The Problem with Households, as Currently Defined

The Census Bureau is tying itself in knots trying to categorize our living arrangements based on the old-fashioned concept of a marriage license. Since 1950, the percentage of households headed by married couples has fallen from 78 to 48 percent, yet a marriage license is still central to the Census Bureau's definition of household types. Here is the distribution of households by type in 2013...

Married couples: 48%
People living alone: 27%
Female-headed families, no spouse present: 13%
Male-headed families, no spouse present: 5%
Male-headed nonfamilies: 4%
Female-headed nonfamilies: 3%

The last four household types on the list (female- and male-headed families/nonfamilies) account for a substantial 25 percent of total households. But here's the problem: many of the households that fall into those four categories are interchangeable and would be recognized as the same type of household, except for the lack of a marriage license.

This is how it works. Let's say two unmarried people, Joe and Ellen, live in an apartment leased in both their names. Depending on who responds to the Census Bureau's Current Population Survey, their household will be categorized as either a male-headed nonfamily household (Joe responds) or a female-headed nonfamily household (Ellen responds). Let's say Joe and Ellen have a biological child. Depending on who responds to the survey, their household will be categorized as either a male-headed family household (Joe responds) or a female-headed family household (Ellen responds). Let's say the child is Ellen's from a previous relationship. If Ellen responds to the survey, then they live in a female-headed family household. If Joe responds, they live in a male-headed nonfamily household.

If Joe and Ellen had a marriage license, all these permutations of their household type would disappear. They would be a married-couple household regardless of who responded to the survey or the paternity of the child. Perhaps it's time to rethink household definitions, remove marriage from the equation, and recognize "couple" households instead.

Fewer Are Moving

Those waiting for an uptick in the nation's geographic mobility rate will have to wait awhile longer. Fewer Americans are moving, according to American Community Survey results. The ongoing decline in mobility casts doubt on rumors of recovery in the housing market.

Between 2007 and 2012, the number of people aged 1 or older who moved from one house to another in the United States fell by nearly 1 million—from 45.7 million to 44.8 million. The percentage who move has declined in every year, falling from from 15.4 percent in 2007 to 14.4 percent in 2012.

To Spank or Not to Spank

The 81 percent majority of Americans believe it is sometimes appropriate for parents to spank their children, according to a Harris survey. An even larger 86 percent say they were spanked by their parents when they were children. Here is the percentage who were spanked as children, by generation...

Millennials: 77%
Generation X: 87%
Baby Boomers: 92%
Older Americans: 88%

Source: Harris Interactive, Four in Five Americans Believe Parents Spanking their Children Is Sometimes Appropriate

Household Income Stable in August 2013

Median annual household income was stable in August 2013, according to the latest monthly update from Sentier Research. The August median of $52,236 was not statistically different from the July median, after adjusting for inflation.  

Sentier extracts its income data from the Current Population Survey, just as the Census Bureau does. From Sentier, we get monthly updates of median household income. From the Census Bureau, we get an annual update—usually released in September of the following year. By tracking Sentier's monthly updates, you pretty much know in advance what the Census Bureau will report in its annual update. Sentier has already revealed trends in median household income through August 2013 (finding no trend at all). With only four months left in the year, it's likely that the Census Bureau's annual update next September will report little to no change in median household income for 2013. 

In fact, according to Sentier, there has been no change in household income since December 2011. "Since December 2011, we have been in a period of income stagnation without any clear trend of direction," says Sentier's Gordon Green. Median household income in August 2013 was 4.4 percent below the median of June 2009, the end of the Great Recession. It was 6.1 percent lower than the median in December 2007, the start of the Great Recession. It was 7.2 percent lower than the median in January 2000. 

The Household Income Index for August 2013 was 92.8 (January 2000 = 100.0). The index compares median annual household income in a given month as a percent of its value in January 2000, after adjusting for inflation. An Excel spreadsheet of the entire household income time series is available from Sentier's web site for $25.00.

Source: Sentier Research, Trends in Household Income: August 2013

Offline Demographics

Fifteen percent of Americans aged 18 or older do not go online or use email, according to a study by Pew Internet and American Life Project. Among adults under age 50, only 2 to 8 percent are offline. The figure rises to a substantial 17 percent among people aged 50 to 64 and peaks at 44 percent among Americans aged 65 or older. The single biggest reason for not being online, say those who are offline, is that the Internet is irrelevant.

These facts and figures are not surprising since the offline lifestyle is a luxury afforded only to those who do not need to earn a living—aka retirees. The offline faction sometimes cheats, however. According to Pew, 44 percent of the offline have asked a friend or family member to complete a task for them on the Internet.

Source: Pew Internet and American Life Project, Who's Not Online and Why

Crime's Repeat Victims

The decline in crime over the past few decades has been unprecedented. Between 1993 and 2010, the rate of nonfatal violent crime fell 76 percent. Not only has the rate of violent crime declined, but so has the percentage of repeat victims. Seventeen percent of violent crime victims in 2010 were victimized more than once during the year. This figure is down from the 23 percent who were repeat victims in 1993.

You read that right. Many crime victims are victimized more than once in a year's time. That's because some of us are more at risk of violent crime because of who we live with, where we live, or where we work. Among those most likely to be victimized repeatedly, the offender is their intimate partner. The 17 percent who experienced multiple victimizations accounted for the 54 percent majority of the victims of violent crime in 2010.

Source: Bureau of Justice Statistics, Measuring the Prevalence of Crime with the National Crime Victimization Survey

Men's Shrinking Incomes

American men had a median income of $33,904 in 2012—fully 10 percent and $3,887 less than their median in 2000, after adjusting for inflation. Men aged 65 or older were the only ones who made gains during those years, with their median rising from $25,881 to $27,612.

Here is the median income of men under age 65 in 2012 by age (and the dollar loss in their median income since 2000; in 2012 dollars)...

Under age 25: $10,869 (-$1,859)
Aged 25 to 34: $34,113 (-$6,226)
Aged 35 to 44: $45,224 (-$5,339)
Aged 45 to 54: $46,466 (-$8,253)
Aged 55 to 64: $42,176 (-$3,409)

Source: Census Bureau, Historical Income Tables

States Losing the Most, 2007 to 2012

Between 2007 and 2012, most states saw their median household income decline. Sixteen states experienced double-digit declines in median household income. The biggest losers were Nevada and Hawaii, each with a 21 percent decline in median household income between 2007 and 2012 after adjusting for inflation.

Delaware, Ohio, Colorado, Louisiana, North Carolina, Arkansas, Indiana, New York, Idaho, New Mexico, Mississippi, Illinois, Georgia, and Arizona each experienced a 10 to 20 percent decline in median household income between 2007 and 2012.

Source: Census Bureau, Historical Income Tables: Households

States with Rising Incomes, 2007 to 2012

In 2012, median household income ranged from a high of $71,836 in Maryland to a low of $36,641 in Mississippi, according to the Census Bureau. Only five states and the District of Columbia experienced an increase in median household income between 2007 and 2012, after adjusting for inflation. Here are those states, their median household income in 2012, and the percent change in their median since 2007 (in 2012 dollars)...

States with growing household incomes, 2007 to 2012
District of Columbia: $65,246 (16.0%)
North Dakota: $55,766 (6.7%)
Wyoming: 57,512 (6.5%)
Vermont: $55,582 (5.9%)
Texas: $51,926 (1.8%)
Oklahoma: $48,407 (1.1%)

Source: Census Bureau, Historical Income Tables: Households

Many Renters Have No Car

Percentage of households without a vehicle by homeownership status...
Total: 9%
Owners: 3%
Renters: 20%

Source: Census Bureau, 2012 American Community Survey

Education and Life Expectancy at Very Old Age

Even at the oldest ages, better educated people live longer. This is the finding of a study published in Demographic Research. In an examination of life expectancy at the advanced ages of 95 and 100, researchers found longer life among more highly educated men and women—although not much longer.

Among women aged 95, those with the highest level of education had a life expectancy of 4.39 years compared with 4.20 years for those with the lowest level of education. This translates into 69 more days of life for the educated women. At age 100, better educated women had 24 more days of life than those with the least education. The pattern was the same among men.

Source: Demographic Research, "Minor Gradient in Mortality by Education at the Highest Ages: An Application of the Extinct-Cohort Method," Roland Rau, Magdalena M. Muszynska, and Paul H.C. Eilers

Who Lives in Group Quarters?

Eight million Americans lived in group quarters in 2012, meaning they live in an institution such as a prison, nursing home, or college dorm.

• 2,252,339 are in adult correctional facilities. This population is 91 percent male, 36 percent non-Hispanic white, and has a median age of 35.
• 1,500,366 are in nursing homes. This population is 65 percent female, 77 percent non-Hispanic white, and has a median age of 82.
• 2,581,035 are in college housing. This population is 54 percent female, 67 percent non-Hispanic white, and has a median age of 20.

Source: Census Bureau, American Factfinder, 2012 American Community Survey

City Incomes Are Growing

Median household income grew strongly in the nation's cities between 2011 and 2012, which may explain why city populations are growing again.

The major cities of the nation's metropolitan areas were one of the few demographic segments to experience a gain in median household income between 2011 and 2012, after adjusting for inflation. In those cities, median household income grew 3.2 percent. In contrast, median income did not change significantly for households in suburbs or nonmetropolitan areas.

Median household income in 2012 (and percent change 2011-12; in 2012 dollars)
Households in the cities of metropolitan areas: $45,902 (+3.2%)
Households in the suburbs of metropolitan areas: $58,780 (+0.5%)
Households outside of metropolitan areas: $41,198 (-0.4%)

Source: Census Bureau, Income, Poverty, and Health Insurance in the United States: 2012

Median Housing Value Still Falling

According to the 2012 American Community Survey (ACS), the median value of owner-occupied homes in the United States is $171,900. Nationally, median housing value has declined by a substantial 20 percent since its 2007 peak of $215,152 (in 2012 dollars). One factor behind the ongoing decline in home values is the drop in the homeownership rate as younger adults opt to rent rather than buy. The ACS data show that the nation's homeownership rate fell from 67.2 percent in 2007 to 63.9 percent in 2012.

Median housing value, 2007 to 2012 (in 2012 dollars)
2012: $171,900
2011: $177,193
2010: $189,419
2009: $198,198
2008: $210,716
2007: $215,152

Source: Census Bureau, American Community Surveys on American Factfinder

Does Buying Local Matter?

Does buying local matter to economic well-being? This question is addressed in a Federal Reserve Bank of Atlanta analysis of county level economic data for the 2000 to 2009 time period. Economist Anil Rupasingha examines the local share of employment by county and compares those numbers to real per capita income growth, employment growth, and poverty.

In most counties, employment in locally-owned businesses surpasses employment in "nonresident-owned" businesses. But the local share varies greatly by county, which you can see at a glance in the county-level maps included in the analysis. The share of employment in locally-owned businesses varies by county from 11 to 87 percent. The share of employment in nonresident-owned businesses varies by county from zero to 85 percent.

"Historically, the most popular local economic development approach was to attract businesses outside a particular municipality, state, or region," says Rupasingha. But theories about local economic development are changing, he says, and "economic development based on local entrepreneurship is increasingly gaining traction." Which approach is better?

Local is better. According to the study's results, the greater the local entrepreneurship, the greater the per capita income and employment growth and the lower the poverty. Also and interestingly, smaller local businesses have a greater effect on local economic performance than larger local businesses. "My results suggest that fostering smaller local businesses may be good local economic development policy," Rupasingha concludes.

Source: Federal Reserve Bank of Atlanta, Locally Owned: Do Local Business Ownership and Size Matter for Local Economic Well-Being?

Demographics of Cell-Mostly Internet Users

Nearly all Americans (91 percent) own a cell phone. Among cell phone owners, 63 percent use the Internet on their phone. But here comes the interesting part: Among those who go online using their cell phone, 34 percent mostly use their phone to go online—rather than a laptop or desktop computer. "The 'cell-mostly Internet user group' represents 21% of the entire cell phone owners population," reports Pew Internet and American Life Project.

The demographic segments that are more likely than average to be cell-mostly Internet users are Hispanics (60%), blacks (43%), aged 18 to 29 (50%), have no more than a high school education (45%), and have a household income below $30,000 (45%).

Source: Pew Internet and American Life Project, Cell Internet Use 2013

Income by Age of Householder, 2012

Median household income by age of householder did not change significantly between 2011 and 2012. That stability is good news, because incomes fell each year between 2007 and 2011. All but the oldest age group have a long way to go before they catch up to where they were five years ago.

Median household income in 2012 (and percent change since 2007, after adjusting for inflation)
Total households: $51,017 (-8.3%)
Under age 25: $30,604 (-13.1%)
Aged 25 to 34: $51,381 (-9.1%)
Aged 35 to 44: $63,629 (-7.5%)
Aged 45 to 54: $66,411 (-8.4%)
Aged 55 to 64: $58,626 (-7.7%)
Aged 65-plus: $33,848 (+8.0%)

Source: Census Bureau, Income, Poverty, and Health Insurance in the United States: 2012

Median Household Income in 2012

The $51,017 median household income of 2012 was not significantly different from the $51,100 of 2011, after adjusting for inflation. The Census Bureau released 2012 income statistics today, revealing little change in most figures including age of householder, race and Hispanic origin of householder, and household type. Of the many demographic segments analyzed by the Census Bureau, only two registered a statistically significant change in median household income between 2011 and 2012: households in the West saw their median household income grow by 3.2 percent, and households in the principal cities of metropolitan areas also experienced a 3.2 percent rise in income.

Nationally, median household income in 2012 was 8.3 percent below the $55,627 median of 2007 and 9.0 percent below the all-time high of $56,080 in 1999, after adjusting for inflation. In every year since 2007, median household income has declined—until this year. The finding of no change in 2012 may not be very interesting, but it is good news.

Median household income, 2007 to 2012 (in 2012 dollars)
2012: $51,017
2011: $51,100
2010: $51,892
2009: $53,285
2008: $53,644
2007: $55,627

Source: Census Bureau, Income Poverty and Health Insurance in the United States: 2012

Another Poll Reveals How Little Americans Know

Yet another poll has been released showing how little Americans know about the Affordable Care Act or the nuts and bolts of their own or anyone else's health insurance.

A Pew Research Center survey finds only 51 percent of the public aware of the health insurance exchanges that will soon be up and running in their state. It all starts on October 1, but awareness is especially low in the 27 states that have refused to set up their own exchange. Only 44 percent of the public in those states knows a health care exchange will soon be available—run by the federal government. In the 24 states with state-run exchanges, a larger 59 percent of the public is aware of what's about to happen. By political party, 40 percent of Republicans and 63 percent of Democrats know that a health care exchange will soon be available in their state.

College Grads Control 53% of Spending

There's something new in the 2012 Consumer Expenditure Survey. The Bureau of Labor Statistics updated the way it accounts for educational attainment. Rather than classifying households by the educational attainment of the reference person (the person responding to the survey), the bureau has adopted a broader approach. Now households are classified by the highest level of educational attainment of any member of the household.

"A major reason for this change is that the highest level of education attained by any member of the consumer unit more accurately reflects the income and spending patterns of a consumer unit than does the education level of the reference person only," explains the BLS. This seemingly minor methodological change is a demographic blockbuster. Take a look...

• More college educated households The percentage of households in which any member has a bachelor's degree is much larger than the percentage in which the reference person has a bachelor's degree (38 versus 30 percent).
• Fewer households with little education The percentage of households without any member with some education beyond high school is much smaller than the percentage in which the reference person has no education beyond high school (30 versus 38 percent).
• Less spending by the less educated Under the new system, the least educated spend less. Here is the comparison, all in 2012 dollars: The spending of households in which no member had a high school diploma was $25,000 in 2012, much less than the $31,000 spent in 2011 by households in which the reference person did not have a high school diploma. The spending of households in which no member had more than a high school diploma was $35,000 in 2012, well below the $41,000 spent in 2011 by households in which the reference person had no more than a high school diploma. The new classification system barely changes the spending of households with college graduates, however. Households in which the reference person had a bachelor's degree spent $70,329 in 2011. Households in which any member had a bachelor's degree spent  $71,151 in 2012.
• College grads control nation's spending The new classification system reveals the importance of the bachelor's degree to the U.S. economy. Households in which at least one member has a bachelor's degree control 53 percent of all household spending. Under the old classification system, they controlled a smaller 42 percent. Conversely, households in which no member has more than a high school diploma control only 19 percent of aggregate household spending, down from 28 percent under the old system. 

Households with college graduates account for $3.4 trillion of the nation's $6.4 trillion in consumer spending. A bachelor's degree is an even bigger deal than we thought.