Monday, December 13, 2004
Is Madison In the House Price Bubble?
by Tom Bozzo
So how does Madison stack up to the big kids? HPI only shows price appreciation within regions, so it tells how price increases here are keeping up with increases elsewhere. The short answer is that we're falling behind highly inflated areas:
(That is, Boston and Washington were more expensive than Madison in early 1995 when these indexes meet, and their HPIs have been growing faster.)
Madison ranks 91st out of 245 ranked MSAs (ranking requires a minimum number of eligible transactions in the area), which puts it below the national average as the most rapidly appreciating MSAs are relatively large.
Boston and Washington both had seen rapid house price inflation in the last bubble-like period and suffered what were considered long hangovers. What's remarkable about Madison is the very consistent and mostly moderate price growth, as seen in the price changes here:
In Boston, where the rate of increase in the '80s was as rapid as some of the suspected bubble areas now, there were a couple of quarters in the "crash" where HPI was down 10 percent over the previous year. As a result, the Boston area didn't regain its 1988-89 HPI peak until 1997. Washington didn't see negative HPI changes of that magnitude, but rather saw an extended period of near-zero average appreciation. I wouldn't want to be buying into the Washington-area housing market now, even with home equity from another well-inflated market to spend. (Sorry, Ezra and Kristin.)
Still, this does show some limitations of the MSA-level data. Within Madison's average, there are locations — mainly in central areas of Madison proper — that have seen very rapid appreciation and have even shown some signs of topping out, such as lengthening times-on-market and even mild deflation for a few properties.
Another methodological issue is that HPI attempts to control for housing quality by using data on repeat transactions. However, the description of the HPI methodology indicates that HPI simply assumes constant quality for any given house, so renovations are a potential 'confounding' factor with pure appreciation. For instance, our old place appreciated by about 9.5 percent per year based on the 2004 and 2000 contract prices, but by a still respectable but considerably lower 6.5-7.5 percent per year if the value of some improvements carried out in the interim were subtracted off the gross appreciation. This would help the don't-panic set a little.
So the answer to the titular question is (drum roll) — maybe. Yes, the economist's classic answer. At least, the history of other highly inflated markets would suggest that we may not have inflated enough to see an extended period of house price deflation in the absence of an especially vicious turn of the housing market fundamentals. (But beware a dollar-interest rate crisis!)
For some more discussion of analytical limitations of efforts to evaluate whether we're in a bubble, see Angry Bear here.
The latest House Price Index release, noted here on Friday, shows national average house prices at a rolling boil. It also offers regional detail down to the Metropolitan Statistical Area (MSA) level — much of it downloadable, a boon to armchair statisticians.
So how does Madison stack up to the big kids? HPI only shows price appreciation within regions, so it tells how price increases here are keeping up with increases elsewhere. The short answer is that we're falling behind highly inflated areas:
(That is, Boston and Washington were more expensive than Madison in early 1995 when these indexes meet, and their HPIs have been growing faster.)
Madison ranks 91st out of 245 ranked MSAs (ranking requires a minimum number of eligible transactions in the area), which puts it below the national average as the most rapidly appreciating MSAs are relatively large.
Boston and Washington both had seen rapid house price inflation in the last bubble-like period and suffered what were considered long hangovers. What's remarkable about Madison is the very consistent and mostly moderate price growth, as seen in the price changes here:
In Boston, where the rate of increase in the '80s was as rapid as some of the suspected bubble areas now, there were a couple of quarters in the "crash" where HPI was down 10 percent over the previous year. As a result, the Boston area didn't regain its 1988-89 HPI peak until 1997. Washington didn't see negative HPI changes of that magnitude, but rather saw an extended period of near-zero average appreciation. I wouldn't want to be buying into the Washington-area housing market now, even with home equity from another well-inflated market to spend. (Sorry, Ezra and Kristin.)
Still, this does show some limitations of the MSA-level data. Within Madison's average, there are locations — mainly in central areas of Madison proper — that have seen very rapid appreciation and have even shown some signs of topping out, such as lengthening times-on-market and even mild deflation for a few properties.
Another methodological issue is that HPI attempts to control for housing quality by using data on repeat transactions. However, the description of the HPI methodology indicates that HPI simply assumes constant quality for any given house, so renovations are a potential 'confounding' factor with pure appreciation. For instance, our old place appreciated by about 9.5 percent per year based on the 2004 and 2000 contract prices, but by a still respectable but considerably lower 6.5-7.5 percent per year if the value of some improvements carried out in the interim were subtracted off the gross appreciation. This would help the don't-panic set a little.
So the answer to the titular question is (drum roll) — maybe. Yes, the economist's classic answer. At least, the history of other highly inflated markets would suggest that we may not have inflated enough to see an extended period of house price deflation in the absence of an especially vicious turn of the housing market fundamentals. (But beware a dollar-interest rate crisis!)
For some more discussion of analytical limitations of efforts to evaluate whether we're in a bubble, see Angry Bear here.
