"The Government Needs Land Price Statistics"
Sarah Laskow has an important post on the difficult task of putting a value on wilderness land. In a related vein, it’s actually surprisingly difficult to put a value on developed land. When you buy a house or a shopping mall or an office building you’re buying, obviously, both a structure and some land. It’s like buying a boat and also buying mooring at a marina. But the transactions happen simultaneously. So how much of the price is the price of the land and how much is the price of the structure?
My curiosity on the subject got me pointed to “Commercial and Residential Land Prices Across the United States” by Stephen D. Oliner, Joseph Nichols, and Michael R. Mulhall of the Federal Reserve Board’s Research & Statistics division. A big part of what I’d like you to take away from this is that the question “what’s the price of land in America” is something the Fed is publishing research papers on (this is from February 2010) not something the Fed’s economists are having interns look up in a database. But it turns out that through the use of hedonic regressions they were able to estimate the prices for 23 metro areas. The broadest conclusion the research gives you is that when you do the decomposition, the implicit fluctuations in the price of land are much bigger than the implicit fluctuations in the price of structures. That makes sense since it doesn’t really make sense to be speculating in the value of structures, but land speculation is a time-honored passtime to you can get herd behavior and bubbles just like with any other commodity. At the same time, the main chunk of the employment impact of the real estate market has to do with structures. People are employed building and upgrading structures, not building land.
So for a few different reasons it seems to me that it would be very useful to policymakers (to say nothing of pundits) for there to be a more deliberate effort for the government to collect systematic data on this issue. Talk about “housing prices” is ambiguous between a few different options.