CREAT: Census Research Exploration and Analysis Tool

Ban-the-Box (BTB) policies intend to help formerly incarcerated individuals find employment by delaying when employers can ask about criminal records. We revisit the finding in Doleac and Hansen (2020) that BTB causes statistical discrimination against minority men. We correct miscoded BTB laws and show that estimates from the Current Population Survey (CPS) remain quantitatively similar, while those from the American Community Survey (ACS) now fail to reject the null hypothesis of no effect of BTB on employment. In contrast to the published estimates, these ACS results are statistically significantly different from the CPS results, indicating a lack of robustness across datasets. We do not find evidence that these differences are due to sample composition or survey weights. There is limited evidence that these divergent results are explained by the different frequencies of these surveys. Differences in sample sizes may also lead to different estimates; the ACS has a much larger sample and more statistical power to detect effects near the corrected CPS estimates.

For the last half-century, it has been a common and accepted practice for statistical agencies, including the United States Census Bureau, to adopt different strategies to protect the confidentiality of aggregate tabular data products from those used to protect the individual records contained in publicly released microdata products. This strategy was premised on the assumption that the aggregation used to generate tabular data products made the resulting statistics inherently less disclosive than the microdata from which they were tabulated. Consistent with this common assumption, the 2010 Census of Population and Housing in the U.S. used different disclosure limitation rules for its tabular and microdata publications. This paper demonstrates that, in the context of disclosure limitation for the 2010 Census, the assumption that tabular data are inherently less disclosive than their underlying microdata is fundamentally flawed. The 2010 Census published more than 150 billion aggregate statistics in 180 table sets. Most of these tables were published at the most detailed geographic level'individual census blocks, which can have populations as small as one person. Using only 34 of the published table sets, we reconstructed microdata records including five variables (census block, sex, age, race, and ethnicity) from the confidential 2010 Census person records. Using only published data, an attacker using our methods can verify that all records in 70% of all census blocks (97 million people) are perfectly reconstructed. We further confirm, through reidentification studies, that an attacker can, within census blocks with perfect reconstruction accuracy, correctly infer the actual census response on race and ethnicity for 3.4 million vulnerable population uniques (persons with race and ethnicity different from the modal person on the census block) with 95% accuracy. Having shown the vulnerabilities inherent to the disclosure limitation methods used for the 2010 Census, we proceed to demonstrate that the more robust disclosure limitation framework used for the 2020 Census publications defends against attacks that are based on reconstruction. Finally, we show that available alternatives to the 2020 Census Disclosure Avoidance System would either fail to protect confidentiality, or would overly degrade the statistics' utility for the primary statutory use case: redrawing the boundaries of all of the nation's legislative and voting districts in compliance with the 1965 Voting Rights Act.

We quantify the effect of education on mortality using a linkage of the full count 1940, 2000, and 2010 US census files and the Numident death records file. Our sample is composed of children aged 0-18 in 1940, observed living with at least one parent, for whom we can construct a rich set of parental and neighborhood characteristics. We estimate effects of educational attainment in 1940 on survival to 2000, as well as the effects of completed education, observed in 2000, on 10-year survival to 2010. The educational gradients in longevity that we estimate are robust to the inclusion of detailed individual, parental, household, neighborhood and county covariates. Given our full population census sample, we also explore rich patterns of heterogeneity and examine the effect of mediators of the education-mortality relationship. The mediators we consider in this study explain more than half of the relationship between education and mortality. We further show that the mechanisms underlying the education-mortality gradient might be different at different margins of educational attainment.

The purpose of this report is to document the important features of Version 7 of the LEHD Origin-Destination Employment Statistics (LODES) processing system. This includes data sources, data processing methodology, confidentiality protection methodology, some quality measures, and a high-level description of the published data. The intended audience for this document includes LODES data users, Local Employment Dynamics (LED) Partnership members, U.S. Census Bureau management, program quality auditors, and current and future research and development staff members.

Housing represents the most important capital asset for most U.S. families. Despite substantial analysis of the intergenerational mobility of income, large gaps in our knowledge of the distribution of housing assets and their transmission over time remain, as housing is generally not reflected by income flows. Using novel linked data that combines survey responses with administrative tax data and information on ownership and valuation from property tax records for over 3.4 million families, we provide new evidence on the intergenerational transmission of housing capital. We find that housing capital is more persistent across generations than labor income. We document important disparities between average housing outcomes for White and Black children. These difference persist even conditional on parent rank in the distribution of housing assets, with the gap growing throughout the parental housing capital distribution. A decomposition shows that average differences in children's labor market outcomes associated with parental assets explain about half of the observed intergenerational persistence (a 'labor income channel'), and that there is also a substantial 'direct channel' ' conditional on children having the same earnings, children of parents with more housing assets have more assets themselves on average. The direct channel is also important for explaining the intergenerational gap in outcomes of Black and White children. Finally, we present quasi-experimental evidence that local housing supply constraints help explain spatial differences in intergenerational persistence across US counties. Our results establish the importance of housing markets, both independently from and jointly with labor markets, in shaping the intergenerational persistence of economic resources.