The Union-Tribune’s latest monthly housing data wrapup features a parade of industry types saying that now is the time to buy. A few of them acknowledge the inventory glut, but not a single one happens to note that homes are still far too expensive in comparison to rents and incomes.
This continued overvaluation is the root of the problem with our local housing market, and now that home prices are no longer being artificially supported by speculative enthusiasm and overly loose lending, there’s little reason to expect them to stop falling until they are back in line with the fundamentals. Yet it would seem that we have people lining up to tell the public that it’s a great time to buy a home.
Moving on, let’s take a look at something else being ignored by the housing cheerleaders: the continued sky-high foreclosure rate.
Would love to understand the
Would love to understand the reasons behind the volatility.
Besides a trailing or rolling average, is there a good way to regress the data points to the mean?
Would love to understand the
Would love to understand the reasons behind the volatility.
Virtually any observed statistic fluctuates. There does not have to be a deterministic reason behind short-term (e.g. month-to-month) changes in observed numbers of events. Behind each NOD and NOT there is a series of events ranging from when people purchased, the prevailing interest rates at that time, specific loan terms, people’s personal financial positions, divorces, jobs losses, people’s decisions when to stop making payments, banks deciding when to pull the trigger, and a host of other events which in aggregate produce random fluctuations.
While it is true that longer term trends in rates, sales, prices, and loan types can be used to explain the long term trends, the month-to-month is pretty random.
Speaking of
Speaking of foreclosures…
This foreclosed house is down at least 50% from peak prices (92057, Oceanside):
http://www.realtor.com/search/listingdetail.aspx?zp=92057&ml=3&bd=4&bth=4&typ=1&sid=aedbecd56d78410dba199e3906f50065&sdir=1&sby=2&lid=1094215290&lsn=1&srcnt=355#Detail
BTW, could the slow-down in foreclosures in November 2007 have anything to do with the fires at the end of October?
While it is true that longer
While it is true that longer term trends in rates, sales, prices, and loan types can be used to explain the long term trends, the month-to-month is pretty random.
Yes, but – the sample size is pretty large in comparison to the month-to-month fluctuations. One would think that month-month fluctuations of 20% would be seen in sample sizes of 10 or 20, not 2,000. Not to say that the drivers could be factored out even if one could identify them, but it would help us understand whether the fluctuations are attributed to fundamentals, or transient noise, like the fires.
Just thinking out loud. Not arguing about the fundamentals one way or the other.
Fearful – I agree that
Fearful – I agree that large short-term movements may sometimes be explained, provided they deviate significantly form the overall monthly noise. One example is the bump in foreclosures in early 2002. Perhaps that event was due to capitulation induced by the slowing economy, coupled with the events of 9/11. We don’t know whether the current volatility is significant (in statistical terms) until we can assess the level of noise around it. More samples does not always mean a reduction in noise. That conclusion is based on the assumption that the samples are independent.
Right you are.
In any
Right you are.
In any event, it is worth remembering that this graph is already population-adjusted – and still it dwarfs what we saw in the early 1990’s.
I stick by my previous analysis of the Case-Shiller inflation adjusted data, indicating that the top third of homes will drop a further 50% on a nominal basis, assuming inflation at 3% a year, by 2012, to get us back to 1997 real prices. Put that in your pipe and smoke it.