Sheila Olmstead, The University of Texas at Austin

The estimated marginal cost of U.S. water pollution control often exceeds its estimated marginal benefits in regulatory impact analysis, and the number of plausibly causal estimates of the benefits of water pollution control in the literature is quite small. Combined, these two facts suggest that some U.S. water quality regulations may be inefficient, but that the studies available to regulators assessing water quality benefits are insufficient to determine whether this is, in fact, the case. Responding to a recent literature using the hedonic property method to value water quality improvements under the U.S. Clean Water Act, we provide intuition, theory and empirical evidence suggesting that this approach may not capture water’s recreational benefits, which are likely to represent a large share of the benefits of water pollution control. Our prior work piloted an approach designed to capture these benefits, along with the local amenity value of water quality improvements, for a single, large coastal U.S. property market Ð the Tampa Bay metro area in Florida. In this work, we first estimate a recreation demand model, which allows us to estimate the marginal contribution of water quality improvements in the region’s major recreational waters to a measure akin to consumer surplus from recreation in the zip codes where property owners reside. In a second stage, we include this consumer surplus measure in a hedonic property model, using sets of fixed effects to control flexibly and comprehensively for unobservables. The hedonic model also includes a variable capturing water quality in the smaller waterbodies closer to each home, so that we can separately identify recreational and amenity values of water quality as capitalized in property prices. The current paper extends the geographic scope of this work, bringing the model to three additional large regions with charismatic, coastal aquatic resources that together encompass several major property markets (Puget Sound, Long Island Sound, and the Texas Gulf Coast). In addition to continuing to develop and refine our approach from the initial application in Tampa, we also collect cellular-device data in an effort to improve upon the traditional recreational visitation data used in most published recreation demand papers in the literature. Preliminary results indicate that homeowners have significant willingness to pay for both local and regional recreational water quality improvements and suggest that prior hedonic studies may underestimate the benefits of water pollution control.