Skip to main content
NCBI home page
The Journals of Gerontology Series B: Psychological Sciences and Social Sciences logoLink to The Journals of Gerontology Series B: Psychological Sciences and Social Sciences
. 2021 Jul 28;77(4):780–789. doi: 10.1093/geronb/gbab141

Impact of the 2008 Recession on Wealth-Adjusted Income and Inequality for U.S. Cohorts

Naomi Zewde 1,✉, Stephen Crystal 2
Editor: Jessica Kelley
PMCID: PMC9122750  PMID: 34320206

Abstract

Objective

To examine the distributional effects of the 2008 recession and subsequent recovery across generational cohorts.

Methods

Using data from the Survey of Consumer Finances (2007–2016), we constructed a measure of economic well-being accounting for income, household size, and annuitized value of assets. We examine trajectories of adjusted income and inequality, using Gini coefficients and income shares by decile, for the overall population and by cohort during the recession and recovery.

Results

Inequality declined temporarily during the recession, but reached new highs during the recovery. During recovery, population-level increases in economic resources were not reflected among below-median households, as the more concentrated financial assets rose while broader-based home equity and employment fell or remained stagnant. Inequality measures increased for cohorts in their primary working years (Generation-X and Baby Boomers), but not among the younger Millennials, who were at early stages of education, workforce entry, and household formation.

Discussion

The study illustrates an integrative approach to analyzing cumulative dis/advantage by considering interactions between historically consistent macrolevel events, such as economic shocks or policy choices affecting all cohorts, and the persistent life-course processes that tend to increase heterogeneity and inequality as cohorts age over time. Although recovery policies led to rapid recovery of financial asset values, they did not proportionately reach those below the median or their economic resource types. Results suggest that in a high-inequality environment, recovery policies from economic shocks may need tailoring to all levels of resources in order to achieve more equitable recovery outcomes and prevent exacerbating cohort inequality trajectories.

Keywords: Generational outcomes, Wealth distribution trends, Wealth inequality

Inequality over the life course reflects interactions between persistent tendencies toward increasing heterogeneity with age (Crystal, 1982; Dannefer, 1987; O’Rand, 1996) and the particular historical circumstances, including economic developments and policy choices, that shape the acquisition of human, social, and health capital over time (Crystal, 2020; Crystal et al., 2017; Crystal & Shea, 1990a). These factors are illustrated by the impacts of the Great Recession of 2008 and the subsequent recovery, which offer an opportunity to examine the effects of economic shock and recovery policies from an integrative cumulative advantage perspective (Crystal, 2020).

The Great Recession drastically affected employment rates and asset values across the U.S. economy. In response, the federal government invested large sums aimed at stabilizing assets held in collapsing financial institutions and stimulating recovery. These policies carry diverse implications for wage earners and asset holders and may interact with ongoing processes of stratification.

Across a generation’s life course, early-life advantages are magnified through disparate occupational and social trajectories that lead to wide late-life disparities in financial and health resources, in a process first termed by Crystal and Shea as one of “cumulative advantage and disadvantage” (CAD; Crystal, 1982, 1986, 2020; Crystal et al., 1992, 2017; Crystal & Shea, 1990b; Dannefer, 1987, 1988). Dannefer (1987) described the trend of increasing inequality over the life course as the “Matthew effect,” applying a biblical dictum first used by Merton (1968), stating that “to he who has much, more is given, and to he who has little, even that is taken.” This ongoing process has also been described as an “obdurate tendency” for increasing inequality over the life course (Dannefer, 2020). In addition to economic well-being, the concept of cumulative advantage has been applied to cumulative disparities in health and other outcomes (Dannefer, 2020; Ferraro & Kelley-Moore, 2003; Hales & Barker, 2001; Lee & Park, 2020).

This persistent tendency toward CAD can be mitigated or exacerbated by policy choices, institutional structures, and temporal economic trends during critical stages of a generation’s life course (Crystal, 1986; Crystal & Shea, 1990a, 1990b). For example, policies on public and private pensions, educational financing, progressiveness of taxation, and health care financing can magnify or buffer processes that would otherwise lead to increasing inequality over the life course. External events such as recessions or wars influence these processes differentially across cohorts experiencing these events at particular ages and developmental stages (Oreopoulos et al., 2012; Pfeffer & Hertel, 2015).

In this study, we examine the interaction between business cycle fluctuation and CAD within and across cohorts during the recession and recovery. We evaluate inequality in asset-adjusted income and highlight the effects of the recession and recovery on inequality trajectories. We anticipated that the adverse effects of the recession would be greatest in cohorts affected during their working years, retarding accumulation of wealth and human capital during critical years of occupational careers.

Background

Economic Context

The historical, political, and economic context of each era shapes the social institutions driving cumulative stratification (Crystal & Waehrer, 1996). In the United States and elsewhere, the context has been one of increasing inequality and instability. The late twentieth and early twenty-first centuries saw a rise in wealth inequality, diverging from the post-World War II decades (Piketty, 2014). In 1978, the wealthiest 0.1% of American households held an outsized 7% of the total wealth; by 2012, they had 22% (Saez & Zucman, 2016). The inequality has only grown since.

Alongside wealth inequality, the labor market has also tilted increasingly in favor of monopsonistic employers. Losses in employee bargaining power, largely credited to substantial declines in unionization, are linked to stagnant wages (Erickson & Mitchell, 2007; Naidu & Sojourner, 2020); declining nonwage benefits, with market-dependent 401k retirement accounts for the advantaged, but weakened guarantees of income and health care for the disadvantaged (Crystal, 2016); and diminishing opportunities for stable employment, particularly for less-educated workers, due to large-scale de-industrialization.

Against this backdrop, the 2008 recession caused a crash in income and asset values followed by an uneven recovery. The recession dramatically reduced employment, earnings, homeownership, and net worth across the economic distribution (Edwards & Hertel-Fernandez, 2010; Famighetti & Hamilton, 2019). The recovery was characterized by faster growth of asset values than either employment rates or real wages, putting labor earners at a disadvantage relative to wealthier individuals deriving income from capital (Kolesnikova & Liu, 2011). Even asset values did not recover uniformly; financial assets grew more than other asset classes (Kuhn et al., 2020). Financial assets are more concentrated than even the characteristically skewed distribution of total wealth (Piketty, 2014). Following the recession, households with financial assets were best-positioned to recover their losses and even realize net improvements in wealth. This article considers distributional outcomes as the shocks of recession and recovery interacted with processes of life-course stratification.

Cohort Histories

Depression/war babies

Though born amidst upheaval, this cohort came of age in a growing postwar economy. Termed by Carlson (2008) the “The Lucky Few,” this smaller group faced less within-cohort competition. Depression and War Babies (or simply “War Babies”) transitioned into adulthood and spent their working years in a growing, high-employment economy. In retirement, they are living longer than their predecessors. As Keister et al. (2019) have shown, they were more likely than members of younger cohorts to receive inheritances and trusts.

Baby boomers

The stable postwar economic environment created the early context for Baby Boomers to gain economic, political, even cultural dominance. Although members of this large cohort faced more intracohort competition (Easterlin, 1961, 1969), they entered adulthood in an era of growth and expanding opportunity, with affordable higher education and rising returns to this education. Many were well situated to realize secure employment and accumulate assets.

As Boomers moved into midlife and retirement, the economic context of the 1980s and later grew increasingly unequal (Crystal et al., 2017). This period coincided with the cohort’s mid and later working years. Boomers thus encountered the recession and recovery from a relatively advantaged occupational and asset position.

Generation-X

This relatively smaller cohort faced less favorable initial conditions than Boomers. They purchased their homes and other assets in higher-priced markets of the 1990s, resulting in more highly leveraged mortgages and vulnerability to becoming underwater in a decline (McCarthy & Peach, 2004). The early aughts saw the initial stages of the accelerating increase in educational debt, which hindered asset accumulation for Gen-X.

Millennials

The Millennial cohort, earliest in the life course, had less-established occupational trajectories going into the recession of 2008. This later cohort faces the full effect of declines in “good jobs” with livable wages, advancement opportunities, job security, and nonwage compensation; many suffer long-term earnings degradation instigated by poor employment prospects at critical early career stages (Jones & Schmitt, 2016; Oreopoulos et al., 2012; Schmitt & Jones, 2012). Much of their career trajectory is yet to be observed, and the recession and recovery took place during a life stage of widespread household formation, making it difficult to distinguish the effect of the recession on the cohort’s longer-term well-being and income distribution.

Millennials were not well positioned to benefit from an asset-centric recovery, having little opportunity to cumulate assets at this early life stage. They realized smaller returns to educational investment than their predecessors. The income returns have remained rather stagnant since 2001 (Naidu & Sojourner, 2020), even while tuition rose nearly 50% (National Center for Education Statistics, 2019). Relatedly, aggregate student debt tripled between 2005 and 2015, with default rates projected to reach 40% by 2023, in a precoronavirus estimate (Scott-Clayton, 2018).

The remainder of the article applies the principles of an integrative CAD model to the economic shock of the 2008 recession and the period of recovery by examining its impact on economic well-being and its distributive effects across differently positioned cohorts.

Design and Methods

Data

This study uses data pooled from 2007 through 2016 waves of the triennial Survey of Consumer Finances (SCF). The SCF, managed by the Federal Reserve Board, serves as an important source for research on inequality, with significant efforts in survey design to capture income and assets in the higher part of the distribution that ultimately reveals more inequality than other survey data (Pfeffer et al., 2016).

The survey captures the upper tail of income and wealth by employing two separate sampling frames. One is an area-probability frame, designed for geographic and demographic representativeness and comprising approximately two thirds of the sample. The other targets wealthy households by analyzing tax records (Kennickell, 2008). The dual frame, a key advantage over other surveys, addresses potential biases due to greater nonresponse of high-wealth households by enabling survey administrators to oversample the hard-to-reach subpopulation. Overall, the SCF reports a response rate of 69% (area-probability sample) and 35% (high-wealth sample; Kennickell, 2017). Missing survey items are imputed prior to public release using a complex set of exogenous benchmarks (Kennickell, 1991).

Because of the complexity of wealth instruments, a summary version of the data is released, which we primarily analyze in this study. The summary file contains microlevel data on U.S. households with higher-order variables, including total household wealth and income, constructed and verified by the Federal Reserve. We merge data on ages and familial relationships of individual household members from the full to the summary data set and retain the survey’s household-level weighting scheme, which adjusts for sample structure.

Each of our four survey waves contains between 22,000 and 33,000 household observations, for a total of 115,810 households representing 306,845 individuals. We exclude individuals born prior to 1933 and those born after 1989, on the basis of our cohort definitions, producing a final sample representing 215,094 individuals.

Measures

We examine life-course trajectories within generational cohorts, namely War Babies (born 1933–1945), Baby Boomers (1946–1964), Generation-X (1965–1979), and Millennials (1980–1989). We terminate the Millennials cohort at 1989 to maintain consistency across waves, as these individuals would be at least 18 years in the first survey wave and thus included in the sampling frame throughout the study period.

Our primary measure of household resources is wealth-augmented household income. Following prior literature on wealth inequality across cohorts, we use this measure to better capture access to consumption over the course of recession, as assets can cushion an economic shock (Crystal et al., 2017; Crystal & Shea, 1990b). Moreover, both wealth and income concentration may reflect and shape inequality of opportunity. Integrating the contribution of income and wealth into a single measure provides a useful summary measure that contributes to understanding trends (Bricker et al., 2016; Wolff et al., 2005).

We build the measure by adding an annuitized portion of household net worth to household income. First, we deduct any income derived from assets, including from trusts or stock dividends. Because our measure of economic resources will include an estimate of income potentially realizable from such assets, this deduction is necessary to avoid double-counting the contribution of assets to household economic resources (Crystal et al., 2017; Wolff et al., 2005). Next, we use annuity tables provided by the Internal Revenue Service (IRS) to determine the portion of net worth that could be sustainably realized as income in a given year (Crystal et al., 2017). We construct the asset component as positive values of net worth (the maximum between net worth and 0) divided by the annuity factor associated with the age of the household head (IRS, 2009; Table S). Finally, wealth-augmented income is constructed as the sum of the annuitized portion of net worth and nonasset income.

In addition to overall wealth-augmented income, we examine components of households’ income and wealth over the study period, including total (nonaugmented) household income and net worth (assets less liabilities). We further divide net worth into broad classes of assets and debts, evaluating liquid financial assets, business-related investments (including stocks), and housing, following broad classes defined in prior literature (Kuhn et al., 2020).

We present each economic outcome as a multiple of the federal poverty level (FPL) in order to scale for household size, which varies systematically across the life course. The poverty level was approximately $12,000 for a single adult ($11,880) and $400 more per additional family member in 2016 such that FPL for a family of four is $24,300 (Assistant Secretary for Planning and Evaluation, 2016). All economic variables are inflation-adjusted to 2016 USD prior to public release by the Federal Reserve.

Outcomes

For each cohort in each wave, we evaluate Gini coefficients of income, a standard measure of inequality ranging from 0 (all income shared equally) to 1 (all income received by one person). As a complementary measure of inequality (Blotevogel et al., 2020), we assess the share of total cohort income realized at quantiles of the distribution. For each year, we find the total (survey-weighted) income realized by the cohort and report the share of aggregate earnings going to the highest-earning 10th and those below the median.

We begin by examining the economic context, with plots of average wealth-augmented income and Gini inequality, followed by trends in the components of income and wealth below the median and for the top tenth over the recession and recovery period. Then, we examine changes to within-cohort inequality over this time, considering both Gini indices of inequality in wealth-augmented income and shares of total income going to the top and to the bottom of each cohort. Last, we take a further look at Millennials, comparing inequality with and without those who live with their parents or other older primary householder to assess the position of this cohort in relation to CAD processes while accounting for household formation

Results

Figures 1 and 2 illustrate the macroeconomic context of the study period, in the years immediately preceding and following the 2008 recession. In Figure 1, we observe a rebounding of average household resources alongside a resurgence and further rise of inequality over the course of the recession and recovery. Mean wealth-augmented income is plotted by the solid orange line and measured along the axis on the left. Mean augmented income dips between 2007 and 2010 (from 4.67 to 4.36 times FPL) but then realizes a net increase from its initial baseline, reaching 4.99 times the FPL by 2016 (Figure 1). For scale, this represents an increase from approximately $106,000 to $121,500 between 2010 and 2016 for a family of four, on average.

Figure 1.

Figure 1.

Open in a new tab

Population trends of mean wealth-augmented income and inequality. Notes: Data from triennial waves of the Survey of Consumer Finances 2007–2016. Gini income (solid line) is presented as a factor of the federal poverty level, which adjusts for household size, and is measured on the left y-axis. Income includes an annuitized portion of household wealth to better represent household resources. Mean inequality (dashed line) is measured on the right y-axis using the same income definition.

Figure 2.

Figure 2.

Open in a new tab

Trajectories of income, net worth, and portfolio composition among below-median and top tenth of wealthiest US Households 2007–2016. Notes: Data are from the 2007–2016 Survey of Consumer Finances. Outcomes are scaled by the federal poverty level for household size. In Panel B, the y-axis reports multiples of the poverty level. In Panel A, 2007 is indexed to one. Wealth-augmented income is constructed as the sum of nonasset income and an annuitized share of household wealth, adjusted for household size and inflation.

Plotted by the solid black and measured on the right-hand y-axis is Gini inequality in augmented income. Inequality falls slightly along with mean income between 2007 and 2010 but by less than 0.01. Gini inequality rose over 5% from 0.53 in 2007 to 0.56 in 2016, representing dramatically high levels of inequality both historically and cross-nationally (Crystal et al., 2017).

Figure 2 plots trajectories in the components of household resources below the median and at the top 10% of the net-worth distribution. Panel A plots average wealth-augmented income (in solid black), and its two components: household income (in the dashed line) and household net worth (in the gray line). The year 2007 is standardized to one, with the following years representing relative changes.

As shown in Panel A, below-median households’ average resources never exceeded their prerecession position as of 2016 (Figure 2). For below-median households, the two income measures are not clearly distinguishable from one another, as wealth augmentation will not meaningfully affect low-wealth households. Both income measures trend together, falling immediately after the recession from 2007 to 2010 and continuing to fall through 2013 before returning to approximately the prerecession baseline by 2016. The average net worth of below-median households falls dramatically over this period, reaching as low as one third of its 2007 level and returning only to approximately half of the baseline by 2016.

By contrast, for the wealthiest 10th of households, wealth-augmented income markedly exceeds income alone in all years, reflecting the contribution of assets to the resources of the top tenth. Household resources fall initially after the recession, begin to return by 2013, and exceed 1.15 times baseline by 2016 (Supplementary Appendix Table 1). Moreover, wealth-augmented income falls by less than income alone in the years following the recession, reflecting the economic protection that wealth offers. Finally, while net worth falls through 2013, as with income, it exceeds its prerecession level by 2016 solely for households in the top tenth.

Panel B breaks down household net worth into its constituent components, further illustrating the persistent postrecession fall in wealth below the median. Panel B plots asset values held in liquid financial assets (black dots), investments (black solid), and housing and nonfinancial assets in black checker. Debts are plotted as negative values in grey: dots for non-housing debt and checker for housing debt. Each component has been scaled for household size and is represented on the y-axis in multiples of the FPL.

Panel B shows that for below-median households, the apparent restoration of net worth between 2013 and 2016 shown in Panel A reflects a combination of consistently falling asset values partially offset by diminishing debts in the postrecession recovery. Nearly all assets held below the median were held in the housing. While home equity continued to fall, so too did outstanding mortgage debts. Overall, debts were nearly as high as assets for households below the median.

Households in the top tenth held only slightly more debt in absolute value, but far more in assets, than those below the median (Figure 2, Panel B). These households held as much or more in financial as in nonfinancial wealth. While home equity never returned to its 2007 baseline, even for the top tenth, financial assets exceeded baseline by 2016, while debts fell slightly, generating the resurgence in net worth observed for the top 10% by 2016.

The recession, driven by falling home values, disproportionately lowered the asset position of below-median households, in large part because they had few other assets and were unable to benefit from resurging financial-asset markets.

Cohort Results

Within this evolving environment, we find that, in line with prior evidence, inequality was generally higher in late life than for individuals at earlier stages of the life course (Crystal et al., 2017; Crystal & Shea, 1990a). Gini inequality in wealth-augmented income was consistently higher for the War Baby and Baby Boomer cohorts than Generation-X and Millennials (Table 1). However, during this period, Baby Boomers surpassed the War Baby generation in Gini inequality. While they began the period at similar levels, War Babies’ inequality fell after the recession and did not regain its initial position by 2016. This cohort was at or very near retirement age by the start of the recession and reached their 70s and 80s by 2016.

Table 1.

Age and Gini Coefficients by Cohort

2007 2010 2013 2016
War Babies Ages 62–74 65–77 68–80 71–83
Gini 0.56 0.50 0.51 0.50
Boomers Ages 43–61 46–64 49–67 52–70
Gini 0.54 0.54 0.55 0.60
Gen-X Ages 28–42 31–45 34–48 37–51
Gini 0.46 0.48 0.51 0.54
Millennials Ages 18–27 21–30 24–33 27–36
Gini 0.50 0.50 0.48 0.47
Open in a new tab

Notes: Data from triennial waves of the Survey of Consumer Finances. Cohorts are defined by corresponding birth years. Gini inequality measured on survey-weighted household income augmented to include annuitized household net worth and adjusted for household size.

Notably, and in line with theories of CAD, the two cohorts in their prime working years experienced consistent, sizable increases in inequality. Boomers were in their 40s and 50s in 2007, on the eve of the recession, with a Gini inequality of 0.54. Their inequality remained steady in the year after the recession and then grew 6 points within the following 6 years as the cohort approached and entered retirement age. Generation-X was in early working years on the eve of the recession, primarily in their 30s. This cohort’s inequality increased 2 points by 2010 and then surged six more points by 2016, for a total increase of 8 points.

In contrast, Millennials’ inequality appears to have decreased over the period, though this may largely reflect shifts in household composition. The cohort’s young age and early career stage suggest that much of the cumulation of financial advantage and disadvantage was yet to transpire. The Gini coefficient declined from 0.50 in 2007 to 0.47 in 2016 (Table 1). Millennials, in their late 20s and early to mid-30s by 2016, ended the period with similar levels of inequality as Gen-X in 2007, a time when Gen-X was at a similar age as today’s Millennials.

Figure 3 further illustrates the notable and consistent increase in inequality realized by the two working-age cohorts. Figure 3 plots the share of wealth-augmented income realized by the highest-earning 10% and the lowest-income bottom half of each cohort between 2007 and 2016. The top left plot shows that in 2007, the top tenth of the War Baby generation realized nearly half (46%) of all cohort resources, while the bottom half realized only 16%, far less than their proportion of the cohort population. The top tenth’s share of resources fluctuated over the following years, ending at 41% in 2016, with the bottom half realizing 18%.

Figure 3.

Figure 3.

Open in a new tab

Cohort-specific income shares. Notes: Data from triennial waves of the Survey of Consumer Finances. Cohorts are defined by birth years (Table 1). Bars represent the share of household income realized by the top 10% of cohort members (top bars) and realized by the bottom half of cohort members (bottom bars) in each panel. Wealth-augmented income is constructed as the sum of nonasset income and an annuitized share of household wealth, adjusted for household size and inflation.

The top right plot of Figure 3 shows this evolution for the Baby Boomer generation. Boomers and War Babies began the period at similar levels of inequality. In 2007, 45% of all income realized by Baby Boomers went to the top 10% of earners, while only 16% went to the bottom half. By 2016, however, inequality surged among Baby Boomers, with approximately half of all income going to the top tenth and only 13% to the bottom half.

Generation-X underwent a trajectory of increasing inequality, similar to that of Baby Boomers but more consistently over the period, with a steadily growing share of all income going to the top 10%. In 2007, the top tenth of Gen-X realized 36% of total income, which grew 1 point in the recession aftermath of 2010, another 4 points by 2013, and three additional points in the next 3 years. By 2016, the top tenth of Generation-X realized 44% of all cohort income (Figure 3).

Again, Millennials appeared to have a contrary trend, with a decreasing share of cohort income realized by the top tenth over time. Figure 3 shows that the top tenth of Millennials realized 38% of income in 2007, but only 36% in 2016. Unlike the other cohorts, the bottom half of Millennials realized an increasing share of income from 18% to 20%.

Millennials: Disparities and Household Formation

The Millennials’ ostensibly egalitarian trend might actually reflect a temporary shift in household composition, as our study coincides with a major period of household formation for the cohort. Historically, 25%–30% of young adults form newly separate households in their 20s and 30s (Furlong, 2016). If the highest-income Millennial individuals were formerly members of their parents’ households, the apparent decline in inequality might represent the immediate, but temporary, effects of household formation rather than the emergence of truly equitable life chances.

Table 2 investigates Millennials’ inequality and household formation more closely, first listing average age for reference, then the proportion of the population in independent households over time in Panel A. More than one third of Millennials formed newly independent households during the study period, as they moved from their mid-20s into the early 30s. By 2016, three in four Millennials headed independent households. Panel B lists the same resource-share trends observed in Figure 3, showing a slight decrease in the share of income realized by the top tenth and a slight increase below the median, between 2007 and 2016.

Table 2.

Income Shares of Millennial Households and Individuals, 2007–2016

2007 2010 2013 2016
Panel A: Millennial characteristics
Average age 24.5 25.9 27.4 31.7
Proportion head or spouse (%) 39.39 40.82 45.76 75.72
Panel B: All Millennial individuals (income shares, %)
Bottom half 18.15 17.72 18.08 19.61
Top tenth 38.20 38.50 36.40 35.60
Panel C: Millennial household heads (income shares, %)
Bottom half 21.29 21.64 21.94 19.51
Top tenth 29.48 28.09 29.89 36.42
Open in a new tab

Notes: Data from triennial waves of the Survey of Consumer Finances. The table lists the share of cohort income realized by the bottom half and the top 10% of income earners within the cohort. Panel B represents household incomes for all Millennial cohort members. Panel C represents the household income of Millennial heads-of-households only. Income reflects total household income augmented to include an annuitized portion of household net worth, adjusted for household size and inflation.

Panel C shows the trend in resource inequality observed among Millennial-headed households only, removing cohort members living with their parents or other older primary householder. Here, inequality increases among Millennial-headed households, contrary to what was observed when including Millennials in shared households. This result indicates the importance of household formation for Millennials during the time surrounding the 2008 recession. By 2016, the income shares observed for Millennial-headed households converged with those observed when including Millennials in shared households as the process of household formation approached completion.

Discussion and Conclusion

Although the immediate effect of the 2008 recession was a temporary decline in inequality, inequality returned full force during the recovery, and by 2016, income and wealth were more unequally distributed than before. First, alongside rebounding per-capita incomes, we observed inequality surging to new heights. Eight years after the housing market crashed, half of the population below the median had just regained their initial income position but never regained the fall in home equity. With almost all their assets held in home equity, the group realized persistent reductions in net worth. The top tenth not only regained but surpassed their prerecession levels of income and net worth, the latter driven by growing values of financial investments that compensated any sustained consequences of the housing market crash.

Next, we observed that the amplified stratification at the population level was most evident among cohorts in their prime working years. While War Babies and Millennials fell or gained only slightly in inequality, Baby Boomer and Generation-X cohorts each experienced a considerable increase in inequality. Millennials were at a young age and early career stage during the decade observed, but underwent much of the process of household formation, and accompanying labor-market entry by the end of the study period. This cohort experienced in 2020 a second economic shock at critical stages of career formation from a new, coronavirus disease 2019 (COVID-19)-induced recession. This COVID-19 recession itself occurs in the midst of an unequal recovery from the prior recession; both factors have the potential to compound the cumulative stratification processes of aging.

Public policy can mitigate such outcomes. For example, given the importance of homeownership to the net worth of below-median households, relief funds following the 2008 recession could have been better targeted at homeowners. Under the Troubled Asset Relief Program (TARP; the official title of the 2008 bank bailouts), the federal treasury spent hundreds of billions of dollars within 1 year, swiftly purchasing risky assets from troubled banks and thereby stabilizing financial markets. By contrast, the TARP funds aimed at homeowners, which had the goal of re-financing 4 million American home mortgages, sat 80% undispersed even several years into the recovery, and reached only 1.5 million households by 2012 (Barofsky, 2013; Department of the Treasury, 2010).

Moreover, given the importance of labor as opposed to asset income for below-median households, recovery policies aimed at wages and bargaining power, such as those implemented during the 1930s in response to the Great Depression, may better target relief to below-median households (Dupor, 2017). As noted by the Federal Reserve Chair Janet Yellen (2016), whereas economic growth following the Great Depression was accompanied by narrowing inequality, the post-2008 recovery presaged levels of inequality “near their highest levels in the past hundred years … probably higher than for much of American history before then (Yellen, 2016).” Yellen highlighted the role of disparate access to economic opportunity in driving this inequality, including child poverty, affordable higher education, and the concentration of business ownership and inheritances at the top of the resource distribution.

The contrast between the distributional impact of economic growth in the recovery following the Great Depression and the recovery following the 2008 recession highlights the observation that recovery does not necessarily increase inequality inexorably. Rather, the effects of recovery on inequality reflect structural factors and political choices in economic stimulus programs aimed at recovery. The programs enacted following the 2008 recession do not appear to have benefited those lower in the income distribution to the same extent as those higher up. In contrast, recovery strategies aimed at lower-income individuals, such as expanded, refundable family tax credits and direct financial assistance, can produce more equitable patterns of recovery. In this regard, it will be of interest to track the distributional impact of the American Rescue Plan enacted in 2021, which includes significant components of direct income support and expanded family tax credits to below-median households.

Limitations

Our study has limitations. First, our study is observational, illustrating the mechanisms for recession-specific CAD processes. We do not attempt, for the most part, to empirically isolate the effects of recession from those of cohort life stage and thus cannot be certain about the causal implications of each. While outside the scope of this study, future research could apply decomposition techniques to estimate the relative importance of life stage and period or examine a natural experiment in which otherwise similar groups face differing economic environments, such as an industry-specific crash, if such an opportunity arises.

Second, we lack longitudinal data on individuals’ experiences. Instead, we examine stratification at repeated points in time, assessing differences between the bottom half and their better-resourced counterparts. Future research could trace individual experiences to explore the mechanisms whereby CAD processes interacted with the financial recession of 2008 and potentially the 2020 recession instigated by the coronavirus. Furthermore, we rely on federal survey data to analyze wealth. While the SCF is considered the best nationally representative data source on U.S. household wealth, with a sample frame carefully designed to minimize bias, still the sensitive nature of wealth data yields higher nonresponse rates than other survey types, particularly among households with substantial assets in various forms and perhaps in various countries (Kennickell, 2008). Any remaining bias caused by nonresponse among high-wealth households would suggest that inequality is greater than what is documented by these data.

Lastly, in our study of inequality trajectories, we have not evaluated variations in the phenomena by race or ethnicity. As subprime mortgages were more likely to be offered to and entered by African American households, and unemployment effects of the recession peaked higher for non-White workers, the resulting stratification may have been heightened across racial and ethnic lines (Calem et al., 2004; Gillespie, 2015). This is an important area for future research.

Conclusion

In this analysis, we consider the impact of the 2008 recession and subsequent recovery from an integrative perspective, considering the interaction of CAD, the persistent phenomenon of stratification over the life course, and historically contingent factors, such as the design of financial recovery policies. During the recovery, population-level increases in economic resources were not reflected among households below the median. Although recovery policies led to rapid recovery of the value of financial assets, they did not proportionately reach those below the median, whose economic resources were more heavily driven by employment and home equity. Assessing the effects of economic shocks, such as recessions, wars, and pandemics, and recovery strategies in their aftermath illustrates how trajectories of inequality unfold within cohorts. Recovery policies often involve critical policy choices in the form of large-scale governmental financial investment that differentially affects those at different places in the income distribution and life stages, making these particularly salient events for studying how such choices affect distributional outcomes and trajectories.

Supplementary Material

gbab141_suppl_Supplementary_Appendix
Click here for additional data file. (232.1KB, pdf)

Funding

S. Crystal is supported in part by the National Institutes of Health, National Center for Clinical and Translational Science, under Clinical and Translational Sciences Center award UL1TR003017.

Conflict of Interest

None declared.

Author Contributions

N. Zewde performed statistical analyses and drafted the manuscript. S. Crystal supervised and assisted in planning data analyses, revised and edited the manuscript.

References

  1. Assistant Secretary for Planning and Evaluation . (2016). Computations for the 2016 poverty guidelines.Office of the Assistant Secretary for Planning and Evaluation. https://aspe.hhs.gov/computations-2016-poverty-guidelines [Google Scholar]
  2. Barofsky, N. (2013). Bailout: How Washington abandoned main street while rescuing wall street. Simon and Schuster. [Google Scholar]
  3. Blotevogel, R., Imamoglu, E., Moriyama, K., & Sarr, B. (2020). Measuring income inequality and implications for economic transmission channels (No. 2020/164; IMF Working Papers). International Monetary Fund. [Google Scholar]
  4. Bricker, J., Henriques, A., Krimmel, J., & Sabelhaus, J. (2016). Estimating top income and wealth shares: Sensitivity to data and methods. American Economic Review, 106(5), 641–645. doi: 10.1257/aer.p20161020 [DOI] [Google Scholar]
  5. Calem, P. S., Gillen, K., & Wachter, S. (2004). The neighborhood distribution of subprime mortgage lending. The Journal of Real Estate Finance and Economics, 29(4), 393–410. doi: 10.1023/B:REAL.0000044020.67401.51 [DOI] [Google Scholar]
  6. Carlson, E. D. (2008). The lucky few: Between the greatest generation and the baby boom. Springer Netherlands. doi: 10.1007/978-1-4020-8541-3 [DOI] [Google Scholar]
  7. Crystal, S. (1982). America’s old age crisis: Public policy and the two worlds of aging. Basic Books. [Google Scholar]
  8. Crystal, S. (1986). Measuring income and inequality among the elderly. The Gerontologist, 26(1), 56–59. doi: 10.1093/geront/26.1.56 [DOI] [PubMed] [Google Scholar]
  9. Crystal, S. (2016). Late-life inequality in the second gilded age: policy choices in a new context. Public Policy & Aging Report, 26(2), 42–47. doi: 10.1093/ppar/prw005 [DOI] [Google Scholar]
  10. Crystal, S. (2020). Linking the levels: Integrating individual trajectories, historical contingency, and social policy choices in cumulative advantage and disadvantage research. The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences, 75(6), 1245–1248. doi: 10.1093/geronb/gbaa059 [DOI] [PubMed] [Google Scholar]
  11. Crystal, S., & Shea, D. (1990a). The economic well-being of the elderly. Review of Income and Wealth, 36(3), 227–247. doi: 10.1111/j.1475-4991.1990.tb00302.x [DOI] [Google Scholar]
  12. Crystal, S., & Shea, D. (1990b). Cumulative advantage, cumulative disadvantage, and inequality among elderly people. The Gerontologist, 30(4), 437–443. doi: 10.1093/geront/30.4.437 [DOI] [PubMed] [Google Scholar]
  13. Crystal, S., Shea, D., & Krishnaswami, S. (1992). Educational attainment, occupational history, and stratification: Determinants of later-life economic outcomes. Journal of Gerontology, 47(5), S213–S221. doi: 10.1093/geronj/47.5.s213 [DOI] [PubMed] [Google Scholar]
  14. Crystal, S., Shea, D. G., & Reyes, A. M. (2017). Cumulative advantage, cumulative disadvantage, and evolving patterns of late-life inequality. The Gerontologist, 57(5), 910–920. doi: 10.1093/geront/gnw056 [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Crystal, S., & Waehrer, K. (1996). Later-life economic inequality in longitudinal perspective. The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences, 51(6), 307–318. doi: 10.1093/geronb/51B.6.S307 [DOI] [PubMed] [Google Scholar]
  16. Dannefer, D. (1987). Aging as intracohort differentiation: Accentuation, the Matthew effect, and the life course. Sociological Forum, 2(2), 211–236. doi: 10.1007/BF01124164 [DOI] [Google Scholar]
  17. Dannefer, D. (1988). Differential gerontology and the stratified life course: Conceptual and methodological issues. In Maddox G. L. & Lawton M. P. (Eds.), Varieties of aging (pp. 3–36). Springer. doi: 10.1007/978-3-662-40050-0_1 [DOI] [PubMed] [Google Scholar]
  18. Dannefer, D. (2020). Systemic and reflexive: Foundations of cumulative dis/advantage and life-course processes. The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences, 75(6), 1249–1263. doi: 10.1093/geronb/gby118 [DOI] [PubMed] [Google Scholar]
  19. Department of the Treasury . (2010). Troubled Asset Relief Program: Two year retrospective. Office of Financial Stability. https://home.treasury.gov/data/troubled-assets-relief-program/reports/tarp-annual-retrospectives [Google Scholar]
  20. Dupor, B. (2017). The Recovery Act of 2009 vs. FDR’s New Deal: Which was bigger? The Regional Economist. The Federal Reserve Bank of St Louis, 6–7. [Google Scholar]
  21. Easterlin, R. A. (1961). The American baby boom in historical perspective. The American Economic Review, 51(5), 869–911. [Google Scholar]
  22. Easterlin, R. A. (1969). Towards a socioeconomic theory of fertility: A survey of recent research on economic factors in American fertility. In S. J. Behrman, Leslie Corsa, Jr., and Ronald Freedman, editors. Fertility and family planning: A world view (pp. 127–156). University of Michigan Press. [Google Scholar]
  23. Edwards, K. A., & Hertel-Fernandez, A. (2010). The kids aren’t alright—A labor market analysis of young workers (No. 258; Briefing Paper).Economic Policy Institute. https://www.epi.org/publication/bp258/ [Google Scholar]
  24. Erickson, C. L., & Mitchell, D. J. B. (2007). Monopsony as a metaphor for the emerging post-union labour market. International Labour Review, 146(3–4), 163–187. doi: 10.1111/j.1564-913X.2007.00012.x [DOI] [Google Scholar]
  25. Famighetti, C., & Hamilton, D. (2019). The great recession, education and home ownership. Kirwan Institute. http://kirwaninstitute.osu.edu/the-great-recession-education-and-home-ownership/ [Google Scholar]
  26. Ferraro, K. F., & Kelley-Moore, J. A. (2003). Cumulative disadvantage and health: Long-term consequences of obesity? American Sociological Review, 68(5), 707–729. [PMC free article] [PubMed] [Google Scholar]
  27. Furlong, F. (2016). Household formation among young adults [FRBSF Economic Letter]. Federal Reserve Bank of San Francisco. https://www.frbsf.org/economic-research/publications/economic-letter/2016/may/household-formation-among-young-adults/ [Google Scholar]
  28. Gillespie, P. (2015). Black unemployment at its lowest since 2008.CNNMoney. https://money.cnn.com/2015/05/08/news/economy/black-unemployment-below-10-percent/index.html [Google Scholar]
  29. Hales, C. N., & Barker, D. J. (2001). The thrifty phenotype hypothesis. British Medical Bulletin, 60, 5–20. doi: 10.1093/bmb/60.1.5 [DOI] [PubMed] [Google Scholar]
  30. Internal Revenue Service . (2009). Actuarial tables. IRS Published guidance. https://www.irs.gov/retirement-plans/actuarial-tables [Google Scholar]
  31. Jones, J., & Schmitt, J. (2016). Trends in job quality for African-American workers, 1979–2011. The Review of Black Political Economy, 43(1), 1–19. doi: 10.1007/s12114-015-9216-3 [DOI] [Google Scholar]
  32. Keister, L. A., Benton, R. A., & Moody, J. W. (2019). Cohorts and wealth transfers: Generational changes in the receipt of inheritances, trusts, and inter vivos gifts in the United States. Research in Social Stratification and Mobility, 59, 1–13. doi: 10.1016/j.rssm.2019.01.002 [DOI] [Google Scholar]
  33. Kennickell, A. B. (1991). Imputation of the 1989 Survey of Consumer Finances: Stochastic relaxation and multiple imputation. Proceedings of the Survey Research Methods Section of the American Statistical Association, 1(10), 41. [Google Scholar]
  34. Kennickell, A. B. (2008). The role of over-sampling of the wealthy in the Survey of Consumer Finances. Irving Fisher Committee Bulletin, 28, 403–408. [Google Scholar]
  35. Kennickell, A. B. (2017). Darkness made visible: Field management and nonresponse in the 2004 SCF. Statistical Journal of the IAOS, 33(1), 101–111. [Google Scholar]
  36. Kolesnikova, N., & Liu, Y. (2011). Jobless recoveries: Causes and consequences. The Regional Economist, 18–19. [Google Scholar]
  37. Kuhn, M., Schularick, M., & Steins, U. I. (2020). Income and wealth inequality in America, 1949–2016. Journal of Political Economy, 128(9), 3469–3519. doi: 10.1086/708815 [DOI] [Google Scholar]
  38. Lee, C., & Park, S. (2020). Examining cumulative inequality in the association between childhood socioeconomic status and body mass index from midlife to old age. The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences, 75(6), 1264–1274. doi: 10.1093/geronb/gbz081 [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. McCarthy, J., & Peach, R. W. (2004). Are home prices the next bubble? Economic and Policy Review, 10(3). [Google Scholar]
  40. Merton, R. K. (1968). The Matthew effect in science: The reward and communication systems of science are considered. Science, 159(3810), 56–63. doi: 10.1126/science.159.3810.56 [DOI] [PubMed] [Google Scholar]
  41. Naidu, S., & Sojourner, A. (2020). Employer power and employee skills: Understanding workforce training programs in the context of labor market power. Roosevelt Institute. https://rooseveltinstitute.org/publications/employer-power-employee-skills-workforce-training-programs-labor-market-power/ [Google Scholar]
  42. National Center for Education Statistics . (2019). Tuition costs of colleges and universities.National Center for Education Statistics. https://nces.ed.gov/fastfacts/display.asp?id=76 [Google Scholar]
  43. O’Rand, A. M. (1996). The precious and the precocious: Understanding cumulative disadvantage and cumulative advantage over the life course. The Gerontologist, 36(2), 230–238. doi: 10.1093/geront/36.2.230 [DOI] [PubMed] [Google Scholar]
  44. Oreopoulos, P., von Wachter, T., & Heisz, A. (2012). The short- and long-term career effects of graduating in a recession. American Economic Journal: Applied Economics, 4(1), 1–29. doi: 10.1257/app.4.1.1 [DOI] [Google Scholar]
  45. Pfeffer, F. T., & Hertel, F. R. (2015). How has educational expansion shaped social mobility trends in the United States? Social Forces, 94(1), 143–180. doi: 10.1093/sf/sov045 [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Pfeffer, F. T., Schoeni, R. F., Kennickell, A., & Andreski, P. (2016). Measuring wealth and wealth inequality: Comparing two U.S. surveys. Journal of Economic and Social Measurement, 41(2), 103–120. doi: 10.3233/JEM-160421 [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Piketty, T. (2014). Capital in the twenty-first century (Goldhammer A., Trans.). Harvard University Press; JSTOR. https://www.jstor.org/stable/j.ctt6wpqbc [Google Scholar]
  48. Saez, E., & Zucman, G. (2016). Wealth inequality in the United States since 1913: Evidence from capitalized income tax data. The Quarterly Journal of Economics, 131(2), 519–578. [Google Scholar]
  49. Schmitt, J., & Jones, J. (2012). Where have all the good jobs gone? CEPR Center for Economic and Policy Research. https://www.cepr.net/documents/publications/good-jobs-2012-07.pdf [Google Scholar]
  50. Scott-Clayton, J. (2018). The looming student loan default crisis is worse than we thought (2:34; Evidence Speaks Reports).https://www.brookings.edu/research/the-looming-student-loan-default-crisis-is-worse-than-we-thought/
  51. Wolff, E. N., Zacharias, A., & Caner, A. (2005). Household wealth, public consumption and economic well-being in the United States. Cambridge Journal of Economics, 29(6), 1073–1090. doi: 10.1093/cje/bei076 [DOI] [Google Scholar]
  52. Yellen, J. L. (2016). Perspectives on inequality and opportunity from the survey of consumer finances. The Russell Sage Foundation Journal of the Social Sciences, 2(2), 44–59. doi: 10.7758/RSF.2016.2.2.0230123834 [DOI] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

gbab141_suppl_Supplementary_Appendix
Click here for additional data file. (232.1KB, pdf)

Articles from The Journals of Gerontology Series B: Psychological Sciences and Social Sciences are provided here courtesy of Oxford University Press

RESOURCES