Background The Evidence-based Policy-Making Act of 2018 requires that Federal agencies use their data to develop statistical evidence to support policy and programmatic decisions.Objective This study assessed Vocational Rehabilitation agency capacity to effectively use their data to inform evidence-based decision-making.Methods The Capacity Survey assessed agency capacity in data management, data visualization and statistical analysis. The survey asked for details about (1) the availability of relevant software programs (e.g., SPSS for statistical analysis), and (2) staff expertise to use that software.Results Results pointed to capacity gaps that would significantly hinder most agencies' application of even a simplified return on investment model. When examining statistical capacity, 60% of agencies responded "not applicable - staff do not have competence in the listed software packages (including SPSS, SAS, R, Python, Stata, and other)" and 71% of respondents said they lacked internal statistical capacity to analyze data using any of the listed programs.Conclusions Results suggested that many state VR agencies lack internal capacity to meet requirements outlined in the Evidence-Based Policy-Making Act of 2018 or regulations related to reporting and performance accountability requirements. Potential solutions to overcome capacity deficits include expanding internal capacity, expanding agency/consultant partnerships, and building cross-agency collaborations.

This article contrasts social and taxpayer return on investment measures of the vocational rehabilitation (VR) program in Virginia. To do this, we use the analyses in prior work which demonstrates substantial social return to Virginia’s VR program. Using this estimated model and administrative data on VR clients in Virginia, we simulate earnings that would be realized with and without VR service receipt by each client and estimate the costs of the services provided to each client. Then, given these simulation results, we compute the taxpayer return on investment. Since most VR recipients have a weak attachment to the labor market (i.e., relatively low employment rates and earnings), the relatively large estimated impact of VR on earnings translates into only a small impact on the taxpayer return. That is, the cost of VR is large relative to the lifetime changes in tax receipt. In particular, we estimate that only 29% of VR recipients have a positive taxpayer return.

We discuss data quality and modeling issues inherent in the use of nationwide property data to value environmental amenities. By example of Zillow's Transaction and Assessment Database, a real estate database covering the United States, we identify challenges and propose guidance for (1) identifying arm's- length sales; (2) geolocating parcels and buildings; (3) identifying temporal links between transaction, assessor, and parcel data; (4) identifying property types, such as single - family homes and vacant lands; and (5) dealing with missing or mismeasured data for standard housing attributes. We review current practice and show that how researchers address these issues can meaningfully influence research findings. (JEL Q51)

This paper provides evidence on how racial differences in the classification of learning and intellectual disabilities bias inferences on labor market outcomes of vocational rehabilitation program clients. Estimates using Rehabilitation Services Administration data from Virginia imply that Whites who have learning disabilities have worse labor market outcomes than non-Whites who have learning disabilities. We argue this unusual finding reflects racial differences in how disabilities are classified. Using an endogenous disability classification model, we find substantial biases in the estimated labor market coefficients. At minimum, the estimated White-Black employment gap is biased down by 3.2% and the earnings gap by 10%.

We construct a structural model of participation in vocational rehabilitation and labor market outcomes for people with vision impairments. There are multiple services to choose among, and each has different effects on employment and earnings. We estimate negative effects for most service types, leading to surprisingly low rates of return to VR service receipt. The negative returns are strongly affected by large administrative costs.

Smart-home technologies have been heralded as an important way to increase energy conservation. While in vitro engineering estimates provide broad optimism, little has been done to explore whether such estimates scale beyond the lab. We estimate the causal impact of smart thermostats on energy use via two novel framed field experiments in which a random subset of treated households have a smart thermostat installed in their home. Examining 18 months of associated high-frequency data on household energy consumption, yielding more than 16 million hourly electricity and daily natural gas observations, we find little evidence that smart thermostats have a statistically or economically significant effect on energy use. We explore potential mechanisms using almost four million observations of system events including human interactions with their smart thermostat. Results indicate that user behavior dampens energy savings and explains the discrepancy between estimates from engineering models, which assume a perfectly compliant subject, and actual households, who are occupied by users acting in accord with behavioral economists’ conjectures. In this manner, our data document a keen threat to the scalability of new user-based technologies.

Introduction: This study describes the characteristics of, services received by, and labor market outcomes of applicants with visual impairments to three state vocational rehabilitation programs. Our objective is to both document cross-state variation in vocational rehabilitation clientele and services and provide new insights on the longitudinal labor market outcomes of clients with visual impairments (i.e., blindness or low vision). This analysis is a first step in assessing the returns to vocational rehabilitation services for this population. Methods: We first created a unique longitudinal data set by matching administrative records on applicants who are visually impaired in state fiscal year 2007 from three vocational rehabilitation agencies to 8 years of employment data from state Unemployment Insurance programs. Using these data, we examined cross-state variation in the descriptive statistics for important client explanatory variables and vocational rehabilitation service categories. We then compared the long-term labor market outcomes of clients receiving services (treated) to untreated individuals. Results: We documented two important findings. First, there were substantial differences in client characteristics, services provided, and costs across the three states. Second, the long-run labor market analysis was consistent with vocational rehabilitation services having no employment effect but a positive earnings effect. Discussion: Labor market results indicate vocational rehabilitation services provided persistent earnings benefits. Yet the substantial cross-state heterogeneity suggests these labor market results might not be generalizable and should be interpreted with caution. We explain what was missing from this analysis and why the results should not be thought of as causal. Implications for Practitioners: This article gives practitioners a sense of a unique new data set on vocational rehabilitation and labor market variables for applicants with visual impairments. We highlight the importance of cross-state variation and linking vocational rehabilitation data to long-term employment measures. The question of how best to inform the efficacy of different vocational rehabilitation strategies for clients with visual impairments is left for future researchers to consider.

This paper discusses issues associated with using readily available administrative data in estimating ROI for vocational rehabilitation services. It starts with a discussion of longitudinal outcomes data. The discussion is divided up into labor market outcomes data, other types of outcomes data, necessary sample sizes (power analysis), and ways to deal with people systematically excluded from the outcomes data. Next, the paper focuses on services data. The topics covered include the need for control groups, using service cohort data, different sources of service, and merging service data with outcomes data. Finally, the paper moves to the need for other controlling explanatory variables including discussions of inclusion of demographic explanatory variables and data from local labor markets. Two online appendices to this paper provide additional details through (a) an example of a power analysis to illustrate sample size issues and (b) a discussion of Institutional Review Board issues associated with conducting empirical investigations using administrative data.

This paper briefly describes and then implements the VR-ROI (Vocational Rehabilitation Return on Investment) Project’s model for applicants in state fiscal year 2007 to Virginia's Department for Aging and Rehabilitative Services (VA DARS) and to Maryland's Division of Rehabilitation Services (MD DORS). We present results that account for differences across disability, agency, VR service type and source, applicant characteristics, and county as well as national economic conditions. This approach provides a rich set of estimates that display considerable heterogeneity within each agency across three disability types (mental illness, physical impairment, and cognitive impairment) and seven service categories (diagnosis & evaluation, training, education, restoration, maintenance, job placement, job supports) within each disability type.

We provide an overview of the basic conceptual issues involved in estimating the return on investment (ROI) of state vocational rehabilitation (VR) programs. Our aim is to highlight some of the key issues in ROI evaluations, especially those associated with estimating the benefits and costs of VR. Finally, we discuss different ways of implementing ROI calculations and suggest that rate of return type analysis is appealing for VR evaluations where there is no widely accepted discount rate.

Current knowledge of the fiscal impacts of alternative land uses comes largely from cost of community services (CCS) case studies, the results of which are viewed skeptically in the literature due to methodological concerns. To address these issues, we develop an econometric approach that allows us to capture both the direct and indirect relationships between a complete accounting of community fiscal measures and the full distribution of acres of land uses in a jurisdiction. Using a novel panel data set, we extensively document empirical correlations that have not yet been formally established in the literature. Our results are inconsistent with the broad conclusions of CCS studies: neither a shift from agricultural to residential land nor a shift to commercial land is associated with a significant change in the budget. We provide support for and insights into our results by extending our analysis to finer revenue/expenditure and land-use subcategories.

When the Deepwater Horizon oil rig exploded in 2010, it resulted in the largest off-shore oil spill in United States history. Economic theory dictates that the oil damage and restitution payments that resulted from the spill should be capitalized into property values. To measure the extent of this capitalization, we create a novel dataset by linking surveys of the location and severity of oil observed along over 4,300 miles of the Gulf Coast to measures of local housing market outcomes. We then perform hedonic-style analysis to determine the net effects of the spill on affected real estate markets. In doing so, we provide the first plausibly causal estimates of the effect of the spill on affected housing markets throughout the Gulf region. Identification comes from a triple difference framework that exploits the random nature of both the spill and the spatial distribution of oil that affected coastal communities, as well as controls for the confounding effects of the housing market crash. Results suggest that on net, the BP oil spill caused a significant decline in home prices of between 4% and 8% that persisted until at least 2015. This implies housing markets capitalized $3.8 billion to $5.0 billion in spill damage inclusive of clean-up and restitution effects. These results are robust to numerous alternative definitions of treatment and control groups.

Supplemental Material, The_Fiscal_Impacts_of_Alternative_Land_Uses-Online_Appendix for The Fiscal Impacts of Alternative Land Uses: An Empirical Investigation of Cost of Community Services Studies by Christopher M. Clapp, James Freeland, Keith Ihlanfeldt, and Kevin Willardsen in Public Finance Review

Federal and state governments spend over $3 billion annually on public-sector Vocational Rehabilitation (VR) programs, yet almost a third of people with disabilities report having inadequate access to the transportation necessary to commute to a job, potentially negating the positive effects of these interventions. We examine this previously understudied connection by assessing the impact access to public paratransit has on measures of VR program effectiveness. To do so, we use the data and estimates from three previously estimated structural models of VR service receipt and labor market outcomes that contain limited information about mobility. We spatially link the generalized residuals from these models to different measures of the availability and efficiency of local paratransit systems to determine whether paratransit explains any of the residual variation in the short- or long-run labor market outcomes of individuals receiving VR services. Results show that access to paratransit is an important determinant of the efficacy of VR services, but that effects are heterogeneous across disability groups. We discuss the policy implications of our findings for VR programs.

Policymakers have been slow to implement price-based congestion policies due in part to how little is known about the effects of policies that influence more than simply an individual's commuting method.An individual can also alter her commute by choosing to travel from a different location.I develop a discrete choice structural model of the joint of preferences.Instead, I use a collective model of the household to account for the fact that spouses rarely commute to the same work location.This allows me to model the interplay between residential and commuting mode choices when spouses consider the proximity of their home to both work locations.I allow family members to have caring preferences, and I treat characteristics of the home as a family public good.The collective model requires observing individual consumption of at least one private good in the household to identify bargaining outcomes, and I use a novel assignable private good: the method and duration of each commute.This work is both an extension of the collective model to the residential choice and travel literatures as well as an application of the collective model to a problem with discrete choices and a rich error structure.

Current knowledge of the fiscal impacts of alternative land uses comes largely from cost of community services (CCS) case studies, the results of which are viewed skeptically in the literature because of numerous methodological concerns. In order to begin to fill the gap in our understanding of these impacts, we provide the first empirical estimates of the relationship between a complete accounting of community fiscal measures and the full distribution of acres of land uses in the jurisdiction. We find evidence in support of the broad conclusions of CCS studies: a shift from agricultural to residential land is associated with a deficit, whereas a shift to commercial land is correlated with a surplus. We provide insights into which revenue/expenditure and land use subcategories are responsible for these results.

Using a panel of Major League Baseball team attendance data for the period 1950 to 2003, the authors determined that after controlling for team quality and other factors, a new modern era ballpark adds 22 to 30 percent to total attendance over a 10-year period and, on average, generated present-value stadium revenues of $272 million for the franchise. Since the construction costs for the group of 14 modern ballparks; averaged $99 million in private money and $198 million in public funds, there were two results with important implications for public finance. First, the revenue estimates were less than the typical cost of most modern stadiums, indicating that the projects generated positive rents for team owners only due to public subsidization. Second, the ratio of recipient benefits to subsidy expenses indicated that public spending on construction of replacement stadiums was a less effective method for subsidizing franchise owners than direct lump-sum payments. Furthermore, the preference for stadium project subsidies over cash subsidies call not be explained by the desire of local officials to improve the quality-of-play of the team. Due to non-complementarity between new stadiums and team success, team profits are maximized when an owner "pockets" increases in revenue rather than reinvesting in the team's level of on-field quality.

Using panel data of MLB team attendance from 1950 to 2002, we determined that the attendance “honeymoon” effect of a new stadium—after separating quality-of-play effects—increases attendance by 32% to 37% the opening year of a new stadium. Attendance remains above baseline levels for only two seasons for multipurpose stadiums built during 1960 to 1974 but for 6 to 10 seasons at newer ballparks. Contrary to expectations, there is no systematic interaction between new venues and team performance on attendance or stadium revenues. This noncomplementarity implies that a profit-maximizing team owner would not use a new stadium’s revenue stream to increase team quality of play.