Someone posted that chart in another thread several days ago.
Anybody who uses that information to make a buying decision for a home in SD/Orange County is going to be in for a rude awakening over the next 3-5 years, aren't they?
Read the whole report, the noise is so high in their signal that they have a 30% error margin. They cant tell the difference between a 575k and a 425k house. 15% on either side of 500k and it is still a fairly valued. Imagine if things fell another 29%. Most people would call that a disaster, they would call it still fairly valued.
Submitted by one_muggle on March 10, 2008 - 1:47pm.
I wouldn't discount these guys out of hand.
These Global Insights reports have been pretty spot-on, at least for the last 6 years that I have been reading them. I am not saying that you should go buy a house because of their report, but before you go attacking the messenger, you should at least consider their analysis.
Up through 2001 they had LA county as slightly undervalued, which I found hard to believe at the time, but it seems right in retrospect.
By 2003 LA county was considered overvalued, and by 2004 it was grossly overvalued (>36%).
They state up front that their over/under valuations are not predictions of future housing prices, but consider overvalued areas to have positive price pressure, and vice-versa. Given the rapid housing price drop, their models might be overcompensating--they are based on several decades of inputs. A rapid disruption is likely to cause short-term ringing.
Also, in the description of their analysis (at least back when I last read it in detail in ~2003) they do not predict future economics either, rather they look at how current prices compare with current economic conditions--but they do take into account historic economic relationships since price to rent ratios are very different from region to region, as is mortgage-to-income.
It it likely that SD will get an "undervalued" rating on this report sometime in the next couple years.
Anyway, I would take this as one more data point, but I think accusing them of being biased a bit unfair, given that they had most metro areas rated as overvalued for much if the last couple years.
There's a reason Rich's trendline analyses start back in 1976 and why they're tied to the historical trends for wages and population as well as the prices themselves. The problem with looking at the pricing and including the recent spike is that it requires assuming that spike was a typical variation rather than a one-time gross distortion.
That's why it skews the "should-be" indicator up a lot farther than it should. If the current pricing is within 1% of what it should be, that means that after excluding the $1,000,000+ luxury condos with the waterfront and penthouse locations, the average sales price of $328,000 for all remaining condos countywide is justified. The reality is that with a 5% downpayment it still takes a $102,000 household income to support that average, and that's about $30,000/year higher than where wages really are.
One problem I see is as alluded to above it takes economic factors as relatively constant.
That 63,000 drop in payrolls last month will have just a minor affect on this study. Other incomes are dropping, and we haven't even seen the next shoe drop in loan resets.
I believe the tightening of credit is going to have a major effect on the price of houses in the near future and this study doesn't account for it. Of course, the study can't account for what the credit markets will do, but I believe that is a major weakness of the study.
Someone posted that chart in another thread several days ago.
Anybody who uses that information to make a buying decision for a home in SD/Orange County is going to be in for a rude awakening over the next 3-5 years, aren't they?
Maybe they meant 0.4x overvalued as opposed to 0.4% overvalued.
If they rely on that kind of analysis they might make a good short, but it looks like you could have made some money already http://phx.corporate-ir.net/phoenix.zhtm...
Their numbers imply that San Diego "fundamentals" improved by at least 25% between Q4/03 and Q4/07.
Forgetting to factor out bubble money in incomes, perhaps?
Read the whole report, the noise is so high in their signal that they have a 30% error margin. They cant tell the difference between a 575k and a 425k house. 15% on either side of 500k and it is still a fairly valued. Imagine if things fell another 29%. Most people would call that a disaster, they would call it still fairly valued.
So the mortgage salesman says it's a good time to buy based on house prices, Income, and population density? Is that what I'm reading here?
I wouldn't discount these guys out of hand.
These Global Insights reports have been pretty spot-on, at least for the last 6 years that I have been reading them. I am not saying that you should go buy a house because of their report, but before you go attacking the messenger, you should at least consider their analysis.
Up through 2001 they had LA county as slightly undervalued, which I found hard to believe at the time, but it seems right in retrospect.
By 2003 LA county was considered overvalued, and by 2004 it was grossly overvalued (>36%).
They state up front that their over/under valuations are not predictions of future housing prices, but consider overvalued areas to have positive price pressure, and vice-versa. Given the rapid housing price drop, their models might be overcompensating--they are based on several decades of inputs. A rapid disruption is likely to cause short-term ringing.
Also, in the description of their analysis (at least back when I last read it in detail in ~2003) they do not predict future economics either, rather they look at how current prices compare with current economic conditions--but they do take into account historic economic relationships since price to rent ratios are very different from region to region, as is mortgage-to-income.
It it likely that SD will get an "undervalued" rating on this report sometime in the next couple years.
Anyway, I would take this as one more data point, but I think accusing them of being biased a bit unfair, given that they had most metro areas rated as overvalued for much if the last couple years.
1998Q1 San Diego was listed as being 27% undervalued and 2005Q3 as being 37% overvalued.
Clearly, these things overshoot in either direction.
There's a reason Rich's trendline analyses start back in 1976 and why they're tied to the historical trends for wages and population as well as the prices themselves. The problem with looking at the pricing and including the recent spike is that it requires assuming that spike was a typical variation rather than a one-time gross distortion.
That's why it skews the "should-be" indicator up a lot farther than it should. If the current pricing is within 1% of what it should be, that means that after excluding the $1,000,000+ luxury condos with the waterfront and penthouse locations, the average sales price of $328,000 for all remaining condos countywide is justified. The reality is that with a 5% downpayment it still takes a $102,000 household income to support that average, and that's about $30,000/year higher than where wages really are.
One problem I see is as alluded to above it takes economic factors as relatively constant.
That 63,000 drop in payrolls last month will have just a minor affect on this study. Other incomes are dropping, and we haven't even seen the next shoe drop in loan resets.
Nice try though.
I believe the tightening of credit is going to have a major effect on the price of houses in the near future and this study doesn't account for it. Of course, the study can't account for what the credit markets will do, but I believe that is a major weakness of the study.