3

Business Register

For a national statistical agency, having a single Business Register that is used as a frame from which to draw the samples for all of its business surveys creates considerable efficiencies. The use of a single register also facilitates coordination and harmonization across an agency’s business data collections. The U.S. Census Bureau’s Business Register is intended to be a universe listing of all U.S.-based businesses, including the U.S. subsidiaries of foreign companies. The completeness and accuracy of the Business Register is the foundation for obtaining quality statistics from the economic censuses and the Bureau’s annual, quarterly, and monthly economic surveys.

The samples for the annual economic surveys are drawn from the Business Register, meaning that one cannot think about redesigning the annual economic surveys without also thinking about the role of the Business Register in carrying out those surveys. At the same time, information collected through the annual economic surveys can be used to improve the accuracy and timeliness of the Business Register, benefitting future years’ annual survey data collections and ultimately the integrated Annual Business Survey System (ABSS) the panel recommends

In this chapter, we begin with background information about the Business Register, to set the stage for our consideration of ways to improve it. In the last part of the chapter we discuss a program for managing interactions with the largest businesses in the register to facilitate both the maintenance of information about them in the register and their response to the various survey requests they receive.

3.1 THE CURRENT CENSUS BUREAU BUSINESS REGISTER

The Business Register is a listing of all legal business entities—incorporated businesses, partnerships, and sole proprietorships—operating in the United States and its territories (island areas) as identified by the U.S. Internal Revenue Service (IRS). It lists businesses that have paid employees (i.e., employer businesses), of which about 5 million have only one location and 160,000 have more than one location. It also lists non-employer businesses, of which there are about 21 million. In somewhat different words, it is the Census Bureau’s master list of businesses.¹

3.1.1 Uses and Units Included

The primary use of the Business Register is to identify target populations for the Census Bureau’s economic statistics programs. Samples for the Bureau’s monthly and quarterly indicator surveys, annual economic surveys, and quinquennial economic censuses are developed on the basis of the information contained in the register. The universe for the economic censuses comprises all non-employer and employer businesses in the register. For some of the annual surveys, both kinds of businesses are included, while for other surveys only employer businesses are included, which are far fewer in number but account for the bulk of economic activity (see Chapter 5).

The Business Register also is a central repository of business-related administrative data, primarily federal tax data, which support a variety of statistical products. These products include the annual County Business Patterns statistics on employment and payroll by industry for U.S. counties (U.S. Census Bureau, 2017); the annual Statistics of U.S. Businesses on the distribution of economic activity by enterprise size and industry (U.S. Census Bureau, 2016); and the Business Dynamics Statistics, which document the job creation and job destruction underlying net changes in employment (Haltiwanger, Jarmin, and Miranda, 2008; Haltiwanger, Miranda, and Jarmin, 2013).

The businesses included on the Business Register have varied structures. Some are simple businesses consisting of a single establishment, some have relatively simple structures but include multiple establishments, and some are complex enterprises that encompass many types of business operations. Two basic types of statistical units are included on the Business Register, establishment and enterprise:

___________________

¹ The material in this section is based on a presentation by Brandy L. Yarbrough, U.S. Census Bureau, “Business Register: Concepts, System, & Maintenance,” during the panel’s December 14, 2015, meeting. See also Davie, Jr. (2015) and Yarborough (2016).

1. An establishment is an economic unit, generally at a single physical location, where business is conducted or where services or industrial operations are performed. The intersection of a physical location and an Employer Identification Number (EIN) is the smallest discrete unit for employer businesses on the Business Register.
2. An enterprise (which is also called a business or a company) is an economic unit comprising one or more establishments under common ownership or control. An enterprise on the Business Register may comprise a single establishment, a U.S. parent and its U.S.-based establishments (but excluding any establishments abroad), or a U.S. subsidiary of a foreign-owned company, together with any associated establishments that are physically located in the United States.

In addition to these standard types of statistical units, the Business Register includes special statistical units that are used for reporting output in the economic censuses. The Census Bureau creates these units when location-based reporting of output is not practical, by consolidating two or more establishments owned by the same enterprise. (The economic censuses still collect employment and payroll by establishment.) Such consolidated units are limited to about 30 industries, including insurance, commercial banking, and wireless telecommunications. The annual economic surveys also may create special units to facilitate reporting by companies, but these may not conform to the special units in the economic censuses. Only the former are maintained on the Business Register.

In addition to information on different types of statistical units, the Business Register includes information on two different types of administrative units, entities identified by EINs and those identified by Social Security numbers (SSNs):

1. An EIN entity is an administrative unit created by the IRS for tax reporting purposes. Each EIN is associated with a particular enterprise, but any given enterprise may have multiple EINs associated with it. For example, a business might use one EIN for income tax reporting purposes and one or more different EINs for payroll tax reporting purposes.
2. An SSN entity is an individual engaged in business as a sole proprietor and identified by the person’s SSN for federal tax reporting purposes. If a person has multiple businesses, his or her SSN may be associated with multiple enterprises. SSN units account for the large majority of nonemployer businesses.

There is great variation in the complexity of business organizations. For a single-establishment employer business, as defined by the Census

Bureau, the enterprise, the establishment, and the EIN unit are the same. The relationships of the units within larger companies can be very complex, in some cases involving thousands of establishments and hundreds of EIN units. Figure 3-1 illustrates the sorts of relationships that may exist among the enterprise, establishments, and EIN units.

3.1.2 Input Data Sources

The Business Register combines data from multiple sources with the goal of providing comprehensive, accurate, and timely coverage of business units. Administrative records are the foundation of the Business Register. The primary data for identifying businesses come from the IRS, which provides information from its Business Master File (BMF), income tax returns, and quarterly payroll tax returns. The IRS provides updates to the Census Bureau for each of these types of records on a weekly basis. The BMF records are a source of information on name, address, and legal form of organization for all of the EIN entities of which the IRS is aware. Tax records provide information on revenues, assets, inventories, payroll, employment, and industry. EIN applications filed with the IRS and processed by the Social Security Administration are shared with the Census Bureau on

a monthly basis and provide North American Industry Classification System (NAICS) codes for new businesses.

The Bureau of Labor Statistics (BLS) is another important source of industry information. Each quarter, the Census Bureau prepares a listing of unclassified or partially classified EINs to refer to the BLS for comparison with its business register. The BLS provides approximately 30 percent of industry codes for EINs that appear on the Census Bureau’s list, mostly for small employers.² In addition to providing data that would otherwise be missing, this operation helps to make the Census Bureau’s register more consistent with the separate register used by BLS for its business surveys.

Over time, businesses may experience changes in their structure and in the nature of their activities. The Census Bureau’s Business Register is updated continuously to provide a set of statistical units that are as complete and accurate as possible for use in conducting the economic censuses and surveys. The sources for these updates include not only income and payroll tax records from the IRS, but also data from two annual surveys that are conducted by the Census Bureau specifically for the purpose of updating the register and data from the quinquennial economic censuses—the Company Organization Survey (COS, formally the Report of Organization Survey) and the Business and Professional Classification Survey (known as SQ-CLASS).

COS is an annual survey designed to collect updated information on the structure of companies on the Business Register. Large multi-establishment companies with 500 or more employees are selected for the COS with certainty. Small multi-establishment and single-establishment companies are selected on the basis of administrative data that indicate there may have been a change in the organization of the enterprise. Roughly 42,000 enterprises are surveyed for the COS each year. The COS collects information on organizational changes, including births, deaths, acquisitions, sales, and changes in payroll tax structure (that is, changes in the EINs the company is using for payroll tax reporting purposes). The intention is to capture all of these characteristics at the establishment level of the company for the companies that are surveyed.

The purpose of SQ-CLASS is to identify a sample of new EIN births that the IRS has newly assigned or recently reactivated to update the Current Retail, Wholesale, and Services surveys. Only EINs that are either unclassified or classified in Retail, Wholesale, and Services are included in the sample. The by-product of this process provides classification and company organization updates to the BR. This survey, which began as a monthly data

___________________

² Based on presentation by David Talan, Bureau of Labor Statistics, “BLS Business Register: Quarterly Census of Employment and Wages—An Overview,” during the panel’s February 5, 2016, meeting (see also Stang, 2016).

collection in 1968, has been administered quarterly since 1976. A new sample of 10,000 to 13,000 eligible EINs with estimated payroll (new and recently activated) is selected each quarter. All eligible EINs with estimated payroll or sales above a specified measure of size, which varies by industry, are selected with certainty. The data collected include 2 months of sales or receipts, principal lines of merchandise, company organization, NAICS codes, wholesale inventories, and other industry-related data.

The Business Register also incorporates information collected in the Annual Survey of Manufactures (ASM), which includes the COS content, but not information collected in any of the other annual economic surveys. For years ending in 2 and 7, the economic censuses replace the COS and ASM as a source of data for updating the Business Register. For a substantial portion of the enterprises included on the Business Register, the economic censuses provide the most complete, or the only, updating of information available.

3.1.3 Content

The Business Register records several dozen attributes for each statistical unit, including organizational information that links associated establishments and EIN entities within an enterprise. Some of the information in the Business Register directly supports statistical program uses, including survey frame construction and access to administrative data. Other information is used by Business Register systems to perform internal management and auditing functions. The core attributes and characteristics recorded on the register include

• primary and secondary business name;
• mailing and physical addresses;
• NAICS codes;
• legal form of organization of the business;
• measures of size for payroll, receipts, and inventories;
• the operational status of the business (whether it is currently active, temporarily inactive, closed, or under new ownership);
• affiliates of the business (e.g., any separate company in which the business has less than 50 percent ownership or, in retail, a separate company that markets its products through the business); and
• linkages to other units.

Address information is important both for contacting a business and for identifying where its activities are occurring. All single-establishment enterprises on the Business Register have mailing addresses derived primarily from IRS tax records. Note, however, that these addresses are for the “per-

son responding” and may not be the physical location of the establishment. The physical location is recorded separately. In some cases the mailing address on file may be for a third party, such as an accountant or payroll processing company. For multi-establishment enterprises, the parent enterprise has a mailing address, and all of its establishments have physical location addresses. The parent mailing address ideally is that of the “person responding” for the company.

3.2 IMPROVING THE BUSINESS REGISTER

Over time, the Census Bureau has made significant improvements to the Business Register, and we understand that planning for a redesign is under way. The panel’s recommendations at the end of this section call for integrating the Business Register more closely with the annual economic surveys. We believe the recommended changes would improve the quality of the information stored on the register and contribute to more efficient production of data from the surveys that better meet the needs of data users.

Specifically, the Business Register would be more useful as a sampling frame for the annual economic surveys if it included information about special reporting units that are used for one or another of the surveys; was updated with a wide range of data sources, including information from the annual surveys themselves; and recorded information about when and for which survey(s) each unit on the register is sampled and responded, which would help in a plan to reduce reporting burden on businesses. In addition, the surveys that are currently used to update the register (COS and SQ-CLASS) could usefully collect information, such as on special reporting units and product classifications, that is needed for the annual economic surveys, and, ultimately, for an ABSS.

3.2.1 Statistical Unit Definition

An important issue in the design of samples for the annual and other economic surveys is the choice of the statistical unit to be surveyed. One limitation of the current Business Register design is that it allows for only two types of basic statistical units—enterprises and establishments. The special reporting units used for the economic censuses are recorded on the register, but not the special reporting units used in the annual economic surveys. Information about divisional headquarters locations or other regional locations at which aggregate data might be available is not systematically collected and recorded. Nor is information recorded on which units are included in one or another of the annual economic surveys.

The standard unit for the collection of data in the economic censuses is an establishment. In contrast, in the majority of the annual business

surveys the panel reviewed, the standard unit for data collection is an enterprise. Whether the desired unit is an establishment or an enterprise, it may not be practical or appropriate for a business to report at the requested level. As already noted, in the economic censuses, special reporting units for multi-establishment businesses are defined that consist of subsets of establishments for which combined data are reported. In the annual economic surveys, reporting units consisting of all of the establishments of a multi-industry business in a particular NAICS industry may be defined. Information about the reporting units used in the annual economic surveys is maintained by the survey staff as part of survey processing, rather than being maintained on the Business Register.

Other statistical organizations have developed methods for maintaining richer information about business structure on their business frames. Models for doing so that may be informative are offered by Statistics Canada (Da Pont, 2014), Statistics Netherlands (2015), and the Dun & Bradstreet Corporation.³ This is also a topic that has been considered by the OECD Wiesbaden Group on Business Registers (Vennix, 2012). To the extent that special reporting units will continue to be used in the annual economic surveys, there would be advantages to maintaining information about them on the Business Register rather than leaving this responsibility to the survey staff. The information could include key contact information, such as the name of the unit, addresses of all the physical locations included in the unit, telephone numbers, and e-mail contact addresses. It also would be desirable to have the name(s) of the individual(s) who routinely provide the data for the reporting unit.

Incorporating information about the special reporting units used in the annual economic surveys into the Business Register should be helpful for sample design and selection, as well as for facilitating coordination across the different surveys. Having a systematic process for adding information collected through the annual economic surveys to the Business Register, such as we recommend below, can help to keep the register information on special reporting units up to date.

3.2.2 Frame Content

For the Business Register to be relevant to the annual economic surveys, it is important that all of the key variables relevant to sampling for the surveys be included on the frame. At present, the main variables used for this purpose are industry classifications, measures of size (e.g., sales or payroll), and geographical identifiers. Looking to the future, we anticipate

___________________

³ Based on presentation by Anthony Scriffignano, Dun & Bradstreet, “Understanding Dun & Bradstreet Data,” during the panel’s February 5, 2016, meeting.

that codes for product information, using the North American Product Classification System, also could become important for sampling purposes. With this information in mind, thought should be given, as part of any redesign of the Business Register, to ensuring that fields for recording product information are available and to including that information in the COS and SQ-CLASS data collection.

3.2.3 Frame Maintenance

Another important future consideration is the possibility that the Census Bureau may turn to new sources of information for the Business Register in addition to those that have been important in the past. The annual COS and quarterly SQ-CLASS have been the primary drivers of register updates in the years between the economic censuses, but the Census Bureau collects a considerable amount of relevant business data through the annual economic surveys. One important innovation could be to make more systematic use of the information collected on these other surveys for updating the Business Register. For example, if new contact information for an enterprise is obtained in the course of conducting an annual survey, that information could be added to the Business Register as a matter of routine. Similarly, if the Census Bureau adopts our recommendation (see below) to create an Account Manager Program for the largest enterprises, account managers also could be an important source of information for updating the Business Register.

A long-standing issue for the federal statistical system has been the fact that the economic and business surveys conducted by the Census Bureau and those conducted by the Bureau of Labor Statistics rely on samples drawn from separate Business Registers, with different strengths and weaknesses (Fairman et al., 2008; Fixler and Landefeld, 2006). Because these registers rely on different underlying source data, the two lists often assign the same establishment to different industries or record the establishment with a different employment level. Employment by geographic area also differs between the two registers, reflecting in part the fact that the BLS collects more comprehensive information about the location of individual establishments through its Multiple Worksite Report than is obtained through the COS (Bureau of Labor Statistics, 2010). In an ideal world, the two agencies would be able to share their register information freely and, together, construct a combined register that reflects the strengths of both agencies’ source data. Although legal barriers currently prevent that ideal world, it would be appropriate for the Census Bureau to anticipate the possibility that, in the future, the BLS register information can be imported into the Census Business Register. Planning now for that possibility could prevent later difficulties.

Even more ambitiously, the Census Bureau may wish to prepare for the possibility that an even broader array of data sources might be used to populate the register. Relevant data sources could include not only administrative and survey data, such as are currently used, but also a variety of private and commercial data sources. For example, credit card companies have data on the location of retail establishments. To the extent that the register expands the number of sources on which it draws, it may be desirable to allow for multiple fields containing information on particular variables, with associated fields for recording the source of the information. Because enterprises are dynamic, with ongoing changes in their structure and activities, it also would be useful to record the date on which information was obtained.

3.2.4 Sample Selection Information

The sample design and designation of sampled units, both certainty and other units is largely uncoordinated across the annual economic surveys (see discussion in Chapter 5). The Census Bureau could more readily develop ways to harmonize the sample designs so as to avoid imposing undue reporting burden on particular businesses if the Business Register were to record information about when and in which survey a statistical unit is included in the sample. Such information will be imperative to include in the register’s data structure when an ABSS is implemented and processes are established to rotate sampled units.

3.2.5 Recommendations on Business Register Redesign

RECOMMENDATION 3-1: In the planned redesign of its Business Register, which the panel supports, the Census Bureau should take steps to make the register more useful for the annual economic surveys and ultimately for an Annual Business Survey System by:

1. 3-1a. creating a database structure that generally supports the maintenance of richer information on corporate structure and specifically supports the inclusion of information on the special reporting units used in the annual economic surveys;
2. 3-1b. enhancing the content of the register with product classification codes once they are developed, as these codes will likely be useful for designing survey samples;
3. 3-1c. modifying the two surveys—Company Organization Survey and the Business and Professional Classification Survey—currently used to update the register to collect the recommended information in points a and b above;

1. 3-1d. adopting processes for routinely incorporating in the register updated information about business structure from the Bureau’s annual economic surveys and other economic data collections; and
2. 3-1e. providing for the possibility, in the future, of broader sharing of information from the Bureau of Labor Statistics, which would permit both agencies to improve their business registers, and of using information from other government, private, and commercial sources to update and enrich the register.

RECOMMENDATION 3-2: As part of the Census Bureau’s planned redesign of its Business Register, it should make provision to record information each time a statistical business unit is sampled and responds in one of the annual economic surveys. This information will assist the Bureau to identify businesses that may be unduly burdened and to determine a cost-effective sample design for an Annual Business Survey System.

3.3 ACCOUNT MANAGER PROGRAM

Given the highly skewed distribution of business size, the largest enterprises play an especially important role in the production of business statistics. They may be asked to respond to a large number of survey requests over the course of a year, and because they account for a disproportionate share of employment and output, it is especially important that their responses be obtained. Thus, survey organizations, such as the Census Bureau, have a reason to identify the most effective ways to communicate with these enterprises and facilitate their reporting of requested information.

Groundbreaking research on these topics was carried out at the Census Bureau in the late 1990s and early 2000s (Sudman et al., 2000; Willimack, Nichols, and Sudman, 2002). A major conclusion from this research was that the largest companies would benefit significantly from some tailoring of the Census Bureau data requests to which they are subject. Common complaints heard from these companies were that survey requests too often were redundant and that the schedule for data collection for all of the surveys a company would be receiving was not generally communicated in advance. It also was not always clear who from the company could provide the information requested on a particular survey.

One of the Census Bureau’s responses to these findings was to establish an Account Manager Program for outreach for the 2012 economic

censuses.⁴ Under this program, individual analysts were assigned responsibility for specified enterprises. The program included telephone interviews and personal visits and is believed to have resulted in higher-response rates from covered enterprises. Although the program was viewed as successful, it was not extended to the balance of the economic statistics programs. More recently, some of the staff working on the annual economic surveys have been asked to assume the role of account manager for one or a small number of enterprises alongside their other responsibilities. Based on information provided to the panel, however, this loosely structured effort appears to have been less than fully successful in achieving its desired objectives.

Statistics Canada offers an example of a different model of how to manage relationships with large and complex business enterprises while at the same time improving the quality and timeliness of the data received. The Statistics Canada Enterprise Portfolio Management (EPM) Program was initiated in 2006. EPM is responsible for managing all aspects of the ongoing relationships with Canada’s 320 largest and most complex businesses (Lavoie, 2015). The EPM staff are responsible for building relationships with the companies and providing them with a single point of contact for survey data collection. Each company in the program is provided a schedule for all of the surveys in which it will be asked to participate during the year, and the redundant content across surveys is identified. EPM staff also are expected to maintain accurate and current information on the legal and operating structures of the companies for which they are responsible on Statistics Canada’s Business Register. New EPM staff members receive training on the Business Register and must demonstrate an appropriate level of mastery of the relevant concepts and procedures before they are authorized to enter information on the register. Establishing a team of dedicated and trained staff to play these roles has provided many benefits to Statistics Canada’s economic statistics program, including high unit and item response rates, improved reconciliation of initially inconsistent responses, and improved timeliness for the data from businesses assigned to EPM staff (Da Pont, 2014; Hughes, 2008; Sear et al., 2007).

Snijkers and Jones (2013, pp. 421-422) discuss special business survey communication procedures for large businesses, noting that many national statistical institutes (including Statistics Canada, Statistics Netherlands, and Statistics Sweden) have established integrated communications with the objective of getting timely, accurate and complete data over all surveys for each designated large business. Specially trained groups of managers, who are sometimes designated as customer relations managers, large enterprise managers, or account managers, handle these communications.

___________________

⁴ Based on presentation by Charles Brady, U.S. Census Bureau, “Annual Survey Account Manager Program,” during the panel’s June 3, 2016, meeting.

Adopting a similar model could be expected to have similar benefits for the Census Bureau. Benefits include the establishment of rapport between a company and a Census Bureau employee who is consistently assigned to that company. This rapport would be expected to facilitate more prompt and accurate responses at a reduced burden to the company and lower data collection cost for the Census Bureau. There also is the potential for the data to be of better quality and for the company to become a user of the data, as well as a provider, and thus a supporter of the Bureau’s economic survey programs. Finally, such a coordinated and managed program would greatly facilitate the implementation of the recommendations of this report for improvements to the Business Register and the annual economic surveys and the development of an integrated Annual Business Survey System.

RECOMMENDATION 3-3: The Census Bureau should establish a centralized and coordinated Account Manager Program that serves as a single point of contact for the largest enterprises with respect to all Census Bureau economic and business data collections. Account managers should have as their primary responsibilities not only the population of the Business Register with up-to-date information about these companies but also the coordination and facilitation of company responses to the Census Bureau’s economic surveys over the course of the year.

3.4 REFERENCES

NOTE: All URL addresses were active as of November 2017.

Bureau of Labor Statistics. (2010). The Multiple Worksite Report. Washington, DC: U.S. Department of Labor. Available: https://www.bls.gov/cew/cewmwr00.htm.

Da Pont, M. (2014). Statistics Canada’s Enterprise Portfolio Management Program: Improving Data Quality for Large Enterprises through Profiling and Coherence Analysis—Benefits for National Accounts. Available: http://ec.europa.eu/eurostat/documents/6501972/6553733/Da-Pont-Luxembourg.pdf/a46eadca-7bf5-4ae0-aa50-d2f4761c648f.

Davie, W.C., Jr. (2015). Overview of the U.S Census Bureau’s Business Register. International Workshop on the Business Register, Economic Census, and Integrated Economic Statistics, Aguascalientes, Mexico, September 30, 2015. Available: https://unstats.un.org/unsd/trade/events/2015/aguascalientes/10.-Panel%20III%20-%20Presentation%202%20-%20US%20Census%20Bureau.pdf.

Fairman, K., Foster, L., Krizan, C.J., and Rucker, I. (2008). An Analysis of Key Differences in Micro Data: Results from the Business List Comparison Project. Available: https://www.bls.gov/osmr/pdf/st080020.pdf.

Fixler, D., and Landefeld, J.S. (2006). The importance of data sharing to consistent macroeconomic statistics. In C. Kuebler and C. Mackie (Rapporteurs), Improving Business Statistics through Interagency Data Sharing: Summary of a Workshop (pp. 91–132). Washington, DC: The National Academies Press. doi: https://doi.org/10.17226/11738.

Haltiwanger, J., Jarmin, R., and Miranda, J. (2008). Business Formation and Dynamics by Business Age: Results from the New Business Dynamics Statistics. Available: https://www.census.gov/ces/pdf/BDS_Business_Formation_CAED_May2008.pdf.

Haltiwanger, J., Miranda, J., and Jarmin, R. (2013). Business Dynamics Statistics Briefing: `Anemic Job Creation and Growth in the Aftermath of the Great Recession: Are Home Prices to Blame? Kansas City, MO: Kauffman Foundation. Available: https://www.census.gov/ces/pdf/BDS_StatBrief8_Home_Prices.pdf.

Hughes, J. (2008). Prioritizing Business Respondents to Target Important NonResponse. Paper presented at the 2008 International Statistics Canada Methodology Symposium: Data Collection: Challenges, Achievements and New Directions, Catalog 11-522-X, Statistics Canada, Ottawa, October 28–31 Available: www.statcan.gc.ca/pub/11-522-x/2008000/article/10984-eng.pdf.

Lavoie, F. (2015). Enterprise Portfolio Management Program–Profiling of Large and Complex Enterprises. Ottawa: Statistics Canada.

Sear, J., Hughes, J., Vinette, I., and Bozzato, W. (2007). Holistic response management of business surveys at Statistics Canada. In Proceedings of the 3rd International Conference on Establishment Surveys (pp. 953-958). Alexandria, VA: American Statistical Association. Available: https://ww2.amstat.org/meetings/ices/2007/proceedings/ICES2007-000218.PDF.

Snijkers, G., and Jones, J. (2013). Business survey communications. In G. Snijkers, G. Haraldsen, J. Jones, and D.K. Willimack (Eds.), Designing and Conducting Business Surveys (Ch. 9). Hoboken, NJ: John Wiley & Sons.

Stang, S. (2016). Data Sharing between the Bureau of Labor Statistics and the Census Bureau—BLS Data Synchronization Project. Available: https://www.census.gov/fedcasic/fc2016/ppt/2_8_Synch.pdf.

Statistics Netherlands. (2015). Annual Report for 2014. The Hague/Heerlen/Bonaire: Statistics Netherlands. Available: https://www.cbs.nl/-/media/cbs/over%20ons/organisatie/jaarverslag2014engels.pdf?.

Sudman, S., Willimack, D.K., Nichols, E., and Mesenbourg, T.L., Jr. (2000). Exploratory research at the U.S. Census Bureau on the survey response process in large companies. In Proceedings of the 2nd International Conference on Establishment Surveys (pp. 327–335). Alexandria, VA: American Statistical Association.

U.S. Census Bureau. (2016). Statistics of U.S. Businesses (SUSB)—Methodology: Universe of Statistics of U.S. Businesses. Washington, DC: Author. Available: https://www.census.gov/programs-surveys/susb/technical-documentation/methodology.html.

U.S. Census Bureau. (2017). County Business Patterns (CBP)—Methodology: Universe of County Business Patterns. Washington, DC: Author. Available: https://www.census.gov/programs-surveys/cbp/technical-documentation/methodology/universe-of-cbp.html.

Vennix, K. (2012). The treatment of large enterprise groups within Statistics Netherlands. In Proceedings of the Fourth International Conference on Establishment Surveys (ICESIV) (pp. 871–880). Alexandria, VA: American Statistical Association. Available: http://www.amstat.org/meetings/ices/2012/papers/301992.pdf.

Willimack, D.K., Nichols, E., and Sudman, S. (2002). Understanding unit and item nonresponse in business surveys. In R.M. Groves, D.A. Dillman, J.L. Eltinge, and R.J.A. Little (Eds.), Survey Nonresponse (Ch. 14, pp. 213–228). New York: John Wiley & Sons.

Yarbrough, B. (2016). The Business Register as a Service Provider: Examples from the U.S. Census Bureau. 25th Meeting of the Wiesbaden Group on Business Registers, Toyko, 8-11 November 2016. Available: http://www.stat.go.jp/english/info/meetings/wiesbaden2016/pdf/2-13_usc_pap.pdf.