In its April/May 2008 publication "Current Issues in Economics and Finance: Second District Highlights," the Federal Reserve Bank of New York presents a study by Andrew Haughwout, James Orr, and David Bedoll on "The Price of Land in the New York Metropolitan Area."

The market value of land, apart from all capital goods (improvements) attached to it, can be measured when vacant land is sold or when a property is sold and the building is demolished. The article begins with an example of demolition: in 2000, a 3.4 acre of land 1.3 miles from the Empire State Building for $345 million, $2300 per square foot. The price of land can also be valued by estimating the value of the building and subtracting that from the total property value.

When real estate prices are rising rapidly, it is really the land value that rises. The price of residential real estate in the New York metropolitan area rose by 160 percent from 1997 to 2007. This price appreciation data excludes the value of better house quality, thus it shows a large increase in the price of land.

The researchers in this article used a data set of estate prices to measure the land values. They separated out over 6000 sales of vacant lots and purchases involving the removal of the existing structures. The authors found that “the price of an acre of raw land near the Empire State Building rose sharply between 1999 and 2006 and was more than $90 million in mid-2006.”

The study sought to correlate the value of land with the distance gradient from the Empire State Building, which they used for the city center. As one would expect, land values drop as the lots become more distant from the center, though with a wide variance.

The study also has a table of average land price in the area from 1999 to 2006. In 1999, the price per square foot was $46.65. It was $88.94 in 2001, dropping to $71.95 in 2002, but then zooming up to 103.71 in 2003 as interest rates fell, and with the real estate boom from 2004 to 2006, the values are $150.63, $248.30, and $366.08.

The authors note that the market value of land depends on the benefits of the neighborhood as well as anticipated changes. Their statistical “regression” analysis controls for the characteristics of the property, such as the past development of the lot, thus capturing the pure increases in land value. In their model, the price of land is the natural logarithm of the price paid per square foot for real estate of a particular type (commercial, industrial, residential) in a particular time period. The adjusted “R squared” is .76, meaning that the model explains about 3/4 of the reason for the prices.

Land that was subject to foreclosure sold at a discount. The county the lot is in also contributes to the pricing differences, since the counties have differing regulations and taxes. The strongly declining distance gradient from the city center is consistent with standard location theory. The high price of Midtown Manhattan reflects its role as the center of commerce not just for New York City but as a leading global center of commerce.

The study also documents the sharp rise of land values, more than doubling from 1999 to 2006, due to low interest rates and the real estate boom. Commercial real estate lagged behind residential real estate, as is usual in a speculative real estate boom. The authors says that the rise in land values reflected the “strength of the area’s economy” and also “a rise in the perceived value of owning vacant parcels as potential building sites to meet future property demands.”

The latter is a restrained scholarly way of saying that land speculators were betting on ever increasing site values. The authors do not attempt to analyze the consequences of this rapid raise of land values especially in Manhattan, but it is evident that the rise has been much greater than the economic value of the neighborhood benefits. At the peak of a land boom, speculators buy because they expect other speculators to buy higher. Eventually, properties become priced too high for actual use, and prices stop rising, and then the speculators seek to sell, and the real estate boom crashes. This has happened first with residential real estate and we are also witnessing commercial real estate crumbling as it too gets squeezed by the credit crunch and excessively high land prices.

A great bonus from this study is the evidence that land values can be calculated apart from the value of buildings. It is common for economists who have never talked to a live real estate appraiser to arm-chair a conjecture that land values cannot meaningfully be separated from the value of buildings. They can be, they are, and this study shows that even economists can do it.