“Breadwinner Moms,” a recently released report from Pew Social & Demographic Trends, suggests, on its face, that gender equality in the labor force is perhaps closer than advocates for women’s rights would have us believe. The authors note that, as of 2011, “a record 40%” of households with children had mom as the primary breadwinner, up from 11% in 1960. There are a number of large-scale social and economic issues reflected in these seemingly straightforward numbers—changing household structures and trends in family formation; increasing female participation in higher education and the labor force; rising costs of living and stagnant wage growth that necessitate multiple earners within a family; and the lingering effects of the recession on labor market participation.

Moving beyond the initial “40%” number shows that there are really two populations being discussed: (1) single mother families and (2) married couple families in which the wife earns more than her husband. Using the term “breadwinner” with respect to women in single-mother families belies the economic realities of their situation. Single moms are the only potential earners in the family; many earn low (or no) wages and rely on public assistance to get by. This topic will be explored in greater depth in Virginia by Annie in our next blog post.

Discussions of the second population, married couples with “breadwinning” wives, gloss over problematic economic issues underpinning this shift, such as the disproportionate impact of the recession in male-dominated industries like construction and manufacturing. While two earner families may create new challenges at home, such as negotiating child care and housework, and yield divergent opinions on what’s best for children, they also reflect a basic economic reality for many American families: one income is not enough. I was also troubled by the use of the term “breadwinner”—traditionally used to describe a household in which a single earner is able to support the entire family unit—to describe two earner households in which one partner is earning more than the other. Moreover, the wage gap between husbands and wives was never made clear; how much more are these “breadwinning” moms actually earning compared to their husbands? Let’s take a quick look at the 2011 American Community Survey data for Virginia.

Two-Earner or “Traditional Breadwinner” Families Most Common

In Virginia in 2011, just over 1.5 million women were married and living with their spouse. Forty-two percent—nearly 623,000 of these married couples—had at least one dependent child under the age of 18. Of these families, 62 percent had both mom and dad in the labor force. The husband was the sole wage earner in 30% of families while the wife was the sole provider in 5% of families. In just under three percent of households, neither parent was employed.

Selected Economic and Family Characteristics by Parental Labor Force Participation, Virginia, 2011

In 2011, the median income for two earner households ($103,000) was substantially higher than other household types, reflecting the earnings power of two (typically full-time) workers. The “traditional breadwinner” families, in which only the husband was working, had the second highest median income, with nearly $79,000, while the female breadwinner households earned substantially less, reporting median incomes of only $57,000.

“Traditional Breadwinner” Families Likely to Transition to Two Earner Households

The mothers and fathers in families in which the husband is the sole provider are younger, on average, than both two earner and female breadwinner families. They also have more children (2.22) and younger children (average age of youngest child is 6) compared to other families. Many of these households may transition into two earner households as their children grow older and enter full-time schooling. In addition, eleven percent of the mothers in these families are currently unemployed, meaning that they are actively looking for work and not staying at home by choice.

Female Breadwinner Households Reflect Economic Necessity

Many families are relying on a single breadwinner because the other partner is unemployed or otherwise unable to work. In the 33,000 households in which mom was the sole provider, forty-two percent of fathers were unemployed. In addition, 19 percent of husbands in female breadwinner households reported a disability—much higher than the overall disability rate of 4.7% among married men with dependent kids

Male/Female Earnings Differences in Two Earner Households

Among the 623,000 married households with kids under the age of 18 in Virginia in 2011, nearly 384,000 were dual-earner families. In the vast majority of these families, husbands earn more than their wives; wives earn more than their husbands in 26 percent of these households. While dual-earner families had the highest median income overall ($103,000), the median income of female breadwinner households is slightly lower ($98,200) than households in which the husband earns the same as, or more than, his wife ($104,250).

Wife’s Income Compared to Husband’s in Two Earner Families with Dependent Children Under 18, Virginia, 2011

When examining how much more wives earn than their husbands in these “female breadwinner” households, a number of women are nearly at parity with their husbands, earning less than $5,000 more than their spouse. A fairly substantial proportion, however, out earn their partners by $10,000 or more, with 4% of wives reporting earnings that are at least $50,000 more than their spouse’s income. In these households, however, overall household income is quite high, with both partners typically earning high salaries.

Summary

 In Virginia, as in the nation, dual-earner families are the norm, reflecting both changing gender roles and increasing costs of living. While traditional breadwinner families are common, with the father the sole provider in nearly one-third of married families with dependent children, these families may soon transition to dual-earner households as their younger children grow older and mom can more readily enter the workforce. Among two earner households, husbands out earn their wives in nearly three quarters of families. In households in which wives earn substantially more than their husbands, both mothers and fathers are earning high incomes.

True “female breadwinner” households, in which women are the sole providers, are a small proportion of married couple households. Analysis suggests that barriers to husband’s employment, such as unemployment and disability, may drive these patterns more than significantly changing gender roles in which mothers “bring home the bacon” and men transition to the role of stay-at-home dad. Compared to two earner and traditional breadwinner families, these female breadwinner households are in an economically precarious position, with lower median incomes and greater reliance on social safety net programs, such as food stamps.

–

Rebecca Tippett is a Research Associate at the University of Virginia’s Weldon Cooper Center for Public Service where she studies household economic well-being and produces population estimates and projections.

For the first time, in a long history of disenfranchisement and suppression, African-American voter turnout surpassed the turnout rate among whites.  2012 was a low-turnout election overall, especially when compared to 2008, and the turnout rates among most of the major racial and ethnic groups went down from 2008 rates.  The turnout rate among blacks in 2012, however, went up.

* Turnout is measured here as total votes divided by the voting-age citizen population. Data are from the CPS microdata for the Voting and Registration Supplement.

There are many possible explanations for this historic turnout rate among African-Americans.  Habit and significance surely have something to do with it.  Once people vote once, they are more likely to do so again, especially if they feel invested in the outcome.  Those blacks who turned out to vote for the first black president in 2008 were more likely to vote for him again in 2012.  The 2012 turnout numbers might also underscore some of the deep political divisions of the county, which are increasingly drawn along racial lines, with the Republican party seen as being out-of-touch by many minority voters.

Another compelling explanation may lie in the black Baby Boomer population.  While other age groups had declining turnout rates, the overall increase in turnout among African-Americans in 2012 is mostly due to increased turnout among the Baby Boomer cohort; these were individuals who grew-up in and lived through the Civil Rights Movement and were perhaps less likely to take their voting rights for granted.  The recent Voter-ID laws, proposals, and sometimes blunt strategic rhetoric on the issue has had many civil rights groups and the NAACP make comparisons to past disenfranchisement efforts.  This may have been a compelling force for the black Baby Boomer Generation to turn out to vote.

There are other striking results from the recently released data; all of them point to a 2012 electorate that looks very similar to the 2008 electorate.  The population age 18 to 29 saw a dramatic decrease in their turnout rate from 2008.  Yet, because many Millennials are just entering voting age, the overall growth in this cohort meant that young voters made-up a similar share of the voting population as in 2008.  A similar dynamic happened among Hispanics and Asians.  Although their turnout rates dropped in 2012, they made up a slightly greater share of the electorate because of their numeric growth in the population.

As I argued in a previous post, the 2012 presidential election provides evidence that a fundamental political realignment has occurred in this country.  The newly released Census Bureau data confirm this and provide greater detail on the demographic forces underpinning this shift.

Now let’s take a look at ground zero for this political realignment:  Virginia…

A year ago, Michele and I published a report on the major demographic trends in Virginia and how they influenced, and will continue to influence, presidential politics in the state. A part of that analysis was gaming-out 2012 election scenarios based on demographic projections.  Now that I have the data, I can see how well we did.

Virginia’s Racial and Ethnic Minority Population and Share of the Electorate

As shown, while we got the total minority population just right (blue bars), we overestimated the growth in the Hispanic citizen population and thus overestimated the share of the voting-eligible population belonging to minority groups (red bars).  However, racial and ethnic minority share in the electorate (green bars) fell right between our two projections based on 2004 and 2008 turnout assumptions, but closer to the 2008 scenario.  Republican pollsters during the campaign tended to believe more in the 2004 assumptions, while Democratic polling firms believed that 2008 patterns were more illustrative of what might happen.  The 2012 election proved the Democrats were more right than the Republicans when it came to polling and election predictions.

In the Virginia case, black turnout was a major contributor for the greater minority share in the electorate.  Just as blacks had higher turnout at the national level, black turnout in Virginia was also up in 2012.  The following figure shows turnout rates between Non-Hispanic whites and African-Americans in Virginia:

White and Black Turnout Rates in Virginia, 1992-2012

The table below shows the young population’s share of eligibles and the electorate in Virginia. As a result of overestimating the growth in the Hispanic population, we also overestimated the growth in the voting-eligible population age 18 to 29 (again, the red bars).  While their share of the pool of possible voters grew, the lower turnout rates among this group ensured that they were a smaller share of electorate (green) compared to 2008 (but still greater than all previous elections since 1996).

Virginia’s 18 to 29 Population as a Share of the Electorate

By looking into the new Census data for Virginia, I cannot help but notice how well the state serves as microcosm for national trends, not only demographic trends but political trends as well.  Stay tuned for more as I continue to delve deeper into the data and look into how demographics are influencing big political changes.

–

Dustin Cable is a Policy Associate at the University of Virginia’s Weldon Cooper Center for Public Service where he conducts research on topics that lie at the intersection of demographics, politics, and public policy.

Despite common sense, official poverty rates do not account for this variability.  The income threshold for poverty for a family living in New York City is the same for a similar family living in Fargo, North Dakota.  More disturbingly, billions of dollars of government benefits and services are distributed to localities and families based on the official poverty measure.  Other public and private organizations also use the official statistics to target their operations.

With so much at stake, a new poverty measure that addresses regional differences in the cost of living is needed.  The Census Bureau has recently developed the Supplemental Poverty Measure (or SPM) to do this at the national level, but states and localities are at a disadvantage.  Only 3-year SPM averages are available for states and none are available at the sub-state level.

Virginia, however, now has a new alternative poverty measure that accounts for regional differences in the cost of living and provides sub-state estimates of poverty rates.  As elaborated in my previous post, the new “Virginia Poverty Measure” (VPM) provides some interesting insights about economic distress in the commonwealth, but perhaps the most striking results are the result of its regional adjustments.

The VPM uses Regional Price Parities published by the Bureau of Economic Analysis for its cost of living adjustments across geography.  These price parities are similar to the Consumer Price Index (CPI), but instead of adjusting dollar values across time, the price parities adjust values across geographic regions.  The price parities are indexed on the national average which is given a value of 100.  Regions with index values above 100 have greater costs of living while those that have values less than 100 experience costs below the national average.

The following table shows price parities for all fifty states across different types of expenditures:

Virginia is 3% more expensive to live in compared to the national average.  However, as shown in the next table, this is primarily because of the extraordinary costs of living in the Washington D.C. metro Area:

Northern Virginia has, on average, an 18.6% higher cost of living compared to the national average, one of the highest Regional Price Parity values in the nation.  Using these to adjust poverty income thresholds for the VPM produces striking differences in poverty rates across Virginia, as shown in the table below from the full VPM report:

These alternative ways of measuring poverty provide surprising insights into the true population in economic distress.  Using them in place of the official poverty measure for certain types of funding allocation schemes or benefit calculations could prove to be a more efficient and effective use of tax dollars.  The Census Bureau stresses that its new SPM, that uses geographic adjustments, is for research purposes only.  However, I hope this kind of research and state experimentation in alternative poverty measures, like the VPM and similar projects in Wisconsin and New York City, lead to replacing the official measure with something that conforms to common sense and better reflects the economic realities of different places.

–

Dustin Cable is a Policy Associate at the University of Virginia’s Weldon Cooper Center for Public Service where he conducts research on topics that lie at the intersection of demographics, politics, and public policy.

Answering that question is not as easy as one may think.  Yes, we do have an official poverty statistic that is produced by the U.S. Census Bureau, but nobody likes it.  Many on the Left think it is too low, failing to capture the full array of expenses that families face.  Folks on the Right think it is too high because it does not account for the effects of many anti-poverty programs and tax credits on family budgets.

Even the Census Bureau is not entirely satisfied with current poverty statistics.  As they continue to produce the official measure, they have recently been releasing alternative statistics through the “Supplemental Poverty Measure” (SPM) program.  These new numbers reflect a more nuanced look into poverty, and are widely believed by researchers and the media to better capture the actual financial circumstances of American families.

But the SPM has its limitations.  Primary among them is that the new measure is designed for the national level.  State estimates are only available as three-year averages, and local-level estimates are not available at all.

This is unfortunate for a state like Virginia, which has wide regional inequalities in terms of economics, education, and even basic demographics.  Because of this, official poverty statistics don’t make sense in Virginia.  A one-size-fits-all measure that defines poverty in Northern Virginia the same as it does in coal country does not work and belies our commonsense understanding of the actual resources and costs families face across regions.  A better method is needed.

Today, the Cooper Center is releasing its work on a new “Virginia Poverty Measure” (VPM) that will provide SPM-like estimates for Virginia and its local regions.

These VPM estimates represent a dramatic improvement upon official statistics including adjustments made for:

• Regional differences in the cost of living. The VPM accounts for regional differences in the cost of major goods and services such as housing, food, and health care. As expected, costs vary tremendously across regions in Virginia, and VPM poverty thresholds are adjusted accordingly.

• Taxes and Credits. Payroll taxes and federal and state income taxes are subtracted from family resources in the VPM. Also, the VPM accounts for important refundable tax credits such as the federal Earned Income Tax Credit.

• Necessary medical expenses. Health care is a growing part of family budgets, and the VPM accounts for these necessary expenses by adding them to the poverty thresholds according to family size, age of household members, and health insurance status.

The full details of the VPM and its major findings are made available through the Cooper Center website, but here are a few of the major highlights:

• Although Northern Virginia counties and cities enjoy some of the highest median incomes in the nation, the VPM shows that the extent of economic deprivation in the region is significantly greater than what official poverty statistics suggest. For example, by capturing the impact of the region’s high cost of housing, the VPM finds many more Northern Virginia residents in or near poverty, particularly those living inside the Beltway.

• The VPM poverty rate for children is dramatically lower than the official rate. Official statistics do not account for the impact of many government programs targeted favorably towards families with young children. By including these tax code provisions and in-kind benefits, the VPM recognizes the full range of resources available for families with young children.

• By including calculations for taxes and adjustments for costs of living, the VPM classifies a greater number of people as “near poor.” However, by including more government programs and subsidies for the poor, the VPM finds fewer Virginians in “deep poverty.”

These results underscore why Virginia needs an alternative way to measure poverty, and the new VPM is a significant step forward towards better understanding the actual population in economic distress in the commonwealth.

You can read the full report here.

–

Dustin Cable is a Policy Associate at the University of Virginia’s Weldon Cooper Center for Public Service where he conducts research on topics that lie at the intersection of demographics, politics, and public policy.

The Wall Street Journal recently released a fantastic interactive graphic that shows trends in monthly food stamp participation, by state, from 1990 through 2013. Most states follow the overall national trend: participation rises in the mid-1990s, gradually declines through the boom years of the late 1990s and early 2000s, flattens slightly through the 2000s, and then sharply increases following 2008.

Some states, like Alaska, have strong seasonal fluctuations in participation rates, although the magnitude of this variation has diminished in recent years.

In other states, program participation spikes dramatically following natural disasters. In Louisiana, SNAP enrollment nearly doubled in late 2005 following Hurricane Katrina. Damaging hurricanes in 2008 (Gustav and Ike) and 2012 (Isaac) led to similar spikes in enrollment.

Interested in exploring more? We have an interactive map of SNAP recipients by county for Virginia in 2010. For even greater detail, the USDA Economic Research Service provides a detailed interactive map of SNAP participation, benefit amounts, and socioeconomic indicators by county for 2006-2010.

–

Rebecca Tippett is a Research Associate at the University of Virginia’s Weldon Cooper Center for Public Service where she studies household economic well-being and produces population estimates and projections.

This map, built on data from the Center for Disease Control’s Behavioral Risk Factor Surveillance System (BRFSS), shows the prevalence rate of adult obesity by state for 1985 to 2010. 

Prior to 1991, no state had an adult obesity rate greater than 15%. Seven years later, in 1998, the highest obesity rates were above 20%. Three years later, in 2001, obesity rates in Mississippi exceeded 25%, and rates in other states soon rose. By 2010, no state had an adult obesity rate less than 20%; adult obesity rates were equal to or greater than 25% in nearly three-quarters (36) of the states.

Data for 2011 shows that adult obesity rates remain high across the nation; though prevalence rates continue to rise, there is some evidence that the rate of increase has slowed in recent years.

Obesity is a major risk factor for many of the leading causes of preventable death, such as heart disease, stroke, and type 2 diabetes, and is associated with significantly higher medical costs. Given the size of the epidemic, even small changes in obesity prevalence would result in substantial medical savings, as well as improvements in longevity.

–

Rebecca Tippett is a Research Associate at the University of Virginia’s Weldon Cooper Center for Public Service where she studies household economic well-being and produces population estimates and projections.

• Virginia 2010

• Fredericksburg City 2010
• Richmond-Pertersburg Metro Area 2010
• Martinsville 2010
• Lynchburg City 2010
• Harrisonburg City 2010
• Staunton-Waynesboro 2010
• Roanoke-Salem 2010

Plus the ones from the previous post:

• Northern Virginia 2010
• Charlottesville City 2010
• Winchester City 2010
• Hampton Roads 2010

–

Dustin Cable is a Policy Associate at the University of Virginia’s Weldon Cooper Center for Public Service where he conducts research on topics that lie at the intersection of demographics, politics, and public policy.

Take the City of Charlottesville as an example: The great thing about dot density maps is that they elegantly convey a lot of data in a small space.  Total population, population density, geographic distribution, and race/ethnicity are displayed in a single visual.  Also, by incorporating the racial and ethnic data, the extent and degree of residential segregation manifests itself.  These maps still work for even the most densely populated areas… Fairfax, the Beltway, and Manassas: Norfolk, Portsmouth, Newport News, and Hampton: However, like the Charlottesville example, some of most interesting maps are for Virginia’s smaller cities and towns. Winchester City:

Despite their utility and beauty, these maps have their limitations.  They are bounded by the highest resolution possible with census data, namely population data by Census Block, the smallest unit of census geography (roughly equivalent to a city block in a urban area).  The dots are randomly placed within Census Blocks so sometimes may not represent actual residences for some larger area and less-populated Census Blocks.

Note:  The racial categories for White, Black, Asian, and Other are all non-Hispanic.  Hispanic dots represent a person of any race, but are usually categorized as Hispanic White or Hispanic Other.

High-resolution images available upon request.

– Dustin Cable is a Policy Associate at the University of Virginia’s Weldon Cooper Center for Public Service where he conducts research on topics that lie at the intersection of demographics, politics, and public policy.

In the spirit of March Madness, the Census Bureau has developed their own bracketology-themed population game. You should take a few minutes and play a round. It’s pretty fun.

You’ll find match-ups of states or metro areas, and you simply pick the one with the larger population. You’ll go through all the pairings until you’ve selected what you think is the state or metro area with the largest population in the country.

The Census Bureau has developed quite a few tools and games like this to showcase their data.  You can find the entire gallery on their webpage: http://www.census.gov/dataviz/

Among those of us who love old maps, the good people at the Atlas of Historical County Boundaries project have digitized and uploaded historical information on the shape of American counties.  With this data one can animate how America’s political boundaries have changed since the founding of the Massachusetts Bay and Virginia Colonies.  The above video shows historic county boundaries from 1630 to 1910 (shortly after Oklahoma and Indian Territory joined to form the State of Oklahoma in 1907).  Please note these boundaries show the creation of government-defined geographic units, not necessarily where population is located.

Another great thing about this data is the level of detail available.  For instance, focusing on the monumental changes that Virginia has gone through is quite interesting:

Note the emergence of many of Virginia’s Independent Cities at the turn of the 20th Century.

Things get more interesting when these county files are merged with historical census data.  Inspired by our previous post on “Every person gets a dot,” I decided to look at county population dot densities from the first United States Census of 1790 to the recent 2010 Census.  Here, every dot represents 5,000 people:

–

Dustin Cable is a Policy Associate at the University of Virginia’s Weldon Cooper Center for Public Service where he conducts research on topics that lie at the intersection of demographics, politics, and public policy.