Defining the Generations

by George Masnick
Fellow

Demographers and other analysts have yet to reach a consensus about how we define post-WWII generations – regarding both naming the generations and defining the age spans that each generation covers.  Yes, there is general agreement that the oldest of these generations are called baby boomers and that they were born between the mid-1940s and mid-1960s.  Some choose 1946 to 1964 to correspond with the period of increasing birth rates, but some choose 1945 to 1964 to produce a cohort that better lines up with typical age groups published in census and survey data, say 15-34 year olds in the 1980 census.

The generations that come after baby boomers are much less consistently defined.  The next youngest was first called the baby bust to identify cohorts born during the period in which birth rates and the number of births were rapidly declining. When this generation became teenagers and young adults, it was discovered that the baby bust would likely behave quite differently from the baby boomers that preceded them in the age structure. Pundits on Madison Avenue and in the media rebranded it Generation X.  Some analysts claim that this generation arrived in the early 1960s (perhaps as early as 1961) and that its youngest members were born in 1981 or 1982. Others chose sometime in the mid-1970s as the date at which the youngest of the baby bust came into this world.

The next youngest generation, called Generation Y by uncreative verbivores, and millennials by a more hip crowd, is even less consistently defined by starting and stopping birth years. Called echo boomers by many demographers, birth years can be anywhere from the mid-1970s when the oldest were born to the mid-2000s when the youngest were. 

The problem with these differing birth year definitions is that either the generations overlap or they cover different age-spans, making data analyses very complicated.  For much of the research done at the Joint Center for Housing Studies, we define the baby boom as the cohort born 1945 to 1964, the baby bust from 1965 to 1984, and the echo boom from 1985 to 2004.  The primary reason for choosing these dates is to have the three generations cover equal 20-year age spans, and have age ranges that line up with typically published age groups.  And, not coincidently, these chosen years line up nicely with levels of annual births (see figure below).  Early baby boom cohorts quickly move to annual birth numbers in the 3.7 million range, and the youngest baby boom members are from a cohort when births returned to approximately this number. The baby bust ends when annual births returned to these same levels.  The echo boom begins and ends with numbers consistently above 3.7 million.

Though it is the number of births used to fix the year of birth of the oldest and youngest members of each generation, the size of each generation moving forward in time is also determined by immigration.  Thus, the baby boom generation arose from approximately 79.3 million births over the 1945 to 1964 period, but by 1985, when the generation was age 20-39, it was over 80 million in number, having been inflated by immigration that more than offset Vietnam War era baby boom deaths.  The baby bust generation contained only 69.7 million persons born in the United States, but high immigration flows from persons born abroad resulted in a cohort size that just surpassed the size of the baby boom when each was age 20-39.  Finally, while the echo boom has about the same number of U.S. births for its base as the baby boom, it will surely further surpass the size of baby boomers as young adults.  The current set of low immigration household projections from the Joint Center are based on population projections that estimate 86.5 million persons age 20-39 year old in 2025. 

Sources: United States Census Bureau and the National Vital Statistics Report

It is not uncommon to hear the claim that the baby boom is being succeeded by a much smaller cohort.  But this is simply no longer true if the comparison is among similar 20-year deep generations. Immigration has backfilled the birth deficit that created the baby bust and it is now larger than the baby boom.  The 2010 decennial census counted 81.5 million baby boomers and 82.1 million baby busters. But perhaps the most relevant comparison generation to aging baby boomers should be the echo boom.  It is this generation that will largely be buying baby boomer housing and making significant contributions to their Social Security and Medicare benefits when all boomers are over the age of 65. After 2030, when the ever-smaller cohort of surviving baby boomers are age 65-84, the echo boom will be age 25-44, and far fewer from this still expanding generation will have died.

The Promise of Big Data for Accelerating ‘Prosperity Development’

by Ben Hecht
Guest Blogger

From time to time, Housing Perspectives features posts by guest bloggers. This post was written by Ben Hecht, President and Chief Executive Officer of Living Cities.  His post reflects thoughts he shared at a Brown Bag Lecture delivered at the Harvard Kennedy School on November 1, 2012.  His talk was entitled "From Community to Prosperity."

Recently, I was asked to contribute a chapter to Investing in What Works for America’s Communities, a book that examines what we can learn from the history of community development and provides a multitude of ideas from diverse perspectives about the future of the field. My essay, entitled From Community to Prosperity emphasizes that powerful forces of change, such as globalization and the internet, have transformed the way that people live, work, and interact; and that the community development industry has failed to adequately adapt to these seismic changes. Today, there is a new 'social operating system’ that is in stark contrast to the one that was built around geography and small tight-knit groups. We must fundamentally re-think how we define “community”, and change the way that we work to reflect the needs and realities of the present and the future. I call this new, broader, focus ‘prosperity development’.  Prosperity development requires us to invest in new models for collective problem-solving that acknowledge that no one institution or sector can solve today’s increasingly interconnected and complex problems alone. It requires overhauling long-broken systems. And, it requires harnessing new technologies in creative ways to accelerate social change efforts.

There is no ‘accelerant’ with greater promise than Big Data, which a recent New York Times article described as “shorthand for advancing trends in technology that open the door to a new approach to understanding the world and making decisions.” Today, more data is being created from more places than ever before. Tweets, clicks, YouTube videos, retailer loyalty cards, cell phones, even sensors on buildings are producing tons of data daily. Trends in public sector data transparency are adding even more valuable data to the mix. In order to turn the promise of Big Data into reality, we must make a real commitment to more evidence-driven decision making, and be willing to challenge entrenched ideas and beliefs. The benefits of such a commitment are significant and have the potential to make cities much better places for all residents:

1. Better Decisions about People and Places. Many people often criticize government for making decisions without full consideration or understanding of the facts. Big Data can really change that. If harnessed right, it can allow us to make better decisions about people and places. On the people side, data collection systems have evolved rapidly over the last decade with more sophisticated and varied sources for capturing information including 311 calls, educational performance and health care. On the place side, more buildings, roads and machines today have sensors that are providing data 24/7 about volume of usage, energy consumption, etc., what people often refer to as the Internet of Things. This proliferation of data enables us to tie an array of decisions to real-time, current and substantial data sets.

2. Opportunities for Accelerating Technology for Civic Change. The growing trend of the public sector to make more of its data open to the public has led to an explosion of innovation and is redefining how citizens participate and interact with their government. To date, ‘civic tech’, or the building of apps based on public data, has focused on improving civic life generally, from real-time bus schedules to virtual land use planning. However, it’s not hard to imagine how civic tech, intentionally applied to the lives of low-income people and communities, could be transformational – from changing the relationship between police and neighborhoods to enabling online appointment scheduling and enrollment for public benefits that now force people to take off work or suffer face-to-face humiliations.

3. Predictive Possibilities. The predictive power of Big Data is being explored and will be a big part of improving fields as diverse as health care, economic development and education. We are already seeing the potential to use big data to predict student performance in states like North Carolina where education leadersare using high-tech data analytics to examine grades, attendance, course failures, declines in grade point average, and disciplinary incidents of elementary school students to predict who might be at risk of falling of track and even failing to graduate high school. Big data’s predictive power is also of increasing interest to police forces that face greater limitations due to municipal budget constraints. New efforts are emerging in cities to map the location and time of crimes to better understand where police need to be and when to provide the greatest benefit to communities. This predictive element of big data can dramatically improve effectiveness and drive efficiency and this will be one of the most exciting areas for cities to focus on especially as more and more of theworld’s population continues to move to metropolitan areas.

As we work to move the community development sector into the future, a fundamental question will be: How can we harness Big Data for common good?

Cities are Growing but Sprawl Continues

by Dan McCue
Research Manager

The 2010 decennial Census provided new data to look at urban growth patterns over the past decade, enabling us to update our views on the latest trends in suburban sprawl and movement ‘back to the cities’.  Growth in our nation’s largest cities has been of particular interest not just because these areas are home to so many people, but because they are the engines of economic growth that often serve as bellwethers of the overall economic health of a region or even the nation as a whole. As has been reported elsewhere, growth in cities has been positive over the past decade, with the majority of the core cities in the nation’s largest metropolitan areas experiencing household growth.  Indeed, our analysis has found that core cities in 72 of the 100 largest metro areas grew in the 2000s. But at the same time, and in contrast to many headlines, our research also shows that the pace of growth in suburbs and exurbs exceeded that in the central cities in all but five metros. Suburban sprawl was alive and well.

Tracking Urban Growth:  Data and Methodology

Using 2000 and 2010 Census data for the top 100 metros, we analyzed the spatial distribution of household growth over the last decade by looking at changes in core urban cities (defined as large cities with populations over 100,000), suburbs (defined as urbanized parts of metros not in large cities), and exurbs (defined as all remaining non-urban tracts).

Our analysis found that, while household growth in core urban cities did occur, on the whole it was slower and smaller than growth in the suburbs and exurbs (See Figure 1).  Overall, the rate of household growth in the core urban cities of these 100 metros was just 6.4 percent, which was much slower than that of suburbs (10.5 percent), and a fraction of the growth rate of the exurban areas (28.3 percent).  In terms of magnitude, the number of households in the top 100 core urban cities grew by just 1.65 million in 2000-2010 while during the same period suburbs increased by 3.0 million households and exurbs grew by 3.2 million.

Source: JCHS Tabulations of U.S. Census Bureau data

Failure to grow as fast or as much as the suburbs and exurbs over the past decade caused the share of households living in the top 100 metropolitan areas’ core urban cities to drop.  Core urban cities were home to fully 39 percent of all households in 2000, but accounted for only 21 percent of all metropolitan area household growth.  Exurban areas, on the other hand, represented only 17 percent of all metropolitan households in 2000 but 41 percent of total metropolitan household growth in 2000-2010. As a result, the share of households living in urban cities dropped from 39 percent in 2000 to 37 percent in 2010, while the exurban share increased from 17 to 20 percent.  The suburban share remained stable at 43 percent of all households.

Looking individually at the distribution of growth within each of the top 100 metros, we found that even in metros where core urban city household growth rates were high, suburban and exurban growth rates were often higher (click here for Excel spreadsheet).  And in the five metros where core urban cities grew faster than the metropolitan areas as a whole, core city growth rates were only 0.5 percentage point higher than overall metropolitan growth rates.  In contrast, in the remaining majority of metros, core urban growth rates were fully 8.8 percentage points slower than the overall metropolitan growth rates.  As a result, many metros saw their share of households living in core urban cities drop significantly (See Figure 2).

Notes: Data include the 100 largest metro areas, ranked by population in 2010. Cores are cities with populations over 100,000. Suburbs are all urbanized areas outside of cores. Exurbs are the remainder of the metro area. Census data do not include post-enumeration adjustments.  Source: JCHS tabulations of US Census Bureau, Decennial Census.

To summarize our top-level findings, data from Census 2010 gives evidence of a ‘back to the cities’ movement in the form of household growth occurring in the majority of core urban cities of the top 100 metropolitan areas.  However, our analyses so far have also provided much more convincing evidence that suburbs and exurbs continued to drive metropolitan household growth over the past 10 years.  Indeed, even in the fastest growing cities, growth typically was neither as fast nor as large as that occurring in the surrounding suburbs and exurbs.

Mind the Gap: The Importance of Over and Under Counts in Interpreting Decennial Census Results

by George Masnick
Fellow

Every ten years the Decennial Census provides a definitive count of the country’s population and households, a critical benchmark of decadal growth trends in these key measures. However, often overlooked in assessments of these trends is that after each census, the Census Bureau conducts a post-enumeration evaluation that provides an estimate of the likely error rate in the Decennial Census counts. When calculating decadal population or household growth using decennial censuses, failure to adjust for differential undercount/overcount between censuses can lead to spurious results. This is especially true when calculating growth between 2000 and 2010 and comparing it to growth between 1990 and 2000, because the magnitude and direction of the estimated errors have varied substantially over these three counts. Taking undercount/overcount estimates into account, growth in both population and households over the last decade was slightly higher than during the 1990s, even though the unadjusted numbers suggest a fairly substantial slowdown in growth.

In the spring the Census Bureau released initial findings from the Post-Enumeration Survey indicating that the 2010 Decennial Census hit the center of the bull’s eye on a true population count.  The Bureau estimated that the 2010 Census overcounted total population by a mere 0.01 percent (1 one hundredth of one percent).  This amounts to only 35,000 people, which is not statistically significant from zero.  In contrast, the Bureau estimated that the 2000 Census overcounted population by 0.49 percent, or 1.37 million persons.  The difference in overcount between the two censuses means that population growth from 2000 to 2010 was actually 1.34 million larger than calculated by unadjusted population counts due to the difference in the estimated overcounts in each year. This estimate is for the overall population, so there are likely some groups (broken down by age or race/Hispanic origin, for example) with relatively greater growth adjustments necessary.

For 2010 the Bureau stated that there was also no significant overcount/undercount of occupied housing units (households). The initial (March 2001) post-enumeration evaluation following the 2000 Census had to be re-done and the final report (March 2003) did not include an estimate of the overcount/undercount for households. But if we assume that households suffered from the same 0.49 percent overcount as population in 2000, then we need to add an additional 500,000 households to the 11.2 million growth calculated for 2000 to 2010 from the unadjusted numbers. At 11.7 million, the 2000 to 2010 census-to-census household growth is actually fairly close to the decadal growth of 12.2 million from estimates derived from the Current Population Survey (based on a rolling 3-year average of households to dampen sampling variation in these estimates).

By adjusting for overcount/undercount, not only was population (and probably household) growth over the past decade greater than results from unadjusted raw census numbers, but also the slowdown in growth from the previous decade’s level was non-existent.  This is because actual growth between 1990 and 2000, adjusting for overcount/undercount, is significantly reduced.  Unadjusted population growth between 1990 and 2000 was 32.2 million, while adjusted growth was substantially less at 26.7 million.  Unadjusted population growth from 2000 to 2010 was 26.4 million, and the adjusted figure is 27.7 million.  So instead of a slowdown in total population growth in the 2000s compared to the 1990s, adjusted figures show a slight increase in population growth (Table 1).  And if household counts were off by the same percentage as the population estimates, household growth between 2000 and 2010 also exceeded growth in the previous decade.

Source: JCHS tabulations of US Census Bureau data

The 11.7 million adjusted household growth estimate for 2000 to 2010 is larger than the 11.5 million adjusted growth number for 1990 to 2000.  Still, growth in the 2000s is well short of the expected household growth due to the impacts of the Great Recession.  Household growth fell off sharply after 2007 because of the huge upheaval in both housing markets and the broader economy, with its attendant hit to household formation rates of those under the age of 35 and a dramatic slowdown in immigration. Together, these forces more than offset the expected increase of household formation by echo boomers who began to turn age 20 in 2005.

Most of the postponed echo boom household formation will be realized in the near-term future as these young adults inevitably form households and so is not permanently lost.  But overall household growth also depends on future levels of immigration, which remains the biggest wildcard.  Immigration flows are very much dependent on future economic trends, both here and in sending countries, as well as on how much future immigration is enabled in a highly uncertain political climate.

The Resurrection of Household Growth: The Missing Link in the Housing Recovery

by Chris Herbert
Research Director

New survey data released by the Census Bureau last week provides more strong evidence that the nascent housing recovery is being built on a solid foundation.  The latest Housing Vacancy Survey (HVS) (which provides quarterly updates on the number of households as well as vacancy and homeownership rates) indicates that as of the third quarter of this year U.S. households were increasing at a rate of just over 900,000 per year, giving a strong boost to overall housing demand.  This represents a substantial increase from 2011 when the same survey found that the country only added about 635,000 households.  While still well below the rate of 1.18 million per year that the Joint Center estimates the U.S. is likely to average over the 2010-2020 decade, the upturn in 2012 indicates that the moderate but steady pace of economic growth is finally translating into increased demand for housing, which bodes well for a sustained recovery.

Much of the attention on housing market woes has focused on the supply side. Certainly, as the housing bubble was bursting at the end of 2006 the excess supply of housing was a significant contributor.  As of that time, the U.S. had added more new homes over the previous 10 year period than at any other point going back to the early 1970s when record keeping for housing completions and mobile home placements began. But after five straight years of housing starts below 1 million units a year—a level last seen in 1945 when World War II was ending—it is hard to argue that the housing market is still suffering the after effects of overbuilding. In fact, by the end of 2011 the total amount of new housing put in place over the previous 10 years was the lowest since record keeping began.

Note: Available data go back to 1974.  Source: JCHS calculations of US Census Bureau data

Instead, the lagging recovery in the housing market has been more attributable to anemic growth in the demand for housing as indicated by weak household growth.  According to the HVS, household growth fell sharply in 2007 as the housing bubble collapsed.  (See Figure 2.)  After averaging 1.35 million from 2000 through 2006, over the next 5 years annual household growth barely exceeded 600,000.  Given the trends of the last five years, the spurt in household growth to an annual rate of 900,000 through the first three quarters of this year is notable.  If the upward trend in household growth continues, housing should see a sustained recovery in 2013.

Note: 2012 is an estimate of annual average growth based on trends through the third quarter of 2012. JCHS low projection assumes that immigration in 2010-20 is half that in the US Census Bureau’s 2008 middle-series (preferred) population projection. Sources: US Census Bureau, Housing Vacancy Survey; JCHS 2010 household growth projections