Matt Yglesias is still trying to figure out the late housing bubble. His latest approach is to separate structures and land, which leads him to this conclusion:
I think it makes more sense to restrict the idea of a “bubble” to speculative asset like land (or stocks or gold or whatever) rather than to the actual building. A building boom may be (indeed probably is) in some sense “unsustainable” but when the boom collapses it’s not like an asset price bubble that leaves nothing in its wake but debt. A boom in structure building leaves you with extra structures. Whether or not this is the most useful thing to have on hand is open to debate, but it’s very much the opposite of “paper wealth” that vanishes when the boom fades. The buildings are still there, and still as useful (or useless) as they ever were.
This is not entirely wrong, but it’s not entirely right either.
While the major part of the movement in real estate value is accounted for by land, not structures, the contribution of building itself to the mid-decade expansion was prodigious. From the end of the recession in the fourth quarter of 2001 through the peak (or something close to it) of the bubble four years later, residential investment accounted for 13% of GDP growth, three times its share of GDP at the outset. Ubiquitous memes to the contrary, consumption contributed slightly less than its share (especially durable goods—so much for that boom in flat-screen TVs), as did nonresidential investment (from office buildings to capital equipment). Military spending contributed almost twice its share, but other categories of government spending were mostly in line with their averages.
In other words, residential investment—meaning the building of new houses and the renovation of old ones—was by far the leading sector in the familiar national income equation (income = consumption + investment + government + exports – imports). That’s not even counting spillovers—the boost to consumption provided by all those contractors, the demand for raw materials and equipment, and the rest. And a major reason people built and renovated so much housing is because prices were rising and were expected to continue doing so until the end of time.
Just as optimism about prices led to rampant overbuilding, now pessimism about prices is suppressing building. Residential investment is now about half its long-term average, and its failure to recover is a major factor in the broad economy’s failure to recover.
And yes, in theory all those new houses could be providing some use value—there are more than a few homeless and underhoused people in the USA who would happily move in tomorrow—but as long as people lack the money to buy or rent them (and banks are unwilling to lend the money to buy them), then they’re as good as useless. In fact, as long as they’re an overhang on the market, causing people to pull back from buying houses for fear that excess inventory will depress prices for years to come, then they’re worse than useless. Because under this delightful capitalist thing, it’s money that matters, as the great political economist Randy Newman once put it. Physical use values often take a distant back seat to the monetary imperative.