Supporting Statement - Section A

1

SUPPORTING STATEMENT
U.S. Department of Commerce
U.S. Census Bureau
Public Employment & Payroll Forms
(Forms E-1, E-2, E-3, E-4, E-5, E-6, E-7, E-8, E-9, E-10)
OMB Control Number 0607-0452
Part A. Justification
1. Necessity of Information Collection
The Census of Governments: Employment and its related program, the Annual Survey
of Public Employment & Payroll, provide a rich source of data on state and local
government employment and payroll in the United States. Data have been collected
annually since 1957. A census is conducted every five years (years ending in ‘2’ and
‘7’). A sample of state and local governments is used to collect data in the intervening
years, with a new sample selected every five years (years ending in ‘4’ and ‘9’). The
survey provides state and local government data on full-time and part-time employment,
part-time hours worked, full-time equivalent employment, and payroll statistics by
governmental function (e.g., elementary and secondary education, higher education,
police protection, fire protection, financial administration, central staff services, judicial
and legal, highways, public welfare, etc.).

Title 13, U.S.C., Section 161 requires the Secretary of Commerce to conduct a
census of governments every fifth year. Title 13, U.S.C., Section 182 allows the
Secretary to conduct annual surveys in other years. Within the Federal Register
notice dated August 28th, 2014, only Title 13, U.S.C., Section 161 was cited as the
legal authority for this information collection. On the advice of legal counsel, we
have added Title 13, U.S.C., Section 182 to that citation.
This information collection request covers the questionnaires needed to conduct the
public employment program for the 2015 Annual Survey of Public Employment &
Payroll, the 2016 Annual Survey of Public Employment & Payroll, and the 2017
Census of Governments: Employment.
The questionnaires for collecting the data are described below. There are ten survey
forms used to collect data on government employment, pay, and hours (see
Attachment 1). Since there are many different types and sizes of governments, each
form is tailored to the unique characteristics of the type and size of government or
government agency to be surveyed.

E-1

State Agencies
State agencies, excluding state colleges and universities

E-2

Institutions of Higher Education
State institutions of higher education colleges and universities

2
E-3

Special Districts and Local Agencies
Dependent agencies of local governments and single function special
district governments

E-4

Municipalities, Counties, Townships
County governments, municipalities, and township governments with
population of 1,000 or more

E-5

Municipalities and Townships
Shortened version of the E-4 form for municipalities and townships with
a population of < 1,000

E-6*

School Systems
Local government operated institutions of education, elementary &
secondary education and/or college & other postsecondary education

E-7

Major Special Districts and Agencies
Multifunction dependent agencies and fire protection agencies for local
governments, and multifunction special district governments

E-8*

Elementary and Secondary Education
Local government operated institutions of elementary and secondary
education

E-9

Police Protection Agencies
State and local government police protection agencies

E-10* College and Other Postsecondary Education
Local government operated institutions of higher education

*For 2012 and prior, form E-6 was sent to all local school systems regardless of
function. As a result of cognitive testing, two new forms were introduced for the
2013 Annual Survey of Public Employment & Payroll. The E-8 form was designed
for elementary and secondary school systems; the E-10 for college and other
postsecondary education school systems. The E-6 now is only sent to governmental
units which may have both functions of education. The units receiving the new form
are receiving a shorter form in place of the existing E-6.
The type of employment and payroll data to be collected by the public employment
& payroll program for the 2015 Annual Survey of Public Employment & Payroll,
the 2016 Annual Survey of Public Employment & Payroll, and 2017 Census of
Governments: Employment are identical to the data collected in recent annual
surveys. The 2015 and 2016 samples support estimates of total local government
employment and payrolls by state by government function (e.g., elementary and
secondary education, higher education, police protection, fire protection, etc.). The
2017 Census of Governments: Employment will collect data for all state and local
governments in the 50 states and the District of Columbia by type of government
and government function. The 2017 Census of Governments: Employment supports

3

tabulations of total local government employment and payrolls by state by
government function, as well as a county area employment and payroll tabulation.
Data are also gathered using means outside of a traditional form canvass. The
Census Bureau also collects data through a web reporting instrument and via central
collection arrangements with state governments.
No changes were made to the form content as currently approved. However,
formatting changes were made to the forms to facilitate data capture using current
technology, Integrated Computer Assisted Data Entry (iCADE), and to clarify
wording and form flow with respondents such as integrating the instruction in bullet
form into the questions. These changes were cognitively tested.

2. Needs and Uses

The Census of Governments: Employment and its related program, the Annual
Survey of Public Employment & Payroll, provide data on state and local
government employment and payroll in the United States. Census Bureau staff apply
a standard set of criteria while classifying government employment activity in order
to provide what is perhaps the only complete and uniform set of data on the
employment activities of governments in the United States.
Statistics compiled from data gathered using these forms are used in several
important Federal government programs. Economists at the Bureau of Economic
Analysis (BEA) use the statistics for developing the National Income and Product
Accounts. According to the Chief Statistician of BEA, “The data obtained from
these forms are critical to BEA for maintaining reliable estimates. Specifically,
BEA uses national, state, local, and type-of-government aggregate data by function
for full-time and part-time employees, payroll, and number of part-time hours
worked to prepare estimates of functional payrolls for the public sector of the gross
domestic product (GDP). BEA also uses these data to prepare estimates of state and
local government compensation of employees in both the annual and benchmark
input-output accounts.” (see Attachment 2)
BEA also uses the Census of Governments and the Annual Survey of Public
Employment & Payroll to derive state-level estimates of the employment and wages
and salaries of students and their spouses who are employed by public institutions of
higher education in which the students are enrolled. There is no other national or
state source for information on student workers at state institutions of higher
education.
The employment data are used for two other data collection efforts currently conducted
by the Census Bureau. The Medical Expenditures Panel Survey (MEPS) collects
data for the Department of Health and Human Services (HHS) on health plans
offered to state and local government employees. The MEPS sample of public
employees is drawn from the Census of Governments: Employment component
universe and employment data from the survey are used in statistical methods for

4

creating national estimates on health plans. The Criminal Justice Employment and
Expenditure program (CJEE), sponsored by the Bureau of Justice Statistics (BJS),
uses employment data to provide employee and payroll statistics on police
protection and correctional activities.
State and local government officials use these employment data to analyze and
assess individual government labor force and wage levels. Both management and
labor consult these data during wage and salary negotiations.
Public interest groups of many types produce analyses of public sector activities
using these data. User organizations representing state and local government
include the Council of State Governments, the National Conference of State
Legislatures, Government Research Association, U.S. Conference of Mayors,
National Association of Counties, National League of Cities, and the National
Association of Towns and Townships. Another category of users, having a more
specific focus on government activities, includes organizations such as the National
School Boards Association and the National Sheriffs Association.
A variety of private sector organizations and individuals make use of these
employment and payroll data. Notable research organizations include the Brookings
Institution and the Nelson A. Rockefeller Institute of Government.
Both public and private universities utilize these data. Instructors, researchers, and
students in schools of public administration, political science, management, and
industrial relations as well as other members of the public also use employment data.
Information quality is an integral part of the pre-dissemination review of the
information disseminated by the Census Bureau (fully described in the Census
Bureau’s Statistical Quality Standards). Information quality is also integral to the
information collections conducted by the Census Bureau and is incorporated into the
clearance process required by the Paperwork Reduction Act.
Beginning with the 1993 annual data series, summary tables, and state & local
individual unit data files have been released on the Internet. At the Internet site,
(www.census.gov/govs), users will find these products and additional survey
documentation.

3. Use of Information Technology
The Census Bureau is using information technology to reduce respondent burden in
the following ways:
a.

Electronic Submissions
All form types can be completed on the Internet. Central collection contacts
are able to upload their electronic data files through the Census Bureau’s
Centurion web instrument (See Attachment 3). Instructions for reporting on

5

the Internet are included in the original mailing as well as the follow-up
mailing.
We estimate that at least 65 percent of respondents will utilize the Census
Bureau’s web instrument or will submit electronic data via a central
collection agreement. This estimate is based on responses for the 2013 and
2014 sample-based surveys, and the 2012 Census of Governments:
Employment. The percentage of responses collected electronically is greater
for sample-based surveys. Electronic responses have shown an upward trend
in recent years as well.
Annually, data for 3,777 state agencies from 46 states are submitted
electronically as a result of central collection agreements with state
governments. The 2015 central collection mail-out includes a letter along
with requesting that data be submitted electronically. Central collection
respondents submit their data primarily in ASCII text or Excel format. The
central collection mail-out letter is included in Attachment 4.
Four state school systems provide their local government school system data
electronically. This is a total of 158 units for the 2015 and 2016 annual
surveys. For the 2017 Census of Governments: Employment, all local
government school systems for these four states are included, increasing the
total units to 616.
b.

Processing technology
Forms are checked in using an automated tracking system that generates and
distributes a daily activity report. This system tracks the movement of forms
through initial review, imaging, data entry, and data delivery to headquarters.
Imaging of the forms reduces the number of hours previously required to sort
and file forms upon receipt and to retrieve them for use by analysts.
Through the use of the Feith Image Retrieval system, analysts have
immediate access to multiple years of data on their desktop.
Starting in 2012, the web collection moved to the Census Bureau’s Centurion
web data collection system. Images for the completed web forms will also
be available in the Feith Image Retrieval system.
Data entry for all forms returned by mail began utilizing the Census Bureau’s
iCADE keying program in 2012. In 2013, we began utilizing an optical
character recognition (OCR) system in iCADE to capture data on the forms.
These processing changes have increased survey efficiency and reduced
costs.

4. Efforts to Identify Duplication

6

The Public Employment & Payroll program does not duplicate any other data
collection. Continuous contact with public interest groups and representatives of
other Federal agencies ensures the absence of duplication. Consultations with staff
of the National Center for Education Statistics (NCES) confirm that the request for
data from Colleges and Universities (Form E-2) does not duplicate NCES data
requirements. In some years, NCES collects data on full-time equivalent positions,
but it does not obtain data on actual numbers of persons employed or corresponding
actual gross pay amounts.
5. Minimizing Burden
The Census Bureau pursues central collection arrangements to obtain necessary data
and reduce respondent burden whenever possible. Through these arrangements, a
single contact agrees to collect and submit the data for their government and its
dependent agencies as one central report. The Census Bureau continues to pursue
electronic reporting arrangements with other governments. Through these
arrangements, government contacts provide an electronic file in a mutually
agreeable format and work with the Census Bureau to create the crosswalk for
functional breakouts of the data. Respondents spend less time in future years
finding and classifying employees into our functional categories.
The Census Bureau has central collection agreements with 46 states and four state
governments on behalf of their local school systems. This eliminates the need to
mail canvass approximately 3,777 state agencies each year, along with 158 school
systems in a sample year (616 in a census year).
Census Bureau staff provide assistance to governments that express difficulty in
completing forms due to limited resources or understanding of how to accomplish
this task. In some cases, Census Bureau staff compile the data from payroll files
submitted by the government.
The Census Bureau also implemented new sample design and estimation techniques
starting in 2009 that minimize total respondent burden. A modified cutoff sampling
methodology and small area estimation reduces the burden on small government
units.
6. Consequences of Less Frequent Collection
The Census Bureau's ability to produce reliable information on government
employment and pay would be significantly impeded if data were collected less
frequently. Government employment and payrolls are key indicators of state and
local government economic activity, providing the most current information about
expenditures by government function in this sector of the economy. These data are
also the basis for additional surveys conducted on an annual basis such as the MEPS
and CJEE.
7. Special Circumstances

7

There are no special circumstances related to this request.
8. Consultations Outside the Agency
The notice inviting comment on plans to submit this request was published in the
Federal Register (Vol. 79 No. 167) on August 28, 2014 on pages 51303 and 51304.
We received no comments in response to the notice. However, we did receive a
letter of support from BEA (Attachment 2). We thank BEA for its continued support
for this collection.
The Census Bureau conducts quarterly meetings with BEA to discuss information
about several surveys including the Survey of Public Employment & Payroll. BEA
uses the national, state, local, and type-of-government aggregate data by function on
full-time and part-time employees, payroll, and number of part-time hours worked to
prepare estimates of functional payrolls for the public sector of the GDP.
The Census Bureau maintains contact with public interest groups representing state
and local governments and groups of public officials and academic organizations.
These groups all use and disseminate data collected in the census and annual
surveys.
All opinions of individuals or groups were provided separately and informally, and
not for the purposes of reaching a consensus. They were not provided as a part of
and did not constitute any Federal Advisory Committee proceedings.
9.

Paying Respondents
Respondents are not paid for providing their data.

10.

Assurance of Confidentiality
The survey does not request any confidential data. These data are subject to
provisions of Title 13, United States Code, Section 9(b) exempting data that are
customarily provided in public records from rules of confidentiality. All forms
contain information regarding this exemption of confidentiality and the voluntary
nature of the survey. The 2015 survey forms are included in Attachment 1.

11.

Justification for Sensitive Questions
No questions of a sensitive nature are asked.

12.

Estimate of Burden Hours
The 16,386 respondents to the 2015 Annual Survey of Public Employment &
Payroll and the 2016 Annual Survey of Public Employment & Payroll are comprised

8

of all 50 state governments and a sample of local governments in the United States.
The 99,492 respondents to the 2017 Census of Governments: Employment are
comprised of all 50 state governments and all local governments in the United
States.
The following tables provide an estimate of the average number of respondents and
burden hours for the individual forms in the 2015 Annual Survey of Public
Employment & Payroll, 2016 Annual Survey of Public Employment & Payroll, and
2017 Census of Governments: Employment. Table 1 figures are based on the 2014
Annual Survey of Public Employment & Payroll. Table 2 figures are based on the
2012 Census of Governments: Employment (actuals of Forms E-3, E-4, E-5, E-7,
and E-9), and the 2014 Annual Survey of Public Employment & Payroll (actuals of
Forms E-1 and E-2, and estimates of Forms E-6, E-8, and E-10).

Table 1: 2015 Annual Survey of Public Employment & Payroll and
2016 Annual Survey of Public Employment & Payroll
Average Number of Respondents Counts And Burden Hours
Form Number

Estimated
Average Minutes
Estimated
Number of
per Response
Total Hours
Responses
E-1
1,532
20
511
E-2
727
60
727
E-3
4,830
20
1,610
E-4
4,627
90
6,941
E-5
311
75
389
E-6
1,517
45
1,138
E-7
1,132
20
377
E-8
1,270
45
953
E-9
190
30
95
E-10
200
45
150
Central Collection from State Agencies and Schools
State Agencies *
46
840
644
Schools **
4
840
56
Total

16,386

13,591

9

Table 2: 2017 Census of Governments: Employment
Average Number of Respondents Counts And Burden Hours
Form Number

Estimated
Average Minutes
Estimated
Number of
per Response
Total Hours
Responses
E-1
1,532
20
511
E-2
727
60
727
E-3
31,510
20
10,503
E-4
21,184
90
31,776
E-5
17,743
75
22,179
E-6
7,565
45
5,674
E-7
11,589
20
3,863
E-8
6,329
45
4,747
E-9
265
30
133
E-10
998
45
749
Central Collection from State Agencies and Schools
State Agencies *
46
840
644
Schools **
4
840
56
Total

99,492

81,562

* Central collection arrangements with 46 state governments are each treated as a
single response with an estimated reporting time of 14 hours (840 minutes) each.
These arrangements reduce the number of state agency forms by 3,777.
** In four states, central collection arrangements with school systems are each
treated as a single response with an estimated reporting time of 14 hours (840
minutes) each. These arrangements reduce the number of school forms by 616 in a
census year and 158 in a sample year.
Based on the tables above, the estimated average annual burden hours are 36,248
and the estimated number of respondents are 44,088. The calculation is an average
of the 2015 and 2016 sample-based surveys, and the 2017 Census of Governments:
Employment.
The Federal Register Notice dated August 28, 2014 had different figures for annual
burden hours and estimated number of respondents. The numbers presented in this
request supercede those earlier estimates and reflect a slight refinement of the
estimated number of respondents.

10

13. Estimate of Cost Burden
The Census Bureau does not expect respondents to incur any costs other than that of
their time to respond. The information requested is of the type and scope normally
carried on government reports or financial statements and no special hardware or
accounting software or system is necessary to provide answers to this information
collection. Therefore, respondents are not expected to incur any capital and start-up
costs or system maintenance costs. Further, purchasing of outside accounting or
information collection services, if performed by the respondent, is part of usual and
customary business practices and not specifically required for this information
collection.
14. Cost to Federal Government
The estimated cost to the Federal Government to conduct the 2015 Annual Survey
of Public Employment & Payroll is $2.2 million. The cost of conducting this survey
for the 2016 will approximate these costs.
The estimated total cost to the Federal Government to conduct the 2017 Census of
Governments: Employment is $7.9 million for its 5-year cycle.

15. Reason for Change in Burden
The current OMB inventory was based on the previous submission, which included
one annual survey year (averaging ~13,500 burden hours) and one Census year
(averaging ~81,500 burden hours). The new OMB submission’s burden statement
cited above is an average of two annual survey years and one Census year; thereby,
decreasing the average burden hours.

11

16. Project Schedule

Table 3: 2015 Annual Survey of Public Employment & Payroll Timetable
Date or Period (Month/Year)
12/2014

Activity
Complete design of mailout/mailback forms

01/2015

Complete web instrument testing

03/2015

Initial request for mailout/mailback forms and
web submissions

05/2015 - 08/2015

Follow-up requests for mailout/mailback forms
and web submissions

03/2015 - 12/2015

Collect and process central collection data

05/2015 - 08/2016

Micro and macro level data review

By 05/2016

Produce initial national estimates for BEA’s
annual NIPA revision

By 09/2016

Release 2015 Annual Survey of Public
Employment & Payroll report and tables.
Release state and local government individual
data files by request.

Table 4: 2016 Annual Survey of Public Employment & Payroll Timetable
Date or Period (Month/Year)
12/2015

Activity
Complete design of mailout/mailback forms

01/2016

Complete web instrument testing

03/2016

Initial request for mailout/mailback forms and
web submissions

05/2016 - 08/2016

Follow-up requests for mailout/mailback forms
and web submissions

03/2016 - 12/2016

Collect and process central collection data

05/2016 - 07/2017

Micro and macro level data review

By 05/2017

Produce initial national estimates for BEA’s
annual NIPA revision

By 08/2017

Release 2016 Annual Survey of Public
Employment & Payroll report and tables.
Release state and local government individual
data files by request.

12

Table 5: 2017 Census of Governments: Employment Timetable
Date or Period (Month/Year)
12/2016

Activity
Complete design of mailout/mailback forms

01/2017

Complete web instrument testing

03/2017

Initial request for mailout/mailback forms and
web submissions

05/2017 - 12/2017

Follow-up requests for mailout/mailback forms
and web submissions

03/2017 - 04/2018

Collect and process central collection data

05/2017 - 02/2019

Micro and macro level data review

By 05/2018

Produce initial national estimates for BEA’s
annual NIPA revision

By 03/2019

Release 2017 Census of Governments:
Employment report and tables. Release state
and local government individual data files by
request.

17. Request to Not Display Expiration Date
The expiration date will appear on all forms.
18. Exceptions to the Certification
There are no exceptions.