Comments: US housing - not dead yet

Don't get too excited though. The figures are seasonally adjusted, and are still down 3.6% year-on-year. The NAR has been caught out in the past playing the little trick of revising last month's numbers in order to make this months seem better than they really are (although they don't seem to have done that too much this time). Their sample also differs from month to month, suspicions about what this could mean are detailed on this very interesting blog:

The rest of that is well worth a read as well.

Prices in the US _are_ falling though, and have been for the past several months. Have a look at the S&P home price index:

I would take press releases and statistics from the National Assoc. of Realtors with a pinch of salt. It's a bit like asking the butcher how the meat industry is doing!

Posted by Minh at March 23, 2007 05:45 PM

Looks like the subprime rot has spread over here too, with Kensington Mortgages in trouble.

Prices are not tumbling - they're in freefall.

Posted by citydrone at March 23, 2007 07:13 PM

Prices aren't tumbling, or in freefall, on either side of the Atlantic and Kensington is a very small operator. I'm not sure I'd regard Robert Shiller as any more independent than the National Association of Realtors, but on the other side of the debate. If I'm reading his index correctly, it shows that house-price inflation has fallen sharply but it doesn't show that prices have fallen much - they're down in some areas up in others. This is an interesting test for the US housing market. As I understand it, and I should say I find this quite hard to believe, prices have never fallen on a calendar year basis.

Posted by David Smith at March 23, 2007 09:40 PM


Well, it was about Feb, that it was starting to look like the Fed had finished raising rates, and the next move would be downwards.

The same thing happened in the UK, in Spring 2005, when it started to look as though the MPC had finished raising rates and the next move would be down.

Basically, wehn interest rates look to have finished rising, people start come out and buy

Posted by kingofnowhere at March 24, 2007 10:03 AM

David, if you scroll across to the far right hand side of the spreadsheet there are two national composite indices which show prices are falling.

Posted by Minh at March 24, 2007 02:07 PM

Well the Economist has been bearish on property since at least mid 2005 and has always argued its case in a sober and even handed fashion.

I am coming around to the view that I believe David and many others take which is that house prices are sticky going down and falls in absolute terms generally only happen if interest rates rise sharply or unemployment heads north - or perhaps both at once.

However the perception factor is important especially if prices rise only slowly or not at all. Rental returns are too low to make sense at the moment if landlords don't make gains from capital appreciation. Otherwise many would be better off selling and investing the money elsewhere - emerging markets or even a solid offshore interest account.

Meanwhile income multiples needed to buy are rising and that eventually limits the room for maneuver of the BoE and other central banks small rate rises will have an amplified effect.

Who knows where the market will go - but one key economic model worth studying is that of the crowded theatre in which someone stands up and shouts fire. Markets can turn suddenly and its often not fundamentals but a decisive shift in sentiment.

We live in interesting times.....

Posted by Jonathan at March 24, 2007 02:28 PM

Re: the Case-Shiller Standard & Poors index. I've had a flick through the methodology and can see no mention of seasonal adjustment. Is it?

Posted by David Smith at March 24, 2007 03:37 PM

Can't find any mention of seasonal adjustment either, but looking at the data I would say so. If you leave aside the last 5 months, the last month-on-month fall in the index was in December 2001 (a fall of 0.08% in fact!). The last 5 months have shown falls of 0.16%, 0.18%, 0.21%, 0.41% and most recently 0.71% in December 2006. So it looks like it is seasonally adjusted, and even if not, the current falls are very much unusual even for this time of year. On current form it should go year-on-year negative within the next 2-3 months.

Posted by Minh at March 24, 2007 04:49 PM

David, it seems like this is all unfolding rather more quickly than expected! Sales of new homes in the US today have fallen to their lowest level since June 2000. The markets have reacted pretty unfavourably to this news. US housing might not be "dead yet", but it seems to be getting "deader" by the day... What's your view of this news?

Posted by Minh at March 26, 2007 05:10 PM

Yes, these were weak numbers, though they are known for their volatility. Housebuilders are carrying more stocks of unsold houses than they want to, and this will act as a drag on the economy.

Posted by David Smith at March 26, 2007 07:10 PM

david smith's musings on U.S. house prices are about as accurate as
his musings on oil prices.

Posted by critic at March 26, 2007 08:24 PM

The US housing market could yet bounce back - there were figures suggesting that lower prices are tempting more buyers into the market - and that could staunch the bleeding. But if prices slip again in absolute terms that's not going to encourage bargain hunters to buy, only to wait.

But at some point the fall in new builds is going to have an impact on the wider economy. Construction companies won't want to keep idle workers on their books indefinitely. Sales of white goods and furnishings may suffer if people don't move houses. Equity withdrawal may become less fashionable and consumer spending could be hit as the wealth effect evaporates.

It may be a case of whether the housing market picks up and confidence is restored before the knock ons kick in - because if that happens we may see a positive feedback loop. That could be uncomfortable.

Posted by Jonathan at March 27, 2007 12:31 AM

"Critic" should try to get out more, or what's on the screen. I said the figures for existing home sales were puzzling. The numbers for new home sales were more in line with housing market sentiment.

As for oil, it has been boosted by the tension over Iran's seizure of UK personnel, which we can hope will ease, and by the fact that OPEC is showing some unusual discipline in cutting back production. The more that housing slows the US economy, the more there will be downward pressure on oil prices but events in the Middle East will always threaten new spikes in prices.

Posted by David Smith at March 27, 2007 10:06 AM

David, just to let you know the case-shiller index has been updated today for Jan 07 and has now gone YoY negative.

Posted by Minh at March 27, 2007 03:52 PM

Not too surprised by that, given that most of the other US house price measures are modestly negative year-on-year.

Posted by David Smith at March 27, 2007 05:44 PM

But you just said above that "As I understand it, and I should say I find this quite hard to believe, prices have never fallen on a calendar year basis."!

Posted by Minh at March 27, 2007 09:08 PM

Yes, but what is meant by economists as a calendar year basis is the average price this year, compared with last year. So for example, average house prices in 2006 were about 1% higher than in 2005. This was the reference I saw, from a piece written last July:

'HOUSE prices are set to drop in the US for the first time on record, US investment bank Goldman Sachs warned this weekend.

Prices in several segments of the market have already started to fall, and the overall market will move into the red even in nominal terms next year, fuelling fears that this will trigger a downturn in consumer spending and hit an already slowing US economy.

Jan Hatzius, economist at Goldman Sachs, said: “The risk is rising that nominal US home prices may be headed for an outright decline in 2007. It would be the first decline in national home prices ever recorded, at least in nominal terms.”

- end of excerpt

....... In fact, checking the data, it seems to me there was a calendar year fall in 1991 ..... and there must have been several calendar year falls in the 1930s

Posted by David Smith at March 27, 2007 10:18 PM

Ah I see, so the average price for the whole year has to be less than the average price for the whole previous year. Well I'm pretty certain 2007 will be, but we'll wait and see. I am guessing we will have a good idea by August or so if this is likely to be the case.

Posted by Minh at March 27, 2007 10:23 PM

I am not sure if I am making sense here...
What industry commentators/experts seem to be most worried about is the building up, in the US, of a large inventory of unsold/repossesse/unfinished houses (in the order of 6 to 8 months worth of sales).
To clear that backlog it was estimated that REAL prices should go back to the level of 2000. If that happens all of a sudden (a crash) or slowly (inflation catches up) that is difficult to say and will be partly determined by decisions of developers/banks. However, the likelyhood of a crash in the US (vs. inflation catch up) is probably mitigated by the low proportion of speculative demand.

Posted by Michele at April 22, 2007 03:33 PM
Post a comment

Remember personal info?