State and Local Government Statistics at a Crossroads (2007)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

1 Introduction T he United States of America owes its existence to an agreement en- tered into over 200 years ago by formerly separate states to form a federal system of shared sovereignty between the states and the new national government. In the U.S. Constitution, the states retain sovereignty over many functions and areas of government, which they, in turn, can devolve to counties, cities, and other local government entities. In the early life of the new republic, officials of the national government recognized the need for information on the expenditures, revenues, employment, and other characteristics of the state and local governments. The 1840 census compiled information on public schools by state, and succeeding censuses asked for increasing amounts of information on state and local govern- ments. Since 1957, the Census of Governments has been compiled every 5 years in conjunction with the nation’s Economic Census. This evolving program of compilations of information on governments is the responsibility of the U.S. Census Bureau’s Governments Division. The division conducts the Census of Governments and related annual and quarterly surveys and maintains and updates a comprehensive directory of state and local governments. The division provides data on major aspects of the finances of the federal government and state and local governments. Its principal role in the economy and for the public sector is the provision of data on the characteristics, finances, and employment of state and local governments. These data are the basis for much needed standardization in the definitions of the structure and activities of state and local governments. The definitions have great value and are extensively used for understanding 11 

12 STATE AND LOCAL GOVERNMENT STATISTICS AT A CROSSROADS state and local government in the American economy. The Governments Division also provides data on federal government finances, presented in important ways that relate to the state and local portion of the U.S. public sector. The division also conducts special surveys for other federal agen- cies on particular aspects of state and local government operations on a reimbursable basis. As part of an internal strategic planning initiative and to respond to a performance assessment of the division by the U.S. Office of Management and Budget, the Census Bureau requested that the Committee on National Statistics establish a group of experts to review the division’s basic program of censuses and surveys. The Panel on Research and Development Priorities for the U.S. Census Bureau’s State and Local Government Statistics Program was established in 2005 and charged to make recommendations of priority areas for research and development to move the state and local govern- ment statistics program forward in the face of several challenges, including constrained budget resources. Topics for the panel to consider include the goals, content, statistical methodology, data quality, and data products from the Census of Govern- ments, the annual surveys of government finances, employment, and public employee retirement systems, and the quarterly survey of government taxes. A key element of the panel’s information gathering was a workshop that convened statistical experts, data users, and Governments Division staff. This report presents the panel’s conclusions and recommendations. IMPORTANCE OF THE STATE AND LOCAL GOVERNMENT SECTOR State and local governments play a major role in the U.S. economy and in the lives of Americans from birth to death (U.S. Census Bureau, 2005). Spending by the nation’s 87,525 state and local governments (states, coun- ties, cities, townships, school districts, and special districts) collectively accounts for 12 percent of gross domestic product and is more than all of the federal government’s nonmilitary expenditures—$1.9 trillion com- pared with $1.5 trillion in fiscal year 2004. State and local governments employ 1 in 7 workers—more than 18 million jobs in all, and seven times as many civilian workers as the federal government employs. Since 2001, state and local governments have employed more workers than the entire manufacturing sector. If governments were ranked along with corporations in the Fortune 500, then every state and 12 local governments (3 counties, 7 cities, and 2 school districts) would be in the Fortune 500 on the basis of their revenues and expenditures, and California and New York would rank in the top 10. 

INTRODUCTION 13 State and local governments touch the lives of Americans in many ways: • State and local governments educate the nation’s children: 5.9 mil- lion public school teachers and other school staff educate 48 mil- lion children in 96,000 public elementary and secondary schools (U.S. National Center for Education Statistics, 2005). • Through Medicaid and related programs, state governments (with federal assistance) provide health care coverage for about one-half of poor children and one-quarter of poor adults. Medicaid finances half of all nursing home expenditures and pays for more than one- third of all births (Holahan et al., 2003). • State and local governments prepare the future workforce, educat- ing about three-quarters of the 17.3 million students in degree- granting institutions of higher education (U.S. National Center for Education Statistics, 2005). • State and local governments protect the public’s safety, financing and operating the nation’s police forces and maintaining custody over 92 percent of the nation’s 2.2 million prison and jail inmates. (U.S. Bureau of Justice Statistics, 2007). • Approximately 97 percent of the public road system is under the control of state and local governments. The state and local share of total public transit funding is about 75 percent (U.S. Department of Transportation, 2007). • States and local governments administer the nation’s public welfare programs. With financial assistance from the federal government, states decide whether and how to provide training, child care, and other assistance that may help the needy find jobs and become self- sustaining, and they decide whether to provide cash assistance and other kinds of benefits. • State and local governments are the front line of homeland secu- rity. In addition to hiring, training, and paying police, they protect water supplies, transit systems, and other networks; they provide emergency response; and they guard the nation’s public health against long-standing and emerging threats. • Finally, in some locations, such services as hospital care, electric power and other utilities, and even alcoholic beverage sales are provided by government entities rather than private companies.   Calculated from U.S. Bureau of Transportation Statistics, Highway Profile, Inventory, for the Year 2000. Available: http://www.bts.gov/publications/national_transportation_statistics/ 2002/html/table_highway_profile.html. 

14 STATE AND LOCAL GOVERNMENT STATISTICS AT A CROSSROADS State and local governments are remarkably diverse. States differ widely in expenditure levels, revenue sources, and expenditure streams: the five highest spending states spend over 40 percent more per capita on state and local government services than the five lowest spending states; Oregon relies on its income tax for 70 percent of its state tax revenue, yet nine states have no broad-based income tax at all; state funding accounts for 90 percent of elementary and secondary education expenditures in Hawaii, but only 30 percent in Nevada. States also exhibit diverse patterns of local government organization: some states are organized primarily into county and city governments, while other states have large numbers of independent townships, municipalities, and school districts. Special districts cover a multitude of functions, such as water and sewer authorities, transportation authorities, and the like, and new forms of special districts keep evolving to meet public needs. State and local government is a growth industry. The federal govern- ment has devolved significant authority to state and local governments in many areas, especially in programs that serve the poor, at the same time that it has scaled back its own direct spending in these areas. The state and local sector has grown dramatically for at least the last five decades, nearly doubling relative to the total U.S. economy between 1950 and 2000. Figure 1-1 shows that state and local government spending on goods and services (including spending of grants from the federal government) now is about the same as direct purchases of goods and services by the federal govern- ment (excluding spending to make grants to state and local governments). ROLE OF THE GOVERNMENTS DIVISION The Governments Division in the U.S. Census Bureau has a dual mis- sion of reporting on the size and scope of the state and local government sector at an aggregate level and reporting on the functions of individual governments. It faces unique challenges in accomplishing this dual mission. Unlike other data collected by the U.S. Census Bureau and other federal statistical agencies, any data obtained on governments through censuses or surveys are in the public domain and are not confidential. Moreover, while U.S. law mandates that individuals participate in the decennial census and that business enterprises participate in the Economic Census, state and local government participation in the Census of Governments is entirely   Per capita expenditure and income tax comparisons are based on analysis of Census Bureau government finance data for 2004. The per-capita spending comparison is for the fifth high- est state relative to the fifth lowest state. If, instead, the median of the five highest states is compared with the median of the five lowest states, then the top five actually spend 70 percent more per capita than the bottom five. The state share of education funding was obtained from U.S. National Center for Education Statistics (2005). 

INTRODUCTION 15 24 22 20 Percent Federal 18 State-Local 16 Federal Share 2004—17.7% 14 State-Local Share 2004—17.3% 12 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 FIGURE 1-1  Government direct general expenditures as a percentage of gross domestic product, FY 1980-2004. NOTE: Grants are counted in the government that finally spends them. SOURCE: Nathan (2006). 1-1 voluntary, in keeping with traditional respect for the rights of state and lo- cal governments. The absence of any requirement to provide information, together with the knowledge that any information provided will undergo public scrutiny, appears to deter state and local government officials from complying fully with requests from the Governments Division. Although they have shortcomings that are detailed in this report, the Governments Division data on state and local governments are the current gold standard for information on government finances and employment: • They are the main source that can be used to describe the whole of government—federal, state, and local—so that decision makers and the public can understand how one level of government relates to the others and how funds flow among the levels. • They are the main source of data that can be used to describe gov- ernmental activities in the United States over long spans of time (decades or even a century) in a reasonably consistent way. • They are the only source of data that can be used to compare large numbers of individual state and local governments in a reasonably consistent way. While other organizations produce some data on government finances, Governments Division data represent the most comprehensive, highest 

16 STATE AND LOCAL GOVERNMENT STATISTICS AT A CROSSROADS quality, and most comparable data source by far. The division’s data are the essential starting point in any comparative analysis of government finances. The Governments Division has two broad groups of users for its data, reflecting its two broad missions. The first group includes the Bureau of Economic Analysis (BEA) and the Federal Reserve Board (FRB), which have great interest in data describing the size and role of the government sector in the national and regional economies but place less emphasis on the finances of specific governments and on specific activities of govern- ment. In turn, many public- and private-sector decision makers, analysts, and agencies use the BEA aggregate data on government revenues and expenditures that are part of the estimates of gross domestic product and other elements in the national income and product accounts (NIPA) and the FRB aggregate data on government assets and liabilities that are part of the Flow of Funds accounts of the United States. The second broad user group includes researchers, analysts, and members of the press and the public who are interested in how government affects the lives of people, place great emphasis on data describing the specific activities of government, and often are interested in the finances of specific governments. Although the interests of these two broad audiences overlap, there is also tension when budget constraints necessitate program cutbacks. For example, in 1992 the Governments Division reduced the scope of the Tax- able Property Values survey and later eliminated it, to the dismay of many in the research community concerned about the loss of detailed informa- tion about the nation’s largest source of state and local government tax revenues. More recently, to save resources, maintain timing, and minimize the loss of quality, the Governments Division reduced the local government sample size for the 2001 and 2003 annual finance surveys. This reduction led to the elimination of these data on local governments for those years and made it difficult for researchers and others to understand how state and local governments responded to the fiscal crises they confronted early in the decade. ISSUES FOR THE PANEL The panel was asked to conduct its review so as to contribute to a strategic planning process, which is being conducted not only for the Governments Division, but also throughout the Census Bureau’s Economic Directorate, of which the division is a part. Early in its review, the panel was informed by Census Bureau management that budget pressures on the Governments Division (and other Census Bureau divisions) are likely to continue and may become more intense. In light of budget constraints, the strategic planning process charges the Governments Division to streamline 

INTRODUCTION 17 programs and to facilitate more accurate and complete reporting by better aligning data collection with the accounting standards and practices of gov- ernmental units. The process also charges the division to rank its activities into four priority categories from most to least important: (1) benchmark measures, (2) principal economic indicators, (3) annual sectoral-level sta- tistics, and (4) the remaining programs and infrastructures. This rank ordering gives most priority to the data items needed for the national income and product accounts and flow of funds estimates, which are key outputs from the federal statistical system that have major conse- quences for public- and private-sector decision making. The panel fully un- derstands the importance of high-quality data for aggregate estimates of the state and local government sector. However, we are concerned that down- grading the importance of collecting and analyzing detailed information on government activities at the level of individual governments undercuts the ability of decision makers and the public to understand the complex web of federal, state, and local government interconnections. Regular, in- depth assessment of these connections is essential for sound policy making in a range of key areas, including health care, transportation, education, public safety, and others. Such assessment is also needed for making sound decisions regarding how to finance the provision of public services. Finally, the availability of disaggregated information is important for evaluating, improving, and maintaining the accuracy of the aggregate information that provides vital inputs to the national income and product accounts and flow of funds estimates. The panel urges the leaders of the Census Bureau to initiate both short-term and long-term studies, as specified in this report, of the priori- ties and balance of the Governments Division’s work. These studies should reflect the distinctive character and purposes of the division’s dual role of providing inputs to national economic indicators and detailed information on state and local government activities. While adequate budget resources may not be available in the near term for the division to serve both roles fully, research and development should address both roles and lay out a plan for implementing improvements to each over the long run. Some of the panel’s recommendations—such as on the need to establish a working group of experts to advise the Governments Division on ways to keep its data as relevant and accurate as possible and the need for proactive dis- semination of the division’s data—are essential to laying the groundwork for the executive branch and Congress to understand and support a strong government statistics program. In keeping with the panel’s charge, the bulk of the panel’s recommen- dations suggest research and development activities designed to produce improvements in the basic government statistics program. In addition, some recommendations call for immediate program improvements that can pro- 

18 STATE AND LOCAL GOVERNMENT STATISTICS AT A CROSSROADS ceed without further research and development. The panel’s findings and recommendations are presented throughout the next five chapters. OUTLINE OF THE REPORT Following this introduction, Chapter 2 provides a brief history of the collection of state and local government data by the federal government, a description of the current base and reimbursable Governments Division program, issues relating to defining and classifying governments (a central role of the division), and a discussion of the effects of constrained budgets on the division’s core or base programs. Chapter 3 discusses the issues addressed by data users from whom the panel obtained input, including the Bureau of Economic Analysis, the Fed- eral Reserve Board, public interest groups, and research institutions. Data users provided feedback on the current uses and potential uses of the data, as well as on their strengths and weaknesses. Chapter 4 examines data accuracy and statistical methodology for the Census of Governments and the annual and quarterly surveys, includ- ing issues of sample frame development and design, data collection, unit nonresponse, editing and imputation, estimation, data processing, revision policies, and cognitive testing of questionnaires. Chapter 5 discusses the dissemination and analysis practices of the Gov- ernments Division, which are the primary concerns of many data users. Chapter 6 addresses strategic issues and challenges facing the Govern- ments Division and its managers in the Census Bureau as they look toward the future. Background materials appear in the appendixes. Appendix A is a brief description of the products issued by the Governments Division. Appendix B is a table describing the current reimbursable programs conducted by the Governments Division. Appendix C reproduces two letters on the Tax- able Property Value Survey. Appendix D summarizes the presentations of public interest and other user groups at the panel’s workshop. Appendix E provides the planning meeting and workshop agendas. Appendix F presents biographical sketches of panel members and staff.