Matt Yglesias, striking that contrarian tone beloved of bloggers (something you’d never find here, of course!), declares that there was no housing boom. Or, more precisely, though there was a boom in house prices, there was no boom in construction.

To make this point, Yglesias uses one on those ubiquitous St. Louis Fed graphs, this one of the history of housing starts since 1970. Sure enough, it sorta supports his point.

But this is only a very partial view. Here’s a fuller one.

First of all, the boom wasn’t just about new houses—there was a lot of renovation and expansion of existing housing stock. That sort of thing, as well as new construction, is included in the category residential fixed investment in the national income accounts. As the graph below shows, this rose strongly in the late 1990s and early 2000s:

But note the blue trendline on the upper graph: housing’s share of the economy, despite cyclical ups and downs, has been in a steady decline since the highs of the early post-World War II years. The late 90s/early 00s boom was a sharp departure from that downtrend. The lower graph shows the relationship of the actual level to that underlying trend—when it’s above the 100% level (dotted horizontal line), it’s above trend. At the peak in 2005, the housing share of GDP was almost 60% above trend, a record by a comfortable margin. It was also above that trendline for a long time—from 1998 to 2007. Most earlier soujourns above trend were less than half that long. It’s since fallen dramatically, to almost 50% below trend, and it’s likely to stay there for some time.

Construction’s share of employment tells a similar story—

—that is, a declining trend from the mid-1950s through the mid-1990s, followed by a long spike, and then collapse. At the peak in 2006, construction’s share of total employment came close to matching its all-time high, set in 1956, during the post-World War II building boom.

And finally, this boom came despite a low level of household formation among those aged 25–44. The boom in younger household formation in the 1950s and 1960s drove that building boom. But this time around, the building wasn’t to accommodate a population bulge, but to satisfy a lust for more space, often far from town. Between 1973 and 2007, the size of the average house increased by almost 50%, while average household size fell by 15%.

So yes, there was a building boom. Between 2001 and housing’s 2006 peak, residential investment accounted for 12% of GDP growth, twice its share of the economy. And it’s led the way down, too. And it’s helping to keep us here.