|
Article Excerpt
Classical economists believed that land rent as a share of total income would increase with economic growth. This belief was important to many 19th and 20th century critiques of capitalism. Land rents as a share of national income apparently declined for most of the 20th century, but increased during the 1990s. In this paper, we develop a model of the classical theory of land rent that allows rent to increase or decrease as a share of national income, depending on several parameters. It seems likely that the long decline in land rent is over and that land rent in the future will slowly increase as a share of national income.
**********
The ordinary progress of a society which increases in wealth, is at all times to augment the incomes of landlords ... independently of any trouble or outlay incurred by themselves. John Stuart Mill Principles of Political Economy, 1872
Real estate has declined in importance over the past century. The total value of U.S. real estate has declined as a fraction of total wealth, and income from rents on real property has declined as a fraction of GDP. This decline would have surprised classical economists of the 19th century, who believed that economic growth would cause rents paid to landlords to increase as a share of the total economy. However, the economic growth of the past decade has been accompanied by extraordinary increases in real estate prices. The relationship between economic growth and the share of the economy controlled by landlords might be more complex than classical economists or their critics have imagined.
In this paper we develop a simple model of the share of land rent in national income and attempt to determine whether rent as a fraction of national income should be expected to rise or fall with economic growth.
Understanding the relationship between real estate and economic growth is important for many reasons. In many developing countries, ownership of land is more concentrated than ownership of other assets, so growth in the share of rent in national income has important implications for the overall concentration of wealth. In the United States, the relationship between economic growth and wealth concentration has received increased attention over the past decade, as concentration levels appear to some analysts to have increased. If the long decline in the importance of real estate is ending, then increasing real estate values might contribute to additional wealth concentration. In particular, rising real estate values might increase the wealth gap between homeowners and renters. The relationship between real estate and economic growth also has implications for the relative returns of real estate versus other investments. Finally, during the latest economic cycle, rising real estate values are believed to have helped to offset the effects of falling values of other assets. If this is the case, then changes in importance of real estate in the economy could have implications for macroeconomic stability.
Demand for land increases with economic development. Thus, if the supply of land is inelastic in the long run, we might expect the share of rental income to increase. On the other hand, changes in the relative productivity or demand elasticities of land and other assets could decrease the share of rental income over time. Thus, it is not at all clear that rent as a proportion of total wealth will increase over time.
The paper is organized as follows. The second section describes different methods of measuring land rents. The third section briefly reviews the Ricardian theory of rent and describes a model based on classical notions of land rent. The model predicts rent as a share of aggregate income. The fourth section examines possible values of the parameters of the model, and the last section summarizes our conclusions.
Measuring Land Rent
Measuring total land rent is a difficult task. As real estate has declined in importance, efforts to collect data on land values and rents have diminished. The only relevant series from the National Income and Product Account data is titled "Rental Income of Persons." These data have many limitations, since they do not include rents received by corporations and partnerships or imputed rent for owner-occupied commercial real estate. If REITs, other corporations and partnerships have increased their share of real estate ownership, then these data may understate rent as a share of national income in recent years. Changes in whether utilities and other costs are included in rent may also affect the accuracy of the data. The data do include imputed rent on owner-occupied housing, so the shift towards individual ownership of residential real estate should not bias the data, although imputed rent may not be measured with a high degree of accuracy. The data also do not include rent on government-owned land and lump return on capital invested in land improvements together with land rent. Keiper, Kurnow, Clark and Segal (1961) point out that the series has been cited as proof that land rent is an insignificant portion of national income, even though the government agencies that publish the data caution that it does not exactly correspond to the economic concept of land rent. They conclude that the series Rental Income of Persons probably understates true land rent. Figure 1 shows this series as a fraction of GDP from 1929 to 2002.
[FIGURE 1 OMITTED]
Guesses can be made about the magnitude of rents that are left out of Rental Income of Persons. Since 1988, the IRS has published data on the net real estate rental income of partnerships. Rapid depreciation for tax purposes understates true net income, but even if reported depreciation is cut in half, net real estate rental income of partnerships has not exceeded 0.7% of national income, and was negative during the years 1998-1991. Net rental income of corporations is not reported, but gross rents received by corporations were approximately 49% of gross rents received by partnerships in 2000, so net rental income of partnerships and corporations combined is unlikely to be much greater than 1% of national income. Rent from owner-occupied commercial real estate is difficult to estimate, but a government survey of buildings (Energy Information Administration 2002) found that 68.7% of commercial buildings in the United States are owner occupied. National Income Accounts data show that nonresidential structures represent 38.3% of the value of all structures in the United States. If 38.3% of all rents are derived from nonresidential land, and if 68.7% of all nonresidential property is owner occupied, then 26.3% of all rents would be from owner-occupied nonresidential property. Finally, if property taxes are paid out of land rent and the services received from government do not contribute to the production of real estate services, then property taxes should be added back to net rental income. In 2000, property taxes represented 3.1% of national income. Rental Income of Persons was 1.8% of national income in 2000. An upper bound of the sum of property taxes, net rental income received by partnerships and corporations and net rental income from owner-occupied nonresidential property is therefore around 6.2% of national income, and so total rent would be, at most, 8% of national income, not including rent on government-owned land. Unfortunately, necessary data are not available to estimate a time series of total land rent. It seems reasonable, however, to expect that the path of total land rent over time may have been similar to that of Rental Income of Persons, although the level of rents would have been somewhat...
|
