. Were house prices in Ireland to fall to the levels Morgan Kelly predicts, then every Tom, Dick and Harry in Ireland (both in Republic of Ireland and Northern Ireland) will be able to afford a holiday house on the Atlantic coast.

    So, regarding Ireland’s housing slump, we should not necessarily expect the slump in housing product and the slump in house prices to end and be reversed at the same time. The two should be treated differently, just as Stephen Kinsella says, While there is no sign yet that the slump in house prices is ending, there are signs that the slump in housing product has. These are not definitive and I don’t wish to exaggerate their significance, but in recent quarters the number of new house starts has increased. In Ireland, there are two statistical indicators of new house starts (a) registrations (b) commencements. The latter includes one-off houses, while the former doesn’t. The latest figures for both are:

    registrations:

    2010 Q1: 479
    2010 Q2: 814

    commencements:

    2009 Q4: 1,353
    2010 Q1: 1,696

    As I say, it takes more than one quarter’s figures to be certain that the turning-point has been reached, so I don’t wish to exaggerate the significance of one quarter’s figures, but it is possible that they signify the turning-point in the housing product slump/boom cycle, although not the house prices slump/boom cycle. Also included in the ‘housing product’ that Stephen Kinsella referred to is trading of houses. This, also, seems to have turned the corner. The Irish Times today reports today a survey showing house sales increasing sharply q-on-q. So, it looks very much as though housing activity (ie starts and trading) has been increasing in 2010, although not house prices.

  19. @anyone

    Is there a source in the State which accurately publishes all property transactions so that you could measure activity levels?

    The IBF publishes mortgage volumes which show the following number of new mortgages for dwelling purchase (the IBF represents 95% + of mortgage lending in the State).

    Q1, 2005 21,371
    Q2, 2005 27,833
    Q3, 2005 29,592
    Q4, 2005 31,699
    Q1, 2006 25,888
    Q2, 2006 28,088
    Q3, 2006 29,884
    Q4, 2006 26,930
    Q1, 2007 20,966
    Q2, 2007 22,086
    Q3, 2007 21,878
    Q4, 2007 19,264
    Q1, 2008 13,299
    Q2, 2008 16,694
    Q3, 2008 14,208
    Q4, 2008 9,415
    Q1, 2009 5,508
    Q2, 2009 6,678
    Q3, 2009 6,804
    Q4, 2009 6,107
    Q1, 2010 4,224

    The government issues Stamp Duty returns doesn’t it which continue to show drops in value but doesn’t tell us about volume.

    The CSO takes mortgage information when producing its average dwelling price and doesn’t appear to produce volume figures.

    Of course the Revenue and Land Registry will hold the information. But is volume information on transaction levels published anywhere?

  20. “All those economists/commentators who say that the economic growth up to 2007 was based on high house prices are talking through their hats. The high house prices (or low house prices now) are not and never were included in real GDP.

    It is generally agreed that an excessive boom in house prices is a bad thing, with very few compensating benefits. A boom in housing product is generally considered a good thing, except for environmental wackos, bringing increases in employment and contribution to GDP.”

    Increases in house prices have a big effect on GDP as equity releases increase domestic demand.

    Equity release is of course accompanied by increased private debt. Therefore unsustainablehousing booms are particularly dangerous.

  21. The only way to avoid a future repeat of the property crash is to insist on multiples of salary around 3.5 times and high deposits in order to qualify for a mortgage. Switzerland had 2 pretty horrific crashes, the last of which ended in 2000 and now anyone getting a mortgage has to stump up a sizeable amount in cash as a deposit before the bank will consider granting a mortgage.

    I also think that there is something in the English language as a vector of bad lending practices/property overheating. Is it any coincidence that the UK, the US and Ireland have all had a property crash in the last 2 years?

    Property taxes seem inevitable.

  22. Housing booms are reflected in the GDP.

    The items produced were produced and costed to expected prices that were inflated in the bubble (boom). The costs in a boom increases more than economically justifiable and almost all that went into producing the overpriced house went in at an inflated price: Wages, raw material.

    In essence most economic activity related to the housing boom were inflated and therefore the housing boom affects the GDP.

    I’m not surprised about the increase in activity in registrations and commencements. NAMA was introduced with mumbling of it placing a floor under the property market: What is the effect of a price floor?

    It guarantees a supply as long as the costs of producing are lower. Sunk costs are gone. The cost of construction is probably (surely?) lower now than during the boom so presto: House building into an oversupplied market can be economically justifiable due to the price floor.

  23. @ Jagdip Singh

    Very interesting data on mortgage volumes you published. That data is especially interesting as it indicates that the peak was in Q4 2005.

    The government imagines that our problems began in 2008 when the economy reversed direction. But it is clear that the bubble was letting out air well before that. One can variously date the peak as being:

    Q4 2005 – peak of mortgage volumes approved
    Spring 2006 – peak of private sector credit growth
    Spring 2007 – peak of Irish bank share prices

    The top was in long before our problems were recognised.

  24. @BOH

    Thanks for the link – there is indeed a paucity of some information.

    What I wanted to get at was volume statistics for sale/purchase transactions. Is it really the case that the State doesn’t publish these? Surely when these pick up, we will know the slump in volume terms is coming to an end (what value do planning permissions and completions have when people are possibly spending lots on extensions and home improvements?).

    I note JTO’s reference to the IAVI report today that sales activity had picked up in Q2. However if you look back to their last report for Q1, they said something similar.

    http://www.iavi.ie/NewsDetail.aspx?rowid=260590

    The mortgage statistics from the IBF show that mortgage lending fell off the cliff in Q1 – look at the volume figures which belie the headline.

    http://www.ibf.ie/gns/media-centre/news/10-05-20/Home_Purchasers_Show_Resilience_in_Market_Slowdown.aspx

    There are 800k mortgages in the State on 2m-odd dwellings (350k-odd are “vacant”), so it might be the case that there are lots of transactions without mortgages to offset the decline in mortgage volumes, but somehow the fact that the IAVI report is based on surveying members whose career prospects depend on a recovery make me wary.

    So anyone? Know of any source of transaction volumes in the State?

  25. @JohntheO & Stephen Kinsella
    “When talking of housing booms/slumps, we need to distinguish two things:

    (a) booms/slumps in product (ie the number of new houses built and houses traded), which is included in GDP

    (b) booms/slumps in house prices, which is not included in GDP”

    I must be missing something.It’s just that I thought that RFI mattered too? As in no. new houses x price = RFI… am I wrong?

    Indeed, I though non-residential fixed investment (apartments, retail, offices, factories) contributed a bigger slice of GDP and that the slowdown in completions and lower prices have also had an effect on GDP?

  26. @Jagdip.

    when he asked me the question, I started by saying something like ‘I agree with Morgan Kelly’ — should probably have left it at that!

  27. @Kevin

    ” ‘I agree with Morgan Kelly’ — should probably have left it at that!”

    So true on so many levels! Did they get the gist of what you said right here

    “it would be better for the economy if house prices here rapidly hit the bottom because otherwise the slump could extend for years”

    And if so what are you suggesting is preventing house prices from hitting the bottom? NAMA or banking/government policy perhaps? If so is this graph the sort of thing you have in mind – and if so would you be prepared to take a stab at the changes to the gradients?

    http://docs.google.com/leaf?id=0B1V6jFjykyK6MTQzODcwY2YtNzUwMi00MTg3LWE5ZjEtZjZiOGMxMDU1NjI2&hl=en

  28. @Jagdip,

    Yes, they did get the gist there.

    All I did was state the obvious (at least to me it is obvious), namely that it is better that prices adjust asap rather than have the thing be dragged out — although the fact that price declines can become self-sustaining after 8 years or so was a striking finding in our analysis and a new reason for wanting those price declines to happen asap.

    I didn’t commit myself to saying ‘NAMA is preventing prices from dropping’ or even ‘something is preventing the prices from dropping’ since I don’t have a basis for saying that; but it would be a very bad thing if NAMA had that effect, and if it did then yes, that graph might make sense (although perhaps the ‘with NAMA’ bottoming out happens even further out than you show, and perhaps it doesn’t even succeed in making the trough shallower in the sufficiently long run).

  29. @Kevin,

    many thanks for the clarification. It would be interesting to see whether NAMA’s loyalty to long term economic value and NAMA’s size in the marketplace would lead to a prolonged slump but alas there is only one future and it may be impossible to untangle the various factors at play.

Comments are closed.