I. INTRODUCTION The United States disposes roughly 60% of the municipal solid waste it generates each year in solid waste disposal facilities, commonly known as landfills. But landfill use has been found to generate external costs. Using plausible assumptions about the number of homes located within close proximity of the landfill, the value of those homes, the quantity of annual waste deposited at the landfill, and the discount rate, Department for Environment, Food and Rural Affairs (2004) estimates the external costs of landfill disposal are between $3.05 and $4.39 for each compacted ton disposed of over the lifetime of the landfill. Implicit to this calculation is the assumption that the reduction in housing values is permanent. That is, external costs of landfill disposal are generated even after the landfill ceases to accept waste and closes the site. This article questions this assumption by estimating whether the closure of a solid waste landfill improves neighboring property values. Neighboring property could remain low if potential home buyers either fear a containment breach would emit odor and toxins into the air and water or simply find the oddly shaped grass-covered hill unattractive. Property values could instead improve with the reduction of garbage trucks on local roads or the elimination of odor permeating from the open face of an operating landfill. If property values increase with the closure of a landfill then estimates of the external marginal cost of landfill disposal such as in Department for Environment, Food and Rural Affairs (2004) would necessarily decrease. Efficient solid waste and recycling policies such as optimal waste taxes or recycling subsidies would be affected. (1) The next section of this article develops the theoretical structure necessary for estimating housing values as a function of a landfill closure. Housing price data collected both before and after a landfill closure are described in Section III. Section IV reports results of this analysis, where a landfill closure is estimated to improve neighboring property values by 10.8%, although this estimate is not statistically different from zero. The repeat-sales estimator used on the same data suggests resales that straddle the landfill closure increase relative to resales that do not. Section V concludes by discussing the policy implications of these results. II. A MODEL OF HOUSING DEMAND Following the logic of Lancaster (1966) and Rosen (1974), assume residents gain utility from consumption (C) and housing services (H) according to the utility function: (1) U = U(C, H), where H is a vector of N individual housing attributes H = H ([q.sub.1], ...,[q.sub.N]) including the geographical distance to a solid waste landfill, structural attributes, and other neighborhood attributes. Assume residents are endowed with income Y. Residents maximize utility subject to the budget constraint: (2) Y = C + P([H.sub.i]), where the price of the composite good is normalized to one and P([H.sub.i]) is a hedonic price function of the N housing attributes (i = 1, ..., N). Residents maximize utility (1) subject to the budget constraint (2) by choosing optimal quantities of the composite consumer good and each of the N housing attributes. The first-order condition for each housing attribute can be simplified to: (3) [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] where subscripts are used to denote partial derivates. The individual chooses each housing attribute, including the distance from a solid waste landfill such that the implicit price of that housing attribute is equal to the marginal rate of substitution between that housing attribute and the composite good. The closing of the landfill could decrease the marginal utility of increasing distance from the landfill and therefore decrease the marginal rate of substitution between the distance to the landfill and the composite good. …