Papers Containing Tag(s): 'Federal Statistical Research Data Center'
The following papers contain search terms that you selected. From the papers listed below, you can navigate to the PDF, the profile page for that working paper, or see all the working papers written by an author. You can also explore tags, keywords, and authors that occur frequently within these papers.
See Working Papers by Tag(s), Keywords(s), Author(s), or Search Text
Click here to search again
Frequently Occurring Concepts within this Search
Viewing papers 61 through 70 of 189
-
Working PaperA Simulated Reconstruction and Reidentification Attack on the 2010 U.S. Census: Full Technical Report
December 2023
Working Paper Number:
CES-23-63R
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. You are reading the full technical report. For the summary paper see https://doi.org/10.1162/99608f92.4a1ebf70.View Full Paper PDF
-
Working PaperCollaborative Micro-productivity Project: Establishment-Level Productivity Dataset, 1972-2020
December 2023
Working Paper Number:
CES-23-65
We describe the process for building the Collaborative Micro-productivity Project (CMP) microdata and calculating establishment-level productivity numbers. The documentation is for version 7 and the data cover the years 1972-2020. These data have been used in numerous research papers and are used to create the experimental public-use data product Dispersion Statistics on Productivity (DiSP).View Full Paper PDF
-
Working PaperOutsourcing Dynamism
December 2023
Working Paper Number:
CES-23-64
This paper investigates the increasing importance of domestic outsourcing in U.S. manufacturing. Under domestic outsourcing, the agency is the employer of record for temporary workers, though they perform their tasks at the client business' premises. On a yearly basis, one in two manufacturing plants hires at least some of its workers through a temporary help agency. Furthermore, domestic outsourcing is becoming increasingly more important: the average share of revenue spent on such arrangements has gone up by 85 percent since 2006. We develop a methodology to transform reported expenses on temporary and leased workers into plant-level outsourced employment counts, using administrative data on the U.S. manufacturing sector. We find that domestic outsourcing is an important margin of adjustment that plants use to modify their workforce in response to productivity shocks. Plant-level outsourced employment adjusts more quickly and is twice as responsive as payroll employment. These micro implications have significant aggregate consequences. Without taking reallocations in outsourced employment into account, the measured pace at which jobs reallocate across workplaces is underestimated. On average, we omit the equivalent of 15 percent of payroll employment reallocations in each year. However, outsourced employment churns at a much higher rate compared to its payroll counterpart. Therefore, the omission of outsourced reallocations can rationalize 37 percent of the secular decline in the aggregate job reallocation rate. Lastly, the extent of mismeasurement varies with the business cycle; falling in downturns and increasing in upturns implying that the speed of economic recovery is underestimated.View Full Paper PDF
-
Working PaperThe 2010 Census Confidentiality Protections Failed, Here's How and Why
December 2023
Working Paper Number:
CES-23-63
Using only 34 published tables, we reconstruct five variables (census block, sex, age, race, and ethnicity) in the confidential 2010 Census person records. Using the 38-bin age variable tabulated at the census block level, at most 20.1% of reconstructed records can differ from their confidential source on even a single value for these five variables. Using only published data, an attacker can verify that all records in 70% of all census blocks (97 million people) are perfectly reconstructed. The tabular publications in Summary File 1 thus have prohibited disclosure risk similar to the unreleased confidential microdata. Reidentification studies confirm that an attacker can, within blocks with perfect reconstruction accuracy, correctly infer the actual census response on race and ethnicity for 3.4 million vulnerable population uniques (persons with nonmodal characteristics) with 95% accuracy, the same precision as the confidential data achieve and far greater than statistical baselines. The flaw in the 2010 Census framework was the assumption that aggregation prevented accurate microdata reconstruction, justifying weaker disclosure limitation methods than were applied to 2010 Census public microdata. The framework used for 2020 Census publications defends against attacks that are based on reconstruction, as we also demonstrate here. Finally, we show that alternatives to the 2020 Census Disclosure Avoidance System with similar accuracy (enhanced swapping) also fail to protect confidentiality, and those that partially defend against reconstruction attacks (incomplete suppression implementations) destroy the primary statutory use case: data for redistricting all legislatures in the country in compliance with the 1965 Voting Rights Act.View Full Paper PDF
-
Working PaperLocal and National Concentration Trends in Jobs and Sales: The Role of Structural Transformation
November 2023
Working Paper Number:
CES-23-59
National U.S. industrial concentration rose between 1992-2017. Simultaneously, the Herfindhahl Index of local (six-digit-NAICS by county) employment concentration fell. This divergence between national and local employment concentration is due to structural transformation. Both sales and employment concentration rose within industry-by-county cells. But activity shifted from concentrated Manufacturing towards relatively un-concentrated Services. A stronger between-sector shift in employment relative to sales explains the fall in local employment concentration. Had sectoral employment shares remained at their 1992 levels, average local employment concentration would have risen by 9% by 2017 rather than falling by 7%.View Full Paper PDF
-
Working PaperOutput Market Power and Spatial Misallocation
November 2023
Working Paper Number:
CES-23-57
Most product industries are local. In the U.S., firms selling goods and services to local consumers account for half of total sales and generate more than sixty percent of the nation's jobs. Competition in these industries occurs in local product markets: cities. I propose a theory of such competition in which firms have output market power. Spatial differences in local competition arise endogenously due to the spatial sorting of heterogeneous firms. The ability to charge higher markups induces more productive firms to overvalue locating in larger cities, leading to a misallocation of firms across space. The optimal policy incen tivizes productive firms to relocate to smaller cities, providing a rationale for commonly used place-based policies. I use U.S. Census establishment-level data to estimate markups and to structurally estimate the model. I document a significant heterogeneity in markups for local industries across U.S. cities. Cities in the top decile of the city-size distribution have a fifty percent lower markup than cities in the bottom decile. I use the estimated model to quantify the general equilibrium effects of place-based policies. Policies that remove markups and relocate firms to smaller cities yield sizable aggregate welfare gains.View Full Paper PDF
-
Working PaperThe Impact of Industrial Opt-Out from Utility Sponsored Energy Efficiency Programs
October 2023
Working Paper Number:
CES-23-52
Industry accounts for one-third of energy consumption in the US. Studies suggest that energy efficiency opportunities represent a potential energy resource for regulated utilities and have resulted in rate of return regulated demand-side management (DSM) and energy efficiency (EE) programs. However, many large customers are allowed to self-direct or opt-out. In the Carolinas (NC and SC), over half of industrial and large commercial customers have selected to opt out. Although these customers claim they invest in EE improvements when it is economic and cost-effective to do so, there is no mechanism to validate whether they actually achieved energy savings. This project examines the industrial energy efficiency between the program participants and non participants in the Carolinas by utilizing the non-public Census of Manufacturing data and the public list of firms that have chosen to opt out. We compare the relative energy efficiency between the stay-in and opt-out plants. The t-test results suggest opt-out plants are less efficient. However, the opt-out decisions are not random; large plants or plants belonging to large firms are more likely to opt out, possibly because they have more information and resources. We conduct a propensity score matching method to account for factors that could affect the opt-out decisions. We find that the opt-out plants perform at least as well or slightly better than the stay-in plants. The relative performance of the opt-out firms suggest that they may not need utility program resources to obtain similar levels of efficiency from the stay-in group.View Full Paper PDF
-
Working PaperTemperature and Local Industry Concentration
October 2023
Working Paper Number:
CES-23-51
We use plant-level data from the US Census of Manufacturers to study the short and long run effects of temperature on manufacturing activity. We document that temperature shocks significantly increase energy costs and lower the productivity of small manufacturing plants, while large plants are mostly unaffected. In US counties that experienced higher increases in average temperatures between the 1980s and the 2010s, these heterogeneous effects have led to higher concentration of manufacturing activity within large plants, and a reallocation of labor from small to large manufacturing establishments. We offer a preliminary discussion of potential mechanisms explaining why large manufacturing firms might be better equipped for long-run adaptation to climate change, including their ability to hedge across locations, easier access to finance, and higher managerial skills.View Full Paper PDF
-
Working PaperAntitrust Enforcement Increases Economic Activity
October 2023
Working Paper Number:
CES-23-50
We hand-collect and standardize information describing all 3,055 antitrust law suits brought by the Department of Justice (DOJ) between 1971 and 2018. Using restricted establishment-level microdata from the U.S. Census, we compare the economic outcomes of a non-tradable industry in states targeted by DOJ antitrust lawsuits to outcomes of the same industry in other states that were not targeted. We document that DOJ antitrust enforcement actions permanently increase employment by 5.4% and business formation by 4.1%. Using an event-study design, we find (1) a sharp increase in payroll that exceeds the increase in employment, meaning that DOJ antitrust enforcement increases average wages, (2) an economically smaller increase in sales that is statistically insignificant, and (3) a precise increase in the labor share. While we cannot separately measure the quantity and price of output, the increase in production inputs (employment), together with a proportionally smaller increase in sales, strongly suggests that these DOJ antitrust enforcement actions increase the quantity of output and simultaneously decrease the price of output. Our results show that government antitrust enforcement leads to persistently higher levels of economic activity in targeted industries.View Full Paper PDF
-
Working PaperAn In-Depth Examination of Requirements for Disclosure Risk Assessment
October 2023
Working Paper Number:
CES-23-49
The use of formal privacy to protect the confidentiality of responses in the 2020 Decennial Census of Population and Housing has triggered renewed interest and debate over how to measure the disclosure risks and societal benefits of the published data products. Following long-established precedent in economics and statistics, we argue that any proposal for quantifying disclosure risk should be based on pre-specified, objective criteria. Such criteria should be used to compare methodologies to identify those with the most desirable properties. We illustrate this approach, using simple desiderata, to evaluate the absolute disclosure risk framework, the counterfactual framework underlying differential privacy, and prior-to-posterior comparisons. We conclude that satisfying all the desiderata is impossible, but counterfactual comparisons satisfy the most while absolute disclosure risk satisfies the fewest. Furthermore, we explain that many of the criticisms levied against differential privacy would be levied against any technology that is not equivalent to direct, unrestricted access to confidential data. Thus, more research is needed, but in the near-term, the counterfactual approach appears best-suited for privacy-utility analysis.View Full Paper PDF
