Externalities and zoning

Abstract

An empirical debate has arisen concerning the existence of external effects in residential property markets. Some studies have indicated that non-conforming land uses which are often regarded as nuisances do not significantly affect residential land values [2, 6]. From this they conclude that externalities are not important. We shall argue that this conclusion may not be warranted, even though their statistical methods and data may be impeccable. Other studies indicate that non-conforming land uses do affect residential property values [9, 11]. Such studies do indicate that externalities matter, though they may underestimate them. Where all of these studies err is in their (sometimes implicit) contention that such results by themselves provide a basis for evaluating the efficiency of municipal zoning. To establish these points, we shall demonstrate that the existence of negative externalities is consistent with either lowering or leaving unchanged the value of residential property, depending on the system of compensation and distribution of property rights. At the same time, we shall show that both conflicting results are consistent with an efficient system of land use controls or an inefficient system. That is to say, neither conclusion about externalities in land values provides sufficient criteria for normative statements about zoning. Finally, we shall discuss the usefulness of the concept of externality in evaluating the efficiency of zoning.' The empirical studies may be characterized as attempting the following experiment. An otherwise homogeneous neighborhood of owner-occupied houses is found to have some non-conforming use such as apartment buildings, commercial establishments, or factories. The question is whether such nonconfirming uses lower the price of land for residential use. The technique is' ordinarily to compare, by multiple regression analysis, residential land values in neighborhoods or communities which have some non-conformers to those with fewer or none. If residential land values are not apparently altered by the non-conforming use, the inference is that externalities are not important and