“All models are wrong, but some are useful” – George Box

Trying to predict the future is a risky business because it is inherently uncertain, but we need to do so in order to plan.

The Cooper Center recently released Virginia population projections for 2020-2040. The projections were developed by extrapolating historical population trends. As part of our comprehensive research efforts, we also study population change in terms of births, deaths, and migration (though our official projection model does not require these input data, historical population trends are driven by trends in components of population change). This blog post focuses on how U.S.A. fertility rates have changed over time and how they might look in the future.

Total Fertility Rates


The total fertility rate (TFR) is the most frequently used statistic to describe birth trends. The TFR for a given year is the number of children a hypothetical woman would have in her lifetime if she followed the age-specific fertility rates (ASFR) observed in that year. The chart above shows U.S.A. TFR grew from about 2 children per woman in the 1930s during depression to more than 3 during baby boom. Currently, the TFR is slightly below 2 children per woman. Read Full Article →

Last month our office released Virginia’s population projections for 2020 until 2040. If you combed through the data tables, then you may have noticed one striking prediction—Whites will make up less than half of Virginia’s population by 2040. Recently, the Census Bureau made a similar projection for the nation as a whole. While these types of predictions about how the U.S. population will change have received a good deal of attention in recent years, there has been little examination of how the U.S. non-White population could manage to grow from 12 percent of the population in 1960 to over 50 percent by the 2040s. The addition of new census race/ethnicity categories in recent decades, such as Hispanic and Two or more races, has played an underappreciated role in the growth of the U.S. minority population. Census Race Projections

Source: Census Bureau U.S. population projections. Aside from Hispanic, all other races are non-Hispanic.

Both our projections and the Census Bureau’s expect Whites to become a minority not because the U.S. White population will shrink considerably but because the U.S. non-White population is expected to grow much faster. Though all non-White racial/ethnic groups are predicted to grow in size, Hispanic Americans are expected to contribute to the bulk of population growth between now and 2040, making up 59 percent of all population growth nationally and 77 percent of all growth in Virginia. But any projections for over 2 decades out that rely on one group maintaining such rapid growth should be treated with some caution. Read Full Article →

Latest population projections show that Virginia will continue to grow steadily over time, and be home to more than 10 million people by 2040. The statewide growth rate is starting to slow down from 13% over 2000-2010 to a little over 9% between 2010-2020, which is consistent with trends observed at the national level. Comparing projections across 50 states shows that Virginia is still poised to become the 10th most populous state in the country by 2040.Historic Census Count & Projection

Read Full Article →

The weatherman

Forecasting weather accurately is rarely possible but the public counts on exactly that – an accurate prediction. And the combination doesn’t always work out well as illustrated in the 2005 film, The Weather Man, when Nicholas Cage, who plays a meteorologist, sometimes gets the weather forecast wrong and is pelted with half-consumed fast food by angry viewers who don’t understand the complexity of forecasting.

Forecasting population change, like forecasting the weather, is complex, requires one to make assumptions about the future, often based on past trends, and is rarely spot on. Because population projections are less familiar to the public, projections are often treated as something closer to a fact, rather than a forecast that can and likely will change. Unfortunately, not understanding population projections can lead to much larger problems than a rained out barbecue. Read Full Article →

Job polarization refers to a situation in the labor force where growth is concentrated among both low-and high-wage jobs, while the number of middle-wage jobs declines. Inspired by a blog post about job polarization in Oregon since the Great Recession, I found that the same trend holds true for Virginia. While the number of low-and high-wage jobs in Virginia has increased compared to pre-recession levels, middle-wage jobs have only recovered about a third of their recessionary losses.

Virginia Employment Change Since 2008 by Wage Category

Source:Bureau of Labor Statistics, 2016 Occupational Employment Statistics.

For purposes of this analysis, I classified the 22 civilian occupational groups of the Bureau of Labor Statistics’ Standard Occupational Classification system into three wage categories—low, middle, and high. Middle-wage jobs were defined as those occupational groups having a 2016 Virginia median wage between 75 and 150 percent of the 2016 Virginia median wage for all occupations (roughly $30,000 to $60,000).

Read Full Article →

I recently needed to research the commuting flows of the Charlottesville workforce and resident population to learn about the daytime population, and I needed to have an idea of what was happening at the neighborhood level. The U.S. Census’ Longitudinal Employer-Household Dynamics Origin-Destination Employment Statistics (LODES) made this level of analysis possible. LODES data include pairs of residence and administrative work locations at the Census Block level for about 95% of all jobs. The 2014 LODES primary jobs data and maps I developed to visualize what’s happening provide a glimpse of where workers are coming from and what the flow exchanges in the area look like.

The Charlottesville Metropolitan Statistical Area includes Albemarle County, Fluvanna County, Greene County, Nelson County, and Charlottesville City. In addition, Louisa and Augusta counties are among the top contributors of workers to Charlottesville. For these maps, I included anyone who lived or worked in the 7 localities mentioned. Workers residing outside but working in the 7 localities are categorized as “Other: in state” or “Other: out of state.”

This bird’s eye view shows where the employed population live:


And this map shows where the employed population work:

1-establishing-shot50 Read Full Article →

“Healthy citizens are the greatest asset any country can have” is a quote often ascribed to the British PM Sir Winston Churchill. In that spirit, and in light of the recent health care reform debate within the US, I wondered what the current picture of health coverage actually looks like.  The Census Bureau offers tools that allow us to see the same data set from different perspectives. Since 2008, the historical face of health insurance has changed significantly as depicted in Fig 1 below. The transition starts gradually and until 2013, most states remain in the brackets with lower insurance rates. But over 2014 and 2015, with the exception of Alaska and Texas, the uninsured rate across all states dramatically dropped below 14%.



Across the US, health insurance rates vary, both by region and by age groups as seen from the map in Fig 2. By 2015, the uninsured percentage of the population was mainly concentrated in the 18-64 ages, with the very young and very old having greater health coverage in comparison. Read Full Article →

This morning the Weldon Cooper Center at the University of Virginia released its 2016 population estimates for Virginia’s counties and cities. The most obvious trend in the population estimates is how much more slowly Virginia and most of its communities are growing this decade. Since 2010, Virginia’s population has grown by a little over 410,000 residents. During the same period of the last decade, Virginia added over 604,000 residents.

Population Change

Source:Decennial Census and 2016 State Population Estimates

As last week’s post noted, Virginia’s population growth is slowing in large part because many of its residents are leaving the state. The last time Virginia grew at a rate that was slow as the rate it is experiencing now was during the 1920s, which was also the last time many Virginians were leaving the commonwealth for other states.     Read Full Article →

In December, the Census Bureau released its annual state population estimates, which showed that Virginia grew by 44,000 residents last year, its smallest numerical gain in population since the 1970s. The main cause of Virginia’s slower growth is that, for the past three years, more people have been leaving Virginia for other states than moving in. Prior to 2013, IRS tax return data (which the Census Bureau uses to track migration) had never recorded out-migration from Virginia since the IRS began publishing the data in 1978.

1990 to 2015 Migration Virginia

*Census migration data (2014-15 net domestic migration). IRS migration data (total net migration) used for all other years. Though the 2014-2015 IRS data showed a similar trend, changes in the way the IRS published the data affected the comparability of 2014-15 IRS migration data with data from previous years. 

The most obvious reason why more people are leaving Virginia than moving in is the state’s economy, which has lagged behind the rest of the country since 2011, especially since the federal budget sequestration began in early 2013. Prior to the budget sequestration, Wells Fargo Securities ranked Virginia, Maryland, and the District of Columbia as the areas most dependent on federal spending.

Since 2013, Northern Virginia in particular has been losing residents to other states. Fairfax County, which contains over a third of Northern Virginia’s population, has experienced a significant rise in residents moving to metro areas with healthier economies, such as Houston or Los Angeles.  Read Full Article →