So lets start with what I think is an important distinction between Texas and the rest of the United States.
Here are real home prices for both since 1975, scaled so that the real price at the beginning of the Great Recession is one.
Why is this important?
Well, first I think there is broad consensus that the Great Recession was in some way related to the boom and bust in the US housing market. The question is how?
Well, in Texas there wasn’t much of a boom or bust. We can see a small rise in real home prices, against a backdrop of falling real home prices over the decades.
However, it isn’t sharp and it didn’t peak until the recession was almost over. I don’t think its crazy to suggest that home prices in Texas were driven down by the Great Recession but did not drive the Great Recession.
Okay, now lets look at housing starts.
Unfortunately I only have 1-unit starts for Texas and they are not seasonally adjusted.
You can see that the growth in the Texas housing stock was faster than the US on average, but that the fall was almost as large. And, timed around the same point as the fall all around the US. Early, 2005.
So in Texas housing starts were falling even as home prices were still rising.
Lets look also at construction jobs
Here the pattern is closer to that of home prices. Texas peaks later and doesn’t fall as far as the rest of the nation.
Yet, housing starts were falling in Texas. . If anything you can see that construction jobs really start going just in Texas around the same time that housing starts are peaking.
This means people must have been building other things besides homes. I really wish I had construction spending by sector for Texas, but I don’t right at my finger tips. My guess is that we are talking about a combination of oil infrastructure and office buildings but I’d really like to know the breakdown, because that would affect the analysis.
Lets look at total non-farm payrolls in Texas vs US
Again, similar pattern. Latter peak, smaller fall, better recovery.
However, notice that the dates of the recovery are similar as is the shape. Lets zoom in.
Interesting. Now I am going to look at my favorite, employment in non-goods non-government. But I loose the ability to index, so I am going to use to different scales.
First zoomed in
Then zoomed out
Again, I think we are looking at similar recovery dates and similar shaped recoveries, just stronger in Texas
My first cut is to say that we are looking at roughly three events.
One occurred around 2005 and was associated with a precipitous decline in housing starts for Texas and the US as a whole. It was also associated with a pause and eventual decline in real home price appreciation in the US, but not Texas. As well as a pause an eventual decline in construction employment for the US but not Texas
A second event occurred during 2008. This event was associate with a decline in construction employment in Texas and overall employment in both the US and Texas.
A third event occurred in mid 2009, when private non-goods employment in both the US and Texas began to trend upward again.
Now I am using event here loosely. It doesn’t have to be one thing. However, the fact that all of these measures are not occurring within the same phase is interesting. Its not as if Texas is just a shifted version of the US. Some things are shifted, some not. Some things come in together and different things come out together.
If we are all now confused on a deeper level about more detailed things, then I hope I can say this post was useful.