Course level: HIST2 History through Reconstruction

Journal of Aging & Social Policy, 22:117–128, 2010 Copyright © Taylor & Francis Group, LLC ISSN: 0895-9420 print/1545-0821 online DOI: 10.1080/08959421003620848

The Impact of the Housing Crash on the Wealth of the Baby Boom Cohorts

DAVID ROSNICK, PhD Economist, Center for Economic and Policy Research, Washington, DC, USA

DEAN BAKER, PhD Codirector, Center for Economic and Policy Research, Washington, DC, USA

The collapse of the housing bubble and the resulting plunge in the stock market destroyed more than $10 trillion in household wealth. The impact was especially severe for the baby boom cohorts who are at or near retirement age. This paper uses data from the Federal Reserve Board’s 2007 Survey of Consumer Finances to compare the wealth of the baby boomer cohorts just before the crash with pro- jections of household wealth following the crash. These projections show that most baby boomers will be almost entirely dependent on their Social Security income after they stop working.

KEYWORDS baby boomers, housing bubble, retirement income, Social Security, survey of consumer finances

INTRODUCTION

The collapse of the housing bubble and the resulting plunge in the stock market have destroyed more than $10 trillion in household wealth between the peak of the bubble in 2007 and the beginning of 2009. This loss of wealth is likely to prove especially painful for the baby boom cohorts who are nearing retirement. This paper uses projections derived from data in the 2007 Survey of Consumer Finance (SCF) to assess the wealth of these cohorts in the wake of the crash. It updates two earlier papers that constructed projections based on the 2004 SCF (Baker & Rosnick, 2008; Rosnick & Baker, 2008).

Received June 15, 2009; revised August 31, 2009; accepted September 8, 2009. Address correspondence to David Rosnick, Center for Economic and Policy Research,

1611 Connecticut Avenue NW, Washington, DC 20009, USA. E-mail: [email protected]

117

118 D. Rosnick and D. Baker

The impact of the collapse of the housing bubble on the wealth of the baby boom cohorts has very important policy implications, first and foremost for the status of Social Security and Medicare. Both programs will face serious financial problems during the retirement years of the baby boom cohorts, leading to substantial pressure for cuts in benefits. The ability of the baby boomers to adjust to lower benefits will depend on the extent to which they are able to generate income outside of Social Security.

This study also has important policy implications for pension policy more generally. If most baby boomers were unable to accumulate sufficient savings to support a decent retirement under the current system, then it suggests a need for improvements in the retirement system. Ideally, private savings plus pensions will be sufficient to allow most workers to maintain their standard of living in retirement.

The first part of this paper briefly reviews some of the prior research on the wealth of the baby boomers. The second section explains the method- ology used to analyze the wealth of the baby boomers in 2007 and makes projections for 2009. The third section presents the analysis and projections. The fourth section is a brief conclusion.

THE WEALTH OF THE BABY BOOM COHORTS: PRIOR WORK

There is extensive literature on the long-term trends in preparedness for retirement. Edward Wolff (2002) concluded that based on analysis of the SCF, most families were poorly prepared for retirement even prior to the crash of the technology bubble at the end of the decade. This study showed that the growth in wealth in the 1980s and 1990s was concentrated at the top. Of course, much of this wealth was lost in the crash of the stock bubble at the turn of the millennium. The crash may have somewhat mitigated the growth in inequality but did not leave middle- and lower-income families better prepared for retirement.

An update of this study (Weller & Wolff, 2005) points to the growing share of Social Security in retirement income, which is largely attributable to the decline of private defined benefit pension plans. Weller and Wolff report that from 1983 to 2001, the median net worth (including defined benefit pensions and Social Security) of those aged 47 to 55 rose only 13.4% in a period in which per capita income rose by 51.0%. The study projected that one in five homeowners and three in five renters aged 47 to 64 would have insufficient wealth to provide an income of more than double the poverty line—even with Social Security benefits included.

Munnell and Soto (2008) examine the effect of the housing boom on homeowners’ balance sheets, although they focus on the issue of equity extraction. They conclude that in 2004 mid-distribution households headed by those aged 50 to 64 consumed an additional $15,000 on account of

Impact of the Housing Crash 119

bubble wealth effects. Munnell, Golub-Sass, Soto, and Webb (2008) find that baby boomers are likely to have insufficient savings to maintain their standard of living in retirement.

Engen, Gale, and Uccello (1999) present a somewhat more optimistic picture. Applying a life cycle consumption model to data from the SCF and the Health and Retirement Survey, the study finds that most families should be able to maintain a consumption level in retirement comparable to that which they enjoyed in their working years.

This paper goes beyond these prior studies in examining the impact of the collapse of the housing bubble and the subsequent fall in the stock mar- ket on the wealth of the baby boom cohorts. These events have substantially reduced their wealth.

METHODOLOGY FOR ANALYZING THE WEALTH OF THE BABY BOOMERS

The analysis is based on the Federal Reserve Board’s (FRB) 2007 SCF. The definition of wealth and weighting of the SCF public data are identical to those used by the FRB in its published summary tables. They are based directly on a Stata translation of the FRB’s SAS program for producing the 2007 SCF Chartbook.1 The one difference is that we define quintiles of net worth by thresholds and not by survey weights.2 Some, notably Edward Wolff, suggest changes to the survey weights to enforce consistency between the FRB income data and the Internal Revenue Service. Wolff also makes additional adjustments to the survey data to match in the aggregate the FRB Flow of Funds data. These adjustments are useful, especially in the context of measuring wealth and inequality, but they are likely to have little impact on a comparative analysis of wealth for households below the top quintile. Therefore, we have not attempted to make these adjustments in this analysis.

Table 1 shows the key assumptions used in projecting wealth forward from September 2007 to September 2009. The projections use three scenarios that are intended to encompass the range of plausible values for housing and stock prices. Scenario 1 is designed to be relatively optimistic. It assumes that nominal house prices do not fall from their March 2009 level. At that point, house prices were falling at a nominal rate of almost 2% per month. It assumes that S&P 500 is at a level of 1100, which would mark a substantial jump from its trough early in 2009. Scenario 2 assumes that nominal house prices decline an additional 9.7% from their March 2009 level to September. It assumes that the S&P is at 1000. Scenario 3 is designed as a pessimistic scenario. It assumes that house prices fall 14.1% from their March 2009 level and that the S&P 500 is at 850.

In all three scenarios, it is assumed that households save out of dis- posable income at the average personal saving rate for 2008. All three

120 D. Rosnick and D. Baker

TABLE 1 Assumptions Used in Asset Projections, 2007–2009

CPI-U-RS Case-Shiller

20-City Index S&P 500

September 2007 306.2 195.69 1,473.96 Latesta 313.2 (+2.3%) 139.99 (−28.5%) 872.74 (−40.8%) September 2009 - Scenario 1 314.6 (+2.8%) 139.99 (−28.5%) 1, 100 (−25.4%) - Scenario 2 314.6 (+2.8%) 127.85 (−34.7%) 1, 000 (−32.2%) - Scenario 3 314.6 (+2.8%) 120.26 (−38.5%) 850 (−42.3%) Note. aTable derived from Case-Shiller as of March, CPI-U-RS as of April, and S&P 500 (open of month) as of May 2009.

projections also assume that interest-bearing assets earned the average yield on 10-year treasury bonds for 2008. Holdings in retirement accounts were assumed to be divided proportionately between stocks, which are assumed to track the S&P 500, and interest-bearing assets.

The tables do not give a full accounting of retirement wealth because they exclude defined-benefit (DB) pensions. This exclusion will matter less for the baby boomer cohorts than it did in prior years; however, a substantial portion of these families still stand to receive at least some payments from DB pensions. Among the 55 to 64 age group, 42.4% will get at least some money from a DB pension. In the middle quintile of this age group, 48.0% of families will receive some income from a DB pension. Among families in the 45 to 54 age group, 34.7% will receive some income from a DB pension. For families in the middle wealth quintile, this figure is 40.0%.

It is unlikely that the inclusion of wealth in DB plans would substantially change this analysis. At the end of 2008, private-sector DB plans had cumu- lative assets of $1,930.5 billion, covering 44 million workers and retirees. This is an average of just under $44,000 per worker.3 The holdings are heav- ily skewed and disproportionately go to older workers and retirees. As a result, the net acquisition of financial assets by these funds has been strongly negative for more than a decade. In 2008, net acquisition of assets was – $91.4 billion, nearly 5% of the assets held in DB accounts.4 Given the sharp rate of decline in private DB pensions, most of the workers in these age groups who have DB plans are not likely to see substantial benefits from them, certainly far less than the $44,000 average holdings. It is worth noting that those with public-sector plans are likely to fare somewhat better.

FINDINGS

Tables 2 and 3 show the median and mean incomes and net worth of people aged 45 to 54 and people aged 55 to 64, respectively, in each of the seven SCFs. The tables also show home equity and the value of holdings in

Impact of the Housing Crash 121

TABLE 2 Family Finances (in Thousands of 2009 Dollars) Age 45 to 54 in Year

Income Net worth Home equity Retirement accounts

Year Median Mean Median Mean Median Mean Median Mean

1989 63.8 102.2 161.5 452.5 71.1 129.1 1.7 47.5 1992 64.0 87.1 126.5 413.7 55.0 97.2 2.7 56.0 1995 56.9 93.9 137.9 420.9 55.6 88.4 7.8 73.4 1998 67.7 92.3 143.1 474.0 54.9 91.0 13.2 70.6 2001 67.9 120.0 168.8 607.0 60.1 118.2 10.2 101.1 2004 70.7 109.7 169.9 626.3 68.8 144.9 25.7 94.5 2007 68.7 121.0 215.1 712.2 88.4 173.4 18.0 110.1

Note. Source: FRB 1989–2007 SCF and author’s calculations.

TABLE 3 Family Finances (in Thousands of 2009 Dollars) Age 55 to 64 in Year

Income Net worth Home equity Retirement accounts

Year Median Mean Median Mean Median Mean Median Mean

1989 46.6 82.0 163.0 523.9 82.7 131.1 0.0 51.7 1992 48.8 78.6 189.1 546.4 74.3 120.6 7.6 61.4 1995 49.8 80.2 180.4 563.7 76.4 115.9 4.2 68.7 1998 53.1 98.2 193.7 725.7 78.5 126.2 9.2 115.9 2001 55.7 113.5 234.8 937.9 91.1 159.1 9.6 150.7 2004 67.2 117.2 310.7 978.6 118.5 225.4 27.1 158.0 2007 59.2 118.3 267.9 991.3 107.9 195.9 23.6 173.5

Note. Source: FRB 1989–2007 SCF and author’s calculations.

defined contribution retirement accounts. The trend in net worth for these age groups is quite different over this period. In 1989, the SCF data indicate that the median family of those between the ages of 45 and 54 had a net worth of $161,500, only 0.9% less than the $163,000 net worth of families of people between the ages of 55 and 64. However, the net worth of the younger age group fell in subsequent surveys, only recovering the 1989 level in 2001. The net worth of the older age group rose in each survey, peak- ing at $310,700 in 2004. Differing patterns in home equity can explain part of the difference, although a fuller explanation would require a more com- plete examination of the changing composition of wealth and also family structures during this period.

In 2007, the median net worth for the younger age group jumped by 26.6% from its 2004 level to $215,100, as shown in Table 3. By contrast, the median net worth for families of people between the ages of 55 and 64 fell by 13.8% from its 2004 level to $267,900. The reason for the divergence is a sharp reported decline in home equity for the older age group, while the

122 D. Rosnick and D. Baker

younger age group had substantially more home equity in 2007 than 2004. Again, the reasons for these changes are beyond the scope of this analysis, but it is important to remember that the data in the tables are showing us different families at the same age, not tracking the same families through time.

Before examining the projections in the three scenarios for 2009, it is worth briefly commenting on the SCF data for 2007. These calculations show the net worth for families very close to or past retirement, and the vast majority of them can anticipate no significant income from a DB pension. In 2007, according to the National Association of Realtors’ existing homes sales data, the median house price was $217,000. This means that the net worth of families of people between the ages of 45 and 54 in that year would have been approximately sufficient to pay off the mortgage on the median house with nothing left over. The median net worth of families of those between the ages of 55 and 64 would have been sufficient to pay off this mortgage, with approximately $50,000 left over.

Table 4 shows the mean and median net worth, along with quintile aver- ages for 2007, and each of the three scenarios projected for 2009, for families of people between the ages of 45 and 54 in 2009. Table 5 shows the same information for families of people between the ages of 55 and 64. There are sharp declines in all three scenarios for families in both age groups, although the projected declines in net worth are much sharper over this 2-year period for the younger age cohort.

In scenario 2, the median family of people between the ages of 45 and 54 is projected to have net worth of $88,000 in 2009, a drop of 59.1% from the median net worth of families in this age group in 2007. In the

TABLE 4 Net Worth: Median, Mean, and by 2007 Quintile of Net Worth (Ages 45 to 54)

Year Median Mean Bottom Second Middle Fourth Top

2007 $215.1 $712.2 $6.3 $77.6 $210.7 $445.1 $2, 823.9 Scenario 1 101.8 473.9 −0.8 26.4 107.3 266.1 1, 972.6 Scenario 2 88.0 444.9 −1.6 19.7 93.4 240.9 1, 874.2 Scenario 3 81.3 419.4 −2.1 15.0 83.4 221.4 1, 781.5 Note. Source: FRB 2007 SCF and author’s calculations.

TABLE 5 Net Worth: Median, Mean, and by 2007 Quintile of Net Worth (Ages 55 to 64)

Year Median Mean Bottom Second Middle Fourth Top

2007 $267.9 $991.3 $6.7 $111.8 $278.5 $591.7 $3, 985.2 Scenario 1 190.3 779.0 3.3 65.3 202.0 429.7 3, 200.5 Scenario 2 168.7 737.2 1.1 55.0 185.7 397.0 3, 052.7 Scenario 3 157.9 697.9 −0.5 47.9 173.1 370.1 2, 904.1 Note. Source: FRB 2007 SCF and author’s calculations.

Impact of the Housing Crash 123

more optimistic scenario 1, the median net worth is $101,800, while it is just $81,300 in the pessimistic scenario 3. In scenario 2, the average net worth projected for families in the bottom quintile in 2009 is negative. For families in the second wealth quintile, net worth is projected to be $19,700. In scenario 2, families in the fourth wealth quintile are projected to, on average, have net worth of $240,900 in 2009, approximately 10% above the median for this age group in 2007.

Table 5 shows that in scenario 2, the median family of people between the ages of 55 and 64 is projected to have $168,700 in net worth in 2009. Even in the more optimistic scenario 1, they are projected to have just $190,300 in net worth, a drop of 29.0% from the median net worth of families in this age group in 2007. In scenario 2, the average net worth projected for families in the second wealth quintile in the 55 to 64 age group is $55,000. For families in the fourth quintile the projected net worth in the scenario 2 is $397,000.

The most obvious explanation for the greater projected decline in net worth for the younger age group between 2007 and 2009 is a sharper pro- jected decline in home equity. Table 6 compares the median and mean holdings in quasi-liquid retirement accounts and home equity in 2007 and the projections for 2009 for families of those between the ages of 45 and 54. Table 7 gives the same information for families of people between the ages of 55 and 64. As can be seen, families in the younger age grouping saw a very sharp decline in both median and mean home equity over this 2-year period. Median home equity is projected to fall in this scenario from $88,400 to $14,800, while mean home equity is projected to fall from $173,400 to

TABLE 6 Balances of “Quasi-Liquid” Retirement Accounts and Home Equity (Ages 45 to 54; in Thousands of 2009 Dollars)

Retirement accounts Home equity

Year Median Mean Median Mean <6% of home price

2007 $25.7 $110.1 $88.4 $173.4 2.4% 2009 scenario 1 18.0 82.0 14.8 74.5 16.9

Note. Source: FRB 2007 SCF and author’s calculations.

TABLE 7 Balances of “Quasi-Liquid” Retirement Accounts and Home Equity (Ages 55 to 64; in Thousands of 2009 Dollars)

Retirement accounts Home equity

Year Median Mean Median Mean <6% of home price

2007 $23.6 $173.5 $107.9 $195.9 1.4% 2009 scenario 1 29.8 149.4 50.4 111.2 13.6

Note. Source: FRB 2007 SCF and author’s calculations.

124 D. Rosnick and D. Baker

$74,500, drops of 83.3% and 67.0%, respectively. The projected decline in the median holding in retirement accounts, from $25,700 to $18,000, is both smaller in percentage terms and far smaller absolutely.

By contrast, the median home equity for families in the 55 to 64 age group is projected to fall from $107,900 to $50,400, a drop of 53.3%. Mean home equity is projected to fall from $195,900 to $111,200, a drop of 43.2%. The steeper decline in home equity among the younger age group can be readily explained by greater leverage. Younger households would typically be less far along in paying off mortgages. This means that they are much more heavily leveraged in their homes. As a result, the same percentage increase or decrease in the house price will typically lead to a much larger change in equity for a younger family than an older family. During the years of the run-up in the housing bubble, the greater leverage allowed younger families to accrue equity at a faster pace. However, when house prices began to fall in 2006, the decline had a proportionately greater impact on the equity of younger families than older families.

The plunge in house prices in the last 3 years is leading to an extraor- dinary situation in which a very high percentage of near-retirement cohorts will lack any equity in their homes. In addition to the large number of homeowners who are underwater, there are also a substantial number of homeowners who have some equity, but not enough to cover the normal costs associated with selling their homes. This means that if they were to sell their homes to move into a smaller house or rental unit, they would have no money from their homes to help with this move and in fact would actually have to put up some amount of money at the closing to pay off the mortgage, realtors’ fees, and other costs associated with selling their homes.

Tables 8 and 9 show, for the 45 to 54 age cohort and the 55 to 64 age cohort, respectively, the share of homeowners in each wealth quintile who had less than 6% equity in their homes in each of the SCF surveys since 1989. It also shows the projected share for 2009 in scenario 2. These are, in effect, the homeowners who will have to pay money to sell their homes.

TABLE 8 Homeowners with <6% Equity in their Homes (Ages 45 to 54)

Year All Bottom Second Middle Fourth Top

1989 0.6% 0.0% 0.0% 0.0% 0.9% 1.2% 1992 2.4 16.5 5.3 0.0 0.6 0.3 1995 3.6 25.8 5.6 2.1 1.6 0.9 1998 4.7 35.8 6.3 6.2 0.9 1.1 2001 3.7 27.5 7.9 1.1 1.2 1.5 2004 3.3 27.1 6.8 1.9 0.1 0.5 2007 3.1 23.5 5.8 1.0 1.5 1.0 Scenario 2 32.4 90.2 56.9 37.1 22.0 13.2

Note. Source: FRB 2007 SCF and author’s calculations.

Impact of the Housing Crash 125

TABLE 9 Homeowners with <6% Equity in their Homes (Ages 55 to 64)

Year All Bottom Second Middle Fourth Top

1989 0.5% 0.5% 2.3% 0.0% 0.9% 0.1% 1992 0.1 1.3 0.0 0.0 0.0 0.0 1995 1.5 8.3 3.5 0.0 0.0 0.2 1998 2.0 22.6 2.4 0.0 0.0 1.1 2001 1.1 11.5 1.0 0.0 0.0 0.0 2004 1.4 15.8 2.5 0.0 0.0 0.1 2007 1.7 13.0 1.8 0.3 1.3 0.6 Scenario 2 19.5 64.0 29.9 15.3 14.0 7.1

Note. Source: FRB 2007 SCF and author’s calculations.

The assumption of 6% sales cost is almost certainly on the low side, since this is the standard realtors’ fee. There are often other costs associated with selling a home, and many state and local governments impose transfer fees that are usually split between buyers and sellers.

As can be seen, historically very few homeowners in this age group fell into this category. This is due to three factors:

1. In prior decades, homeowners typically had to pay a substantial down payment when they bought their homes, so they would generally start their period of homeownership with more than 6% equity in their home.

2. By the time homeowners reached these age groupings, they would have typically been in their homes for some period of time and therefore would have paid down much of the principle on their mortgages.

3. The price of homes typically rose, roughly in step with inflation.

However, the current group of near retirees is in a very different situa- tion. Many of them bought homes with little or no money down. In many cases they were encouraged to move during the housing boom earlier in the decade, so they have not had years of paying down the principle on mort- gages, even if they may have been homeowners for decades. In addition, many families who had accumulated equity borrowed against it during the peak years of the housing bubble. Finally, the sharp plunge in house prices eliminated much of the equity that these families had accumulated during the boom years.

As a result, 32.4% of the homeowners between the ages of 45 and 54 will have to bring cash to their closing based on their projected equity in scenario 2. In the older age group, 19.5% will be forced to bring cash to their closing. By comparison, fewer than 4.0% of homeowners in the younger age grouping would have needed to bring cash to a closing based on the prior SCF surveys, except in 1998, when 4.7% would have needed to bring cash to a closing. For the older age group, 1998 was also the previous peak year,

126 D. Rosnick and D. Baker

with 2.0% of homeowners between the ages of 55 and 64 having to bring cash to their closing.

While the lower wealth quintiles will disproportionately be in the sit- uation of needing money to sell their homes, in scenario 2, 56. 9% of homeowners in the middle wealth quintile will have to bring cash to a closing. Even among families in the fourth wealth quintile, 22.0% will have to bring cash to a closing. For the older age grouping, 15.3% of homeowners in the middle wealth quintile and 14.0% of homeowners in the fourth wealth quintile will have to bring cash to a closing.

DISCUSSION

The projections for 2009 in this study suggest that the baby boom cohorts will be very poorly prepared for retirement. While there have been grounds for some time for concerns about the adequacy of their retirement savings, the prospects have become much more dire following the collapse of the housing bubble and the resulting plunge in the stock market.

While there is some prospect that both markets will recover some of their lost ground, it is certainly not reasonable to assume that there will be anything resembling a complete recovery in the years before the heads of these families reach retirement age. Certainly, the markets are not anticipat- ing a full recovery or this would already be reflected in asset prices. While some of these families have already retired or are at the edge of retirement, the younger age groups in this analysis will still have a substantial period of time remaining in the labor force. This can allow them to raise their savings considerably beyond the levels shown in these projections.

However, the recent economic projections show the economy remain- ing below its capacity until 2014. The implied weakness in the labor market will make it more difficult for workers to find well-paying jobs and, possibly, to find work at all. As a result, many baby boomers will find it unusually difficult to build up their savings over the next few years. Given the limited time that they have left in the labor force, the projected weakness in the labor market over the next 4 years could have a substantial impact on their retirement security.

With the vast majority of families in these age groups expecting little or no money from traditional DB pensions and having accumulated little money in defined contribution accounts, families in these age groups are likely to be hugely dependent on Social Security as a source of retirement income. Policy makers will have to take this into account when considering proposals to restructure the program. For a least a decade into the future, new retirees are likely to be more dependent on Social Security than otherwise would have been expected because of the distortions created by the housing bubble.

Impact of the Housing Crash 127

Finally, the loss of wealth due to the plunge in house prices has created an extraordinary situation for the homeowners among these age groups. In a very high percentage of cases, instead of being assets, their homes are likely to prove to be financial albatrosses. If they decide to move into homes that may be more appropriate for their retirement, they will find that they will have to draw down wealth from other assets to cover the costs. The lack of home equity is likely to make these age cohorts less mobile in retirement than ordinarily would be expected from families in these age groups.

CONCLUSION

This study presented a set of wealth projections for the baby boom cohorts based on the 2007 SCF and subsequent movements in stock and housing prices. The projections show that the baby boomers’ wealth fell sharply as a result of the bursting of the housing bubble and the subsequent plunge in house prices. As a result, the baby boom cohorts are much less well- prepared for retirement than the cohorts that preceded them. Most of the families in these age groups are likely to be heavily dependent on Social Security in their retirement.

NOTES

1. See http://www.federalreserve.gov/pubs/oss/oss2/bulletin.macro.txt. 2. For example, the FRB draws the line between the bottom and second quintiles of net worth

by sorting all observations (including repetitions) by net worth and including in the second quintile the first observation that would, if included in the bottom quintile, result in a total survey weight of at least 20%. Thus, observations with identical levels of net worth may be placed in different quintiles. We define quintiles of net worth as per Stata, where the all observations of a given net worth are added as blocks to the bottom quintile until the total survey weight is at least 20%. Thus, observations with identical levels of net worth will be in the same quintiles of net worth, but the total weight in each quintile may be unequal.

3. Data on the assets of private defined benefit plans come from the FRB’s Flow of Funds, Table L.188b., Line 1, available at http://www.federalreserve.gov/releases/z1/Current/z1r-6.pdf. The number of people enrolled in these plans is taken from the Pension Benefit Guarantee Corporation’s Web site at http://www.pbgc.gov/.

4. Flow of Funds Table F.118.b, Line 1.

REFERENCES

Baker, D., & Rosnick, D. (2008). The housing crash and the retirement prospects of late baby boomers. Washington, DC: Center for Economic and Policy Research.

Engen, E., Gale, W. & Uccello, C. (1999). The adequacy of retirement savings. Brookings Papers on Economic Activity, 30, 65–188.

128 D. Rosnick and D. Baker

Federal Reserve Board. (2007). 2007 Survey of Consumer Finances: Definitions of variables given by SAS program. Federal Reserve Board. Retrieved from http://www.federalreserve.gov/pubs/oss/oss2/bulletin.macro.txt.

Munnell, A. H., Golub-Sass, F., Soto, M., & Webb, A. (2008). Do households have a good sense of their retirement preparedness? Issue in Brief, 8–11. Chestnut Hill, MA: Center for Retirement Research at Boston College.

Munnell, A. H., & Soto, M. (2008). The housing bubble and retirement security. Working Paper, 2008–2013. Chestnut Hill, MA: Center for Retirement Research at Boston College.

Rosnick, D., & Baker, D. (2009). The wealth of the baby boom cohorts after the collapse of the housing bubble. Washington, DC: Center for Economic and Policy Research.

Weller, C., & Wolff, E. N. (2005). Retirement income: The crucial role of social security. Washington, DC: Economic Policy Institute.

Wolff, E. N. (2002). Retirement insecurity: The income shortfalls awaiting the soon- to-retire. Washington, DC: Economic Policy Institute.

Copyright of Journal of Aging & Social Policy is the property of Taylor & Francis Ltd and its content may not

be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written

permission. However, users may print, download, or email articles for individual use.