By William H. Frey
Sunday, February 14, 2010; B03
Every 10 years, we have to count people. At least that's what Article 1, Section 2 of the Constitution says. It doesn't sound too complicated. But it is. Who gets counted, and how, determines not only congressional representation but how funding is distributed for a slew of federal programs that affect all of us. As we prepare to stand and be counted in 2010 -- and the U.S. Census Bureau is spending a lot of advertising money to make sure that everyone is -- let's note a few misconceptions about our population and the efforts to tally us up.1. Immigration is the biggest force behind the nation's racial and ethnic diversity.
If immigration stopped today, we would still see substantial gains in our minority populations for decades to come. Recent Census Bureau projections showed that under a "no further immigration" scenario, the minority share of our population would rise from about 35 percent today to 42 percent in 2050. The preschool (under age 5 ) population would become minority white. The greater minority presence would arise from higher natural-increase rates for minorities than for the aging white population. This momentum is already in place: Since 2000, natural population increase accounted for 62 percent of the growth of Hispanics, the country's largest minority group, with immigration responsible for the rest.
Already, the District and four states (Hawaii, New Mexico, California and Texas) are minority white, and in six more, whites are less than 60 percent of the population. Minorities now make up more than 30 percent of the residents in half of the nation's congressional districts, compared with a quarter in 1992.
The census will tell us more about the dispersal of Hispanics and other groups to traditional white enclaves -- suburbs and the country's midsection. A majority of all Hispanic, black and Asian residents of major metro areas now live in the suburbs. And since 2000, according to recent estimates, the fastest Hispanic growth occurred in South Carolina, South Dakota, Tennessee and Arkansas.
Color lines within our population are blurring in a different way, too, with people who identify with more than one race. The number of mixed-race married couples more than doubled since 1990, and they make up nearly 8 percent of all marriages.2. The country is getting uniformly older.
As a baby boomer, I am part of a demographic mob. As we age over the next 20 years, the nation as a whole will see a surge in senior citizens. But different parts of the country will be aging at different rates, largely because selective "younging" is going on. This is evident from census estimates showing that during the first nine years of this decade, 25 states -- mostly in the Northeast, Midwest and Great Plains -- and the District exhibited absolute declines in their child populations, while 25 others, led by Nevada and Arizona, showed gains.
This variation in where families and children live is poised to shape a young-old regional divide that could intensify over time. Census projections for 2020, made earlier this decade, showed median ages over 40 in Maine, West Virginia and Pennsylvania, compared with below 36 in Utah, Texas, Georgia and California.3. Big states will keep getting bigger -- especially in Congress.
For much of the postwar period, the Sunbelt megastates of California, Florida and Texas just kept growing: They led all other states in adding congressional seats based on censuses since World War II. But the economic turbulence of this past decade will affect their political fortunes. Florida was one of the nation's growth leaders for the first half of the decade and was poised to gain as many as three congressional seats after the 2010 Census, tying or overtaking New York's congressional delegation. But the mortgage meltdown led to an unprecedented exodus from the state in the past two years. Florida's likely gain of one seat will be its smallest addition since the 1940 Census.
California is not positioned to gain any seats for the first time since statehood in 1850. Despite its status as an immigration magnet, the Golden State lost large numbers of people fleeing high housing costs during the bubble years. California might have even lost a seat had that bubble not burst.
Of the three Sunbelt behemoths, Texas will take the biggest prize, probably four congressional seats -- its largest increase since the 1880 Census. It was largely immune from the housing crisis late in the decade, while it gained Katrina-driven migrants from Louisiana.4. The census is the main source of information about our population.
Not as much as before. Unlike previous censuses, the 2010 count will provide only bare-bones information that does little more than fulfill its constitutional mandate. The questions will include the age, sex, race, Hispanic origin and household relationship status of each individual, and the size and homeownership status of each household.
What happened to all the rich data on poverty, income, ancestry, immigration, marital status and some 30 other categories we have come to expect from the census? Those "long form" questions have been given to a sample of census respondents in every count going back to 1940 -- but they won't be handed out this year. The queries have been diverted to the Census Bureau's American Community Survey.
In 2005, the bureau began administering the ACS to 3 million households each year to elicit the same kind of information that was previously available only every 10 years. This large and sophisticated survey has already provided important and timely insights on changing poverty, immigration and migration patterns in this economic roller coaster of a decade.5. New technology gives us much more demographic data than the census can.
Not true. Technological developments and data collected via the Internet do give us new ways of looking at the population, and complex surveys and estimates conducted by the Census Bureau and other organizations allow us to monitor change over the decade -- but there is no substitute for counting everyone. Aside from the census's constitutional mandate to provide the basis for congressional apportionment, a national headcount also allows us to know how many people live in the nation's cities, suburbs and neighborhoods and to break them down according to race, age and gender.
There are plenty of examples of a decennial census surprising the experts. The 2000 Census, for instance, discovered sharp population surges in many old, large cities. This was unanticipated for Chicago, which had experienced decades of decline. And the spread of the nation's Hispanic population into new states such as North Carolina far exceeded expectations.
Many government and private surveys, including the ACS, rely on the decennial census to make sure their work accurately reflects the population as a whole.
This census will also tell us more about small but growing groups, such as same-sex married partners and multiracial populations, whose presence and interests can change laws and public policies.
The Census Bureau's ad campaign urges Americans to answer "10 Questions in 10 Minutes" -- and those are still 10 very important questions, whose responses will guide us for the next 10 years.
William H. Frey is a demographer and senior fellow in the Metropolitan Policy Program at the Brookings Institution and a research professor at the University of Michigan's Population Studies Center.