Exploring Differences in Employment between Household and Establishment Data
April 2009
Working Paper Number:
CES-09-09
Abstract
Using a large data set that links individual Current Population Survey (CPS) records to employer-reported administrative data, we document substantial discrepancies in basic measures of employment status that persist even after controlling for known definitional differences between the two data sources. We hypothesize that reporting discrepancies should be most prevalent for marginal workers and marginal jobs, and find systematic associations between the incidence of reporting discrepancies and observable person and job characteristics that are consistent with this hypothesis. The paper discusses the implications of the reported findings for both micro and macro labor market analysis
Document Tags and Keywords
Keywords
Keywords are automatically generated using KeyBERT, a powerful and innovative
keyword extraction tool that utilizes BERT embeddings to ensure high-quality and contextually relevant
keywords.
By analyzing the content of working papers, KeyBERT identifies terms and phrases that capture the essence of the
text, highlighting the most significant topics and trends. This approach not only enhances searchability but
provides connections that go beyond potentially domain-specific author-defined keywords.
:
payroll,
report,
survey,
employee,
employ,
employed,
labor,
recession,
job,
estimates employment,
employment estimates,
discrepancy,
employing,
worker,
labor statistics,
employment count,
unemployment rates,
employment measures,
employment statistics,
employee data,
census employment,
unemployed,
employment trends
Tags
Tags are automatically generated using a pretrained language model from spaCy, which excels at
several tasks, including entity tagging.
The model is able to label words and phrases by part-of-speech,
including "organizations." By filtering for frequent words and phrases labeled as "organizations", papers are
identified to contain references to specific institutions, datasets, and other organizations.
:
Internal Revenue Service,
Bureau of Labor Statistics,
Social Security Administration,
Center for Economic Studies,
Current Population Survey,
Survey of Income and Program Participation,
Social Security,
Social Security Number,
Longitudinal Employer Household Dynamics,
PSID,
Protected Identification Key,
Quarterly Census of Employment and Wages
Similar Working Papers
Similarity between working papers are determined by an unsupervised neural
network model
know as Doc2Vec.
Doc2Vec is a model that represents entire documents as fixed-length vectors, allowing for the
capture of semantic meaning in a way that relates to the context of words within the document. The model learns to
associate a unique vector with each document while simultaneously learning word vectors, enabling tasks such as
document classification, clustering, and similarity detection by preserving the order and structure of words. The
document vectors are compared using cosine similarity/distance to determine the most similar working papers.
Papers identified with 🔥 are in the top 20% of similarity.
The 10 most similar working papers to the working paper 'Exploring Differences in Employment between Household and Establishment Data' are listed below in order of similarity.
-
Working PaperThe Consequences of Long Term Unemployment: Evidence from Matched Employer-Employee Data*🔥
January 2016
Working Paper Number:
CES-16-40
It is well known that the long-term unemployed fare worse in the labor market than the short-term unemployed, but less clear why this is so. One potential explanation is that the long-term unemployed are 'bad apples' who had poorer prospects from the outset of their spells (heterogeneity). Another is that their bad outcomes are a consequence of the extended unemployment they have experienced (state dependence). We use Current Population Survey (CPS) data on unemployed individuals linked to wage records for the same people to distinguish between these competing explanations. For each person in our sample, we have wage record data that cover the period from 20 quarters before to 11 quarters after the quarter in which the person is observed in the CPS. This gives us rich information about prior and subsequent work histories not available to previous researchers that we use to control for individual heterogeneity that might be affecting subsequent labor market outcomes. Even with these controls in place, we find that unemployment duration has a strongly negative effect on the likelihood of subsequent employment. This finding is inconsistent with the heterogeneity ('bad apple') explanation for why the long-term unemployed fare worse than the short-term unemployed. We also find that longer unemployment durations are associated with lower subsequent earnings, though this is mainly attributable to the long-term unemployed having a lower likelihood of subsequent employment rather than to their having lower earnings once a job is found.View Full Paper PDF
-
Working PaperA New Measure of Multiple Jobholding in the U.S. Economy
September 2020
Working Paper Number:
CES-20-26
We create a measure of multiple jobholding from the U.S. Census Bureau's Longitudinal Employer-Household Dynamics data. This new series shows that 7.8 percent of persons in the U.S. are multiple jobholders, this percentage is pro-cyclical, and has been trending upward during the past twenty years. The data also show that earnings from secondary jobs are, on average, 27.8 percent of a multiple jobholder's total quarterly earnings. Multiple jobholding occurs at all levels of earnings, with both higher- and lower-earnings multiple jobholders earning more than 25 percent of their total earnings from multiple jobs. These new statistics tell us that multiple jobholding is more important in the U.S. economy than we knew.View Full Paper PDF
-
Working PaperComparing Measures of Earnings Instability Based on Survey and Adminstrative Reports
August 2010
Working Paper Number:
CES-10-15
In Celik, Juhn, McCue, and Thompson (2009), we found that estimated levels of earnings instability based on data from the Current Population Survey (CPS) and the Survey of Income and Program Participation (SIPP) were reasonably close to each other and to others' estimates from the Panel Study of Income Dynamics (PSID), but estimates from unemployment insurance (UI) earnings were much larger. Given that the UI data are from administrative records which are often posited to be more accurate than survey reports, this raises concerns that measures based on survey data understate true earnings instability. To address this, we use links between survey samples from the SIPP and UI earnings records in the LEHD database to identify sources of differences in work history and earnings information. Substantial work has been done comparing earnings levels from administrative records to those collected in the SIPP and CPS, but our understanding of earnings instability would benefit from further examination of differences across sources in the properties of changes in earnings. We first compare characteristics of the overall and matched samples to address issues of selection in the matching process. We then compare earnings levels and jobs in the SIPP and LEHD data to identify differences between them. Finally we begin to examine how such differences affect estimates of earnings instability. Our preliminary findings suggest that differences in earnings changes for those in the lower tail of the earnings distribution account for much of the difference in instability estimates.View Full Paper PDF
-
Working PaperSocial, Economic, Spatial, and Commuting Patterns of Dual Jobholders
April 2007
Working Paper Number:
tp-2007-01
Individuals who hold multiple jobs have complex working lives and complex commuting patterns. Economic and spatial information on these individuals is not readily available in standard datasets, such as the 2000 Decennial Census Long Form, because the survey questions were not designed to collect details on multiple jobs. This study takes advantage of firm-based data from the Unemployment Insurance administrative wage records, linked with the Census Bureau's household-based data, to examine multiple jobholders - and specifically a sentinel group of dual jobholders. The study uses a sample from Los Angeles County, California and examines the dual jobholders by their demographic characteristics as well as their economic, commuting, and spatial location outcomes. In addition this report evaluates whether multiple jobholders should be included explicitly in future labor-workforce analyses and transportation modeling.View Full Paper PDF
-
Working PaperThe Recent Decline of Single Quarter Jobs
January 2015
Working Paper Number:
CES-15-05
Rates of hiring and job separation fell by as much as a third in the U.S. between the late 1990s and the early 2010s. Half of this decline is associated with the declining incidence of jobs that start and end in the same calendar quarter, employment events that we call 'single quarter jobs.' We investigate this unique subset of jobs and its decline using matched employer-employee data for the years 1996-2012. We characterize the worker demographics and employer characteristics of single quarter jobs, and demonstrate that changes over time in workforce and employer composition explain little of the decline in these jobs. We find that the decline in these jobs accounts for about a third of the decline in the fraction of the population that holds a job in the private sector that occurred from the mid 2000s to the early 2010s. We also find little evidence that single quarter jobs are stepping stones into longer-term employment. Finally, we show that the inclusion or exclusion of these single quarter jobs creates divergent trends in average earnings and the dispersion of earnings for the years 1996-2012. To the extent that administrative records measure the volatile tail of the employment distribution better thanconventional household surveys, these findings show that measurement of short duration jobs matters for economic analysis.View Full Paper PDF
-
Working PaperSocial, Economic, Spatial, and Commuting Patterns of Informal Jobholders
April 2007
Working Paper Number:
tp-2007-02
A significant number of employees within the United States can be considered "informal" or "off-the-books" workers. These workers, who by definition do not appear in administrative wage records, are distinct from the larger group of private jobholders who do appear in administrative records. However, while socioeconomic and spatial information on these individuals is readily available in standard datasets, such as the 2000 Decennial Census Long Form, it is not possible to identify the informal workers by only using such data because of the lack of accurate, formal wage records. This study takes advantage of firm-based data that originates in Unemployment Insurance administrative wage records linked with the Census Bureau's household-based data in order to examine informal jobholders by their demographic characteristics as well as their economic, commuting, and spatial location outcomes. In addition this report evaluates whether informal jobholders should be included explicitly in future labor-workforce analyses and transportation modeling. The analyses in this report use the sample of workers who lived in Los Angeles County, California.View Full Paper PDF
-
Working PaperSocial, Economic, Spatial, and Commuting Patterns of Self-Employed Jobholders
April 2007
Working Paper Number:
tp-2007-03
A significant number of employees within the United States identify themselves as selfemployed, and they are distinct from the larger group identified as private jobholders. While socioeconomic and spatial information on these individuals is readily available in standard datasets, such as the 2000 Decennial Census Long Form, it is possible to gain further information on their wage earnings by using data from administrative wage records. This study takes advantage of firm-based data from Unemployment Insurance administrative wage records linked with the Census Bureau's household-based data in order to examine self-employed jobholders - both as a whole and as subgroups defined according to their earned wage status - by their demographic characteristics as well as their economic, commuting, and spatial location outcomes. Additionally, this report evaluates whether self-employed jobholders and the defined subgroups should be included explicitly in future labor-workforce analyses and transportation modeling. The analyses in this report use the sample of self-employed workers who lived in Los Angeles County, California.View Full Paper PDF
-
Working PaperErrors in Survey Reporting and Imputation and Their Effects on Estimates of Food Stamp Program Participation
April 2011
Working Paper Number:
CES-11-14
Benefit receipt in major household surveys is often underreported. This misreporting leads to biased estimates of the economic circumstances of disadvantaged populations, program takeup, and the distributional effects of government programs, and other program effects. We use administrative data on Food Stamp Program (FSP) participation matched to American Community Survey (ACS) and Current Population Survey (CPS) household data. We show that nearly thirty-five percent of true recipient households do not report receipt in the ACS and fifty percent do not report receipt in the CPS. Misreporting, both false negatives and false positives, varies with individual characteristics, leading to complicated biases in FSP analyses. We then directly examine the determinants of program receipt using our combined administrative and survey data. The combined data allow us to examine accurate participation using individual characteristics missing in administrative data. Our results differ from conventional estimates using only survey data, as such estimates understate participation by single parents, non-whites, low income households, and other groups. To evaluate the use of Census Bureau imputed ACS and CPS data, we also examine whether our estimates using survey data alone are closer to those using the accurate combined data when imputed survey observations are excluded. Interestingly, excluding the imputed observations leads to worse ACS estimates, but has less effect on the CPS estimates.View Full Paper PDF
-
Working PaperEmployer-to-Employer Flows in the United States: Estimates Using Linked Employer-Employee Data
September 2010
Working Paper Number:
CES-10-26
We use administrative data linking workers and firms to study employer-to-employer flows. After discussing how to identify such flows in quarterly data, we investigate their basic empirical patterns. We find that the pace of employer-to-employer flows is high, representing about 4 percent of employment and 30 percent of separations each quarter. The pace of employer-to-employer flows is highly procyclical, and varies systematically across worker, job and employer characteristics. Our findings regarding job tenure and earnings dynamics suggest that for those workers moving directly to new jobs, the new jobs are generally better jobs; however, this pattern is highly procyclical. There are rich patterns in terms of origin and destination of industries. We find somewhat surprisingly that more than half of the workers making employer-to-employer transitions switch even broadly-defined industries (NAICS supersectors).View Full Paper PDF
-
Working PaperNOISE INFUSION AS A CONFIDENTIALITY PROTECTION MEASURE FOR GRAPH-BASED STATISTICS
September 2014
Working Paper Number:
CES-14-30
We use the bipartite graph representation of longitudinally linked em-ployer-employee data, and the associated projections onto the employer and em-ployee nodes, respectively, to characterize the set of potential statistical summar-ies that the trusted custodian might produce. We consider noise infusion as the primary confidentiality protection method. We show that a relatively straightfor-ward extension of the dynamic noise-infusion method used in the U.S. Census Bureau's Quarterly Workforce Indicators can be adapted to provide the same confidentiality guarantees for the graph-based statistics: all inputs have been modified by a minimum percentage deviation (i.e., no actual respondent data are used) and, as the number of entities contributing to a particular statistic increases, the accuracy of that statistic approaches the unprotected value. Our method also ensures that the protected statistics will be identical in all releases based on the same inputs.View Full Paper PDF
