Prof. Morris Davis Outlines Land Price Research for Beijing Visitors

---

In mid-June, a delegation from the Ministry of Land and Resources from the People’s Republic of China in Beijing, along with Professor Lou Jianbo of the Center for Real Estate Law at Beijing University, visited the UW School of Business to learn more about our activities and to discuss areas of potential cooperation.

The visit was the latest in a series of connections between the two universities. Last fall, senior lecturer Rod Matthews made a presentation to the Ministry in which he discussed a recent article by Professor Steve Malpezzi on the price issues of real estate. During the spring semester, he strengthened the connection with weekly internet video sessions among students at Beijing University and UW.

As part of the June visit, Morris Davis, former Federal Reserve economist and now UW-Madison professor of real estate and urban land economics, summarized his current research in a presentation for the Chinese visitors. Full treatment of his research can be found here.

“My work describes changes to the price and value of land in the United
States from 1950 to 2006; proposes rational explanations for these changes; and suggests three policy implications.

“I start by defining land as an input into production. I argue that land, broadly speaking, has three productive uses: Farm land produces farm goods; residential land, when combined with a structure, produces shelter and other services; and commercial land, when combined with a structure, provides shelter for inventories (warehousing) and serves as a focal point for exchange (retail). Land has value because it has productive uses, and because its quantity is limited.

“The U.S. government collects detailed statistics on the price and value of farm land. Summarizing these data for the 1950-2006 period, the inflation-adjusted price-per-acre of farm land has increased, on average, by 1.9 percent per year; the number of acres used for farm production has declined by 0.5 percent per year; and, the value of farm land as a percent of GDP has gradually declined from about 25 percent in 1950 ($78 billion) to about 13 percent in 2006 ($1,772 billion). Based on these data, I conclude that although the price and value of farm land has increased over time, it has not increased at an especially remarkable rate. This is likely consistent with the fact that the inflation-adjusted price of food, although volatile, has remained constant on average since 1950.

“I then note that the U.S. government does not collect data on either
commercial land or residential land. As far as I know, we have no good data on the price or value of land used for commercial purposes. For residential land, I have developed, with coauthors, a procedure to measure changes to its price and value.

[See Davis and Heathcote (2006), "The Price and Quantity of Residential Land in the United States" and Davis and Palumbo (2006), "The Price of Residential Land in Large U.S. Cities."]

“This measurement is not straightforward, since houses are usually bought and sold where people want to live, and not land. Another way of stating this is to note that the "best" residential land - that is, land with good schools and a reasonable commute to a downtown - already usually comes bundled in a sale with a residential structure.

“After describing the procedure my coauthors and I use to measure the price and value of residential land, we describe our results for the aggregate United States. The first result is that land's share of home value has been gradually increasing over the post-war period, from about 10 percent in 1950 to about 45 percent in 2006. That is, if an average house sold in 1950 for $10,000, about $1,000 would be attributable to the value of land; if an average house sold in 2006 for $200,000, about $90,000 would be attributable to the value of land. The second result is that, after adjusting for consumer-price inflation, the price of residential land has increased by nearly 20 times since 1950: The average growth rate of the real price of
residential is 5.1 percent per year. Finally, we document that the value of residential land as a percent of GDP has been steadily increasing since 1950, from 10 percent in 1950 ($31 billion) to over 80 percent in 2006 ($11,370 billion).

“The table below summarizes the key findings:

Value of Land, In $billions | Farm | Residential

1950	$78	$31
1980	$763	$1,052
2006	$1,772	$11,370

“In 2002, 940 million acres were used in farms and 60 million acres were used for residential and commercial. Yet, in 2006, the value of all residential land was 6.4 times the value of all farm land.

“The rest of my talk focused on explanations for why the price of residential land has increased so rapidly, and specifies reasons that policy-makers should care.

“Why has the price of residential land increased so rapidly? We know that the price of a good rises, in general, if it becomes more scarce relative to the typical basket of goods that people consume. So, the question that I ask is: Why has residential land become scarce?

“To answer this question, I appeal to a standard model of urban economics. In this standard model, development of new housing, over time, occurs further and further away from city centers. The land on this new housing has associated with it a longer commute to the downtown area than previously built-on land, and thus is less "desirable" and, by extension, of lower quality.

‘As cities have increased in population from 1950-2006, the new developments have, in general, occurred further and further away from the city centers. The price of existing homes on the more desirable land with the shorter commutes was bid-up by the new residents desiring shorter commutes.

“The above is a story about the demand for land increasing. The question is: Why didn't supply of good land increase to bring land prices back down? In the context of an urban-economics model, an increase in supply occurs when the commute time associated with any given parcel is reduced. Such a reduction could be the outcome of investments in commuting infrastructure, or improvements in transportation technology. I note that it could be the case that supply of land has increased. However, in the past 50 years, changes to supply has not kept pace with the surge in demand, and the real
price of land has marched steadily upwards.

“I conclude with what I think are the three major policy implications of my research on the price of residential land.

“The first is that it is not obvious the U.S. will take steps to reduce the price of land: Homeowners vote, and any steps that may increase the supply of available land (improvements in infrastructure, building permits, etc.) will likely reduce the price of land and thus homeowners' wealth. It would be an especially altruistic outcome if homeowners were to vote to reduce their own wealth. Professor Ortalo-Magne has a paper explaining this result in the context of a formal model.
[See Ortalo-Magne and Prat (2006), "The Political Economy of Housing Supply."]

“Second, as land's share of home value continues to rise, house prices will behave more and more like land prices. Since land prices have big swings up and down, they are "volatile" and thus land is a "risky" asset. This implies that housing will be viewed by future generations as a riskier asset than in previous generations. Households do not like risk, and so they will take steps -- perhaps right now -- to manage this risk. Another byproduct of the increase in land's share of housing is that property tax revenues will become more volatile and thus local governments will have to be willing to more actively manage property tax rates to keep tax revenues smooth.

“The final point is to mention that the increase in the price of land everywhere may have implications for migration into and out of big and small cities. If people are more productive in big cities, as is commonly assumed, then productivity is maximized when most people live in the biggest cities. Continued increases to the price and value of land everywhere may force a good number of households away from big cities (where land is most expensive) and towards smaller cities (where land is less expensive), with possible consequences for aggregate productivity and GDP.”