Dementia: the 3rd Leading Cause of Death?

Dementia is a major cause of death. We know that. The government's mortality reports show Alzheimer's disease to be the 6th leading cause of death in the United States. In 2017, Alzheimer's disease killed 121,000 Americans.

But Alzheimer's disease accounts for only a portion of dementia deaths. A much larger 262,000 people died of dementia in 2017, according to a report by the National Center for Health Statistics. If all types of dementias were considered a single cause of death (as they are in some countries), then dementia would be the third leading cause of death in the United States, following heart disease and cancer.

The NCHS report provides a detailed look at dementia deaths by type. Alzheimer's disease is most common, accounting for 46 percent of dementia deaths in 2017. Vascular dementia deaths are another 6 percent, and other types or unspecified dementias account for the rest. Regardless of the type, all dementias have one thing in common—they are debilitating for the individual and devastating for the family. One table in the report reveals the debilitation and devastation: place of death. Most dementia deaths occur in a nursing home, long-term care facility, or hospice. In other words, most patients and families are unable to deal with the consequences of dementia on their own or at home. Among all deaths in the U.S., just 27 percent occur in a nursing home, long-term care facility, or hospice. Among dementia deaths, the figure is 60 percent.

Source: National Center for Health Statistics, Mortality Data, Dementia Mortality in the United States, 2000–2017

Average Household Spending on Travel Tops $1,800

Americans are spending more on travel than ever before. The average household spent $1,852 on travel in 2017, according to the Bureau of Labor Statistics' Consumer Expenditure Survey. This is 7 percent more than it spent in 2007, after adjusting for inflation. A 7 percent increase doesn't sound like much, but consider this: travel spending fell 15 percent between 2007 and 2010, in the aftermath of the Great Recession. Between 2010 and 2017, average household spending on travel grew 26 percent.

Average household spending on travel, 2007 to 2017 (in 2017 dollars)
2017: $1,852
2010: $1,468
2007: $1,735

The single biggest item in the average household's travel budget is lodging, which accounts for 27 percent of total travel spending. Airline fares are second at 24 percent of the budget, and restaurant meals (17 percent) are third. Together, these three items account for two-thirds of household spending on travel. The remaining one-third of the budget is accounted for by a range of items listed here in rank order: recreational expenses on trips, gasoline on trips, alcohol on trips, ship fares, groceries purchased while traveling, local transportation on trips, train fares, luggage, vehicle rentals, parking fees and tolls on trips, and intercity bus fares.

Between 2010 and 2017, average household spending increased on all but two travel budget items, after adjusting for inflation. Average household spending on gasoline fell 18 percent because of lower gas prices. Spending on vehicle rentals fell by a larger 30 percent because of competition from ride-sharing services such as Uber and Lyft.

One of the biggest gains in travel spending was experienced by the category "local transportation on trips," which includes spending on ride-sharing services. Between 2010 and 2017, average household spending on local transportation on trips climbed 45 percent, after adjusting for inflation. In 2010, the average household spent 52 percent more on rented vehicles than on local transportation when traveling. Ride-sharing has reversed this pattern. In 2017, the average household spent 36 percent more on local transportation than on rented vehicles when traveling.

Source: Demo Memo analysis of the Consumer Expenditure Survey

Driving Alone to Work: Top and Bottom Metros

Most Americans drive alone to work, according to the Census Bureau's American Community Survey. Nationwide, the percentage of workers aged 16 or older who commute alone in an automobile stood at 76 percent in 2017. The figure varies by metropolitan area.

The five metros with the most lone drivers
The five metropolitan areas with the largest share of lone drivers are all in the South: Dothan, AL (89 percent); Wheeling, WV (89 percent); Owensboro, KY (88 percent); Huntsville, AL (88 percent); and Florence-Muscle Shoals, AL (88 percent). One reason for these above-average figures is a lack of public transportation. The percentage of workers in these metros who use public transportation to get to work ranges from 0.0 to 0.5 percent.

The five metros with the fewest lone drivers
All five metropolitan areas with the smallest share of lone drivers are in the Northeast or West: New York (50 percent); San Francisco (57 percent); Ithaca, NY (58 percent); Boulder, CO (64 percent); and Corvallis, OR (66 percent). In New York and San Francisco, the availability and popularity of public transportation is the biggest factor reducing the percentage of workers who drive alone. In the New York metro, 31 percent of workers commute on public transportation. In San Francisco, the figure is 17 percent. Among metropolitan areas, New York and San Francisco are the ones with the highest use of public transportation.

Driving to work alone is relatively low in the other three metropolitan areas for different reasons.

• Ithaca, NY, distinguishes itself as the metropolitan area with the largest share of workers who walk to work—12.5 percent did so in 2017. Additionally, a relatively large 10 percent of Ithaca workers work at home. 
• Boulder, CO, is the metropolitan area with the largest share of workers who work at home—13.6 percent. Also, Boulder ranks second among metropolitan areas in the percentage of workers who commute to work by bicycle (4.6 percent). 
• Corvallis, OR, is the metropolitan area with the largest share of workers who bicycle to work—6.8 percent in 2017. Additionally, it has a relatively large share of workers who walk to work (7.6 percent) or who work at home (8.7 percent). 

Source: Demo Memo analysis of the 2017 American Community Survey

"No Religious Preference" Now and in Childhood

Only 9 percent of Americans were raised without a religious preference, according to the General Social Survey. But a substantial 22 percent now say they have no religious preference. In each generation, the percentage who currently have no religious preference is at least twice as large as the percentage who were raised without a religious preference...

Percent with no religious preference today (and in childhood)
iGeneration: 28% (11%)
Millennials: 32% (13%)
Gen Xers: 21% (10%)
Boomers: 15% (5%)
Older: 11% (3%)

Note: In 2016 the iGeneration was 18 to 21, Millennials were 22 to 39, Gen Xers were 40 to 51; Baby Boomers were 52 to 70, and older Americans were 71 or older.

Source: Demo Memo analysis of the 2016 General Social Survey

Baby Food Spending Plunges

The ongoing baby bust is being felt in at least one grocery store aisle. Average household spending on baby food fell 64 percent between 2007 (the year births peaked in the U.S.) and 2017, according to the Bureau of Labor Statistics' Consumer Expenditure Survey.

Average household spending on baby food (in 2017 dollars)
2017: $18.16
2010: $40.19
2007: $50.05

Beyond the decline in births, the other factor dragging down spending on baby food is the greater propensity of parents to make their own rather than relying on the store bought variety. That helps to explain why spending on baby food fell more than twice as much as spending on infants' clothes during the past decade. Between 2007 and 2017, average household spending on clothes for children under age 2 fell from $110 to $77, after adjusting for inflation—a 30 percent decline.

Source: Demo Memo analysis of the Consumer Expenditure Survey

Slowdown Ahead for College Enrollment

The nation's colleges should prepare for slower growth, according to a new set of projections by the National Center for Education Statistics. Enrollment in post-secondary institutions grew 9 percent between 2007 and 2017. Between 2017 and 2027, the gain should be only 3 percent. Enrollment growth will slow in every age group and for both men and women.

Some enrollment declines are forecast as well. NCES projects enrollment by race and Hispanic origin only for U.S. residents and not for foreign students. Among U.S. residents enrolled in college, NCES projects a decline in non-Hispanic Whites and growth for Asians, Blacks, and Hispanics in the decade ahead...

Percent change in college enrollment of U.S. residents, 2017 to 2027
Asians: 8.7%
Blacks: 5.8%
Hispanics: 13.3%
Non-Hispanic Whites: –6.9%

By 2027, non-Hispanic Whites will account for 48 percent of the nation's college students, down from 53 percent in 2017.

Source: National Center for Education Statistics, Projections of Education Statistics to 2027

Characteristics of Nursing Home Residents

More than 1.3 million Americans lived in the nation's 15,600 nursing homes in 2015–16, according to the National Center for Health Statistics' National Study of Long-Term Care Providers. Despite the aging of the population, the nursing home population is shrinking. There were a larger 1.5 million nursing home residents in 2000 compared to the 1.3 million counted by the latest survey. Here are the characteristics of nursing home residents in 2015–16...

• 65 percent are women
• 84 percent are aged 65 or older, and 65 percent are aged 75 or older
• 75 percent are non-Hispanic White, 14 percent Black, and just 5 percent Hispanic
• 48 percent have been diagnosed with Alzheimer's disease or another dementia
• 60 percent need help eating, 87 percent need help transferring in and out of a chair or bed, 89 percent need help toileting, and more than 90 percent need help bathing, dressing, and walking
• 62 percent are dependent on Medicaid to pay their nursing home costs

Source: National Center for Health Statistics, Long-Term Care Providers and Services Users in the United States, 2015–2016

Highest Paying Occupation by Education, 2017

Want your children or grandchildren to make a lot of money? The Bureau of Labor Statistics has suggestions for them, depending on how long they want to stay in school. It has identified the occupations with the highest annual wage for each educational attainment category. Not only that, the BLS has projected the number of job openings for those occupations during the 2016 to 2026 time period.

Doctoral or professional degree: Anesthesiologist
Mean annual wage: $265,990
Job openings 2016–26: 1,400

Masters degree: Nurse anesthetist
Median annual wage: $165,120
Job openings 2016–26: 2,800

Bachelor's degree: Chief executive
Median annual wage: $183,270
Job openings 2016–26: 20,000

Associate's degree: Air traffic controller
Median annual wage: $124,540
Job openings 2016–26: 2,400

Postsecondary nondegree award: Electricial repairer, powerhouse, substation, and relay
Median annual wage: $78,140
Job openings 2016–26: 2,100

High school diploma: Nuclear power reactor operator
Median annual wage: $93,370
Job openings 2016–26: 500

No formal educational credential: Mine shuttle car operator
Median annual wage: $56,890
Job openings 2016–26: 100

Note that many of these occupations do not have a lot of openings projected for the decade ahead. Not to worry. The BLS has alternatives. Here are the occupations near (but not at) the top of the pay scale in each educational attainment category that will have the most job openings in the decade ahead, from highest educational attainment to lowest: family practitioner (5,600 openings, $208,560); physician assistant (10,600 openings, $104,860); financial manager (56,900 openings, $125,080); dental hygienist (17,500 openings, $74,070); aircraft mechanic and service technician (10,900 openings, $61,020); detective (7,500 openings, $79,970); service unit operator, oil, gas, and mining (6,400 openings, $48,290).

Source: Bureau of Labor Statistics, High-Wage Occupations by Typical Entry-Level Education, 2017

First-Time Homebuyer Watch: 4th Quarter 2018

Homeownership rate of householders aged 35 to 39, fourth quarter 2018: 58.9%

After a month's delay because of the government shutdown, the Census Bureau has released the 4th quarter 2018 homeownership statistics. They show an uptick in the homeownership rate of younger adults. The homeownership rate of 35-to-39-year-olds—the nation's first-time home buyers—increased in the fourth quarter of 2018, rising above 58 percent for the first time since 2011. Post Great Recession, the homeownership rate of the age group dipped as low as 54.6 percent in 2015. It peaked at 65.7 percent in 2007. Clearly, there is an upward trend in the homeownership rate of this age group, likely due to the full-employment economy.  

What about their younger counterparts, householders aged 30 to 34, who were once the nation's first-time home buyers? Their homeownership rate rose to 48.4 percent in the fourth quarter of 2018, up from 47.1 percent a year earlier. Before the Great Recession, 30-to-34-year-olds were the nation's first-time home buyers (defined as the age group in which the homeownership rate first surpasses 50 percent). But their rate fell below 50 percent in 2011 and has been stuck there ever since. With the recent gains, 30-to-34-year-olds may be on their way to reclaiming first-time homebuyer status.


Nationally, the homeownership rate was 64.8 percent in the fourth quarter of 2018, up from 64.2 percent one year earlier. The difference is not statistically significant.

Source: Census Bureau, Housing Vacancy Survey

Median Household Income Rises in January 2019

Median household income continues to rise, according to Sentier Research, climbing to $63,688 in January 2019. This was 4.6 percent higher than the January 2000 median, after adjusting for inflation. It was the highest median yet measured by Sentier's household income series, which began in January 2000. Sentier's estimates are derived from the Census Bureau's Current Population Survey and track the economic wellbeing of households on a monthly basis.

The January 2019 median was 3.0 percent higher than the January 2018 median, after adjusting for inflation. It was 15.7 percent higher than the post-Great Recession low reached in June 2011 ($55,038)—a bottom hit two years after the official end of the Great Recession.

Sentier's Household Income Index in January 2019 was 104.6 (January 2000 = 100.0). To stay on top of these trends, look for the next monthly update from Sentier.

Source: Sentier Research, Household Income Trends: January 2019

You Don't Want to Belong To This 1%

The average American incurred $5,006 in medical expenses in 2016, according to the federal government's Medical Expenditure Panel Survey. But an unlucky few spent much more than average. Just 5 percent of the population accounted for half of all medical spending in 2016. Here is the average health care spending incurred per person in 2016 at selected percentiles of spending...

Average spending per person by percentile of health care spending, 2016
Top 1 percent of spending: $110,003
Top 5 percent of spending: $50,077
Top 10 percent of spending: $33,053
Top 50 percent of spending: $9,735
Bottom 50 percent of spending: $276

Not surprisingly, people aged 65 or older were the largest share (43 percent) of those in the top 1 percent of medical spending, and 45-to-64-year-olds accounted for another 34 percent.

Medicare was a big source of payment for those in the top 1 percent, covering 36 percent of their health care costs. Private insurance paid another 37 percent of their costs. The top 1 percent paid 5.5 percent of its health care bills out-of-pocket.

Source: Medical Expenditure Panel Survey, Concentration of Health Expenditures and Selected Characteristics of High Spenders, U.S. Civilian Noninstitutionalized Population, 2016

More Evidence of Census Concerns

The public is feeling uneasy about the 2020 census. A Census Bureau survey of attitudes toward the 2020 census found only 68 percent of the public reporting that they would be very or extremely likely to answer the census. This figure is considerably smaller than the 86 percent who felt that way in a survey prior to the 2010 census. And that's not all. A recent Public Religion Research Institute (PRRI)/Atlantic survey probed the public's attitude toward the census with two questions—one about whether the census would be used to check people's immigration status and the other about how the potential citizenship question would affect the census count. The public's answers raise alarm bells.

Question 1: "Do you think the government will use...Census records to check on an individual's immigration status, or do you think this information will only be used for counting the population?"

39% said they don't know how census records will be used
33% said census records would be used to check immigration status
just 26% said census records would be used only to count the population

Question 2: "For the first time, the 2020 Census may include a new question that asks individuals if they are legal citizen of the United States. In your opinion, how likely do you think it is that the Census will NOT get an accurate count because some people will be worried about answering this question?"

53% said it would be very likely that the census would not be accurate
23% said it would be somewhat likely the census would not be accurate
11% said they didn't know how it would affect the count
only 10% said it was somewhat/very unlikely that the census count would not be accurate

Interestingly, Republicans and Democrats are in agreement about the harm of a citizenship question —81 percent of Republicans and 77 percent of Democrats believe it is somewhat/very likely to result in an inaccurate count. Let's hope the Supreme Court has the same concern when it hears arguments about the citizenship question later this year.

Source: PRRI/The Atlantic 2018 Pluralism Survey, American Democracy in Crisis: The Fate of Pluralism in a Divided Nation

Feelings of Discrimination Linked to Income

Feelings of perceived discrimination are linked to income for both whites and blacks, according to an analysis of survey data by Ana Hernåndez Kent of the Federal Reserve Bank of St. Louis. Examining data from two surveys—the American Identity and Representation Survey (Resource Center for Minority Data) and the Americans' Changing Lives survey (National Institute on Aging), Kent compared the household incomes of whites and blacks after controlling for feelings of racial discrimination. While her analysis confirmed the fact that whites typically have higher household incomes than blacks, the within-race income comparisons offered some surprises...

• Whites who thought they had been discriminated against because of their race had a lower household income than whites who did not feel discriminated against, according to the results of the American Identity Survey. In fact, the "whites who did feel they were discriminated against had a median range of household income equivalent to the typical black who also felt discrimination," reports Kent. But the second survey—Changing Lives—showed no difference in the household incomes of whites based on perceived discrimination.
• The black pattern was different. "The pattern of results for blacks in both surveys turned these findings on their head," says Kent. "Blacks who felt they were discriminated against because of their race actually had higher median incomes than blacks who didn't feel racial bias."

"Looking at only blacks, felt discrimination appears to be related to improved financial outcomes," Kent concludes.

Source: Federal Reserve Bank of St. Louis, Perceived Bias and Income Patterns Differ by Race

Bored Teenagers

What is the single most commonly shared feeling among today's teens? Boredom, according to a Pew Research Center survey. Fully 40 percent of teens aged 13 to 17 say they feel bored every day or almost every day. Another 44 percent say they feel bored "sometimes." Boredom is far more common among teens than feeling tense or nervous, with a smaller 29 percent of teens saying they feel tense or nervous every day or almost every day.

When teens are asked about the personal pressures they face, the biggest by far is pressure to get good grades. The 61 percent majority of teens say they feel a lot of pressure to do well in school. Few teens feel pressure to be sexually active, drink alcohol, or use drugs. In fact, most teens say they feel no pressure at all to engage in those activities.

Percent of teens who feel "a lot" of pressure
61% to get good grades
29% to look good
28% to fit in socially
21% to be involved in extracurricular activities
13% to help family financially
8% to participation in religious activities
8% to be sexually active
6% to drink alcohol
4% to use drugs

Source: Pew Research Center, Most U.S. Teens See Anxiety and Depression as a Major Problem among Their Peers

42% of New Mothers Have a Bachelor's Degree

More than 40 percent of women who gave birth in 2017 had a bachelor's degree or more education, according to the National Center for Health Statistics. The 41.7 percent of of new mothers with at least a bachelor's degree in 2017 is nearly twice the 22.8 percent of 1997.

The percentage of new mothers with a bachelor's degree or more education varies by race and Hispanic origin...

Percentage of new mothers with at least a bachelor's degree, 2017
67.9% of Asians
51.3% of non-Hispanic Whites
24.6% of Blacks
18.9% of Hispanics
13.2% of Native Hawaiians, other Pacific Islanders
12.7% of American Indians, Alaska Natives

Source: National Center for Health Statistics, Educational Attainment of Mothers Aged 25 and Over: United States, 2017

Most Whites Do Not Perceive Local Discrimination

When Americans are asked whether Blacks in their community are treated less fairly than Whites, attitudes are strikingly different depending on whether the respondent is Black or White, according to a Gallup survey. The majority of Blacks think Blacks are treated less fairly than Whites in a variety of local situations. Few Whites agree or perhaps are unaware of local problems. Here are the percentages who Blacks and Whites who think Blacks are treated less fairly than Whites in their community by type of situation...

Dealing with police, such as traffic incidents
Blacks: 77%
Whites: 45%

On the job or at work
Blacks: 60%
Whites: 22%

In stores downtown or the shopping mall
Blacks: 59%
Whites: 25%

In restaurants, bars, theaters or other entertainment places
Blacks: 50%
Whites: 22%

In getting healthcare from doctors and hospitals
Blacks: 49%
Whites: 17%

In neighborhood shops
Blacks: 48%
Whites: 22%

Source: Gallup, Americans Less Satisfied with Treatment of Minority Groups

32% of Americans Are Financially Insecure

Despite the robust job market, many Americans are financially insecure—a substantial 32 percent of the nation's adults in 2017, according to the Urban Institute. This estimate comes from the The Well-Being and Basic Needs Survey, a new effort by the Urban Institute to measure financial insecurity.

The Urban Institute classified survey respondents as financially insecure if they had experienced at least one of three things in the past 12 months: 1) they were not sure they could come up with $400 for an unexpected expense; 2) they had missed a credit card or nonmortgage loan payment; and/or 3) they had been contacted by a debt collector. Fully 32 percent of adults were found to be financially insecure—22 percent were not confident they could come up with $400 for an unexpected expense, 14 percent had been contacted by debt collector; and 13 percent missed a loan payment. Here are some of the demographics of financial insecurity...

• Financial insecurity does not vary much by age, with 27 percent of 50-to-64-year-olds, 34 percent of 35-to-49-year-olds, and 36 percent of 18-to-34-year-olds financially insecure. 
• Race has more of an impact on financial insecurity, with Blacks most likely to be insecure (52 percent), followed by Hispanics (40 percent), and non-Hispanic Whites (27 percent). 
• Education matters even more than race. Fully 51 percent of people with less than a high school education and 38 percent of those with a high school diploma/some college were financially insecure. Among college graduates, the figure was a much smaller 17 percent. 
• Income matters a bit more than education. Among people with household incomes below poverty level, 58 percent were financially insecure. Among those with household incomes above 400 percent of poverty level, the figure was 14 percent. 

Despite the improving economy of 2017, concludes the Urban Institute, "many Americans are facing financial distress and struggle to keep up with their bills and cover unexpected expenses."

Source: Urban Institute, Financial Distress among American Families: Evidence from the Well-Being and Basic Needs Survey

Occupations that Grew Faster than Projected, 2006–16

In 2006, the Bureau of Labor Statistics projected that occupational employment would grow 10.4 percent by 2016. Boy, were they wrong. The actual growth during those years was just 3.6 percent, thanks to the Great Recession. When the BLS produces labor force projections, it assumes full employment. Employment was anything but full during and in the aftermath of the Great Recession.

But the BLS got a lot of things right. It correctly projected which occupations would grow or decline 75 percent of the time, according to its Occupational Projections Evaluations. (Yes, the BLS periodically evaluates its own projections for accuracy.) It projected which occupations would grow faster than the economy as a whole 54 percent of the time.

The Bureau projected especially rapid growth in the 2006 to 2016 time period for a number of occupations that ended up growing even faster than forecast—despite the ravages of the Great Recession. Here are some of those occupations, along with their actual and projected growth from 2006 to 2016...

Computer software engineers, applications
Actual growth: 64%
Projected growth: 45%

Mental health counselors
Actual growth: 58%
Projected growth: 30%

Personal financial advisors
Actual growth: 54%
Projected growth: 41%

Veterinary technologists/technicians
Actual growth: 43%
Projected growth: 41%

Pharmacy technicians
Actual growth: 41%
Projected growth: 32%

Source: Bureau of Labor Statistics, The 2006–16 Projections: How Did Fast-Growing Occupations Fare?

Millennials: The College Divide

A lot has been written about the struggles of Millennials. The generation had the misfortune to enter the job market in the midst of the Great Recession, an economic setback that has been throwing shade on them for more than a decade. But some in the generation are doing better than others, according to a Pew Research Center analysis that compares the wellbeing of Millennials to that of older generations at the same age (the 25-to-37 age group was used for the analysis). The dividing line between Millennial Haves and Have Nots is the bachelor's degree. Millennials with a bachelor's degree are doing as well or better than Gen Xers, Boomers, or older Americans (the Silent Generation) at the same age. Millennials without a bachelor's degree are doing worse.

Among Millennials who work full-time, those with at least a bachelor's degree earned a median of $56,000 in 2017, according to Pew. This was about the same as college-educated Gen Xers earned when they were aged 25 to 37, and it was more than college-educated Boomers or older Americans earned as young adults. The opposite is true for Millennials without a college degree. Those with only some college earned a median of $36,000 in 2017, less than their older counterparts at the same age. Millennials with no more than a high school diploma earned just $31,300 in 2017, also less than equally-educated Gen Xers, Boomers, or the Silent Generation when they were young adults.

The rising fortunes of Millennial college graduates and the declining fortunes of Millennials without a college degree have resulted in a growing gap in the median household income of young adults by educational attainment. Millennials with at least a bachelor's degree had an (adjusted for household size) median household income of $105,000 in 2017 versus $49,000 for those with no more than a high school diploma—a gap of $56,000. The gap was $54,000 for Gen Xers at the same age, $41,000 for Late Boomers, $29,000 for Early Boomers, and just $20,000 for the Silent Generation.

Source: Pew Research Center, Millennial Life: How Young Adulthood Today Compares with Prior Generations

Everyday Newspaper Readers Are Disappearing

Only 20 percent of Americans aged 16 or older say they read the newspaper every day, according to the 2016 General Social Survey. This figure is likely even lower today. In 2000, more than one-third (37 percent) of the public read the newspaper every day. In 1990, more than half (53 percent) were daily newspaper readers. Here is the percentage of Americans by generation who were daily newspaper readers in 2016...

Read a newspaper every day
Millennials: 10%
Gen Xers: 15%
Boomers: 30%
Older: 38%

A substantial 48 percent of Millennials, 39 percent of Gen Xers, 28 percent of Boomers, and 22 percent of older Americans say they never read the newspaper.

Note: In 2016, Millennials were aged 22 to 39; Generation Xers were aged 40 to 51; Baby Boomers were aged 52 to 70; older Americans were aged 71 or older.

Source: Demo Memo analysis of the General Social Survey

Giving Money Away in 2016

Forty-eight percent of American households gave money to persons or organizations outside the household in 2016, according to a Bureau of Labor Statistics' analysis of the Consumer Expenditure Survey. Among those who did, the average amount given was $4,298. Here is the percentage of households that gave money away during the year by type of recipient (and average amount given by donors)...

25.16% gave money to churches/religious organizations ($2,970)
16.54% gave cash gifts to family/friends outside the household ($2,391)
16.11% gave money to charitable organizations ($2,462)
  3.15% paid child support ($7,142)
  2.84% gave money to support students in college ($3,580)
  2.30% gave money to political organizations ($837)
  2.26% gave money to educational institutions ($2,716)
  0.46% paid alimony ($20,754)
  0.15% gave gifts of stocks/bonds/mutual funds to family/friends ($25,717)

Source: Bureau of Labor Statistics, The Relationship between Cash Contributions, Pretax Income, and Age

37% Retire Earlier than Planned

A substantial 37 percent of Americans retire before their planned retirement age, according to the Center for Retirement Research. The Center's researchers came to this conclusion after examining longitudinal data from the Health and Retirement Study for the years 1992 to 2012 to determine how many older Americans ended up retiring before their planned retirement age—one of the questions asked by the survey. The older the age at which people plan to retire, the more likely they are to retire before they planned...

Percent retiring earlier than planned
20% of those who planned to retire at age 61 or younger
26% of those who planned to retire at age 62
38% of those who planned to retire at ages 63 or 64
42% of those who planned to retire at age 65
55% of those who planned to retire at age 66 or older

What accounts for all these early retirements? Of the four factors considered by the researchers (health, employment, family, and financial), the most important is health. Absent health problems or a change in health status, the percentage who retire earlier than planned would drop from 37 to 32 percent, the researchers report. That's not much of a decline. In fact, the four factors considered by the researchers can explain only one-quarter of early retirements. What accounts for the rest? Perhaps "soft" factors, say the researchers, "like the lure of leisure time."

Source: Center for Retirement Research at Boston College, Retiring Earlier than Planned: What Matters Most?

19% Have Experienced Identity Theft

Millions of Americans have been the victims of identity theft, according to the Bureau of Justice Statistics. Nearly one in five (19.4 percent) people aged 16 or older has been a victim at least once in his or her lifetime, and 10 percent have been a victim in the past year. These findings come from the 2016 Identity Theft Supplement to the National Crime Victimization Survey.

Identity theft is not only common, but affects a growing share of the population. The 10 percent who were victims in 2016 is greater than the 7 percent measured in 2014. The Bureau of Justice Statistics defines identity theft as "fraud that is committed or attempted using a person's identifying information without authority." The three types of identity theft are the misuse of an existing credit card, bank, or other account; the opening of a new account in a person's name; and the misuse of personal information for fraudulent purposes—such as to get medical care.

The most common type of identity theft is misuse of an existing credit card account. In 2016, a substantial 4.3 percent of Americans aged 16 or older—or 11 million people—were the victims of credit card fraud. How did the victims discover the identity theft? Most were alerted to suspicious activity by a financial institution, the Bureau of Justice Statistics' reports. Only 7 percent of victims reported their identity theft to police, while 88 percent reported it to a credit card company or bank.

Actions taken by identity theft victims in the past 12 months
75.6% checked bank or credit statements
67.5% shredded documents with personal information
44.3% checked their credit report
36.8% changed passwords on financial accounts
16.2% used an identity-theft security program on a computer
11.7% purchased identity-theft insurance or credit monitoring service
4.7% purchased identity-theft protection

While the above list makes it look as though identity theft victims have learned their lesson and are getting serious about protecting their personal information, in fact the percentages who undertake these security enhancing activities are about the same for identity theft victims as they are for nonvictims.

Source: Bureau of Justice Statistics, Victims of Identity Theft, 2016

E-Cigarette Use: Young Adults by State, 2017

Young adults are the biggest users of e-cigarettes but their use varies by state, according to the CDC's Behavioral Risk Factor Surveillance System. Here are the five states in 2017 with the largest percentages of 18-to-24-year-olds who use e-cigarettes...

States with largest percentage of 18-to-24-year-olds who use e-cigarettes
1. Oklahoma: 15.4%
2. Wyoming: 15.2%
3. Arkansas: 14.7%
4. Tennessee: 14.5%
5. Missouri: 13.9%

Maryland had the smallest percentage of 18-to-24-year-olds who use e-cigarettes, with only 5.4 percent doing so in 2017. Vermont (6.1 percent) and California (6.4 percent) followed.

Source: CDC, BRFSS Prevalence and Trends Data

Median Household Income Stable in December 2018

Median household income in December 2018 was stable at $63,517, according to Sentier Research. This was not significantly different from the November 2018 median, after adjusting for inflation. The December 2018 median was 2.7 percent higher than the December 2017 median. Sentier's estimates are derived from the Census Bureau's Current Population Survey and track the economic wellbeing of households on a monthly basis. 

Median household income in December was 15.4 percent higher than the post-Great Recession low reached in June 2011 ($55,051)—a bottom hit two years after the official end of the Great Recession.

Sentier's Household Income Index in December 2018 was 104.3 (January 2000 = 100.0). To stay on top of these trends, look for the next monthly update from Sentier.

Source: Sentier Research, Household Income Trends: December 2018

More People Aged 65-Plus Are at Work

Two decades ago, the percentage of older men and women in the labor force was negligible. Marketers could safely ignore them and target only retirees in their messages to the 65-plus age group. Not so anymore. More than one-third of men aged 65 to 69 and one-fourth of those aged 70 to 74 are still working. More than one in four women aged 65 to 69 has a job...

Labor force participation rate of people aged 65-or-older, 1998 and 2018

	2018	1998
Men
Aged 65 or older	24.0%	16.5%
Aged 65 to 69	37.6	28.0
Aged 70 to 74	23.8	16.5
Aged 75 or older	11.9	7.5
Women
Age 65 or older	15.9%	8.6%
Aged 65 to 69	28.9	17.8
Aged 70 to 74	15.8	9.3
Aged 75 or older	6.4	2.9

Between 1998 and 2018, the number of workers aged 65 or older more than doubled (up 161 percent) because of the double whammy of rising labor force participation rates and the aging of the baby-boom generation. These increases will continue, according to the Bureau of Labor Statistics. The labor force participation rate of men aged 65 or older is projected to rise to 25.9 percent by 2026, and women's rate should climb to 18.3 percent. The number of workers aged 65 or older will expand by another 46 percent between 2018 and 2026.

Source: Demo Memo analysis of the Bureau of Labor Statistics' Labor Force Statistics from the Current Population Survey

The 6 Most Commonly Purchased Groceries

During an average week, nearly every household spends money on groceries. But shopping carts are disproportionately filled with only a handful of items. These are the 6 items purchased by at least one-third of households during the average week of 2017...

Percent of households buying item in an average week
1. Fresh fruit: 56%
2. Fresh vegetables: 56%
3. Milk: 47%
4. Bread: 44%
5. Cheese: 39%
6. Prepared food from the supermarket deli: 38%

Bananas are the most frequently purchased fruit (34 percent). Tomatoes are the most frequently purchased vegetable (22 percent). During an average week, nonwhite bread is added to more shopping carts (40 percent) than white bread (35 percent).

Source: Demo Memo analysis of the 2017 Consumer Expenditure Survey

Only 5.0% of Workers Have More than One Job

Really? If we are to believe the Bureau of Labor Statistics, which collects monthly employment figures through the Current Population Survey, then only 5.0 percent of workers had two or more jobs during an average week of 2018. But there is growing evidence that this number is way too low.

A National Bureau of Economic Research study by economists Lawrence F. Katz and Alan B. Krueger raises serious doubts about the 5.0 percent figure. Not only do Katz and Krueger think the number is too low, they are also skeptical of Current Population Survey data that show a decline in multiple job holding over the years—from a peak of 6.2 percent in 1996 to the 5.0 percent of today. So they designed an experiment to test the accuracy of the CPS's multiple jobs question.

Using a sample of Amazon Mechanical Turk participants, many of whom are multiple job holders, Katz and Kreuger asked their sample the standard Current Population Survey question about multiple jobs ("Last week did you have more than one job or business, including part time, evening or weekend work?") to see how many said yes. They also probed the sample about any additional work they did in the past week ("Did you work on any other... small paid jobs last week that you did not include in your response to the previous question?")

Among those who reported having only one job on the CPS question, fully 61 percent said they had failed to report another small job they had done in the reference week. Among those who reported having multiple jobs on the CPS question, an additional 38 percent reported having even more work than was captured by the CPS.

"The MTurk sample is highly non-representative," the authors note, "but this survey experiment demonstrates that the standard multiple job holding question in the basic monthly CPS is susceptible to underreporting." Interestingly, the Bureau of Labor Statistics agrees. While the BLS disputes the notion that its surveys have missed the rise of the gig economy, it admits that the CPS may undercount multiple job holders. For more on this, see the Monthly Labor Review article, Measuring Labor Market Activity Today: Are the Words Work and Job too Limiting for Surveys?

Source: National Bureau of Economic Research, Understanding Trends in Alternative Work Arrangements in the United States, Working Paper 25425 ($5)

State-to-State Student Migration

Among high school graduates who enroll in a four-year institution within 12 months of receiving their high school diploma, most enroll in a school in their home state. Fully 74 percent of students choose an in-state school, according to 2016 data collected by the National Center for Education Statistics. But the rate ranges from lows of 39 percent in New Hampshire and 42 percent in Connecticut to highs of 89 percent in West Virginia and 90 percent in Utah.

The net migration of first-year four-year college students also varies greatly by state. Most states gain from student migration, but 13 states lose more students to other states than they attract to their schools. Here are the five states with the biggest net migration streams, both positive and negative...

States with the biggest net gain of first-year college students
Pennsylvania: 15,525
Indiana: 9,656
Alabama: 8,532
Arizona: 7,361
Iowa: 7,358

States with the biggest net loss of first-year college students
New Jersey: –27,262
Illinois: –19,171
California: –14,164
Texas: –12,059
Maryland: –8,564

Other states that lose more first-year four-year college students than they gain are Alaska, Connecticut, Hawaii, Minnesota, Nevada, New Mexico, Washington, and Wyoming.

Source: National Center for Education Statistics, Digest of Education Statistics 2017

Homicides at the Nation's Schools

Hundreds of children have been murdered while at school over the past few decades, according to a grim study by the CDC. Nearly 400 children (393) were murdered in single-death incidents while at school from 1994 to 2016. Another 121 children were murdered in multiple-death incidents from 1994 to 2018. There are differences in these two types of crimes, according to the study findings...

Sex of victims and perpetrators: The victims in single-death incidents were mostly male (77 percent), while the victims in multiple-death incidents were evenly split between males and females. Males were the great majority of perpetrators in both types of incidents, with the proportion reaching as high as 98 percent in multiple-death incidents.

Race of victims: Blacks were a much larger share of victims in single-death (53 percent) than multiple-death incidents (12 percent). Non-Hispanic Whites were a much larger share of victims in multiple-death incidents (69 percent) than single-death incidents (23 percent).

Age of victims: Fully 78 percent of victims in single-death incidents were aged 15 to 18, another 19 percent were aged 10 to 14, and 3 percent were aged 5 to 9. In multiple-death incidents, a smaller 54 percent of victims were in the 15-to-18 age group, 23 percent were aged 10 to 14, and 23 percent were aged 5 to 9.

Cause of death: Firearms were the cause of death in 63 percent of single-death incidents. Stabbing accounted for another 24 percent. In multiple-death incidents, fully 95 percent of deaths were caused by firearms.

Source: CDC, Characteristics of School-Associated Youth Homicides—United States, 1994–2018

Online vs Paper 2020 Census Response

The 2020 census, for the first time, will allow the public to respond online rather than fill out a paper form. According to a 2018 Census Bureau survey of attitudes toward the census, most Americans would prefer to respond to the census online or don't care whether the form is online or paper. Overall, 40 percent would prefer online to paper, 28 percent do not have a preference, and 32 percent would prefer paper. Young adults and Asians are two of the groups who would most prefer to answer the census online...

More than half of young adults would prefer the online option: Among people aged 18 to 34, a substantial 56 percent would prefer to fill out the census online, as would 54 percent of those aged 35 to 44. The preference for an online form falls to 39 percent among 45-to-64-year-olds and to just 19 percent among people aged 65 or older. Fully 56 percent of the oldest age group would prefer a paper form.

Most Asians would prefer the online option: The 59 percent majority of Asians would prefer an online form, with only 20 percent expressing a preference for paper. Blacks are least likely to prefer the online option (29 percent) and most likely to prefer paper (44 percent).

Having the option to answer the 2020 census online may boost census response rates, helping communities better attract the government funding they are due because of a more complete count of residents. Those who would most prefer an online form—young adults and Asians—are also the demographic segments least enthusiastic about the census. Just 54 percent of young adults and 55 percent of Asians report being "very" or "extremely" likely to fill out a census form. This compares with a larger 67 percent of all adults, 69 percent of non-Hispanic Whites, and 73 percent of people aged 65 or older.

Source: Census Bureau, 2020 Census Barriers, Attitudes, and Motivators Study (CBAMS) Final Survey Report

Number of Homeless Has Declined since 2007

Every January in communities across the country the homeless are counted in a one-night census. The recently released 2018 report on the homeless has both good news and bad. The good news is that the number of homeless has fallen considerably since 2007. The bad news is that there are still more than half a million people living in shelters or on the street and the number has increased for the past two years.

Number of homeless in the U.S.
2018: 552,830
2017: 550,996
2016: 549,928
2015: 564,708
2010: 637,077
2007: 647,258

The nationwide effort to count the homeless divides the homeless population into individuals and families with children. Of the 372,417 homeless individuals in 2018, 52 percent were in shelters and the rest were on the street. Seventy percent of homeless individuals were men and 30 percent were women, transgender, or gender non-conforming, according to the Housing and Urban Development report to Congress. Among homeless individuals, 54 percent were White, 35 percent were Black, and 19 percent were Hispanic. Among homeless families with children, Blacks accounted for 51 percent.

Not surprisingly, the most populous state also has the largest number of homeless. California accounted for 30 percent of homeless individuals in 2018 and nearly half (49 percent) of the nation's unsheltered homeless. Among cities, Los Angeles had the largest number of homeless—42,079 in 2018. New York City was second with 33,391 homeless.

The 2019 count of the homeless is happening this week—it is conducted during the last 10 days of January. The 2019 report on the homeless population will be released by the end of the year. It will show us whether the number of homeless ticked upward for the third year in a row.

Source: US Department of Housing and Urban Development, The 2018 Annual Homeless Assessment Report (AHAR) to Congress

Student Loans May Explain Rural Population Decline

Rural counties have been losing population in recent years and urban counties have been growing. One factor behind rural population loss is the migration of young adults from rural to urban areas. What's driving young adults away from their rural homes? Student debt may be a factor, according to a study by the Federal Reserve Board. In an analysis of how student debt affects the migration of young adults in rural areas, the Feds examined Equifax/Federal Reserve Bank of New York Consumer Credit Panel data, comparing student debt levels of young adults in rural areas with changes in their census tract of residence over a six-year time period. These are the findings...

• Young adults with student loans were less likely to remain in rural areas than those without student loans—52 percent of those with student loans were still in a rural area six years later versus 66 percent of those without student loans.
• Young adults with the greatest student loan debt were least likely to remain in rural areas—37 percent of those in the highest quartile of student loan debt were in a rural area six years later versus 73 percent of those in the lowest quartile of debt.
• Young adults who moved to metropolitan areas did better than those who stayed in rural areas. They paid down their loans faster, had higher credit scores, and were more likely to have mortgage debt (own a home).

"With students borrowing at higher rates and in larger amounts to pursue postsecondary education," conclude the Fed researchers, "student loan debt may play an increased role in the dynamics of urban-rural migration."

Source: Federal Reserve Board, Consumer and Community Context, January 2019, "Rural Brain Drain": Examining Millennial Migration Patterns and Student Loan Debt," (PDF)

How Overweight Are Americans?

When the General Social Survey is taken every two years, GSS interviewers who talk to respondents in person are asked to rate respondents and their home and neighborhood on a variety of measures. How clean is their home, and how safe is their neighborhood? How attractive is the respondent, how well groomed, and how overweight? The answers to these questions can reveal interviewer bias and, over time, perhaps changes in the health and wellbeing of the public. Here is what GSS interviewers thought about the weight of Americans (the nationally representative sample of respondents to the GSS) in 2016...

How would you rate the respondent's weight?
  5.7% very overweight
26.7% overweight
60.2% about the right weight
  6.6% slightly underweight
  0.9% very underweight

GSS interviewers clearly don't know what overweight looks like since they report that only 32 percent of respondents are overweight or very overweight. Measurements of a nationally representative sample of the population, taken by the federal government’s National Health and Nutrition Examination Survey, show that a much larger 70 percent of the public is overweight and 40 percent is obese. GSS interviewers report that most respondents (68 percent!) are the right weight or underweight, but in reality just 30 percent of the public is normal or underweight.

It’s no surprise that GSS interviewers are poor judges of weight. The public is bad at judging weight too—especially their own. Only 43 percent of obese Americans know they’re obese, according to one study. The rest think they are about the right weight or only overweight and not obese. A Gallup survey finds that just 35 percent of men and 40 percent of women think they are overweight, well below the 70 percent measured by the government.

Source: Demo Memo analysis of the 2016 General Social Survey

How Clean Are American Homes?

More about this little known fact: when the General Social Survey is taken every two years, GSS interviewers who talk to respondents in person are asked to rate respondents and their home and neighborhood on a variety of measures. How clean is their home, and how safe is their neighborhood? How attractive is the respondent, how well groomed, and how healthy? The answers to these questions can reveal interviewer bias and, over time, perhaps changes in the health and wellbeing of the public. Here is what GSS interviewers thought about the cleanliness of the homes of Americans (the nationally representative sample of respondents to the GSS) in 2016...

How clean was the interior of the housing unit?
32.4% very clean
44.4% clean
15.0% so-so
  5.3% not very clean
  2.8% dirty

Opinions on cleanliness vary by age of respondent. The homes of younger respondents are perceived to be less spotless than the homes of their older counterparts. Among interviewers whose respondent was aged 18 to 29, only 23 percent reported that the respondent's home was very clean. Among interviewers whose respondent was aged 65 or older, a larger 46 percent reported that the respondent had a very clean home.

Source: Demo Memo analysis of the 2016 General Social Survey

Furloughed Workers Are Owed $4.7 Billion

The government shutdown is now in its 34th day, with an estimated 800,000-plus furloughed workers missing their paychecks. Just who are these federal workers? Sentier Research, which produces a monthly median household income series, analyzed the Census Bureau’s American Community Survey to determine the characteristics of furloughed federal government employees. Here are a few of the findings...

• 56% of furloughed workers are men, 44% are women.
• Furloughed workers have median earnings of $67,000.
• 38% of the households of furloughed workers include children.
• Among the 74% of households with a furloughed worker who own their home, 82 percent have a mortgage. Those with a mortgage have an average monthly mortgage payment of $1,600. The nation's furloughed workers’ cumulative monthly mortgage payment is $752 million.

Furloughed workers are now owed $5,600 each, Sentier estimates, or a total of $4.7 billion.

Source: Sentier Research, Furloughed Workers

How Attractive Is the American Public?

Little known fact: when the General Social Survey is taken every two years, GSS interviewers who talk to respondents in person are asked to rate respondents and their home and neighborhood on a variety of measures. How attractive is the respondent, how well groomed, and how healthy? How clean is their home and how safe their neighborhood? The answers to these questions can reveal interviewer bias and, over time, perhaps changes in the health and wellbeing of the public. Here is what GSS interviewers thought about the attractiveness of the American public (the nationally representative sample of respondents to the GSS) in 2016...

How attractive was the respondent?
  7.0% very attractive
31.0% attractive
55.4% about average
  4.9% unattractive
  1.8% very unattractive

Interviewer opinions about attractiveness vary by the age of the respondent being interviewed. Younger respondents are perceived to be more attractive than their older counterparts. Among interviewers whose respondent was aged 18 to 29, the 53 percent majority reported that the respondent was attractive or very attractive, 42 percent said average, and 5 percent said unattractive or very unattractive. Among interviewers whose respondent was aged 65 or older, a smaller 24 percent reported that the respondent was attractive or very attractive, 68 percent said average, and 8 percent said unattractive or very unattractive.

Source: Demo Memo analysis of the 2016 General Social Survey

Is There an Economic Bar to Marriage?

Is there an "economic bar" to marriage? To answer this question, try this thought experiment. Who would you rather marry: someone who lives on a couch in his/her parents' basement, or someone who has a job with health insurance and a home of his/her own? If you would rather marry the person with health insurance and a home, then yes, Virginia, there is an economic bar to marriage. We can all feel the bar, but there have been few empirical studies to determine whether facts support our feelings.

Until now. A recent study in Demography sets out to describe the economic bar to marriage and tests whether those who meet the bar are more likely to marry than those who do not. Researchers defined the bar as these seven accomplishments: 1) having private health insurance; 2) experiencing earnings growth in the past 15 months; 3) homeownership; 4) having a bank account; 5) being employed; 6) not experiencing any material hardship in the past year (hardship is defined as not being able to pay rent or mortgage, having utilities cut off, and/or being evicted); and 7) not receiving public assistance in the past month. Those who scored a four out of seven were defined as meeting the economic bar.

Analyzing three waves of the Building Strong Families dataset, the researchers determined how many unmarried parents in their 20s got married after a 15 and 36 month time period, controlling for whether or not they met the economic bar. Among study respondents, one or the other parent passed the bar (had a score of 4 or higher) in 67 percent of the couples. Only the mother passed the bar in 28 percent of couples, and only the father in 48 percent of couples. Both parents passed the bar in 14 percent of couples.

Those who met the bar were more likely to have gotten married after 15 and 36 months, the study found. "At both time points, using the either parent definition of the bar, couples who met the bar were significantly more likely to marry than couples who did not meet the bar." The study also found that "meeting the both-parent bar was associated with larger increases in marriage than meeting the either-parent bar." Meeting the mother-only bar also boosted marriage, although not as much as meeting the father-only bar. "Our findings suggest that father's economic contributions may be more important than mother's in determining marriage entry. Nevertheless...the bar's association with marriage was not driven solely by the father's contribution," report the researchers.

"Our results suggest that the [academic] enthusiasm for the marriage bar is warranted," conclude the researchers. "Couple-level economic progress may play a role in marriage formation."

Source: Demography, "His" and "Hers": Meeting the Economic Bar to Marriage, Volume 55, No. 6 ($39.95)

Rise in Student Loan Debt Accounts for 20% of Homeownership Decline among Young Adults

If student debt had not increased between 2005 and 2014—both in prevalence and in the amount owed—there would be 400,000 additional homeowners in the 24-to-32 age group. This is the finding of a Federal Reserve Board study of the factors behind the steep decline in homeownership among young adults.

The homeownership rate of 24-to-32-year-olds fell from 45 to 36 percent between 2005 and 2014, report the Fed researchers, an 8.8 percentage-point decline. This was much greater than the 3.9 percentage-point decline for the total population. At the same time, the share of the age group that had student debt climbed from 30 to 40 percent, and the average amount owed per capita doubled from $5,000 to $10,000. The researchers calculated how much these increases reduced homeownership, estimating that there would have been 400,000 additional homeowners in the age group if student debt had remained at the 2005 level.

But the increase in student debt accounts for only 2 percentage points of the 8.8 percentage-point decline in the homeownership rate of young adults (20 percent). What accounts for the rest? The Fed researchers suggest that student loan debt affected the credit scores of young adults in the aftermath of the Great Recession. Lower credit scores made it harder for young adults to qualify for a mortgage, resulting in lower rates of homeownership.

Source: Federal Reserve Board, Consumer and Community Context, January 2019, Can Student Loan Debt Explain Low Homeownership Rates for Young Adults? (PDF)

Opioid Prescriptions Are More Common in Rural Areas

Rural areas have been hit hard by the opioid epidemic. The injury is, in part, self-inflicted. Primary care physicians in rural areas are more likely than those in urban areas to prescribe opioids for their patients, reports the CDC in a study of opioid prescribing rates.

CDC researchers examined electronic prescriptions written by 31,422 primary health care providers during the January 2014 to March 2017 time period. Their findings document the big differences in opioid prescription rates by a county's urban-rural status...

Percent of patients receiving opioid prescriptions, January 2014—March 2017 average
9.6% in rural counties
9.4% in micropolitan counties
7.7% in small metro areas
6.7% in medium metro areas
5.6% in large fringe metros
5.2% in large central metros

The CDC undertook the study to determine whether the 2016 release of its Guideline for Prescribing Opioids for Chronic Pain had helped to reduce opioid prescriptions. It did. In every type of county, the opioid prescription rate declined over the time period. In rural counties, the percentage of patients who received opioid prescriptions fell from 10.3 percent in 2014 to 9.0 percent during the March 2016 to March 2017 time period. In large central metropolitan counties, the rate fell from 5.4 to 5.0 percent.

"As less densely populated areas appear to indicate both substantial progress in decreasing opioid prescribing and ongoing need for reduction," the researchers conclude, "community health care practices and intervention programs must continue to be tailored to community characteristics."

Source: CDC, Morbidity and Mortality Weekly Report, Opioid Prescribing Rates in Nonmetropolitan and Metropolitan Counties among Primary Care Providers Using an Electronic Health Record System—United States, 2014–2017

Below-Replacement Fertility in 48 States

The average woman in the United States will have 1.8 children in her lifetime, according to the age-specific fertility rates of 2017. This calculation, called a "total fertility rate" (TFR), is well below the 2.1 children the average woman would have based on age-specific fertility rates in 2007—the year the number of births in the U.S. peaked. A TFR of 2.1 is considered replacement level—meaning the population will neither grow nor decline, absent immigration. A TFR below 2.1, if extended for decades, will eventually lead to population decline.

Every state had a lower TFR in 2017 than it did a decade earlier. Here are comparisons between then and now...

• In 2017, only 2 states had at least replacement-level fertility (South Dakota and Utah). In 2007, more than half the states (27) had a TFR of 2.1 or higher. 
• In 2017, the lowest TFR was in Washington D.C., at 1.4. This was well below the lowest TFR in 2007, which was 1.7 in Vermont. 
• The highest TFR in 2017 was 2.2 in South Dakota, followed by 2.1 in Utah. In 2007, the highest TFR was 2.6 in Utah.  
• Between 2007 and 2017, the biggest declines in TFR (more than 20 percent) occurred in eight states: Nevada, Arizona, California, New Mexico, Colorado, Oregon, Georgia, and Texas. 
• Between 2007 and 2017, the smallest declines in TFR (below 10 percent) occurred in six states: Iowa, Ohio, Kentucky, Michigan, South Dakota, and North Dakota. 

In 2017, non-Hispanic White women had a TFR below the replacement level of 2.1 in every state, according to the National Center for Health Statistics. Black women had a TFR at or above replacement level in 12 states, Hispanic women in 29 states.

Source: National Center for Health Statistics, Birth Data, Total Fertility Rates by State and Race and Hispanic Origin: United States, 2017 and Births: Final Data for 2007

44% of Americans Are "Revolvers"

How many Americans carry a credit card balance, and how much do they owe? The answers can be found in a study by economist Joanna Stavins of the Federal Reserve Bank of Boston. She wanted to determine the unique characteristics of credit card users and in particular those who are "revolvers" — meaning they owe a balance on their credit cards.

Stavins examined findings from the Survey of Consumer Payment Choice, comparing revolvers' self-reports of their credit card balances with the Equifax credit bureau records of individual respondents. One of her goals was to determine the accuracy of self-reported balances, and she discovered they aren't all that accurate. Self-reported credit card balances are significantly lower than the balances reported by Equifax. Overall, 44 percent of adults in 2015–16 were revolvers (carrying a credit card balance). The average balance on the credit cards of revolvers was $6,597—25 percent greater than what respondents self-reported. Here are the percentages of Americans who are credit card revolvers by age, and their average credit card balance based on Equifax records...

Credit card revolvers by age, 2015–16 (and Equifax credit card balance)
Total adults: 44% ($6,597)
Under age 25: 26% ($2,913)
Aged 25 to 34: 44% ($4,472)
Aged 35 to 44: 49% ($7,192)
Aged 45 to 54: 51% ($8,336)
Aged 55 to 64: 48% ($7,493)
Aged 65-plus: 35% ($6,261)

Actual credit card balances are higher than self-reported balances in every age group. The biggest difference is among people aged 65 or older, who report an average balance of $3,795 while Equifax data show an average balance of $6,261.

Source: Federal Reserve Bank of Boston, Credit Card Debt and Consumer Payment Choice: What Can We Learn from Credit Bureau Data?

How Many Americans Have Ever Been Homeless?

More than half a million Americans are homeless on a given night, according to the Department of Housing and Urban Development. An annual survey, which is taken on a January night each year, counted 552,830 people homeless or in shelters in the United States in 2018.

Half a million people is a small fraction of the population on any one night. A much larger share of Americans experience homelessness at some point in their life. Among Boomers, the figure is 6.2 percent, according to a study in Demography.

The Demography study analyzed data from the Health and Retirement Study—a longitudinal survey of people aged 50 or older, which included a question about homelessness ("Have you ever been homeless or lived in a shelter?") in 2012 and 2014. The researchers calculated the lifetime prevalence of homelessness for people born from 1946 through 1964 (Boomers) as a whole and by race and Hispanic origin...

Percentage of Boomers who have ever been homeless
Total: 6.2%
Blacks: 16.8%
Hispanics: 8.1%
Non-Hispanic Whites: 4.8%

After controlling for socioeconomic characteristics, the difference in the experience of homelessness between Hispanics and non-Hispanic Whites disappears, report the researchers. Not so the difference between Blacks and non-Hispanic Whites. Black Boomers are three times as likely as non-Hispanic White Boomers to have experienced homelessness at some point in their lives, the researchers find.

"Do experiences of homelessness contribute to racial disparities in health?" they ask. "Do health disparities contribute to differences in the prevalence of homelessness? Are the two mutually reinforcing, or do they covary as products of social discrimination or economic inequalities? Future research might aim to better understand these complex pathways."

Source: Demography, Racial and Ethnic Disparities in the Prevalence of Homelessness in the United States, Vol. 55, No. 6 ($39.95)

36% Have Used Ride-Sharing Services

The percentage of Americans who use ride-sharing services such as Uber and Lyft more than doubled between 2015 and 2018, according to a Pew Research Center survey, climbing from 15 to 36 percent. Just 3 percent of the public had not heard of these services in 2018, down from a substantial 33 percent in 2015. Not surprisingly, young adults are most likely to have ever used a ride-sharing service...

Percent who have ever used a ride-sharing service, 2018
Aged 18 to 29: 51%
Aged 30 to 49: 43%
Aged 50-plus: 24%

Among those who use ride-sharing services, 10 percent say they use them at least weekly, 22 percent monthly, and 67 percent less often.

Source: Pew Research Center, More Americans Are Using Ride-Hailing Apps

40% of Women Under Age 30 Want to Leave U.S.

"Ideally, if you had the opportunity, would you like to move PERMANENTLY to another country, or would you prefer to continue living in this country?" Sixteen percent of Americans aged 15 or older say they would like to move to another country, larger than the share who wanted to move during the Obama (10 percent) or George W. Bush (11 percent) administrations.

Politics are a big factor in the desire to leave the United States. Among those who disapprove of Trump, 22 percent want to move. Among those who approve, only 7 percent want to get out.

There are also big differences by age and sex. The percentage who would like to move out of the U.S. is higher among women (20 percent) than men (13 percent). It is higher among adults under age 30 than older adults...

Desire to leave the U.S. by age
Aged 15 to 29: 30%
Aged 30 to 49: 19%
Aged 50-plus: 8%

Among women under age 30, fully 40 percent say they would like to leave the United States versus 20 percent of their male counterparts.

Source: Gallup, Record Numbers of Americans Want to Leave the U.S.

Household Growth Will be Slower, According to New Projections

Between 2018 and 2028, the number of households in the United States is projected to increase by 12.2 million, according to the Joint Center for Housing Studies (JCHS). This figure is 1.4 million less than the household growth projected by JCHS a few years ago.

Why the decline? The primary reason for the decline is that the Census Bureau issued new population projections, updating its 2014 vintage with a 2017 series. The Census Bureau's 2017 projections forecast slower population growth than the earlier series, says Daniel McCue of the JCHS and author of the report. The bureau's population projections form the basis of the JCHS household projection series.

Change in number of households by age of householder, 2018 to 2028 (in 000s)

	number (in 000s)
Total households	12,174
Under age 25	21
Aged 25 to 34	64
Aged 35 to 44	2,855
Aged 45 to 54	-401
Aged 55 to 64	-1,513
Aged 65 to 74	4,361
Aged 75 or older	6,787

As shown above, the greatest growth in the number of households in the decade ahead—accounting for more than half the increase—will occur in the 75-plus age group. Households headed by people aged 45 to 64 will decline as the small Generation X passes through. Very little growth is forecast in the number of households headed by adults under age 35.

Behind the Census Bureau's scaled back 2017 vintage projections—and behind the slower household growth forecast by JCHS—is lower immigration. Rather than gaining a net of 1.27 international immigrants per year during the coming decade (the assumption of the bureau's 2014 projections), the annual gain will be just 1.0 million. Fewer immigrants will slow the growth of the Asian and Hispanic populations and reduce the number of young adults establishing households in the years ahead.

Because of lower immigration, the JCHS projects that the total baseline demand for new housing in the 2018-to-2028 decade will be 1.51 million units a year, down from 1.69 million a year projected in its earlier series. The report cautions, however, that the new projection may not be conservative enough: "Given the strong steps taken by the Trump administration to curtail immigration, it remains to be seen whether there will be further declines going forward."

Source: Joint Center for Housing Studies of Harvard University, Updated Household Growth Projections: 2018–2028 and 2028–2038

Most Older Workers Experience Involuntary Job Loss

If you think you've got a retirement plan, think again. A study by the Urban Institute finds that more than half of older full-time workers—seasoned employees—are likely to lose their job before they turn 65, with dire consequences for earnings, household income, and retirement savings.

Examining Health and Retirement Study data from 1992 to 2016, the Urban Institute researchers tracked full-time workers aged 51 to 54 who had been with their current employer or self-employed for at least five years. Respondents were followed from their early 50s until at least age 65 to determine how many experienced an involuntary job separation—defined as an employer-related separation that resulted in at least six consecutive months of nonemployment or that reduced weekly earnings by 50 percent or more for at least two years.

Most of these seasoned older workers lost their job at some point during those years, with some losing a long-term job more than once. Fully 56 percent experienced at least one employer-related involuntary job separation. Demographics do not explain these derailments. There were few differences in the percentage of workers who experienced an employer-related job separation by sex, race, Hispanic origin, education, industry, or region of the country. Losing a steady job appears to be the norm for workers as they age.

The consequences of this kind of job loss are ugly. Only 10 percent of those who lost their job ever again earned as much as they had on the job, report the researchers. Median household income fell 42 percent after the job separation, with little difference in the extent of decline by demographic characteristic. At age 65, those who had experienced a job separation had a significantly lower household income than those who did not have a job separation, whereas the incomes of the two groups at ages 51 to 54 were essentially the same.

"Employment becomes increasingly precarious as workers age," conclude the researchers. "The steady earnings that many people count on in their 50s and 60s to build their retirement savings and ensure some financial security in later life can vanish, upending retirement expectations and creating economic hardship."

Source: Urban Institute, How Secure is Employment at Older Ages?

Attitudes toward Public Transit by Metropolitan Area

How many metro residents are happy with their public transit system? It depends on where they live, according to the 2017 American Housing Survey, which was fielded nationally and in 15 metropolitan areas. One of the questions asked by the survey is whether residents agree or disagree that their neighborhood has "good bus, subway, or commuter bus service." Here is the percentage of respondents who agree by metropolitan area...

Neighborhood has good bus, subway, or commuter bus service, 2017 (percent agreeing)
75.4% in San Francisco
75.4% in New York City
67.3% in Los Angeles—Long Beach
65.1% in Boston
65.1% in Washington, DC
64.5% in Chicago
62.1% in Riverside—San Bernardino
61.6% in Philadelphia
59.7% in Seattle
56.2% in Miami
47.6% in Phoenix
40.4% in Detroit
36.5% in Dallas
35.6% in Houston
32.7% in Atlanta

Attitudes toward public transit vary not only by metro area but also by homeownership status within metros. Near the top of the list is New York, for example, where 87 percent of renters and 64 percent of homeowners say public transit in their neighborhood is good. At the bottom of the list is Atlanta, where only 50 percent of renters and 22 percent of homeowners give their public transit a thumbs-up.

Nationally, 42 percent of Americans think their neighborhood has good bus, subway, or commuter bus service—57 percent of renters and 35 percent of homeowners.

Source: Demo Memo analysis of the 2017 American Housing Survey

Wireless-Only Now the Majority in 45-to-64 Age Group

The 55 percent majority of Americans aged 18 or older live in a wireless-only household, according to January-June 2018 estimates by the National Center for Health Statistics. Another 37 percent have both cell phones and a landline phone in their household. Just 5 percent are landline-only, and 3 percent are phoneless.

The percentage of adults who live in a wireless-only household became the majority in 2016. The wireless-only share grows each year as older age groups abandon landlines through generational replacement and/or technological surrender. Here are the percentages of adults by age who lived in a wireless-only household in the first half of 2018 (and 2015)...

Adults who live in a wireless-only household, January-June 2018 (and 2015)
Total, 18-plus: 55.2% (46.7%)
Aged 18 to 24: 65.0% (59.4%)
Aged 25 to 29: 77.3% (71.3%)
Aged 30 to 34: 77.3% (67.8%)
Aged 35 to 44: 65.7% (56.6%)
Aged 45 to 64: 50.7% (40.8%)
Aged 65-plus: 29.2% (19.3%)

Source: National Center for Health Statistics, National Health Interview Survey, Wireless Substitution: Early Release of Estimates from the National Health Interview Survey, January—June 2018

Median Household Income Rises in November 2018

Median household income in November 2018 climbed to $63,554, reports Sentier Research. This is the highest median recorded by Sentier since the January 2000 start of its monthly household income series. The November 2018 median was 3.2 percent higher than the November 2017 median, after adjusting for inflation. Sentier's estimates are derived from the Census Bureau's Current Population Survey and track the economic wellbeing of households on a monthly basis. 

"We are at a point now where real median household income is 4.3 percent higher than January 2000, the beginning of this statistical series," reports Sentier's Gordon Green. "Not an impressive performance by any means over a period spanning almost two decades, but the trend line has been positive for about seven years." More impressive is the 15.4 percent rise in median household income since the post-Great Recession low reached in June 2011—two years after the official end of the Great Recession.

Sentier's Household Income Index in November 2018 was 104.3 (January 2000 = 100.0). To stay on top of these trends, look for the next monthly update from Sentier.

Source: Sentier Research, Household Income Trends: November 2018

10 Questions: An Update (Part 2)

Two years ago Demo Memo presented 10 vital demographic questions and asked how many answers to these questions we would have once we had more data in hand. Two years later, the same questions are still of great importance. We have more data. So how much more do we know? Questions 1 through 5 were examined in this post. Here's a look at the rest...

6. Is the average American getting richer? With the benefit of hindsight, the answer to this question is yes and no. The wealth of American households plunged in the aftermath of the Great Recession. Median household net worth fell from $139,700 in 2007 to a post-Great Recession low of $83,700 in 2013, then climbed to $97,300 in 2016, after adjusting for inflation—still 30 percent below the 2007 peak. An analysis by the Federal Reserve Bank of St. Louis finds that the wealth of Americans born in the 1950s and earlier has recovered from the Great Recession losses, while the wealth of those born in the 1960s, 1970s, and 1980s has not.

7. Who voted in the 2016 election? This question was answered by the Census Bureau's survey of voting and registration, released in the spring of 2017. We now know that the number of older non-Hispanic White voters surged in 2016. Largely because of the aging of the baby-boom generation, 2.8 million more non-Hispanic Whites aged 65 or older voted in 2016 than in 2012. This trend is only going to intensify as the baby-boom generation continues to fill the 65-plus age group. While minorities will become the majority of the population in 2044, they will not become the majority of voters until 2064.

8. Are we back to square one with health insurance? Although Republican efforts to repeal the Affordable Care Act have not been successful, this question still matters after a federal judge in Texas declared the entire Affordable Care Act invalid—a case that may be headed for the Supreme Court. Meanwhile, a growing share of the public has a favorable view of the ACA, the figure rising from 43 percent in November 2016 to 53 percent in November 2018. This battle is ongoing.

9. How big is the gig economy? Are gig workers a tiny and stable fraction of the workforce, or are they an enormous and growing share of workers—24 percent according to one study and 31 percent according to another? We still don't know. In the past year, the BLS failed in its attempt to measure the gig economy, but nevertheless claimed gig workers to be few, far between, and not growing as a share of workers. Researchers scoffed at the BLS findings, theorizing that the Current Population Survey's labor force questions failed to capture gig work. The BLS fired back with a defense of the CPS. As the dust settles from this kerfuffle, all we know is that the size of the gig economy ranges from negligible to enormous.

10. Are we over the automobile? The evidence is building that we are past the point of peak transportation spending. The percentage of the household budget devoted to transportation is well below the all-time high of 19-plus percent of the mid-1980s and early 2000s. In 2017, transportation consumed a smaller 15.9 percent of the household budget. With transportation the second biggest expense for the average household, helping Americans cut their transportation costs is a no-brainer for both businesses and governments. It also helps explain the appeal of cities: urban households spend much less than their rural counterparts on transportation.

10 Questions: An Update (Part 1)

Two years ago Demo Memo presented 10 vital demographic questions and asked how many answers to these questions we would have once we had more data in hand. Two years later, the same questions are still of great importance. We have more data. So how much more do we know? Let's take a look...

1. When will the baby bust end? We now know it might not end. The decline in births since the 2007 peak is as great as the decline in the aftermath of the Great Depression (a 10.7 percent drop). But this may be more than a dip in the road. An analysis by the Center for Retirement Research suggests the fertility decline may be permanent, driven by structural changes in childbearing patterns. The United States appears to be adopting the low-fertility regime common in other developed countries.

2. Why is life expectancy declining? In two of the past three years, life expectancy in the United States has declined. We now know that the declines in 2015 and 2017 were due in large part to rising mortality rates among people under age 65—primarily from drug overdoses and suicides. The biggest increases in drug overdoses and suicide rates have occurred in rural areas—see Question 4, below.

3. Will homeownership make a comeback? The homeownership rate hit a post-Great Recession low of 63.4 percent in 2016, then climbed to 63.9 percent in 2017—the first statistically significant rise in more than a decade. While the increase is encouraging, the homeownership rate is unlikely to return to the highs of the early 2000s. The Millennial generation—now at the age of first-time home buying—is burdened by student loan payments. Student loans delay homeownership by 7 years, according to the National Association of Realtors. Consequently, the age of first-time home buying (the age at which the homeownership rate first surpasses 50 percent) has shifted from the early to the late thirties.

4. What will save small town and rural America? Corporate America will not be coming to the rescue, judging by Amazon's choice to place its second headquarters in two of the largest and richest metropolitan areas in the country. The continuing allure of urban areas makes this question more important than ever—and we don't yet have any answers. But we may know more about why rural areas are falling behind. A statistical concept called "gambler's ruin" explains it, says economist and New York Times columnist Paul Krugman. The gambler (small town or rural area) who starts with the fewest pennies (economic opportunities) is the one most likely to go bankrupt. Small towns and rural America have a shrinking pile of pennies to play with and are increasingly likely to face ruin because of it. We still don't know how to solve this problem or even if we can.

5. When will Millennials marry? The median age at first marriage continues to rise, reaching a new record high in 2017 of 27.8 years for women and 29.8 years for men. Why are Millennials delaying marriage? One reason is that a much larger share of them go to college, which delays marriage and childbearing. Another reason is student loan debt, which not only delays homeownership but also marriage, according to the National Association of Realtors.

To be continued tomorrow...