The U.S. Census will determine the population of the United States as of April.
Every 10 years, the United States counts its population.
Figures from the decennial census, an event that has occurred every 10 years since 1790, are used in calculations to determine the number of U.S. House seats apportioned to each state and the boundaries of each state’s congressional and state legislative districts and many local government districts. Also, the General Accountability Office found that 2000 Census figures influenced the distribution of 85 percent of federal obligations in grants to state and local governments.
With population increases and demographic shifts both nationwide and within Texas, changes in the sizes and boundaries of districts are likely. The stakes are high, with potential effects on federal funding and on which political party controls Congress and many statehouses.
This report reviews the purpose and methods of the U.S. Census and discusses its implications for Texas, and describes issues surrounding how people are counted and the potential effects of 2010 Census results on the redistricting of congressional and state legislative seats.
Use of Census Data
The U.S. Constitution, under Art. 1, sec. 2, and in the 14th Amendment, sec. 2, requires a census every 10 years in order to distribute among the states the 435 voting seats of the U.S. House of Representatives. Each state is guaranteed at least one seat. The remaining 385 are allocated through a mathematical formula using population figures derived from the census. Because the standard number of seats in the U.S. House has been 435 since 1913 – a number set in statute by the Permanent Apportionment Act of 1929 – distributing congressional seats among the states is a zero-sum game with winners and losers.
Population changes can affect the distribution of House seats and the allocation of almost $600 billion a year in population-based funding to state and local governments. Following the 2000 Census, 18 states saw a change in the size of their congressional delegations, with 12 seats being reallocated. Federal funding to states and local governments for transportation, economic development, emergency preparedness, public works, public health, public education, and other areas is distributed by formulas that are at least partially dependent on population figures.
Census workers have long tried to resolve errors in the final count, which can be caused by people who were never counted or who were counted more than once. In anticipation of the 2000 Census, policy debates focused on ways to produce a more accurate count of the U.S. population. Legislators, policy experts, and census officials offered different approaches, with some debate on which adjustment proposals would produce the most accurate results. Adjusting raw census data through statistical formulas and models and representative sampling to count certain populations were widely discussed. However, most of the controversy centered on whether or not such adjustments were allowable under the U.S. constitutional requirement to apportion U.S. House seats among the states based on population, which the U.S. Supreme Court has said requires an unadjusted count of the population. The Census Bureau uses alternative approaches to generate data for some purposes, but not for the reapportionment of congressional seats.
Effects of Population Shifts
The traditional census head count in 2010 is projected to yield a resident population in the United States of 309 million, reflecting growth of more than 9 percent since 2000. The ongoing U.S. population shift to the South and West and to urban and suburban areas is expected to continueand to be reflected in 2010 census data. Bucking this trend is Louisiana, whose population still has not recovered from losses due to dislocation from hurricanes Katrina and Rita. As a result, as many as 14 states could see changes in the sizes of their congressional delegations, according to the National Conference of State Legislatures.
The Texas population now exceeds 24 million and is second only to the population of California, which has more than 36 million people. The Texas Department of State Health Services predicts that Texas will surpass 25 million during 2010. The state’s largest areas likely will see the most growth, with urban counties expected to continue growing rapidly. While some predict that Texas may not have grown as much as previously expected, due to slower international and domestic migration into Texas brought on by the recent recession, others predict that population growth in Texas is likely to be robust regardless of this trend because of highbirth rates.
During the last decade, most growth in Texas has occurred in South Texas, along the Interstate-35 corridor, or next to major metropolitan counties. The Panhandle and rural West and rural East Texas grew more slowly. Only a few of Texas’ 254 counties are expected to have lost population or not to have grown at all. According to the Census Bureau, 10 of the 25 fastest-growing U.S. counties are in Texas: Harris, Tarrant, Bexar, Collin, Dallas, Travis, Fort Bend, Denton, Williamson and Hidalgo. The recent recession has slowed the growth of suburban areas somewhat by decreasing movement away from large, established cities.
Two factors are behind Texas’ population growth: net migration into the state and a high birth rate. The biggest contributing factor to Texas’ growth is net migration, both domestic and foreign, to Texas. Texas also has a high rate of natural growth, which means more births than deaths. Almost three-fourths of Texas’ recent population growth is attributable to increases in racial and ethnic minorities, who account for most of the migration and Texas’ high birth rate, according to the Texas State Data Center. The State Data Center estimates that 60 percent of all Texas births are Hispanic/Latino.
The Texas Legislative Council expects the ideal, or average, district populations for U.S. congressional, state legislative, and State Board of Education (SBOE) districts in Texas to increase significantly, especially since the number of legislative and SBOE districts will not change. An ideal district population is determined by dividing the population of the state by the total number of a certain kind of district. Because some districts are expected to have higher levels of growth than other areas, district boundaries will have to be redrawn in order to equalize their populations. High-growth districts will have to relinquish constituents to low- or no-growth districts and to loss districts. Geographic regions with the most growth may gain new districts, while other regions may lose them. Such boundary changes necessarily will affect even districts with constituent populations close to the new ideal. Some regions that have seen little or no growth in population might be folded into districts covering other regions.
Texas, which gained three U.S. House seats in 1990 and two in 2000, is expected to gain either three or four in 2010, according to the National Conference of State Legislatures.
Taking Census 2010
The 2010 Census will determine the population of the United States as of April 1, which is known as Census Day. The Census Bureau must complete the national and state counts by December 31 and report them to the president. Within the first week of the next session of Congress that follows completion of the census, the president must report to Congress each state’s population and the number of representatives to which each state is entitled. Each state is awarded at least one congressional seat. Since 1941, the bureau has apportioned the remaining 385 seats by means of a complex formula that works to ensure that congressional seats represent as equal a number of people as possible. This formula is enshrined by federal law, and the wide discretion of Congress to allocate seats has been upheld by the Supreme Court in Department of Commerce v. Montana, 503 U.S. 442 (1992).
Counting people. The goal of the census is to tie each person to the geographic location of their usual residence as of April 1, 2010. A usual residence is where a person resides for more than six months of the year. People without such a residence will be counted wherever they are on April 1, 2010. According to the Texas State Demographer, each person not counted in Texas will result in a loss of $13,500 in federal funds over the next decade to Texas state and local governments.
The Census Bureau has counted illegal immigrants since 1790, as the Constitution requires a full count of all persons. The homeless also are counted where they are on April 1. The census has a service-based enumeration program for those without conventional housing. Census workers will visit shelters, known outdoor camps, soup kitchens, and mobile food units to determine who is tied to those locations and services.
The census has special rules and procedures for counting people who live in group quarters, such as college dorms, nursing homes, prisons, and military barracks. People who live in group quarters will be counted as residing there. The administrators of these facilities will fill out the necessary information and submit it to the census on behalf of the residents. Group quarters can be difficult because they often give rise to over-counts, where a person is counted twice – once at the place they consider a permanent address and again at their group home. A common example is college students who are reported by their college dormitory administrators and then reported again by their parents at their homes. The census has targeted these groups for educational outreach to reduce the number of over-counts.
In 2001, the 77th Texas Legislature considered but did not enact HB 2639 by Dutton. The bill would have mandated that the comptroller adjust U.S. Census figures so that people incarcerated in local, state, and federal facilities be counted at the last address before incarceration or admission instead of at the institution where they lived on April 1 of a census year, and that the adjusted figures be used for redistricting purposes.
Supporters of HB 2639 said that inmates’ hometowns should not be penalized when their residents must leave to be incarcerated for a criminal offense. In the same way, the benefits of inmates’ census participation should not follow them to another location whose population is inflated artificially by the presence of a prison. Opponents of the bill argued that Texas should not deviate from the long-standing federal standard that allocates resources and representation to where people actually are present. The idea of extrapolating inmates to their permanent residences has even less justification in light of longer sentences and reduced parole since it may be years before inmates return to their previous homes. HB 2639 was reported favorably by the Elections Committee, but failed to pass on second reading in the House.
At the direction of Congress in 2006, the Census Bureau studied the issue of counting prisoners at their permanent home of record rather than the location of their incarceration. It concluded that such an adjustment would result in substantially higher costs and decreased accuracy, especially concerning prisoners with no valid home address.
Federal employees abroad will be counted as part of their home state for purposes of apportionment. Other Americans abroad are not counted. The General Accountability Office and the Census Bureau have studied different ways of counting Americans abroad. In 2004, the Census Bureau designed and planned test counts that would have counted Americans living abroad in France, Kuwait and Mexico. Congress stopped the effort because of concerns about feasibility, data quality and cost. Other efforts in Congress have sought to require that specific groups besides federal employees be counted. Some of these groups are so numerous that they could affect redistricting outcomes. For example, lawmakers from Utah argue that the state missed out on an additional congressional seat because the approximately 11,000 Mormon missionaries abroad were not counted in the 2000 Census.
Preparation. The Census Bureau is tasked with counting everyone and tying each person to the correct location so that congressional representation and federal funds can be divided fairly. Doing so requires long-term advance preparation. In 2006, the Census Bureau conducted tests in Austin, Texas, and on the Cheyenne River Sioux Reservation in South Dakota. These tests provided valuable data on attempts to update address lists and on the effectiveness of bilingual questionnaires. Census workers also were able to test hand-held GPS devices as they updated addresses. In 2008, the Census Bureau conducted dress rehearsals in which census workers directly gathered data from households by going door to door in Fayetteville, N.C., and San Joaquin County, Calif. In the past, such practice runs were used to study under- and over-counts of populations. This time, the Census Bureau has used them primarily to test new data-gathering techniques and technologies.
A major part of census preparation is maintaining a master file that lists the addresses of all living quarters in the United States. Congress enacted legislation in 1994 to allow the Census Bureau to obtain postal address lists and combine them with the 1990 census address list for residences with city-style number and street-style addresses. Non-city-style addresses are incorporated by hiring temporary census workers to locate, record, and place each residence on a map. U.S. Postal Service letter carriers were to verify the accuracy of these lists about three months before the census.
In Texas, census officials invited 1,422 local governments to help in this process through the Local Update of Census Addresses program, and 499 local governments responded. These state and local officials are helping the Census Bureau to locate and map hard-to-count areas. Officials in the Census Bureau’s Dallas Regional Office, the Texas Attorney General’s Office, the Texas State Data Center, and various cities and counties are working to identify and target these areas in hopes of improving coverage.
Texas also is cooperating with the Census Bureau on the Redistricting Data Program. In Phase 1 of the program, local and state officials suggested topographic and geographic features, such as major roads and bodies of water, that would define census blocks, the smallest unit of census geography. In Phase 2, begun in January 2009, the state identified legislative districts and voting tabulation districts (census blocks approximating election precincts) and now is verifying district boundaries on census maps. This phase should be completed by May 2010. Phase 3 is the delivery of census redistricting data – also known as “P.L. 94-171 data,” for the federal law requiring its release – and is scheduled to occur by April 1, 2011.
Hiring Census Workers. The U.S. Census Bureau hopes to recruit 1.2 million temporary workers to gather and compile census data. The workforce will have a slogan for the 2010 Census: “It’s in our hands.” Across the nation, census workers will be paid between $8 and $24 an hour, based on position and the prevalent wage scales in the area. The temporary jobs started in March and will last between five and 10 weeks. In Texas, the Census Bureau has established 39 regional offices. The Census Bureau has budgeted for 84,000 temporary-worker positions in Texas. Most of them will work as enumerators, going door to door to gather data on Texas households.
The primary qualifications for census work are knowledge of the local area and availability. Most of the positions are for door-to-door census takers, known as enumerators, who follow up on residents who do not return the initial mailed questionnaire. Because the enumerators have the best chance of reaching people at home in the evenings, most shifts are for late afternoon and evening work. All applicants must successfully pass an FBI background check, and all hired workers will be fingerprinted in order to protect the public. In most cases U.S. citizenship is required, although some exemptions are made for documented workers who have special skills, such as a needed second language like Spanish.
Collecting data. In March of 2010, the Census Bureau mailed a 10-question form to each household in every county. For Census 2010, the Census Bureau will not use a “long form.” In the past, one in six homes received the long form, a multi-page questionnaire that gathered detailed information about people living in the United States. Census 2010 will use only a 10-question short form in the hope that a shorter questionnaire will be easier to fill out and will be more likely to be sent back. The Census Bureau estimates that it should take about 10 minutes to complete the form. This change is intended to increase the response rate, and thus the accuracy, of the census. Replacement questionnaires will be sent to low- and medium-response areas, with visits by census workers to non-responding households starting in late April and early May.
Because census data quickly become outdated, the Census Bureau has moved toward continuous measurement to fill statistical gaps more often than every 10 years. The Census Bureau now collects detailed information about households through the American Community Survey, rather than through the long form. The American Community Survey consists of annual and multi-year estimates of demographic and economic data generated by monthly sampling of the populace.
The Census Bureau touts the American Community Survey for its increased sampling options and flexible design and content, as well as for providing for more frequent evaluation of data. Specifically, the bureau hopes that the American Community Survey will:
· allow more effective targeting of neighborhoods that require assistance, such as those with a certain percentage of non-English speakers;
· simplify census data collection and processing by replacing the long form;
· improve coverage by continuously updating address lists through regular interaction with local officials;
· spread the decennial census budget more evenly over the decade;
· apply newer and improving sampling and statistical techniques to fresher information for more useful and accurate data; and
· establish a cadre of professional, experienced field representatives in hard-to-count areas.
American Community Survey data can be accessed on-line at http://factfinder.census.gov/.
Census 2010 will not collect data from households online. While the Census Bureau is exploring ways to collect data or have households submit their census forms electronically, it has not yet devised a way to do so that would adequately protect both the integrity of the data and the privacy of households being surveyed. The Census Bureau also has determined in initial experiments that online reporting does not increase the response rate, nor does it lead to cost savings when compared to traditional data-gathering techniques.
Outreach. Census 2010 will employ a massive language outreach program, with bilingual forms in English and Spanish being sent to about 13 million households. They will be sent to groups of households where a certain percentage of the population is considered to be linguistically isolated. A household is determined to be linguistically isolated when someone in the household over the age of 5 does not speak English at home. Households will be able to request a census form in one of six languages: English, Chinese-Simplified, Korean, Russian, Spanish and Vietnamese. Promotional outreach campaigns will be conducted in 28 languages. Language assistance guides to help with completing the census form will be published in 59 languages. Finally, Census 2010 will directly offer or will partner or contract with organizations that offer language assistance in 101 languages. Census 2010 will use paid advertising, public-private partnerships, cooperation with national and local community organizations, and dial-in call centers to reach as many hard-to-count households as possible.
Census 2010 is strongly encouraging the formation of state, regional, and local “complete count committees.” A complete count committee is a group of community and government leaders who work to build awareness of the census and its importance. The Census Bureau hopes that complete count committees will provide the cultural and community insights necessary to build awareness of the 2010 Census by working to educate both the community and the local census office. It also is hoped that when local leaders espouse the importance of being counted, more households will return their census forms. Most major Texas cities and counties have complete count committees. While 37 states have statewide committees, Texas does not.
Census 2010 also has partnered with private groups to raise awareness of the census. A Spanish language telenovela, or soap opera, is incorporating census themes into its stories. The PBS children’s television program Sesame Street has produced segments on the census. A “Census in the Schools” curriculum will be offered to school districts to educate school-age children about the census. The goal of the program is to provide children with a basic understanding of what the census is, why their household should fill out and return its census form, and the confidentiality of the data.
Adjusted and traditional counts. The U.S. Supreme Court ruled in 1999 in Department of Commerce v. House of Representatives, 525 U.S. 316, that only an unadjusted head count may be used in reapportioning the 435 U.S. House seats among the states. The counts used for apportionment cannot reflect the use of statistical sampling to adjust for over-counting or under-counting. Alaska, Arizona, Colorado, and Kansas have enacted legislation barring the use for redistricting of census figures that have been statistically adjusted to account for groups presumed to have been undercounted.
Sampling currently is used to study general demographics and population trends, but the federal government has not determined whether to use the adjusted data for funding purposes in upcoming budgets. According to an October 2007 report by the U.S. General Accountability Office (GAO), 2010 Census, Population Measures Are Important for Federal Funding Allocations, Texas and several other states would stand to gain millions of dollars a year in federal funding if adjusted population figures, rather than a head count, were used in spending allocation formulas to account for those not counted by the census.
For example, the GAO examined allocations of Social Services Block Grants, which are distributed solely on the basis of population. Under statistical correction, Texas would stand to see a .51 percent increase in Social Services Block Grants from the federal government. The District of Columbia would see the largest increase with a gain of 2.05 percent, while Minnesota would see the largest decrease, with a loss of 1.17 percent. The GAO points out that, under statistical correction, only $4.2 million or .25 percent of the $1.7 billion allocation of Social Services Block Grants would have been reallocated.
Post-census redistricting of congressional and state legislative districts has become commonplace for state legislatures since a series of U.S. Supreme Court rulings in the 1960s established the principle of “substantial equality of population” among electoral districts.
The Census Bureau plans to ship each state its specific census data by April 1, 2011. Once legislative and executive leaders have confirmed their receipt of the data, the bureau will post the data on its American Fact Finder site on the Internet, at http://factfinder.census.gov.
Texas lawmakers will have access to redistricting data and software through the Legislative Council. Legislative offices will be able to access the redistricting application (RedAppl) through their Capitol offices’ computers and by appointment in the Legislative Council offices. RedAppl can generate maps that have detail at the level of census blocks tabulated by race, Hispanic surname, and voting age. The Legislative Council also provides a program called DistrictViewer that displays and overlays interactive maps and related information. DistrictViewer is available to the general public on the Internet, at http://gis1.tlc.state.tx.us/. These programs will provide a better visual representation of how given geographic areas are affected by different redistricting plans and also allow for easier comparisons of different criteria and layouts.
Legislative deadlines. When the 82nd Texas Legislature convenes in January 2011, it will be tasked with redrawing boundaries for legislative, congressional, and State Board of Education districts. P.L. 94-171 requires the Census Bureau to supply the states with detailed redistricting data no later than 12 months after the census is taken. For Census 2010, that will be April 1, 2011.
Art. 3, sec. 28 of the Texas Constitution requires completion of legislative redistricting during the regular session following publication of the census. The courts have interpreted this provision to mean the regular session during which census data are released (see Attorney General Opinion DM-6, February 27, 1991). The regular session in 2011 must end by May 30, 2011.
Under Art. 3, sec. 28, if the Legislature does not enact a valid redistricting plan for the House or the Senate during the regular session, then the task falls to the Legislative Redistricting Board, which includes the lieutenant governor, the speaker of the House, the attorney general, the comptroller, and the land commissioner. The board must meet within 90 days of the end of the regular session and, within 60 days of convening, must adopt its own House or Senate plan. The Legislative Redistricting Board’s duty to redraw district boundaries is invoked not only if a House or Senate plan is not enacted by the Legislature during the regular session, but also if an enacted plan is vetoed or held invalid by a court or the U.S. Department of Justice under the federal Voting Rights Act.
The Legislative Redistricting Board has no jurisdiction over congressional redistricting, which is not addressed in the Texas Constitution. U.S. constitutional “one person, one vote” equal-population requirements and the need to accommodate any additional U.S. House seats apportioned to Texas will mean that congressional districts must be re-drawn using the 2010 census figures in time for the 2012 elections.
Unlike state House and Senate redistricting, the Legislature may consider congressional redistricting during a special session should it fail to enact a valid redistricting plan during the regular session or if a legislatively enacted plan subsequently is invalidated. The governor decides whether to call a special session for that purpose.
If the Legislature does not enact a valid redistricting plan in time for the 2012 elections, then a federal court likely would draw the plan. In 2001, when the Legislature did not enact a congressional redistricting plan during the regular session and the governor did not call a special session for the Legislature to complete that task, a three-judge federal panel set the congressional district boundaries used for the 2002 election. After much controversy, the Legislature in 2003 replaced the federal-court plan and, using 2001 census figures, drew its own congressional redistricting plan, which after partial modification by the federal courts remains in effect today.
Because Texas is a jurisdiction that falls under the pre-clearance provisions of Section 5 of the federal Voting Rights Act, any redistricting plan the state enacts must withstand scrutiny by the U.S. Justice Department or a three-judge federal court panel in the District of Columbia. The plan also must meet all the requirements of the state and U.S. constitutions and the federal Voting Rights Act, particularly section 2 of the act, or else be subject to challenge in state or federal court. These requirements are primarily concerned with ensuring districts meet constitutional “one person, one vote” equal-population requirements and that any proposed changes do not have the purpose or the effect of denying or abridging the right to vote on the basis of race or language group. H – By Tom Howe
This excerpt was reprinted with permission from the House Research Organization, a nonpartisan department of the Texas House of Representatives that examines state issues and analyzes legislation being considered by the Texas Legislature. To view the report in full, go to http://www.hro.house.state.tx.us/focus/Census81-10.pdf.