LEHD Supporting Statement A (1220-0012) 2017 FINAL

Supporting Statement A - Justification

Longitudinal Employer-Household Dynamics

0607-XXXX

1. Necessity of the Collection

A 21st century statistical system must provide information about the dynamic economy quickly, using data assets efficiently while minimizing the burden of collecting and providing data and fully preserving confidentiality. The Census Bureau’s Longitudinal Employer-Household Dynamics (LEHD) program has demonstrated the power and usefulness of linking multiple business and employee data sets with state-of-the-art confidentiality protections to build a longitudinal national frame of jobs.

The authority to conduct the LEHD program is 13 U.S.C. Section 6; confidentiality is assured by 13 U.S.C. Section 9. This program supports the Department of Commerce plan to improve American competitiveness and measures of innovation. It provides federal, state, and local policymakers and planners, businesses, private sector decision makers, and Congress with comprehensive and timely national, state, and local information on the dynamic nature of employers and employees. The LEHD program significantly reduces the overall effort for the generation of its quarterly data product by:

• Leveraging existing federal administrative and state data

• Avoiding costs required to expand existing surveys to collect the information directly

• Reducing respondent burden by limiting the number of required resources to the owners of the required data

2. Use of Information

The LEHD program relies upon the LED (Local Employment Dynamics) partnership between the US Census Bureau and the Labor Market Information (LMI) agencies from 49 states, the District of Columbia, and the territories of Puerto Rico and the Virgin Islands. This partnership supports the development, promotion, and distribution of the following data products:

• QWI Public Use - The flagship data product of the LEHD program is the QWI Public Use which provides 32 statistical indicators such as those on employment, job creation and destruction, accessions (e.g., hires and recalls), and separations (e.g. exits and layoffs). These statistics are released for the following by-groups for all quarters for which data are available for each partner state:

• County, metropolitan, and workforce investment area

• Age, sex, race, and ethnicity categories

• Detailed industry (i.e., type, firm age, frim size)

• LEHD Origin Destination Employment Statistics (LODES) - LODES data provide detailed spatial distributions of workers' employment and residential locations and the relation between the two at the Census Block level. LODES also provides characteristic detail on age, earnings, industry distributions, and local workforce indicators.

• Job-to-Job Flows (J2J) - Job-to-Job Flows (J2J) is a new set of statistics on worker reallocation in the United States constructed from the LEHD data. The initial release of national data distinguishes hires and separations associated with job change from hires from and separations to non-employment. Future releases will be published at more detailed levels of aggregations, and will tabulate the origin and destination job characteristics of workers changing jobs.

These data products highlight state and local labor market dynamics that cannot be learned from other statistical sources and are therefore used in many different arenas. For example, the QWI can be used as local-labor-market controls in regression analysis; to identify long-term trends; to provide local context in performance evaluations, and a host of other applications. Additional indicators will be developed as the program expands. Measures of individual earnings dynamics across consecutive quarters are already in the internal system but have not been released. Additional indicators of cross-state flows and inter-industry mobility are in development.

2. Use of Electronic Collection Methods

The data products created by the LEHD program are not generated by a traditional survey. Rather, all data required is collected electronically as follows:

• State Unemployment Insurance (UI) and Quarterly Census of Employment and Wages (QCEW) are provided from each state LMI agency via secure File Transfer Protocol (FTP) directly to the Census Bureau. This transfer of data is governed by a Memorandum of Understandings (MOUs) with each state partner.

• Federal and Census Administrative data are acquired from other directorates or divisions within the Census Bureau where an internal agreement has been established for the use of the data.

• Public Use data sets are acquired from public source websites or public FTP servers.

4. Efforts to Identify Duplication

The LEHD program is not directly responsible for the collection of the information provided in the various data files used to create our data products. Therefore, our activities generate no duplication. However, while the States only provide the UI data to the Census Bureau, they do provide the QCEW data to the Bureau of Labor Statistics (BLS).

5. Impact on Small Businesses

The key requirement for the development of the QWI infrastructure and public use data products is not to burden respondents to participate in a traditional survey. As such, the collection of the required data has no impact on the day-to-day operations of small businesses.

6. Consequences of Less Frequent Collection

The LEHD program relies on data that have already been collected for other administrative purposes. Because these source data are updated quarterly, the QWI data products are designed for quarterly output. Specifically, the LEHD infrastructure and its data products use these quarterly data to portray the most recent information in employer/employee dynamics as well as an evolving history of this interaction.

The quarterly nature of our core data allows for more meaningful statistics to be made available to the public. Changing the frequency of obtaining these data from state partners will not impact the frequency of collection by states (as law requires the quarterly collection of Unemployment Insurance information) but will decrease the usefulness of data products we produce using these administrative data. It could also put at risk the analysis that is currently being performed by those relying on the quarterly LEHD statistics.

7. Special Circumstances

There are no special circumstances for the collection of the required input data sets

8. Federal Register Notice and Consultation Outside the Agency

On December 12, 2016 the general public and other Federal agencies were, via the Federal Register (Citation Number: 81 FR 88662), invited to provide comments regarding the LEHD program, its methods for collection of required data, and efforts to be in compliance with the Paperwork Reduction Act of 1995.

This notification can be found at:

https://www.federalregister.gov/documents/2016/12/08/2016-29395/proposed-information-collection-comment-request-longitudinal-employer-household-dynamics-lehd

Comment period closed on February 6, 2017 with one comment received (Attachment A) and Census Bureau response (Attachment B).

The LED partnership, as described in Section 2 of this document, is governed though a steering committee which is made of up regional representatives of all state partners and subject matter experts from the LEHD program (i.e., administration, project management, research, and dissemination). It is through this committee that the following occurs:

• Recommendations for and evaluations of new data products

• Recommendations for and evaluations of enhancements to the current data products

• Consultation on and review of changes to the data products

• Consultation on promotion of training on the data products and supporting web applications

• Consultation on the planning and execution of the annual LED Partnership Workshop which provides a venue to promote use of the data products and supporting web applications

The LEHD program also gains input on its data products by regular consultations with the academic community and other Federal agencies.

9. Payment to Respondents

No payment is made to any agency which provides data to the LEHD program

10. Assurance of Confidentiality

All tabulations from the LEHD data may be released only if they follow a procedure approved by the U.S. Census Bureau’s Disclosure Review Board (DRB), the primary body governing the release of any public use data. The partnership MOUs also specify guidelines governing the release of data at the state or sub-state level, such as a requirement of state approval for use of their data for specific research projects. There are, broadly, two types of activities that require researchers to go before the DRB which are:

• Regular production of one of LEHD’s data products such as the Quarterly Workforce Indicators, OnTheMap/LEHD Origin-Destination Employment Statistics (LODES), etc. For these, procedures are approved in advance of the regular production of the statistical data product. The data product is approved for release thereafter provided the protocols the DRB approves are followed.

• Published tabular summaries of linked employer-employee data usually use a job frame (e.g., statutory employer linked to a specific employee) but include characteristics of both the individual (employee) and workplace (employer establishment). Formal privacy protection of these characteristics requires identifying the sensitivity of the published statistic to variation in a single individual or a single workplace (establishment).

11. Justification for Sensitive Questions

The LEHD program only uses administrative data that have been compiled by other Federal and State agencies. As such, no questionnaire is used that would contain any sensitive questions.

12. Estimates of Respondent Burden

The Program’s use of administrative data was designed to minimize the level of effort (LOE) by partner agencies when providing their data. Some of the data sets used are from public sources or from within the Census Bureau. Loading of these data sets is accomplished by employing commercially-available technology for secure file transfer. The amount of time needed to collect and transfer data from state partners is also small. It is estimated that the LOE needed for this transfer is less than 4 hours per state partner per quarter. However, additional effort may be necessary by a state agency when there is a discrepancy such as an incorrect address or a missing observation in the data set that needs correction. In this case, the LEHD analysts work with the sending agencies to evaluate the error and, if necessary, provide a recommended method for remediation.

Overall we estimate that the total number of annual burden hours to be approximately 1,664 man hours. This estimate is based on the following:

• Each state partner (52) will spend an average of 8 man hours per annual quarter to prepare and send their data to the Census Bureau as well as resolve any data quality issues (i.e., resending of data). This would be approximately 1,664.

• Based on an estimated average loaded per hour labor rate of $50.00, the estimated costs for the States to provide their wage record data to the Census Bureau is $83,200.

13. Estimate of Cost Burden to Respondents

Minimal additional costs to respondents are anticipated as processes for moving data onto appropriate agency data sharing servers and uploading of data to Census secure servers in fairly automated.

14. Cost to the Government

The estimated costs to generate the LEHD infrastructure files and subsequent data products are best defined in the following table:

Product	Description	Level of Effort Annual	Estimated Cost Annual
QWI Infrastructure	Generate the intermediate micro-data files used to create the public use data products as well as investigate enhancements and implement changes to the SAS code.	9 FTE	$ 1,423,000
QWI	Generate the quarterly public use data product including population of the web application data base.	1.5 FTE	$ 240,000
LODES	Generate the annual release of this public use data product. Includes review of privacy protection schemes, inclusion of new data time lines, and population of the supporting web application data base. Note: Updates can occur during the year to address found issues with the data.	2 FTE	$ 320,000
Job-to-Job Flows	Generate the quarterly release of this public use data product including review and incorporation of enhancements.	4 FTE	$630,000

15. Changes in Respondent Burden

Since the LEHD program uses administrative data and the process for the access to this data is a standard process, there are no real or significant changes to time required by a respondent (i.e., more or less burden that what was originally planned) to provide the required data. In order to promote participation by the state partners, the LEHD program continues to review and employ methods that would reduce the burden for providing the required data.

16. Publication Plans

The LEHD data products and supporting web applications are for public use and can be accessed from the LEHD website: https://lehd.ces.census.gov.

17. OMB Expiration Date Display

As the LEHD program is not a traditional survey which requires extensive development of various documents (e.g., respondent survey), the program is requesting OMB approval for permission to suppress the expiration date as there are no specific documents to which this information is attached.

18. Exceptions to Certification Statement

There are no exceptions to the “Certification for Paperwork Reduction Act Submissions.”

9