DOCUMENTATION OF FEDERAL FUNDS DATA FISCAL YEARS 1994 - 1997 By Samuel Calhoun, Faqi Bagi, and Richard Reeder U.S. Department of Agriculture, Economic Research Service SUMMARY This documentation describes 4 raw data files created by the Economic Research Service using data from the Census Bureau's Consolidated Federal Funds Reports (CFFR). For each Fiscal Year 1994 - 1997, four data files are provided: 1. the county file 2. the program file 3. the State file 4. the State/program file. NOT ALL OF THE DATA ARE ACCURATE AT THE COUNTY LEVEL. PLEASE USE CAUTION WHEN USING THESE FILES. Although all four files contain data for all federal programs on them, the county level data reported in the first two files (the county file and the program file) are accurate only for those programs with zero disposition code. For programs with non-zero disposition codes, it is best to use State level data (the State file and the State/program file). File 1. The county file has individual counties as observations, sorted by the 5-digit FIPS State/county code. The variables include county name, county population and income, ERS county typology codes, Beale codes, and amounts received under each Federal program, including aggregate amounts by function and object, and for each function/object combination (see attached documentation for description of functions and objects and combinations). File 2. The county/program file has individual programs as observations, sorted by program ID-code. The variables include program disposition, object and function codes, and program totals (and per capita amounts) aggregated by type of county (Beale typology and ERS typology) and region (Census' 4 major regions). File 3. The State file has individual States as observations, sorted by State two-digit FIPS code. The variables include State population and income, our State typology codes (see documentation for how we defined these), and the amounts received by the State for each Federal program, and for each aggregation of programs by function, object, and function/object combination. File 4. The State/program file has programs as observations, sorted by program ID. The variables include totals and per capita amounts for each type of State and region. DATA SOURCES AND INTERPRETATION FEDERAL FUNDS DATA The principal data source we used to construct these files is the Consolidated Federal Funds Reports (CFFR) data from the U.S. Department of Commerce, Bureau of the Census, Governments Division. We usually refer to these data as the Federal Funds data. Census collects these data annually from each Federal Department or agency. We aggregated the data to the county, State, region, and national level for each Federal program for Fiscal Years 1994 (FY94) through 1997 (FY97). The Census data covered 1,196 individual programs in FY94, 1,214 programs in FY95, 1,146 programs in FY96, and 1,256 programs in FY97. Not all of these programs had reliable data at the county level. Each program has individual characteristics that affect the way the data can be used to interpret geographic patterns. For example, funds for many programs go directly to State capitals or regional centers that redistribute the money or program benefits to surrounding areas. Examples include block grant programs and some procurement programs that involve a substantial degree of subcontracting. Census screens the data to identify such programs, and we have added our own screen to identify those programs which allocate 25 percent or more of their funds to State capitals. We ended up with 719 programs (zero disposition code programs) which we believe are fairly accurate to the county level for FY94, 750 such programs for FY95, 695 such programs for FY96 and 816 such programs FY97. For the screened-out programs (disposition codes 1, 2, 3), we believe it is only meaningful to indicate geographic variations among States but not among counties. The benefits of Federal programs do not all go to the places that receive funds. For example, money spent on National parks benefits all who visit the parks and not just those who live where the parks are located. Rural electric loans are reported to the location of the company headquarters and do not indicate other places that receive subsidized electricity. Such "spillover" benefits are present in almost all Federal programs and are not reflected in the Federal funds data. In addition, different programs affect communities in different ways and have different "multiplier" impacts on local income, employment, and community wellbeing. Thus, even if the reported funding dispersion is considered to be an accurate depiction of where the funds end up being spent, care is required when interpreting the data as program impacts. Federal Funds data may represent either program expenditures or program obligations, depending on the form of the data provided to Census. Direct loans and loan guarantees are reported according to the volume of loans obligated, and do not take into account interest receipts or principal payments. Consequently, these data do not always correspond to program totals reported in government budget documents, such as budget authority, outlays, or obligations (see definitions). OTHER DATA The 1994-1997 county population estimates from the Bureau of the Census are found in the respective county files, and these were also used to compute per capita amounts by type of county in the program files. The 1994-1997 State population estimates from the Census were used in the State file and for per capita estimates in the State/program file. State and county income data are from the Bureau of Economic Analysis (BEA). The 1994 BEA State income data was not available at the time the data set was compiled, we used the data for the fourth quarter of 1993 and the first three quarters of 1994. The county income data used was for the year 1992. The 1995 file has county income for 1994. 1996 and 1997 both use county income data for 1995. FIPS CODES Five-digit Federal codes are used to identify specific geographic areas. The first two digits represent the State, the last three digits represent the county. If the county is unknown, the last three digits are 999; if the State is also unknown, the first two digits are also 99. DEFINITIONS TYPOLOGIES: Classification systems developed and periodically revised by ERS to group counties and States by economic and policy-relevant characteristics. The county economic and policy typology codes used in this issue are those described in Peggy J. Cook and Karen L. Mizer, The Revised ERS County Typology: An Overview, RDRR 89, U.S. Department of Agriculture, Economic Research Service, Dec. 1994. Because many Federal programs do not have accurate county level data, we developed a State typology to use State level data to differentiate among types of States. First we categorized States into three groups (Rural, urban, and other) based on the percent of a State's population residing in urban parts of metropolitan (metro) areas. We defined four other types of States: farming dependent, persistent poverty, retirement destination, and Federal lands. In each case, we used the same kinds of measures that were used to construct ERS's county typologies, however the cutoffs were lowered because States have more internal socioeconomic diversity than most counties. COUNTY ECONOMIC TYPES (mutually exclusive, a county may fall into only one economic type): Farming-dependent--Farming contributed a weighted annual average of 20 percent or more of total labor and proprietor income over the 3 years from 1987 to 1989. Mining-dependent--Mining contributed a weighted annual average of 15 percent or more of total labor and proprietor income over the 3 years from 1987 to 1989. Manufacturing-dependent--manufacturing contributed a weighted annual average of 30 percent or more of total labor and proprietor income over the 3 years from 1987 to 1989. Government-dependent--Federal, State, and local government activities contributed a weighted annual average of 25 percent or more of total labor and proprietor income over the 3 years from 1987 to 1989. Service-dependent--Service activities (private and personal services, agricultural services, wholesale and retail trade, finance and insurance, real estate, transportation, and public utilities) contributed a weighted annual average of 50 percent or more of total labor and proprietor income over the 3 years from 1987 to 1989. Nonspecialized--Counties not classified as a specialized economic type over the 3 years from 1987 to 1989. The farming-dependent, mining-dependent, manufacturing-dependent, government-dependent, service-dependent, and nonspecialized counties are abbreviated, respectively, as FM, MI, MF, GV, TS, and NS. Each of these County Economic Types is represented by a dummy variable. For example, when a county is identified as farming-dependent then FM=1, and otherwise 0. The same holds for other economic types. COUNTY POLICY TYPES (overlapping, a county may fall into any number of these types): Retirement-destination--The population aged 60 years and older in 1990 increased by 15 percent or more during 1980-90 through inmovement of people. Federal lands--Federally owned lands made up 30 percent or more of a county's land in the year 1987. Commuting--workers aged 16 years and over commuting to jobs outside their county of residence were 40 percent or more of all the county's workers in 1990. Persistent poverty--persons with poverty-level income in the preceding year were 20 percent or more of total population in each of 4 years: 1960, 1970, 1980, and 1990. Transfer dependent--Income from transfer payments contributed a weighted annual average of 25 percent or more of total personal income over 3 years from 1987 to 1989. The retirement-destination, federal-lands, commuting, persistent-poverty, and transfer-dependent counties are abbreviated, respectively, as RT, FL, CM, PV, and NS. Each of these County Policy Types is represented by a dummy variable. For example, if a county is identified as retirement-destination then RT=1, and otherwise 0. The same holds for other policy types. STATE TYPES (overlapping, a State may fall into up to 5 of these types, including only 1 of the first three types): Rural/Small Town--45 percent or less of the State's population in 1993 resided in urban portions within metro counties. Rural/Small Town States include Alaska, Arkansas, Idaho, Iowa, Kentucky, Maine, Mississippi, Montana, Nebraska, New Hampshire, North Carolina, North Dakota, South Dakota, Vermont, West Virginia, Wyoming. Urban/Metro--70 percent or more of the State's population in 1993 resided in urban portions of metro areas. Urban/Metro States include Arizona, California, Colorado, Connecticut, Delaware, District of Columbia, Florida, Hawaii, Illinois, Maryland, Massachusetts, Nevada, New Jersey, New York, Rhode Island, Texas, Utah. Other (neither urban nor rural)--more than 45% but less than 70% of the State's population in 1993 resided in urban portions of metro areas. Other States include Alabama, Georgia, Indiana, Kansas, Louisiana, Michigan, Minnesota, Missouri, New Mexico, Ohio, Oklahoma, Oregon, Pennsylvania, South Carolina, Tennessee, Virginia, Washington, Wisconsin. Farming-dependent--4% or more of the total labor and proprietor income came from farm labor and proprietor income in 1991-93. Farm-Dependent States include Arkansas, Idaho, Iowa, Kansas, Montana, Nebraska, North Dakota, South Dakota, Wyoming. Persistent Poverty--15% or more of a State's population had income below poverty in 1960, 1970, 1980, and 1990. Poverty States include Alabama, Alaska, Arkansas, District of Columbia, Georgia, Kentucky, Louisiana, Mississippi, New Mexico, South Carolina, South Dakota, Tennessee, West Virginia. Retirement-destination--a State's aged (over 60) population in 1990 increased by 5% or more due to net inmigration from 1980 to 1990. Retirement-Destination States include Arizona, Florida, Hawaii, Idaho, Nevada, New Mexico, North arolina, Oregon, South Carolina, Utah, Washington. Federal Lands--28% or more of State land is owned by the Federal government. Federal Lands States include Alaska, Arizona, California, Colorado, Idaho, Montana, Nevada, New Mexico, Oregon, Utah, Washington, Wyoming. Each of the rural, urban, other, farming-dependent, persistent-poverty, retirement-destination, and federal-lands States is represented by a dummy variable. For example, if a State is identified as retirement-destination then it is equal to 1, and otherwise 0. BEALE COUNTY TYPES: The Beale typology distinguishes between different types of metro and nonmetro areas, according to a county's urban population size and adjacency to metropolitan areas. The county types are formed using 1993 definitions of metropolitan areas, 1990 population data, and use the following definitions: Metro areas: Metropolitan Statistical Areas (MSA's) as defined by the Office of Management and Budget, include core counties containing a city of 50,000 or more people or have an urbanized area of 50,000 or more and a total area population of at least 100,000. Additional contiguous counties are included in the MSA if they are economically integrated with the core county or counties. For most data sources, these designations are based on population and commuting data from the 1990 Census of Population. The Current Population Survey data through 1993 categorizes counties as metro and nonmetro based on population and commuting data from the 1990 Census. Throughout this publication, "urban" and "metro" have been used interchangeably to refer to people and places within MSA's. Nonmetro areas: Counties outside of metro area boundaries. Throughout this publication, "rural" and "nonmetro" are used interchangeably to refer to people and places outside of MSA's. Adjacency: Adjacency to metro areas is determined by a combination of physical adjacency (shares a border) with the metro area and significant (at least two percent of the county's employed labor force) commuting to the metro area. Urban population: refers to population within Census-defined urban places, which are generally incorporated communities with 2,500 population or more. BEALE CODE The Beale Code (BC) variable (0-9) identifies 10 different urban-rural types of counties as follows: Metro counties (BC=0-3): Large Metro Central: Central counties of metro areas of 1 million population or more (BC=0). Large Metro Fringe: Fringe counties of metro areas of 1 million population or more (BC=1). Medium Metro: Counties in metro areas of 250,000 to 1 million population (BC=2). Small Metro: Counties in metro areas of fewer than 250,000 population BC=3). Nonmetro counties (BC=4-9): Urbanized Adjacent: Urban population of 20,000 or more, adjacent to a metro area (BC=4). Urbanized nonadjacent: Urban population of 20,000 or more, not adjacent to a metro area (BC=5). Less Urban Adjacent: Urban population of 2,500 to 19,999, adjacent to a metro area (BC=6). Less Urban Nonadjacent: Urban population of 2,500 to 19,999, not adjacent to a metro area (BC=7). Rural Adjacent: Completely rural or less than 2,500 urban population, adjacent to a metro area (BC=8). Rural Nonadjacent: Completely rural or less than 2,500 urban population, not adjacent to a metro area (BC=9). For more information on the Beale typology, see Margaret A. Butler and Calvin L. Beale, "Rural-Urban Continuum Codes for Metro and Nonmetro Counties, 1993." Staff Report No. AGES 9425. U.S. Department of Agriculture, Economic Research Service, September 1994. CENSUS REGIONS--We used the conventional 4 Census-defined regions as follows (Region code in parenthesis): Northeast (1): Connecticut, Maine, Massachusetts, New Hampshire, New Jersey, New York, Pennsylvania, Rhode Island, and Vermont. Midwest (2): Illinois, Indiana, Iowa, Kansas, Michigan, Minnesota, Missouri, Nebraska, North Dakota, Ohio, South Dakota, and Wisconsin. South (3): Alabama, Arkansas, Delaware, District of Columbia, Florida, Georgia, Kentucky, Louisiana, Maryland, Mississippi, North Carolina, Oklahoma, South Carolina, Tennessee, Texas, Virginia, and West Virginia. West (4): Alaska, Arizona, California, Colorado, Hawaii, Idaho, Montana, Nevada, New Mexico, Oregon, Utah, Washington, and Wyoming. DISPOSITION CODES Disposition codes were produced by ERS screening of data from CFFR to determine the extent to which the data may reflect funds that are tracked only to the State capital city or county. Programs with funding shown for counties but with less than 25 percent of funding going to State capitals, nationwide, were considered to be accurate at the county level (disposition code 0). Those with 25 percent or more funding going to State capitals or undistributed among counties (disposition codes 1, 2, and 3) were considered to be inaccurate at the county level because a significant amount funding that was reported as going to State capitals may ultimately be passed-through to local areas in one form or another. Each individual program has one of these four disposition codes (0, 1, 2, and 3). For variables or observations representing totals of groups of programs (such as function/object combinations), we assigned a disposition code of 4. Disposition Code 0. Less than 25% of funds reported as going to State capitals. Hence data considered accurate at the county level, for all states. Disposition Code 1. All data reported to the county level, but 25% or more of the funds went to State capitals. Hence, data are not accurate at the county level. Disposition Code 2. Federal Funds data are available at the county level for some States, but not for other States, for which only State totals are available. Hence, not generally available or accurate at the county level. Disposition Code 3. No county level data were available. Data were available only at the State or national level. Disposition Code 4. Assigned to function/object combinations; not applicable to individual programs. OBJECT CODES CFFR data are classified by broad object categories in accordance with the CFFR act. These correspond, in general, to the classification used in annual Federal Budget. Two alphabetic letters are used to designate the broad object category of Federal expenditure. These include: SW=Salaries and Wages. These include salaries and wages paid to Federal Government employees, for example, to Defense Department military employees--both active and inactive--and civilian employees of the Defense Department, U.S. Postal Service, U.S., Coast Guard, Federal Bureau of Investigation, and all other Federal Government civilian employees. PC=Procurement Contracts. Procurement contracts of the Defense Department, U.S. Postal Service, and all other civilian contracts. DR=Direct Payments for individuals for retirement and disability. DO=Direct Payments to individuals, other than for retirement and disability. Examples include unemployment compensation benefit payments, Federal payments for excess earned income tax credits, vocational rehabilitation for disabled veterans, public safety officers death benefits, Medicare hospital insurance, Medicare supplementary medical insurance, food stamps, and unemployment compensation benefit payments (Federal to State). DX=Direct Payments other than to individual. Examples include government payments to the U.S. Postal Service, Federal Employee Life and Health Insurance Premium Payments--employee share, Legal Service Corporation payments, farm payments, Department of Transportation operating-differential subsidies, all volunteer force educational assistance, veterans education assistance, rural rental assistance payments, and interest reduction payments for rental and coop housing for low income families. GG=General Grants. Include block grants, formula grants, project grants, and cooperative agreements. DL=Direct Loans. Examples include commodity loans and purchases, emergency loans, farm ownership loans, farm operating loans, soil and water loans, irrigation system rehabilitation and betterment loans, intermediary relending programs, economic injury disaster loans, physical disaster loans, loans for small businesses, direct investment loans, water and waste disposal systems for rural communities, community facilities loans, rural economic development loans. GL=Guaranteed/Insured Loans. Include farm operating loans, farm ownership loans, soil and water loans, business and industrial loans, small business investment companies, small business loans, state and local development company loans, bond guarantees for surety companies, certified development company (504) loans, foreign investment guarantees, water and waste disposal systems for rural communities, community facilities loans, rural electrification and rural telephone loans and loan guarantees, rehabilitation mortgage insurance, mortgage insurance of homes especially of low and medium income families and veterans homes, and nursing homes, higher education insured loans. Individual program data were reported mainly for grants, direct payments and loans. Salaries and procurement data were generally not available by Federal program. Instead, these expenditures were broken down into subcategories for the Department of Defense and selected other agencies. Each subcategory is treated as if it were a separate Federal program, with a program ID Code beginning with the relevant Object Code. Several programs have more than one type of assistance, such as loans and grants. These programs are incorporated in this data set as if each type of assistance qualified as a separate program. For example, the commodity loan and purchase program is broken up into two components: loans and purchases. FUNCTION CODES Each Program is assigned a function code, and a subfunction code, developed by ERS as follows: Agriculture and natural resources function (100). Subfunctions include agricultural assistance (110), agricultural research and services (120), forest and land management (130), and water and recreational resources programs (140). Community resources function (200). Includes business assistance (210), community facilities (220), community and regional development (230), environmental protection (240), housing other than veterans (250), housing for veterans (260). Native American programs (270), and transportation programs (290). Defense and space function (300). Includes aeronautics and space (310), defense contracts (320), and defense payrolls and administration (330). Human resources function (400). Includes elementary and secondary education (410), food and nutrition (420), health services (430), social services (440), and training and employment (450). Income security function (500). Includes medical and hospital benefits (510), public assistance and unemployment compensation (520), retirement, disability, and survivors social security payments (530). National functions (600). Includes criminal justice and law enforcement (610), energy (620), higher education and research (630), all other federal funds programs excluding insurance programs (640). Each year, ERS reviews the programs to assign function codes to new programs in the data set, and sometimes existing programs receive a new function code reflecting a change in function. PROGRAM IDENTIFICATION CODES Each program in the CFFR has a program identification code patterned after its Catalog of Federal Domestic Assistance (CFDA) program number. This is a six-character code. The first two characters identify the Federal department or agency that administers the program, followed by a decimal and three numeric characters identifying the program. The two-digit codes representing Federal departments and agencies are as follows: 1994 Federal Departments and Agencies 10 =Department of Agriculture 11 =Department of Commerce 12 =Department of Defense 14 =Department of Housing and Urban Development 15 =Department of the Interior 16 =Department of Justice 17 =Department of Labor 20 =Department of Transportation 21 =Department of Treasury 23 =Appalachian Regional Commission 30 =Equal Employment Opportunity Commission 43 =National Aeronautics and Space Administration 44 =National Credit Union Administration 45 =National Foundation on the Arts and the Humanities 47 =National Science Foundation 57 =Railroad Retirement Board 59 =Small Business Administration 64 =Department of Veterans Affairs 66 =Environmental Protection Agency 70 =Overseas Private Investment Corporation 72 =Action 77 =Nuclear Regulatory Commission 81 =Department of Energy 82 =United States Information Agency 83 =Federal Emergency Management Agency 84 =Department of Education 86 =Pension Benefit Guaranty Corporation 89 =National Archives and Records Administration 93 =Department of Health and Human Services 1995 Federal Departments and Agencies 10 =Department of Agriculture 11 =Department of Commerce 12 =Department of Defense 14 =Department of Housing and Urban Development 15 =Department of the Interior 16 =Department of Justice 17 =Department of Labor 20 =Department of Transportation 21 =Department of Treasury 23 =Appalachian Regional Commission 30 =Equal Employment Opportunity Commission 34 =Federal Mediation and Conciliation Service 43 =National Aeronautics and Space Administration 45 =National Foundation on the Arts and the Humanities 47 =National Science Foundation 53 =President's Committee on Employment of People with Disabilities 57 =Railroad Retirement Board 59 =Small Business Administration 64 =Department of Veterans Affairs 66 =Environmental Protection Agency 70 =Overseas Private Investment Corporation 72 =Action 77 =Nuclear Regulatory Commission 81 =Department of Energy 82 =United States Information Agency 83 =Federal Emergency Management Agency 84 =Department of Education 85 =Harry S. Truman Scholarship Foundation 86 =Pension Benefit Guaranty Corporation 89 =National Archives and Records Administration 91 =United States Institute of Peace 93 =Department of Health and Human Services 94 =Corporation for National and Community Service 96 =Social Security Administration 1996 and 1997 Federal Departments and Agencies 10 =Department of Agriculture 11 =Department of Commerce 12 =Department of Defense 14 =Department of Housing and Urban Development 15 =Department of the Interior 16 =Department of Justice 17 =Department of Labor 19 =Department of State 20 =Department of Transportation 21 =Department of Treasury 23 =Appalachian Regional Commission 30 =Equal Employment Opportunity Commission 34 =Federal Mediation and Conciliation Service 43 =National Aeronautics and Space Administration 45 =National Foundation on the Arts and the Humanities 47 =National Science Foundation 57 =Railroad Retirement Board 59 =Small Business Administration 64 =Department of Veterans Affairs 66 =Environmental Protection Agency 70 =Overseas Private Investment Corporation 77 =Nuclear Regulatory Commission 81 =Department of Energy 82 =United States Information Agency 83 =Federal Emergency Management Agency 84 =Department of Education 85 =Harry S. Truman Scholarship Foundation 86 =Pension Benefit Guaranty Corporation 89 =National Archives and Records Administration 91 =United States Institute of Peace 93 =Department of Health and Human Services 94 =Corporation for National and Community Service 96 =Social Security Administration Program names and their CFDA numbers sometimes change from year to year, reflecting program changes and shifts from agency to agency. Programs may also be in the data set one year and not in the next due to unavailability of data. If a CFDA program number did not exist, a pseudo CFDA code was assigned by the agency submitting the data. This pseudo code consists of two numeric characters (representing the agency prefix in the CFDA), followed by the decimal and three alphabetic characters. Pseudo codes are also used for some other items that do not fall under officially recognized programs such as Federal retirement payments, wages and salaries, etc. These pseudo codes begin with two alphabetic characters, followed by three numeric characters. EXCLUDED PROGRAMS Twelve insurance programs that were present in the original data provided by\ the Bureau of the Census are excluded from the Federal Funds data files because the reported amounts substantially exceed the program payoffs during Fiscal Years 1994-1997. For example, we excluded several insurance programs that reported amount insured rather than the amount of payments or claims. The Federal Funds data also exclude a few programs that were targeted exclusively at U.S. Territories. GRAND TOTALS, U.S. TOTALS, AND OTHER TOTALS For every Federal Funds program, we computed two totals across all counties in the United States: a grand total and a U.S. total. The grand total includes all expenditures reported for a given Federal funds program including State undistributed and U.S. undistributed funds. The U.S. total includes only the funds that are reported accurately at the county level, and thus it excludes all State undistributed, and the U.S. undistributed funds. In other words, grand total includes data on all disposition codes (0, 1, 2, 3), while U.S. total includes data only disposition codes 0. The Program File (one of the county files) has the individual programs as the observations (that is, the rows), followed by object codes, function codes, and the combinations of object and function codes. Funding totals by type of county, region and U.S. total are supplied for each object code, each function, and combinations of each object code and function. These totals only included programs for which data are accurate and reliable at the county level (Disposition Code 0). COUNTY FILE STRUCTURE According to the Geographical Information System (GIS) there are 3,141 counties in the fifty States, but our Federal Funds data are reported for 3085 counties. This reduction of fifty-six counties in the CFFR data was required in order to match the CFFR data with other variables on the data sets. These 56 counties, which are separate in the GIS system, have been combined with other counties in our data. For example, some of the GIS counties in Alaska have been combined to form 14 county groups; 38 independent cities in Virginia have been combined with the surrounding counties, resulting in only 98 counties for Virginia; 5 New York boroughs are combined into just one county. In each of the three states, Hawaii, Montana, and Wisconsin, two GIS counties have been combined into one county. In the county file, the rows consist of individual counties, while the individual programs are the columns, followed by 19 more columns which have information about county IDs, county types, regions, income, and population. In the rows, after all the counties in a given State, there is one additional row that contains a FIPS Code ST999 (ST varies by State), which corresponds to amounts that are not allocated to individual counties within the State. For more information about the file structure for individual data files, see "*.TXT" files corresponding to each data file.