Purchase - $5 NBER Subscribers Download Add Paper to My Library Share: Permalink Using these links will ensure access to this page indefinitely Copy URL Using Administrative Data to Impute Income Non-Response in Household Surveys NBER Working Paper No. w30420 21 Pages Posted: 5 Sep 2022 Last revised: 9 Sep 2022 See all articles by V. Kerry SmithV. Kerry SmithArizona State University (ASU) - Economics Department; National Bureau of Economic Research (NBER)Michael WelshSaint Petersburg College - Independent ConsultantRichard T. CarsonUniversity of California, San Diego (UCSD) - Department of EconomicsStanley PresserUniversity of Maryland Date Written: September 2022 Abstract Income is simultaneously one of the most important variables used by economists and the variable most likely to be missing due to item non-response. While observations that are missing income responses are often dropped from analyses, such treatment is usually inappropriate. More appropriate solutions rely on imputation based on either covariates (e.g., age and education) measured in the survey or on spatial estimates (most often for zip codes) from the American Community Survey. We describe a new spatially-based alternative using publicly available Internal Revenue Service tax data that allows estimates of zip code's income distribution.Institutional subscribers to the NBER working paper series, and residents of developing countries may download this paper without additional charge at www.nber.org. Suggested Citation: Suggested Citation Smith, V. Kerry and Welsh, Michael and Carson, Richard T. and Presser, Stanley, Using Administrative Data to Impute Income Non-Response in Household Surveys (September 2022). NBER Working Paper No. w30420, Available at SSRN: https://ssrn.com/abstract=4210007 V. Kerry Smith (Contact Author) Arizona State University (ASU) - Economics Department ( email ) Tempe, AZ 85287-3806United States National Bureau of Economic Research (NBER) 1050 Massachusetts AvenueCambridge, MA 02138United States Michael Welsh Saint Petersburg College - Independent Consultant ( email ) Russia Richard T. Carson University of California, San Diego (UCSD) - Department of Economics ( email ) 9500 Gilman DriveLa Jolla, CA 92093-0508United States619-534-6319 (Phone)619-534-7655 (Fax) Stanley Presser University of Maryland ( email ) College ParkCollege Park, MD 20742United States Purchase - $5 NBER Subscribers Download Do you have negative results from your research you'd like to share? Submit Negative Results Paper statistics Downloads 5 Abstract Views 118 PlumX Metrics Related eJournals NBER Working Paper Series Follow NBER Working Paper Series Subscribe to this free journal for more curated articles on this topic FOLLOWERS 12,270 PAPERS 32,498 Feedback Feedback to SSRN Feedback (required) Email (required) Submit If you need immediate assistance, call 877-SSRNHelp (877 777 6435) in the United States, or +1 212 448 2500 outside of the United States, 8:30AM to 6:00PM U.S. Eastern, Monday - Friday.

The 2010 Deepwater Horizon oil spill in the Gulf of Mexico was the largest ever in U.S. waters, eclipsing the 1989 Exxon Valdez spill in terms of the sheer quantity of oil released and the scale and scope of activities impacted. We developed a recreation demand model to monetize economic damages associated with lost shoreline recreational user days attributable to the spill. The unprecedented magnitude of the spill disruption led to a variety of innovations. We estimate a model of shoreline recreation trips to the Gulf Coast region from the general population of the contiguous U.S., combining single and multiple-day trips, calculating travel costs that incorporate detailed information on flying costs and transportation mode choice, and using alternative-specific constants to control for site characteristics. Losses per recreational user day are assessed using utility adjustments that reproduce the decline in recreation observed through onsite counts. Sensitivity analyses demonstrate our lost user day value is robust to changes in income imputation, nesting structure, site aggregation and spill calibration, and show the importance of accounting for flying as a mode choice. Estimated losses from the primary shoreline study are $520 million (±166) out of the total recreational damages of $661 million (2015$).

Stated-preference research supports $17.2B in protections

In the wake of a massive offshore oil spill, federal and state agencies undertook studies to determine the reduction in outdoor recreational trips along the Gulf of Mexico. This paper describes three of those efforts, which gauged the impact of the oil spill on beach visits, boating, and fishing. Together the studies represented an enormous undertaking, involving nearly half a million aerial photographs, 35,000 onsite counts, and 129,000 interviews over a period of three years. For each activity, samples of time-place combinations were quickly developed and systematic data collection covering hundreds of miles of the most affected areas began within six weeks of the start of the spill and soon expanded to cover more than 1,000 miles of shoreline. The research used a mix of aerial and onsite counts and interview data to estimate the reduction in recreation due to the spill. "Baseline" estimates were developed that represented what the level of recreation would have been in the absence of the spill. These estimates were derived from the observed levels of recreation at a later period when the spill was unlikely to have had continuing effects on recreation, with adjustments for differences in the weather between the baseline period and the period immediately after the spill. Overall, the results suggest that there were about 12.3 million fewer visits to the beach because of the spill. In addition, 215,000 fewer boaters went out in the Gulf and anglers made 144,000 fewer fishing trips. We conclude with a discussion of the lessons learned from the study.