Balance Sheets of Early Boomers: Are They Different from Pre-boomers?
J Fam Econ Iss (2006) 27:542–561
DOI 10.1007/s10834-006-9026-7
O R I G I N A L P A P E R
Balance Sheets of Early Boomers: Are They Different from Pre-boomers?
Michael S. Finke Æ Sandra J. Huston Æ
Deanna L. Sharpe
Published online: 12 August 2006
Ó
Springer Science+Business Media, Inc. 2006
Abstract Nationally representative data are used to compare the magnitude, composition, and distribution of accumulated wealth of early boomers (born between 1946 and
1957) and pre-boomers (born between 1934 and 1945) at the same life cycle stage (ages
44–45, occurring in 1989 for pre-boomers and in 2001 for early boomers). Early boomers have accumulated greater mean wealth than pre-boomers at the same age, however median wealth did not change and net worth among lower-middle wealth groups declined. Multivariate analysis identified demographic change among those in the wealth distribution tails that indicates the increasing importance of education and race as predictors of wealth.
Keywords Cohort wealth Æ Baby boomers Æ Net worth
As baby boomers near retirement, interest in their financial behavior has increased. Some call boomers a generation that has ‘‘saved less’’ and ‘‘taken on more debt’’ as compared with previous generations, asserting that ‘‘striking differences’’ exist between the saving and spending patterns of boomers and previous generations (Francese,
). Anecdotal observation of and media attention on overt materialism among boomers suggests this generation is simply unwilling to moderate spending and save for future retirement. But other explanations of the financial situation of boomers may exist.
Boomers’ financial decisions have been made in substantially different financial markets than the ones their parents faced. Shifts away from defined benefit plans have made boomers relatively more responsible for their financial future. Growth of mutual fund options and tax shelters has widened choice of investment instruments and facilitated market investment.
M. S. Finke (
&
) Æ S. J. Huston Æ D. L. Sharpe
Department of Personal Financial Planning, University of Missouri-Columbia, 239 Stanley Hall,
Columbia, MO 65211, USA e-mail: FinkeM@missouri.edu
123
J Fam Econ Iss (2006) 27:542–561 543
Relative to other generations, have boomers accumulated a lower level of financial assets, or have they set aside more than might be readily apparent through such venues as defined contribution plans or increases in home equity? Did the large stock market gains of the 1990s accrue to a relative few equity holders while increased credit access and rising housing prices pushed other boomer families toward financial instability?
Comparing the magnitude, composition, and distribution of accumulated wealth of the baby boom generation with that of an older birth cohort that has experienced similar economic circumstances during the latter half of the 20th century can shed light on the financial resource allocations of boomers as a generation. To date, no such study has been done. To address this gap in the literature, this study uses nationally representative data to compare the characteristics and the asset and liability holdings of two generations: early boomers, defined in this study as those born between 1946 and 1957 and ‘‘pre-boomers’’
(Verma & Lichtenstein,
2003 ) defined in this study as those born between 1934 and 1945.
This research compares the demographic characteristics of the two generations, the inflation-adjusted asset and debt holdings of the two generations (over time for each generation and at a point in time between generations), and the inflation-adjusted equity value, capital gains, and debt service of the two generations (over time for each generation and at a point in time between generations). In addition, significant factors associated with being in the tails of the wealth distribution are identified for each generation.
Review of the Literature
Recent studies of boomer financial status have focused on three aspects of that status: retirement income adequacy, wealth or net worth, and debt levels.
Boomer Retirement Income Adequacy
As the baby boom generation has aged, both academic researchers and the media have questioned whether or not boomers have adequate retirement savings. Media headlines such as ‘‘More Workers Can’t Afford to Retire’’ (Dugas,
2003 ) or ‘‘Boomers are Good at
Saving, but Not Good Enough’’ (Svensson,
) paint a dismal picture of boomers who either must continue working past the conventional retirement age of 65 or drastically ratchet down consumption during retirement to have a chance to make ends meet. Conclusions about retirement income adequacy by the research community are somewhat mixed, however. Results depend on the comparison group selected, the measure of adequacy used, assumptions about real rates of return and life expectancy, and the data set utilized (Congressional Budget Office,
One line of research has found that, except for those with low levels of education or nonhomeowners, boomers have more financial resources than their parents, however the savings rate for both generations is about the same (Easterlin, Schaeffer, & Macunovich,
;
Sabelhaus & Manchester,
1993 ). Easterlin et al. ( 1993 ) cite lower fertility rates and wo-
men’s increased labor force participation as the main reasons for boomer financial gains.
Other studies have compared projected income for boomers at retirement age against the income and asset levels of current retirees (AARP,
1994 ; Kotlikoff & Auerbach, 1994
).
These studies present a relatively optimistic picture of boomers who, taken as a group, will generally be able to maintain or exceed their current consumption levels at retirement. For post retirement income, lower income persons will rely heavily on Social Security while higher income individuals will rely much more on pensions and income generating assets.
123
544 J Fam Econ Iss (2006) 27:542–561
Still other studies judge sufficiency by comparing projected boomer financial status at retirement against a calculated standard of need. The theoretical basis for this work is the permanent income hypothesis, which presumes that people act to maintain a certain level of consumption across time, saving excess in times of plenty and drawing down savings in times of lack. This set of studies has defined ‘‘adequate income’’ variously as being above the official poverty threshold (Toder et al.
), as replacing at least 80% of pre-retirement income (Gist, Wu, & Ford,
1999 ; Gustman & Steinmeier, 1998 ; Warshawsky &
Ameriks,
) or as maintaining pre-retirement consumption levels (Moore & Mitchell,
; Yuh, Montalto, & Hanna,
). The statistical models used in these studies are sensitive to assumptions about real rates of return and life expectancy.
Results of this line of research have been mixed. Toder et al. (
) projected that poverty rates among boomers will fall. Chances of maintaining pre-retirement levels of income and consumption are somewhat more pessimistic, however. In a series of studies completed between 1993 and 1997, Bernheim utilized data obtained from phone surveys of around 2,000 Boomer households to obtain measures of expected retirement income from
Social Security and pensions as well as from private savings (See Bernheim,
,
,
1997 ). The collected data were then put into a statistical model that projected future
levels of income and wealth. He decided that boomers were saving only about one-third of the amount needed to adequately fund their retirement. Bernheim’s work has since been
criticized for assuming low real rates of return and excluding home equity. Gale ( 1997 )
concluded that boomer prospects at retirement were more optimistic when home equity was included in boomer wealth. At least a third of boomers, according to Gale (
), were adequately prepared for retirement by any measure, a third were ready by some standards but not others, and the remainder were not adequately prepared.
Reaching a similar conclusion, Moore and Mitchell ( 1997 ) asserted that around 30–40%
of boomer households can preserve consumption levels (assuming retirement at age 62 or later). Yuh et al. (
), on the other hand, found that a little over half of households headed by someone of the early boomer age in 1995 were on track for retirement. Gist
et al. ( 1999 ) recommended that boomers who currently are not prepared for retirement
increase their savings from 3% to 29% of annual income, depending on presence or absence of a pension plan and on assumptions about longevity. In general, this line of research suggests that, while high-income boomers can likely retire at age 62 or later and maintain pre-retirement levels of income or consumption, other boomers are at risk of falling short of that goal. These boomers will have to either work longer or adjust postretirement consumption levels downward.
Boomer Wealth
Using Survey of Consumer Finances (SCF) data from 1989 to 2001, Gist and Wu ( 2004 )
documented rising wealth inequality among the boomer generation. Between 1989 and
2001, the three bottom wealth quintiles experienced minimal change in the proportion of wealth held. At the same time, the share of wealth held by the top 10% of the wealth distribution increased by almost seven percentage point and that of the top 5% by over 10 percentage points. At every measure between 1989 and 2001, the top 1% of the wealth distribution held more wealth than the bottom 80%. The top-heavy nature of boomer wealth inequality is echoed in studies of the broader population, indicative of the influence of the boomer cohort (Danzinger & Gottschalk,
; Karoly,
; Ryscavage,
;
Wolff,
).
123
J Fam Econ Iss (2006) 27:542–561 545
Rising value of previously purchased investment assets or acquisition of investment assets may increase wealth. The strong bull market of the 1990s likely contributed to both events for many investors. Yao, Hanna, and Lindamood (
) noted that between 1983 and 2001, financial risk tolerance appeared to increase in response to increases in stock
returns and decrease in the wake of declines in stock value. Finke and Huston ( 2003 ) also
note a positive relationship between risk tolerance and level of net worth and financial assets. If investors acted consistently with reported risk tolerance, increased stock purchase during the bull market would be expected.
Using Panel Study of Income Dynamics data for 1984, 1989, and 1994, Gittleman &
Wolff (
) concluded that whites devote a greater portion of their income to saving. But, racial differences in savings rate were not significant once income differences were controlled.
Gist, Wu, and Verma (
) used SCF data and various measures of wealth to explore the wealth distribution of boomers. They note a rise in wealth holdings over time for all wealth levels. Median net worth of boomers tripled between 1989 and 2001, rising from
$35,951 to $107,150 in 2001 dollars. Excluding housing equity from the net worth measure drops the dollar value of wealth, but it still triples from $13,410 to $50,700. They found higher levels of net worth for whites, men, married couples, older persons, families with children, homeowners, and defined benefit plan holders. Using multivariate analysis, they discovered that having a defined benefit plan resulted in lower savings rates, ceteris paribus , suggesting that defined benefit plans substituted for rather than augmented other
forms of retirement savings. Tracy, Schneider and Shan ( 1999 ) noted that while rapid
growth of the stock market during the 1990s fueled the perception that stocks dominated household portfolios, actually residential real estate continued to represent the lion’s share of most household portfolios. Baek and DeVaney (
2004 ) noted that nearly two-thirds of
baby boomers have met the recommended investment assets to net worth threshold and three of four boomers have a debt-to-assets ratio beneath the recommended level.
Retirement Debt Trends
Recently, concern has risen regarding debt levels of the nation’s seniors. The National
Association of Consumer Bankruptcy Attorneys (NACBA) noted that over a quarter million Americans aged 55 and older declared bankruptcy in 2001. Over the past 10 years, the number of Americans 65 and older filing for bankruptcy has tripled, making this the fastest growing age group in the bankruptcy courts. Indications are that the boomer generation is likely to follow this same trend (AccountingWeb,
). The main factors leading to bankruptcy are job loss (leading to income loss), medical costs, marital disruption, and cost of caring for family members (Warren,
) rather than excessive spending on credit cards for non-necessity items. Bankruptcy risk is especially high for boomers with high levels of mortgage and consumer debt.
Conceptual Framework
According to the lifecycle income hypothesis (Ando & Modigliani,
), during youth and young adulthood, dissavings occurs as education is completed and an independent household is established. During mid-life, peak earning years are reached, facilitating repayment of earlier debt and savings for future retirement. At retirement, dissavings
123
546 J Fam Econ Iss (2006) 27:542–561 occur again as labor income ceases and consumption is sustained by withdrawals from previous savings and investment earnings. Thus, income generates the means to save and invest.
This study focuses on the first cohort of the baby boom generation, born between 1946 until the peak year of the baby boom in 1957 (early boomers) and on the immediately prior generation born between 1934 and 1945 (pre-boomers). Using the lifecycle hypothesis as our conceptual framework, we compare the two generations at ages 44 to 55 to capture peak earning years and to permit comparison of various aspects of the inflation-adjusted financial status of two contiguous generations at the same lifecycle point. We expected to see similar patterns of asset accumulation for the two groups, given the similarity in age and, consequently, lifecycle stage. Recognizing the dynamic nature of financial resource allocation and accumulation over time, we also examined inflation-adjusted net worth components, equity value, capital gains and debt service for every three years between
1989 and 2001. Use of the same time period allowed us to control for variation in economic and market performance. Use of the 12-year span allowed us to observe progressive changes in wealth for each generation over time. Given the life cycle hypothesis, we expect assets of the older birth cohort might decline between midlife (44–55) to early retirement
(56–67), while the early boomer cohort might accumulate assets between early adulthood
(32–43) and midlife (44–55). Multivariate analysis was used to identify demographic differences between those in the highest and lowest wealth categories for each generation at midlife (age 44–55). Correlates in the model of wealth accumulation were chosen from prior literature including within cohort age, education, work status, presence of children in the home, homeowner and marital status, existence of a pension, race
1
, and log income
(Kennickell,
Description of Data
Data for this research are from the 1989, 1992, 1995, 1998, and 2001 Surveys of Consumer
Finances (SCF). Primarily designed to measure wealth of American households, the SCF also collects detailed financial and demographic characteristics of U.S. families.
The Board of Governors of the Federal Reserve System and the U.S. Department of the
Treasury have sponsored triennial data collection for the SCF since 1983. The dual frame design of the SCF combines a standard, multi-stage area-probability sample with a stratified list sample derived from income tax data that disproportionately includes wealthy families (excluding those Forbes identifies as the 400 wealthiest people in the U.S.). The complex sample design makes nonresponse an important issue in the SCF. Missing data are multiply imputed, resulting in five implicates for each year of data; weights are used to compensate for both sample complexities and nonresponse patterns. Although the SCF is a cross-sectional data set, substantial consistency of survey questions since 1989 makes the data highly comparable over time.
The number of families interviewed was 3,803 in 1989, 3,906 in 1992, 4,299 in 1995,
4,309 in 1998 and 4,449 in 2001. Out of this group, this study selects household heads born between 1934 and 1957 to compare the early part of the baby boom (head born between
1946 and 1957) with the prior generation (head born between 1934 and 1945).
1
Race proxies possible differences in risk tolerance, inherited wealth, and asset allocation. Race of respondent is assumed to be representative of the household.
123
J Fam Econ Iss (2006) 27:542–561 547
Variable Measurement
Balance Sheet Variables
Household balance sheet information is measured through several asset (financial and nonfinancial) and debt variables defined primarily by the SCF and outlined in Federal Reserve
Bulletins. Proportion of households with a particular asset is reported. Financial assets include: liquid assets (sum of transaction accounts including checking, savings, money market, and call accounts at brokerage houses), certificates of deposit, mutual funds (stock mutual funds, tax-free bond mutual funds, government bond mutual funds, other bond mutual funds, combination and other mutual funds not including money market mutual fund accounts), stocks, bonds, retirement quasi-liquid assets (sum of the current value in
2001 dollars of IRA/Keogh accounts, thrift-type accounts including 401k, 403b, thrift, savings, SRA, any other account with the option to borrow or withdraw and future pensions held by the head and/or spouse/partner), savings bonds, cash value of life insurance policies, other managed assets (sum of the value of trusts, annuities, and managed investment accounts in which the household has equity interest) and other financial assets not previously included (e.g. loans from the household to another party, future proceeds, royalties, futures, non-public stock, deferred compensation, and oil/gas/mineral investments).
There are six categories of non-financial assets: vehicles, principal residence, other residential real estate, net value of non-residential real estate, business interests, and other non-financial assets, including precious metals, gems, art, and other market-valued collectibles.
Debt is the outstanding balance on housing, other lines of credit, other residential real estate, credit cards, installment loans not classified elsewhere, and other debt not previously captured. Housing debt includes all mortgages, home equity loans, and home equity lines of credit. Other residential real estate debt includes land contracts, residential property other than the principal residence, vacation property and installment debt reported for cottage/vacation properties. Other debt can include items such as loans against pensions, loans against life insurance, margin loans, and any other miscellaneous debt. Total assets are the sum of total financial assets and total non-financial assets. Net worth is total assets minus total debt.
Additional created variables include equity in principal residence, equity of assets in stock holdings, capital gains
2
(on principal residence, stocks, businesses), and total monthly debt payment amount. Equity of assets held in stock is calculated using the net value of directly-held stock, stock mutual funds (full value if described as stock mutual fund, half the value of combination mutual funds), IRAs/Keoghs invested in stock (full value if mostly invested in stock, half the value if split between stocks/bonds or stocks/ money market, a third of the value if split between stocks/bonds/money market), other managed assets with equity interest (full value if mostly stock, half value if split between stocks/mutual funds and bonds/CDs, or indicated ‘‘mixed diversified’’ in the survey, onethird the value if respondent indicated ‘‘other’’), and thrift-type retirement accounts invested in stock (full value if mostly stock, half the value if split between stocks and interest earning assets). Equity of assets directly held in stock is the sum of the net value of directly held stock, stock mutual funds, and half the value of combination mutual funds. Equity of
2
Capital gains on principal residence is the current value less original purchase price and less improvements, on businesses is current value less tax basis—active and non-active businesses, on stocks and mutual funds is the current value less gain/losses.
123
548 J Fam Econ Iss (2006) 27:542–561 retirement assets held in stock is the net value of IRAs/Keoghs invested in stock plus any thrift-type retirement accounts invested in stock.
The debt payment to income ratio is calculated by multiplying the mean total monthly debt payments by 12, and then dividing the product by mean annual household income.
3
The stock equity to income ratio is calculated as equity of assets held in stock divided by annual household income.
4
Data from 1989 and 2001 are used to create wealth deciles
(descriptive analysis) and quintiles (multivariate analysis) for evaluating wealth inequality.
Other Variables
Other variables in this study are education, marital status, saving status, retirement as a reason for saving, race, income, presence of children in the household, employment status, homeowner status, and household pension status. Education is classified as no high school, high school, some college, and college. Marital status is 1 if married or living with partner,
0 otherwise. Proportion of savers (households reporting spending less than income) is reported for the early boomers cohort in 2001, but was not available for 1989. The proportion of households indicating at ages 44–55 that retirement/old age was a reason for saving is given for both cohorts.
Race is equal to 1 if white, non-Hispanic, 0 otherwise. Children live in household is 1 if true. Pension status is coded 1 for ‘yes’ if either the household head or spouse/partner has any type of pension (defined benefit and/or account-based plan) on their current or former job(s). Homeowner is coded 1 for ‘yes’ if the household owns its primary residence.
Employment status categories indicate that the household head is either an employee of someone else; self-employed or in business partnership; retired or otherwise not currently employed and not expecting to work soon (homemaker, student, disabled); or not in the labor force. Income is the total household income for the previous year. In the multivariate analysis, the logarithm of income is used as a non-linear predictor of accumulated wealth.
Method
This study uses the five implicates for each of the Survey of Consumer Finances public data sets for 1989, 1992, 1995, 1998, and 2001. Procedures recommended by the Federal
Reserve are used to calculate descriptive statistics, including use of weights and RII.
5
Given the dual frame design, despite weighting, outliers or skewness may overly influence financial data. Consequently, medians are reported as well as means for many of the continuous balance sheet variables. The Consumer Price Index (CPI) is used to adjust all dollar values to 2001 dollars.
3
Debt payments include the monthly obligation amount on all loan payments. Household annual income is the total income for the previous year.
4
Financial assets held in stocks include value of directly held stock, mutual funds (full value if described as stock mutual fund, ½ value of combination mutual funds), IRAs/Keoghs invested in stock (full value if mostly invested in stock, ½ value if split between stocks/bond or stocks/money market,
1 = 3 value if split between stocks/bonds/money market), other managed assets w/equity interest (annuities, trusts, MIAs: full value if mostly invested in stock, ½ value if split between stocks/MFs & bonds/CDs, or ‘‘mixed/diversified’’,
1 = 3 value if other), thrift-type retirement accounts invested in stock (full value if mostly invested in stock, ½ value if split between stocks and interest earning assets). Income measure for 1989 is the total income for the previous year and income for 2001 is total normal income as reported by the respondent.
5
RII—repeated imputation inference http://www.federalreserve.gov/pubs/oss/oss2/scfindex.html
123
J Fam Econ Iss (2006) 27:542–561 549
Logistic regression is run on the five implicates for a given year using unweighted data and RII to compute parameter estimates and test statistics (Yuh & Montalto,
). Each logistic regression uses the same set of independent variables; averages from the five implicates are used to compute the max-rescaled R
2 values and the percent change in the odds ratios.
Results and Discussion
Comparison of Demographics of Household Heads for Each Generation
Table
reports demographic characteristics of both generations at ages 44–55 (1989 for pre-boomer and 2001 for early boomer). Findings reflect social and economic changes over the 12-year span between 1989 and 2001. Mean educational attainment is higher for early boomers (13.6 years vs. 12.9). A larger proportion of early boomers have attended some college (20% vs. 16%, respectively) or have a college degree (40% vs. 29%, respectively).
The proportion of married or white is about equal for both groups. Almost one and a half times more early boomers as pre-boomers gave ‘save for retirement’ as a reason for saving.
Early boomers have higher levels of mean ($95,598 vs. $80,174) and median inflationadjusted income ($54,479 vs. $50,245) than pre-boomers, with most of the gains coming from those with annual incomes >$100,000. Interestingly, mid-range of the income distribution is similar in both generations. Higher means than medians indicates presence of high-income outliers in each generation’s income distribution. Early boomers are more likely than pre-boomers to be employed (and equally likely to be self-employed) and less likely to have children in the household or a pension.
Table 1 Demographics of pre-boomer and early boomer household heads at ages 44–55
Pre-boomers (1989) Early boomers (2001)
Education (mean)
No high school
High school
Some college
College
Married
Savers
Save for retirement
White
Mean household income
Median household income
<10,000
10,000 to <25,000
25,000 to <50,000
50,000 to <100,000
>100,000
No children in household
Employee
Self-employed
Retired
Unemployed
Households with pension
75
$80,174
$50,245
10
12
29
65 n/a
28
12.9
18
37
16
26
34
18
39
68
16
11
6
70
78
$95,598
$54,479
6
15
40
64
61
41
13.6
12
29
20
24
31
23
48
71
16
9
5
66
Note . Results are in terms of 2001 dollars and all numbers are proportions unless otherwise indicated
123
550 J Fam Econ Iss (2006) 27:542–561
Trends in Inflation-adjusted Balance Sheet Data between 1989 and 2001 for Each
Generation
For this comparison, the time frame and price level are held constant for both generations and we observe change in the inflation-adjusted balance sheets as pre-boomers advance through the life cycle from mid-life (age 44–55 in 1989) to retirement (age 56–67 in 2001) and early boomers progress from young adulthood (age 32–43 in 1989) to mid-life (age
44–55 in 2001). Observed changes are likely generated by such things as investment earnings (e.g., increasing cash value of life insurance over time for both generations), purchase of more or better quality use assets (e.g., increased value of vehicles), shifting needs/wants over the lifecycle (e.g., increased investment in other residential real estate), or other factors not controlled.
Perhaps the most striking change for each generation as it progressed through the life cycle is the increase in real mean and median financial assets. Real mean financial assets of pre-boomers tripled from $112,340 in 1989 to $345,100 in 2001; their real median assets rose 204.5% from $20,600 to $47,500 (see Table
2 ). Results for the early boomer gen-
eration are even more dramatic. For this group, real mean financial assets in 2001 are four and a half times larger than they are in 1989 ($234,000 vs. $51,400) and their real median financial assets increased 254.5% from $11,000 to $39,000 (see Table
). For both generations, the investment market drove these substantial increases. For pre-boomers, the bulk of the increase in real mean financial assets comes from retirement quasi-liquid assets
($75,600 larger in 2001 vs. 1989), stocks ($56,600 higher in 2001 vs. 1989), other managed assets ($38,000 higher in 2001 vs. 1989), mutual funds ($28,900 higher in 2001 vs. 1989), and liquid assets ($15,100 higher in 2001 vs. 1989). The same set of assets contributed to the increase in real mean financial assets of early boomers, but in different rank order and amount as compared with pre-boomers. Early boomers realized a $65,400 gain in average retirement quasi-liquid assets, a $44,500 gain in average stock holdings, a $28,300 gain in average mutual fund holdings, a $17,500 gain in average liquid asset balance, and an
$8,800 gain in average other managed assets.
For pre-boomers and early boomers alike, increase in the real value of home and business assets contributed substantially to increase in real mean and median non-financial assets. Preboomers’ average home value rose 24% (from $133,500 to $165,600) while that of early boomers rose 62% (from $97,000 to $157,500) over the same 12 years, perhaps due to a relatively larger boomer generation bidding up the price of existing housing stock.
Interestingly, the 167% increase in real value of business interest for early boomers far exceeded the 52% increase in real business value experienced by the pre-boomers over the
12 year period.
Gains in financial and non-financial assets are not equally shared. In both generations, a relatively small proportion of the population held the liquid and non-liquid financial assets that yielded the largest gains over time, although the proportion in each generation holding retirement quasi-liquid, mutual fund, and stock assets increased somewhat between 1989 and 2001. Spread between real mean and median figures increased over time within each generation as well, yet another indicator of increasing wealth inequality. In 2001, preboomers had average total real assets of $755,300 as compared with $424,700 in 1989; real median assets were $220,900 and $173,400, respectively. Early boomers have average total real assets equal to $571,400 in 2001 (real median $208,100) as compared with $226,900
(real median $107,900) in 1989.
123
J Fam Econ Iss (2006) 27:542–561 551
Table 2 Mean value of assets and debts held by pre-boomers between 1989 and 2001
Age in survey year 1989
44–55
1992
47–58
1995
50–61
1998
53–64
2001
56–67
Financial assets (% have)
M
Mdn
Liquid
Certificates of deposit
Mutual funds
Stocks
Bonds
Retirement quasi-liquid
Savings bonds
Cash value of life insurance
Other managed assets
Other financial assets
Non-financial assets (%)
M
Mdn
Vehicles
Primary residence
Other residential real estate
Net non-residential R.E.
Business interests
Other non-financial assets
Total assets (% have)
M
Mdn
Liabilities
Housing
Other lines of credit
Other residential real estate
Credit card
Installment loans
Other debt
Total debts (% have)
Total debts M
Total debts Mdn
Mean net worth
Median net worth
91
112.8
20.6
19.8 (88)
6.1 (20)
7.3 (9)
15.6 (22)
7.0 (5)
35.6 (51)
1.0 (22)
7.7 (41)
6.9 (3)
5.9 (16)
93
311.9
132.1
16.3 (90)
133.5 (76)
29.0 (20)
40.6 (15)
86.5 (15)
6.1 (14)
95
424.7
173.4
32.9 (58)
0.8 (4)
5.1 (10)
1.4 (49)
8.0 (58)
1.5 (8)
85
49.6
22.0
375.1
134.3
93
133.5
21.5
19.2 (91)
5.0 (18)
7.2 (10)
23.2 (21)
8.9 (5)
49.1 (52)
1.3 (24)
7.3 (42)
7.0 (5)
5.2 (11)
93
284.5
108.6
12.9 (90)
119.4 (76)
27.3 (18)
39.2 (16)
81.4 (17)
4.2 (8)
96
417.9
155.0
37.7 (56)
0.5 (3)
7.3 (9)
1.6 (47)
4.4 (45)
1.4 (12)
82
52.8
21.0
365.1
112.0
92
183.7
29.5
27.2 (89)
6.5 (15)
21.6 (17)
28.2 (17)
9.9 (4)
62.1 (55)
1.7 (23)
10.7 (40)
10.1 (5)
5.5 (13)
94
306.1
124.3
16.3 (89)
122.1 (81)
27.0 (19)
26.5 (14)
109.3 (17)
5.1 (9)
97
489.8
171.2
38.0 (54)
0.4 (1)
6.8 (9)
1.7 (47)
5.2 (41)
2.7 (10)
80
54.7
18.5
435.1
128.2
96
268.2
42.8
25.6 (94)
9.3 (17)
29.4 (16)
65.9 (24)
12.7 (3)
89.1 (60)
1.5 (18)
8.9 (34)
21.4 (7)
4.4 (9)
94
360.5
132.6
18.4 (89)
139.8 (80)
34.4 (21)
34.1 (11)
126.5 (15)
7.2 (9)
98
628.6
214.7
42.5 (51)
0.2 (2)
6.4 (8)
2.6 (48)
6.6 (41)
3.6 (9)
79
61.9
23.1
566.8
142.4
36.4 (45)
0.6 (2)
5.7 (7)
1.6 (40)
5.0 (36)
2.6 (8)
73
51.9
10.9
703.4
185.8
Note . Results are in 000s of 2001 dollars, numbers in parentheses represent proportion of households holding the line item
95
345.1
47.5
34.9 (93)
7.2 (17)
36.2 (21)
72.2 (26)
18.0 (6)
111.2 (57)
0.9 (13)
11.6 (39)
45.7 (14)
7.1 (10)
94
410.2
141.1
19.1 (89)
165.6 (82)
38.2 (17)
47.5 (12)
131.2 (15)
8.5 (8)
98
755.3
220.9
Home mortgage accounted for over 70% of average liabilities in both generations in
2001. In 2001, the amount owed by early boomers is over one and a half times higher than that owed by pre-boomers. Real level owed remains relatively constant for pre-boomers
($32,900 vs. $36,400) but early boomers faced a 41% increase in mortgage level between
1989 and 2001. Pre-boomers, nearing retirement in the years observed, have likely paid down their mortgages in anticipation of retirement, while early boomers have more years left to pay on their mortgages and may be facing attempts to upgrade housing in a tight and competitive market.
Given the criticism that boomers have received for being spendthrifts, it is fascinating to see that the percent of average total debt due to credit card liability is exactly the same for
123
552 J Fam Econ Iss (2006) 27:542–561
Table 3 Mean value of assets and debts held by early boomers between 1989 and 2001
Age in survey year 1989
44–55
1992
47–58
1995
50–61
1998
53–64
2001
56–67
Financial Assets (% have)
M
Mdn
Liquid
Certificates of deposit
Mutual funds
Stocks
Bonds
Retirement quasi-liquid
Savings bonds
Cash value of life insurance
Other managed assets
Other financial assets
Non-financial assets (%)
M
Mdn
Vehicles
Primary residence
Other residential real estate
Net non-residential R.E.
Business interests
Other non-financial assets
Total assets (% have)
M
Mdn
Liabilities
Housing
Other lines of credit
Other residential real estate
Credit card
Installment loans
Other debt
Total debts (% have)
M
Mdn
Mean net worth
Median net worth
89
51.4
11.0
8.9 (84)
2.8 (15)
1.4 (6)
4.3 (15)
3.7 (3)
16.7 (47)
0.7 (32)
5.5 (38)
4.8 (3)
2.5 (14)
90
175.4
90.4
12.1 (88)
97.0 (64)
9.7 (10)
7.4 (11)
41.6 (15)
7.5 (14)
95
226.9
107.9
43.6 (57)
0.4 (5)
3.3 (6)
1.4 (49)
9.2 (66)
0.7 (8)
88
58.8
33.0
168.1
54.6
90
57.0
10.0
10.4 (86)
2.0 (10)
4.5 (10)
6.1 (19)
4.3 (3)
18.3 (49)
0.9 (28)
5.5 (36)
2.5 (3)
2.5 (11)
92
182.7
92.9
11.8 (89)
93.8 (65)
11.4 (12)
10.7 (7)
51.8 (16)
3.1 (9)
97
239.7
117.4
47.0 (56)
0.1 (3)
4.3 (7)
1.7 (51)
5.9 (58)
1.2 (11)
87
60.1
30.9
179.6
57.2
92
90.3
17.4
10.4 (88)
3.7 (10)
10.0 (13)
10.5 (16)
4.4 (3)
35.2 (56)
1.2 (29)
8.5 (31)
3.4 (3)
2.9 (11)
91
206.2
108.1
14.9 (88)
102.1 (69)
14.5 (13)
9.1 (9)
61.5 (14)
4.2 (10)
96
296.5
135.8
49.1 (58)
0.2 (2)
4.6 (7)
2.4 (58)
6.2 (58)
1.2 (11)
86
63.7
33.6
232.8
74.7
95
157.0
34.3
18.2 (92)
3.9 (12)
22.5 (21)
29.4 (22)
5.1 (3)
49.2 (60)
1.1 (24)
12.8 (32)
11.6 (6)
3.1 (11)
93
261.5
120.8
16.7 (87)
122.7 (74)
17.4 (14)
17.5 (10)
83.6 (17)
3.6 (8)
97
418.5
170.3
55.3 (61)
0.2 (3)
5.0 (6)
2.5 (53)
8.5 (51)
2.7 (10)
87
74.3
42.3
344.2
102.2
61.4 (61)
0.4 (2)
4.4 (7)
2.2 (50)
7.4 (47)
1.9 (8)
85
77.6
39.5
493.7
134.8
Note . Results are in 000s of 2001 dollars, numbers in parentheses represent proportion of households holding the line item
94
234.1
39.5
26.4 (92)
4.3 (16)
29.7 (20)
44.8 (22)
11.0 (3)
82.1 (64)
1.6 (22)
14.3 (30)
13.5 (6)
6.5 (9)
95
337.2
135.0
19.3 (90)
157.5 (77)
25.0 (15)
18.9 (10)
110.9 (17)
5.6 (7)
98
571.4
208.1
early boomers as it is for pre-boomers, 3%. While the average balance of $2,200 for early boomers in 2001 is higher than the average $1,400 balance in 1989, it represents a decline of $300 as compared with 1998. These results lend little support to the idea that boomers, on average, are recklessly consuming today at the expense of tomorrow.
Average net worth of pre-boomers is $703,400 in 2001 (real median $185,800) as compared with $375,100 in 1989 (real median $134,300). Real net worth of early boomers in 2001 ($493,700, real median $134,800) is lower than that of pre-boomers. But, early boomers have experienced a 194% increase in real average net worth between 1989 and
2001 as compared with the 88% increase in real average net worth among pre-boomers for the same time period.
123
J Fam Econ Iss (2006) 27:542–561 553
Comparison of Asset and Debt Holdings of Each Generation at Ages 44 to 55
Table
contains inflation-adjusted balance sheet components for pre-boomers and early boomers when each generation is at the same stage in life (age 44 to 55). Results of t -test comparisons indicate that the real mean net worth of early boomers aged 44–55 ($493,700) is significantly higher than the real mean net worth of pre-boomers ($375,100) at comparable ages.
Midlife early boomers have more than double the real mean financial assets of preboomers ($234,100 vs. $112,800). Much of the difference between early boomers and preboomers is in value of mutual funds ($29,700 vs. $7,300), stocks ($44,800 vs. $15,600),
Table 4 Mean values of assets and debts held by pre-boomer and early boomers at ages 44–55
1989 (pre-boomer) 2001 (early boomer)
Financial assets (% have)
M
Mdn
Liquid
Certificates of deposit
Mutual funds
Stocks
Bonds
Retirement quasi-liquid
Savings Bonds
Cash value of life insurance
Other managed assets
Other financial assets
Non-financial assets (%)
M
Mdn
Vehicles
Primary residence
Other residential real estate
Net non-residential R.E.
Business interests
Other non-financial assets
Total assets (% have)
M
Mdn
Liabilities
Housing
Other lines of credit
Other residential real estate
Credit card
Installment loans
Other debt
Total debts (% have)
Mn
Mdn
Mean net worth
Median net worth
Net worth bottom quintile limit
Net worth upper quintile limit
91
112.8
20.6
19.8 (88)
6.1 (20)
7.3 (9)
15.6 (22)
7.0 (5)
35.6 (51)
1.0 (22)
7.7 (41)
6.9 (3)
5.9 (16)
93
311.9
132.1
16.3 (90)
133.5 (76)
29.0 (20)
40.6 (15)
86.5 (15)
6.1 (14)
95
424.7
173.4
32.9 (57)
0.8 (5)
5.1 (6)
1.4 (49)
8.0 (66)
1.5 (8)
85
49.6
22.0
375.1
34.3
20.8
380.4
61.4 (61)
0.4 (2)
4.4 (7)
2.2 (50)
7.4 (47)
1.9 (8)
85
77.6
39.5
493.7
134.8
22.7
486.6
Note . Results are in 000s of 2001 dollars, numbers in parentheses represent proportion of households holding the line item. Bold, italicized results indicate statistically significant mean difference at 90% or greater
123
94
234.1
39.5
26.4 (92)
4.3 (16)
29.7 (20)
44.8 (22)
11.0 (3)
82.1 (64)
1.6 (22)
14.3 (30)
13.5 (6)
6.5 (9)
95
337.2
135.0
19.3 (90)
157.5 (77)
25.0 (15)
18.9 (10)
110.9 (17)
5.6 (7)
98
571.4
208.1
554 J Fam Econ Iss (2006) 27:542–561 and quasi-liquid retirement funds ($82,100 vs. $35,600). Median financial assets for early boomers ($39,500) are nearly double that of pre-boomers ($20,600).
Early boomers also have significantly greater non-financial assets than pre-boomers have at the same age; however the gap is not as wide as with financial assets. Worth of vehicles and home is somewhat higher for early boomers than for pre-boomers at the same age ($19,300 vs. $16,300 and $157,500 vs. $133,500, respectively). At midlife, preboomers have a significantly larger interest in net non-residential real estate ($40,600 vs.
$18,900), however. Total asset holdings are 35% greater for early boomers ($571,400) than for pre-boomers ($424,700), and median total assets are 20% higher ($208,100 vs.
$173,400).
While early boomers hold greater assets than pre-boomers did at the same age, they also hold greater liabilities. Total liabilities for midlife early boomers are $77,600, compared with $49,600 for pre-boomers at the same life stage. This $28,000 disparity in total liabilities is essentially a result of the $28,500 higher mean housing debt for early boomers.
‘‘Free spending’’ midlife boomers held $9,600 in credit card and installment debt in 2001, while pre-boomers have $9,400 in credit card and installment debt at the same age. Mean assets minus liabilities for early boomers are 32% greater than for pre-boomers ($493,700 vs. $375,100), however median net worth had only differed by $500 between the two cohorts at ages 44–55 (from $134,300 to $134,800).
The limit of the 20th percentile for pre-boomers at ages 44–55 is $20,785; the same percentile limit for early boomers is 9% higher at $22,670. Similarly, the 80th percentile limit for pre-boomers aged 44–55 is approximately $380,000 and for early boomers at the same age, the limit is 28% higher at approximately $487,000.
Equity Value, Capital gains, and Debt Service Trends Between 1989 and 2001 for Each
Generation
Changes in the equity in financial and non-financial assets, as well as changes in monthly debt burden and the proportion of stock equity to income are presented in Tables
(preboomers) and
(early boomers). Both generations experienced a substantial gain in equity of assets held in stock between 1989 and 2001 (392% for pre-boomers and 829% for early boomers) and in capital gains on stocks and mutual funds (228% for pre-boomers and
1027% for early boomers). Despite a decline in real estate equity in the early to mid 1990s, both generations experience a gain in principal residence equity over time. The 66%
Table 5 Mean values for equities, capital gains, and debt service of pre-boomers
1989 1992 1995 1998
Age in survey year
Principal residence equity
Equity of assets held in stock
Equity in directly held stock
Equity in ret. stock assets
Capital gains on residence
Capital gains on business
Capital gains on stocks/MF
Total monthly debt payments
Debt PMT to income ratio
Stock equity to income ratio
44–55
100.5
38.1
19.3
13.3
68.1
45.1
6.6
0.8
12%
48%
Note . Results are in 000s of 2001 dollars
123
47–58
81.8
56.0
26.8
25.1
51.7
52.6
6.4
0.8
14%
82%
50–61
84.1
79.0
42.5
30.0
54.1
68.0
12.1
0.7
12%
108%
53–64
97.4
154.7
88.2
56.2
57.4
84.7
34.8
0.9
13%
194%
2001
56–67
129.2
187.5
96.0
63.5
78.7
83.7
21.7
0.7
10%
226%
J Fam Econ Iss (2006) 27:542–561 555
Table 6 Mean values for equities, capital gains, and debt service of early boomers
1989 1992 1995 1998 2001
Age in survey year
Principal residence equity
Equity of assets held in stock
Equity in directly held stock
Equity in ret. stock assets
Capital gains on residence
Capital gains on business
Capital gains on stocks/MF
Total monthly debt payments
Debt PMT to income ratio
Stock equity to income ratio
Note . Results are in 000s of 2001 dollars
32–43
53.4
14.7
5.5
5.8
32.2
19.8
17%
24%
1.1
0.9
35–46
46.9
17.6
8.5
8.0
25.1
36.6
18%
30%
2.1
0.9
38–49
53.0
39.0
17.2
20.5
26.4
31.7
16%
60%
3.7
0.9
41–52
67.5
87.5
46.2
32.8
33.9
55.4
13.0
1.0
16%
118% increase in capital gains on residence of early boomers between 1989 and 2001 is greater than the modest 16% gain of pre-boomers. Debt payment to income is somewhat smaller for pre-boomers than for early boomers (10% vs. 12%). For both generations, levels of debt payment to income dropped 2 to 3 percentage points between 1989 and 2001.
44–55
96.2
136.6
69.3
58.6
53.6
72.8
12.4
1.0
12%
143%
Equity Value, Capital Gains, and Debt Service of Each Generation at Age 44–55
Midlife early boomers have significantly more equity in directly and indirectly held stock and a much larger stock equity to income ratio than midlife pre-boomers, but their capital gains on residence are significantly below what pre-boomers enjoy (see Table
). Total monthly debt payments for early boomers are somewhat higher than for pre-boomers (1.0
vs. 0.8) but the debt payment to income is not significantly different between the two generations.
Wealth Distribution Among Pre-boomer and Early Boomers
To examine wealth dispersion within and between the generations at the same lifecycle stage (age 44–55), the distribution of financial assets, non-financial assets, and net worth for each generation is sorted in ascending order and divided into deciles or percentiles (see
Table
Table 7 Mean values of equities, capital gains, and debt service of pre- and early boomers at ages 44–55
1989 (pre-boomer) 2001 (early boomer)
Principal residence equity
Equity of assets held in stock
Equity in directly held stock
Equity in ret. stock assets
Capital gains on residence
Capital gains on business
Capital gains on stocks/MF
Total monthly debt payments
Debt PMT to income ratio
Stock equity to income ratio
100.5
38.1
19.3
13.3
68.1
45.1
6.6
0.8
12%
48%
96.2
136.6
69.3
58.6
53.6
72.8
12.4
1.0
12%
143%
Note . Results are in 000s of 2001 dollars except where percentages are indicated. Bold, italicized results indicate statistically significant mean difference at 90% or greater
123
556 J Fam Econ Iss (2006) 27:542–561
Table 8 Asset and net worth deciles of pre-boomers and early boomers at ages 44–55
Deciles Financial assets Non-financial assets Net worth
Pre-1989 Early 2001 Pre-1989 Early 2001 Pre-1989 Early 2001
10
20
30
40
50
60
70
80
90
95
Mean
0.0
1.7
4.7
10.4
20.6
32.8
53.5
110.3
240.0
409.5
112.8
0.3
2.3
9.0
21.6
39.5
69.6
116.3
217.5
478.5
877.5
234.1
Note . Results are in 000s of 2001 dollars
2.8
19.5
57.8
96.0
132.1
177.6
246.4
309.7
552.0
985.5
311.9
4.1
20.0
67.0
98.6
135.0
178.2
238.9
343.6
582.6
1,098.0
337.2
2.8
20.8
44.9
87.6
134.3
178.7
264.3
380.4
739.0
1,317.2
375.1
3.2
22.7
43.9
81.9
134.8
198.5
295.2
486.6
943.4
1,517.5
493.7
Within each decile, early boomers have greater mean financial assets than pre-boomers; each group between the 30th and 95th percentile held roughly double what pre-boomers have at the same age. Likewise, within each decile, early boomers hold greater nonfinancial assets than pre-boomers with the exception of the 70th percentile. Total net worth is slightly lower for the 30th and 40th percentiles among early boomers than for preboomers at the same age. Early boomers have slightly higher net worth when comparing the lower and middle percentiles, with the greatest difference observed at the 80th percentile and the 90th percentile (both 28% higher). These results suggest that across the wealth distribution, midlife early boomers fared better than midlife pre-boomers, although it is clear that the disparity between generations is much greater at the top deciles as compared with the lower deciles.
Multivariate logistic regression is used to identify significant factors associated with the odds of either midlife pre-boomers or midlife early boomers being in the bottom (see
Table
) quintile of the net worth distribution. As noted in Table
inflation-adjusted quintile limits for the bottom 20% and top 80% of net worth are higher for early boomers aged 44–55 as compared with the pre-boomer cohort at the same life cycle stage. Results for the max-rescaled R
2 indicate the four models account for 60% to
70% of the variation in the dependent variables.
Relatively few factors are significantly associated with odds of midlife pre-boomers being in the bottom net worth quintile (see Table
). Significantly reduced odds are associated with having a college education as compared with less than high school, being a homeowner, and having high levels of income. Significantly higher odds are associated with working for someone else or being retired as compared with being self-employed.
Among midlife early boomers, considerably more factors are significant (see Table
).
Odds of being in the bottom net worth quintile are significantly lower for those with at least a high school degree, those closer to age 55 than age 44, those with no children at home, homeowners, married, those with any pension, those who are white, non-Hispanic, and those who have high levels of income. As for the pre-boomers, those working for someone else or retired as compared with being self-employed have significantly higher odds of low net worth.
Higher education and employment are key factors for both generations to escape the bottom net worth quintile. College educated pre-boomer cohort heads are almost twice less likely (81% odds decrease) to be in the bottom net worth quintile as compared with those
123
J Fam Econ Iss (2006) 27:542–561 557
Table 9 Logistic regression results–odds of being in the bottom net worth quintile
Variable 1989 Pre-boomers ( n =729) 2001 Early boomers ( n =1255)
Coefficient % Change in odds ratio
Coefficient % Change in odds ratio
Education
High school (HS)
Some college
College
Age
Work status
Employee
Retired
Unemployed
No children in the home
Homeowner
Married
Any pension
White
Income (log)
Max-rescaled R
2
)
)
)
)
0.346
0.672
1.663***
0.058
0.991*
1.254*
0.784
)
0.151
)
2.953***
)
0.569
)
0.223
)
0.381
)
0.486***
0.606
–
–
)
81.0
3
–
170.9
256.3
–
–
)
94.8
1
–
–
–
)
38.5
2
0.705
)
)
)
)
0.933**
0.816*
2.044***
0.073*
1.997***
2.788***
1.436**
)
0.332
)
3.297***
)
0.694**
)
1.167***
)
0.863***
)
0.410***
)
)
)
60.5
55.6
87.0
)
7.0
639.6
1536.3
328.0
2
–
)
96.3
1
)
49.6
)
68.7
)
57.4
)
33.6
3
Note .
1,2,3
Rank order based on standardized regression coefficients from the five implicates. (Note: for preboomers 3 of 5 implicates have the rank order indicated above, the remaining 2 regressions ranked homeowner first, college second, and income third)
* p <.10; ** p < .05; *** p < .01
Table 10 Logistic regression results—odds of being in the top net worth quintile
Variable 1989 Pre-boomers ( n =729) 2001 Early boomers ( n =1255)
Coefficient % Change in odds ratio
Coefficient % Change in odds ratio
Education
High school (HS)
Some college
College
Age
Work status
Employee
Retired
Unemployed
No children in the home
Homeowner
Married
Any pension
White
Income (log)
Max-rescaled R
2
)
0.019
0.623
1.085***
0.008
)
1.377***
)
1.107
)
1.898*
0.065
1.132***
0.464
)
0.684*
0.524
1.728***
0.692
–
–
192.2
3
–
)
74.7
2
–
)
84.4
–
277.1
–
)
49.1
–
470.3
1
0.687
0.289
0.767*
1.629***
0.070***
)
2.084***
)
0.766*
)
0.818*
)
0.230
1.119***
0.218
0.122
0.690**
1.078***
Note .
1,2,3
Rank order based on standardized regression coefficients from the five implicates
* p < .10; ** p < .05; *** p < .01
–
116.8
412.5
3
7.2
)
86.7
2
)
53.5
)
55.8
–
206.6
–
–
100.3
193.9
1 without a high school degree. All levels of education above no high school degree are significant factors for midlife early boomers. Early boomer heads with high school and some college are more than half less likely (60% and 56% odds decreases, respectively)
123
558 J Fam Econ Iss (2006) 27:542–561 and, like pre-boomer heads, college educated early boomer heads are almost twice less likely (87% odds decrease) to be in the bottom net worth quintile. Self-employed heads in either generation are much less likely to be in the bottom net worth quintile but the magnitude of this effect is much greater for early boomers. Retired pre-boomer heads are about three and half times more likely to be in the bottom net worth quintile while retired early boomer heads are more than 16 times more likely to be in the bottom quintile as compared with their respective self-employed counterparts.
For pre-boomer cohort heads, marriage, race, and pension status have no impact on bottom quintile membership. For early boomer heads, being married, white, and having a pension in the household reduced the odds of being in the bottom net worth quintile. Not surprisingly, a negative relationship is found between income and bottom quintile net worth membership.
Reasonably, the factors significantly associated with increased odds of being in the top wealth quintile generally reflect the inverse of factors associated with being in the bottom wealth quintile (see Table
10 ). High school education, having children in the home, and
marital status are not statistically significant predictors of top net worth quintile membership for either generation.
Age is not a significant factor in wealth acquisition for pre-boomers, but is for early boomers with those nearer age 55 having greater odds of being in the top wealth quintile.
Having at least a high school education, being self employed, being a home owner, white non-Hispanic, having high income are all associated with greater odds of being in the top wealth quintile. College educated pre-boomer heads are almost three times as likely and college educated early boomers more than five times as likely to be in the top net worth quintile as compared with their counterpart heads with less than a high school education.
Self-employed heads from both generations have higher odds of being in the top net worth quintile compared to heads that are employees, retired, or unemployed. Pre-boomer homeowners are almost four times as likely and early boomers are three times as likely to be in the top wealth quintile as compared with non-homeowner counterparts. Interestingly, pre-boomer heads with a pension have lower odds of top net worth quintile membership, but having a pension is not a significant factor for early boomers. White early boomer heads are twice as likely to be in the top net worth quintile, but race is not a significant factor for pre-boomers. A positive relationship exists between income level and top net worth quintile membership for both cohorts.
Rank order of standardized regression coefficients assessed over the five implicates identified which variables in each logistic regressions are most influential in explaining variance in quintile membership. Analysis of the membership in the bottom wealth quintile indicated that homeownership ranks first in importance among the independent variables for both generations. Odds of being in the bottom net worth quintile for non-homeowners are almost twice as high as the odds for homeowners. For midlife pre-boomers, income and college education rank second and third in importance, respectively, as explanatory variables. For midlife early boomers the second and third ranked independent variables are college education and working for someone else in contrast to being self-employed.
Rank order of variables associated with membership in the top wealth quintile indicates that importance of income level is identical for the two cohorts. Interestingly, the magnitudes of these effects are similar for work status, but reversed for income and education.
College educated pre-boomers have almost three times higher odds while college educated early boomers have just over five times higher odds of being in the highest net worth quintile. These results point to the importance of education, especially for early boomers,
123
J Fam Econ Iss (2006) 27:542–561 559 not only for its role in income generation, but also in the financial sophistication required for making one’s wealth grow.
Conclusion
At $493,700, the inflation-adjusted mean net worth of early boomers is 32% higher for early boomers than it is for pre-boomers at the same age ($375,100), making it easy to conclude, as others have (Easterlin et al.,
; Keiser,
2000 ; Sabelhaus & Manchester,
), that the boomer generation as a whole is better off than the prior generation. But, averages do not capture individual experience.
Financial assets gains, in particular, are not equally distributed. Real financial asset levels of midlife early boomers at and above the 80th percentile are about 100% larger than such holdings of midlife pre-boomers. Among early boomers, financial assets at the 95th percentile are 22 times larger than the 50th percentile and 2,925 times larger than the lowest decile, suggesting that early boomers in the mid- to lowest-deciles may face difficulty funding retirement, especially if they have relatively less opportunity to be in a defined benefit plan.
Education and race are significantly linked to wealth disparity among early boomers an emerging policy issue of interest to family economists. Interestingly, self-employment is a significant correlate of being in the top wealth quintile. These findings underscore the importance of obtaining higher education and considering entrepreneurial opportunity as a means of improving financial status. They also point to the possibility of racial differences in saving and investing behavior. Targeting financial education to minority races may be beneficial.
Far from their caricature as materialistic overspenders (Francese,
), early boomers, on average, accumulated no more credit and installment debt than the previous age cohort and, in fact, have a debt to income ratio identical to the ratio that the pre-boomers have at the same age. A smaller proportion of early boomer households have a pension plan, but a larger proportion are saving for retirement at midlife as compared with their predecessors.
Although the top of the wealth distribution enjoy disproportionately larger gains from financial asset growth, the proportion of early boomers holding virtually every major financial asset, especially equity based assets (e.g. mutual funds), increases markedly between 1989 and 2001. More than previous generations, early boomers are relying on the stock market to support their retirement. Early boomers have a slightly larger portfolio at midlife than pre-boomers have, but it is a riskier portfolio weighted toward volatile financial asset categories. Greater reliance by boomers on private, tax-sheltered accounts to fund retirement underscores the increasing importance of investor education since a decline in equity prices could increase the social risk of supplementing retirement income and reduce the possibility that boomers can retire comfortably.
Acknowledgements The authors would like to thank the Certified Financial Planner Board of Standards and the American Council on Consumer Interests (ACCI) for recognizing an earlier version of this research for the ACCI 2005 Certified Financial Planner Board of Standards Financial Planning Research Award.
References
AccountingWeb (2004). New faces in bankruptcy court: The middle aged. Retrieved September 9:2004 from http://www.accountingweb.com
123
560 J Fam Econ Iss (2006) 27:542–561
American Association of Retired Persons (AARP) (1994).
Aging baby boomers: How secure is their economic future?
Washington, D.C.: AARP.
Ando, A., & Modigliani, F. (1963). The ‘‘life cycle’’ hypothesis of saving: Aggregate implications and tests.
The American Economic Review, 53 (1), 55–84.
Baek, E., & DeVaney, S. A. (2004). Assessing the baby boomers’ financial wellness using financial ratios as a subjective measure.
Family and Consumer Sciences Research Journal, 32 (4), 321–348.
Bernheim, B. D. (1993).
Is the baby boom generation preparing adequately for retirement? Summary report . Princeton, N.J.: Merrill Lynch.
Bernheim, B. D. (1995).
The Merrill Lynch Baby Boom retirement index, update ‘95 . Princeton, N.J.: Merrill
Lynch.
Bernheim, B. D. (1996).
The Merrill Lynch Baby Boom retirement index, update ‘96 . Princeton, N.J.: Merrill
Lynch.
Bernheim, B. D. (1997).
The adequacy of personal retirement saving: Issues and options . In: D. A. Wise
(Eds.), Facing the age wave . Publication no. 440. Stanford, CA: Hoover Institution Press, pp. 30–56.
Congressional Budget Office (2003). Baby boomers’ retirement prospects: An overview. Washington, DC:
Congressional Budget Office.
Danzinger, S., & Gottschalk, P. (1995).
America unequal . New York: Russell Sage Foundation.
Dugas, C. (2003). More workers can’t afford to retire. USA Today. Retrieved October 13:2004, from http:// www.usatoda.com/money/perfi/retirement/2003-04-10-noretire_x.htm
Easterlin, R. A., Schaeffer, C. M., & Macunovich, D. J. (1993). Will the baby boomers be less well off than their parents? Income, wealth, and family circumstances over the life cycle in the United States.
Population and Development Review, 19 (3), 497–522.
Finke, M. S., & Huston, S. J. (2003). The brighter side of financial risk: Financial risk tolerance and wealth.
Journal of Family and Economic Issues, 3 (24), 233–256.
Francese, P. (2001). Big spenders.
American Demographics, 23 (9), 29.
Gale, W. G. (1997). The aging of America: Will the Baby Boom be ready for retirement?
Brookings Review,
15 (3), 4–9.
Gist, J. R., & Wu, K. (2004).
The inequality of financial wealth among boomers . Washington, D.C.: AARP
Public Policy Institute.
Gist, J., Wu, K., & Verma, S. (2004).
The distribution of financial wealth among boomers . Washington,
D.C.: AARP Public Policy Institute.
Gist, J.R., Wu, K.B., & Ford, C. (1999).
Do baby boomers save and if so, what for?
Washington, D.C.:
AARP Public Policy Institute.
Gittleman, M., & Wolff, E. (2004). Racial differences in patterns of wealth accumulation.
Journal of Human
Resources, 39 (1), 193–227.
Gustman, A. L., & Steinmeier, T. L. (1998). Effects of pensions on savings: Analysis with data from the
Health and Retirement Study. Working paper no. 6681. National Bureau of Economic Research. Retrieved on October 5:2004 from http://www.nber.org/papers/w6681
Karoly, L. (1993). The trend in inequality among families, individuals, and workers in the United States: A twenty-five year perspective. In: S. Danzinger, & P. Gottschalk (Eds.), Uneven tides: Rising inequality in America (pp. 19–97). New York: Russell Sage Foundation.
Keister, L. (2000).
Wealth in America: Trends in wealth inequality . New York: Cambridge University Press.
Kennickell, A. (1999). Using income data to predict wealth. Federal Reserve Board. Retrieved on April
22:2005 from http://www.federalreserve.gov/pubs/oss/oss2/papers/wealth.income.7.web.pdf.
Kotlikoff, L., & Auerbach, A.J. (1994). U.S. Fiscal and savings crises and their impact for baby boomers. In:
D.L. Salisbury N.S. Jones (Eds).
Retirement in the 21st century: Ready or not?
Washington, D.C.:
Employee Benefit Research Institute, pp. 85–126.
Moore, J. F., & Mitchell, O. S. (1997). Projected retirement wealth and savings adequacy in the Health and
Retirement study. Working paper no. 6240, National Bureau of Economic Research. Retrieved on
October 10 from http://www.nber.org/papers/w6240
Ryscavage, P. (1999).
Income inequality in America . Armonk: New York: M.E. Sharpe.
Sabelhaus, J., & Manchester, J. (1993). Baby boomers and their parents: How does their economic wellbeing compare in middle age?
Journal of Human Resources, 30 (4), 791–806.
Svensson, P. (2003). Boomers are good at saving, but not good enough. Associated Press. Retrieved October
13:2004 from http://www.taxpolicycenter.org/news/boomers_save.cfm
Toder, E., Thompson, L., Favreault, M., Johnson, R., Perese, K., Ratcliffe, C., et al. (2002).
Modeling income in the near term: Revised projections of retirement income through 2020 for the 1931–1960 birth cohorts, final report . Washington, D.C.: Urban Institute.
Tracy, J., Schneider, H., & Shan, S. (1999). Are stocks overtaking real estate in household portfolios?
Federal Reserve Bank of New York Current Issues in Economics and Finance, 5 (5), 1–6.
123
J Fam Econ Iss (2006) 27:542–561 561
Verma, S., Lichtenstein, J. (2003).
In brief: Retirement plan coverage of baby boomers and retired workers:
Analysis of 1998 SIPP data research report . Washington, D.C.: AARP Public Policy Institute.
Warshawsky, M. J., Ameriks, J. (2000). How prepared are Americans for retirement? In: O. S. Mitchell, P.
B. Hammond, & A. M Rappaport (Eds.), Forcasting retirement needs and retirement wealth . Philadelphia: University of Pennsylvania Press, pp. 33–67.
Warren, E. (2003). Consumer bankruptcy: Issues summary. Harvard Law School.
Wolff, E. (2002).
Top heavy . New York: The Century Fund.
Yao, R., Hanna, S. D., & Lindamood, S. (2004). Changes in financial risk tolerance, 1983–2001.
Financial
Services Review, 13 (4), 249–266.
Yuh, Y., & Montalto, C.P. (1998, August). SAS code for RII logit estimation. Retrieved February 2:2005 from http://hec.osu.edu/people/shanna/nrsas.htm
Yuh, Y., Montalto, C. P., & Hanna, S. (1998). Are Americans prepared for retirement?.
Financial Counseling and Planning, 9 (1), 1–13.
123
