Center for Retirement Research https://crr.bc.edu Mon, 18 Mar 2024 13:24:41 +0000 en-US hourly 1 https://wordpress.org/?v=6.0.7 Homeownership in Retirement: an Asset or a Burden? https://crr.bc.edu/homeownership-in-retirement-an-asset-or-a-burden/ https://crr.bc.edu/homeownership-in-retirement-an-asset-or-a-burden/#comments Thu, 14 Mar 2024 13:35:08 +0000 https://crr.bc.edu/?p=46918 The conventional wisdom for older workers heading into retirement is that owning a house is a good thing. The property has no doubt appreciated over many years, adding to their wealth. 

But new research finds that homeowners often strain to a pay a mortgage that is larger than what they can realistically afford. The essence of the problem for retirees, who usually rely on savings to help with living expenses, is that many do not have enough to make their monthly payments comfortably.

This problem has become more pressing over the years. Half of the retired homeowners who were born in the early years of the baby boom wave are still making mortgage payments. They are in a very different situation than their parents’ generation when the majority of retirees owned their homes free and clear.

Buying a house has also become increasingly expensive, and the size of the typical retired household’s mortgage has more than doubled compared with their parents, to about $108,500 in inflation-adjusted dollars.

Retired Black and Hispanic homeowners have even less to fall back on. Their financial assets are under $20,000, compared with $134,000 for Whites.

Although White homeowners tend to be in better financial shape, being more heavily leveraged is also an issue for them. The researchers found that households that don’t have a mortgage have the most financial assets, while those with the most housing debt – because, in many cases, they bought a relatively expensive house – have the least in savings.

Older people struggling to pay a mortgage face difficult choices. One option is to delay retirement, and the most leveraged households retire a full year later than those that don’t have a mortgage payment. “High housing leverage,” the researchers said, “may induce later retirement.”

To make ends meet, the largest mortgage holders also spend less money on non-housing expenses, whether food or travel. Retirees naturally spend less as they age. But consumption spending between ages 65 and 81, for example, falls by 39 percent if the household has a larger mortgage than is typical. This compares with a 28 percent drop for homeowners with relatively small mortgages.

While a house is an asset, it is not a liquid asset. Retirees who can no longer afford a mortgage payment may be forced to sell to cash in on the equity and relieve their financial stress. “Unsustainable mortgage borrowing,” the researchers said, is an even bigger issue for Black and Hispanic retirees.

It turns out that homeownership has financial value only for some – the most privileged retirees, who are mostly White. Using a model that simulates financial decisions, the researchers find that well-off retirees can benefit from selling the house and cashing in on the equity to fund nursing home care or an assisted living facility, which can cost $55,000, $100,000, or more a year.

For them, a house is a form of “informal long-term care insurance,” they concluded.

To read this study by Leora Friedberg, Wei Sun and Anthony Webb, see “The House: is it an Asset or a Liability?”

The research reported herein was derived in whole or in part from research activities performed pursuant to a grant from the U.S. Social Security Administration (SSA) funded as part of the Retirement and Disability Research Consortium.  The opinions and conclusions expressed are solely those of the authors and do not represent the opinions or policy of SSA, any agency of the federal government, or Boston College.  Neither the United States Government nor any agency thereof, nor any of their employees, make any warranty, express or implied, or assumes any legal liability or responsibility for the accuracy, completeness, or usefulness of the contents of this report.  Reference herein to any specific commercial product, process or service by trade name, trademark, manufacturer, or otherwise does not necessarily constitute or imply endorsement, recommendation or favoring by the United States Government or any agency thereof.

]]>
https://crr.bc.edu/homeownership-in-retirement-an-asset-or-a-burden/feed/ 4
Can a ‘Nudge’ Reduce Credit Card Debt? https://crr.bc.edu/can-a-nudge-reduce-credit-card-debt/ https://crr.bc.edu/can-a-nudge-reduce-credit-card-debt/#respond Wed, 13 Mar 2024 14:05:31 +0000 https://crr.bc.edu/?p=46976 New experiment finds progress more elusive than expected.

Our recent study about debt holdings of older Americans focused our attention on how credit card debt could get people in trouble.  We, like other observers, bemoaned people’s tendency to pay the minimum required amount, and, like other observers, blamed these minimum payments on the prominence of this option on the credit card company’s billing statement.  If the full amount due came first and the minimum required payment appeared in a secondary slot, we were sure that many more people would increase their payments.  In short, the credit card debt problem seemed like one that could be easily remedied by a “nudge.” 

Wrong.  A recent paper by David Laibson and many co-authors documents how a nudge that reduces the anchoring of credit card payments to the minimum failed to reduce credit card debt.  Let me briefly describe the experiment, the details of the results, and the authors’ explanations for why the initiative failed. 

When customers in the UK open a new credit card online, they are presented with the option to sign up for a version of Autopay – a Fin Tech feature that automatically transfers money from the cardholder’s bank account to the credit card company.  If they decide to opt-in, they are faced with three payment options: 1) Autopay Min – minimum amount; 2) Autopay Fix – the higher of the minimum or a fixed dollar amount; and 3) Autopay Full – full amount due.  Cardholders enrolled in Autopay can also make manual non-Autopay payments either by phone or online.   

The experiment involved removing Autopay Min as a visible anchor when individuals open their card.  Of course, Min remains an option operationally if participants choose a very low Autopay Fix amount.  The outcomes for individuals in this experimental group were then compared to those of the control group whose options included Autopay Min.  The notion is that removing Autopay Min increases the salience of Autopay Fix, which was expected to increase automatic payments and thereby reduce debt and interest costs. 

The headline results were dramatic.  The de-anchoring reduced Autopay Min enrollment from 36.9 percent to 9.6 percent.  (Participants in the experimental group were no more likely to pay the full balance than those in the control group.)  Importantly, however, after seven payment cycles, the researchers found no difference, on average, in credit card debt, spending, total payments, or borrowing costs.  Thus, the nudge was ultimately ineffective. 

Why didn’t the nudge work?  The researchers offer three reasons.  Perhaps the most important is that nudged cardholders set up Autopay Fix amounts that were only slightly higher than the minimum amount due.  In fact, in the long run they were no higher because the minimum rises automatically over time as card balances increase.  Second, cardholders in the experimental group were less likely to sign up for Autopay than those in the control group, which resulted in more missed payments.  Third, those cardholders in the experimental group who did enroll in Autopay made lower supplemental manual payments.  All these offsetting behaviors reflect the fact that many participants had very limited liquid cash balances. 

The fact that nudges are not the answer to staggering amounts of credit card debt is both discouraging and intuitive – upon reflection.  In our earlier study, the largest group (33%) of at risk debtholders consists of “financially constrained” households, which have low levels of wealth, are often overleveraged, and struggle with the essentials.  This group is borrowing just to get by.  They do not have the cash to pay for new tires for their car so they can get to work. They have to use a credit card to cover an emergency expenditure.  And, given their financial constraints, they cannot pay more than the minimum.  They may be charged 25 percent or more on the unpaid balance, and accumulate enormous interest costs over many years.  

This seems like a terrible arrangement.  People need some way to cope with emergencies without being socked with sky-high borrowing costs and no way to get out.  Solving the problem – at least for vulnerable groups – is going to require much more than a nudge.

]]>
https://crr.bc.edu/can-a-nudge-reduce-credit-card-debt/feed/ 0
Caregivers Share their Stresses and Joys https://crr.bc.edu/caregivers-share-their-stresses-and-joys/ https://crr.bc.edu/caregivers-share-their-stresses-and-joys/#respond Tue, 12 Mar 2024 13:16:30 +0000 https://crr.bc.edu/?p=46859 Jacquelyn had finally snared her dream job as an assistant to a television writer in New York City. Riding the subway one Saturday night, she got a call that changed all that.

Her mother, who was living with her grandmother, was in trouble back home. Jacquelyn returned to find rotting food in the refrigerator and a house on the brink of foreclosure for past due mortgage payments. Her mother had Alzheimer’s disease.

Jacquelyn said she had to quit her job and never returned to New York. “Caregiving requires a restructuring of who you are,” she said.

Undervalued, stressed, exhausted, guilty and even resentful – these are some of the feelings that unpaid caregivers, who are mostly women, experience on the emotional roller coaster they ride every day. In a survey sponsored by the pharmaceutical company Otsuka, two-thirds of caregivers say it is the most stressful job they’ve ever had. 

The joy came, Jacquelyn said, when her mother would dance in public when she heard music. At first, she was embarrassed but after awhile that no longer mattered.

She is one of three women who shared their fears, happy moments, and the mental health challenges of caring for a parent in a set of videos, also by Otsuka, which has developed a drug for treating agitation from Alzheimer’s.   

Our readers who are caregivers might find comfort in hearing from others about what they’ve gone through. (Warning: watching the videos requires sitting through a few seconds of annoying ads before being allowed to skip over them.) 

One undercurrent in their stories is that caregiving is a lonely job. Even when others are supporting them, the person with the primary caregiving responsibility carries most of the weight.

Jessica, who is in her 30s and constantly under stress taking care of her mother, came to appreciate the help she received from the women at her church and the people she met on social media who understood the stress she was under.

She also realized she needed to take time for herself. To be able to help her mother, she said, “you have to have a life.”

Finding this time has been very difficult for Bailey, who cares for the three people who live with her and depend on her – her daughter, mother and father. “I leave myself hanging to make sure everyone else is okay,” she said.

That sounds like a mantra other caregivers can relate to.

Squared Away writer Kim Blanton invites you to follow us @SquaredAwayBC on X, formerly known as Twitter. To stay current on our blog, join our free email list. You’ll receive just one email each week – with links to the two new posts for that week – when you sign up here.  This blog is supported by the Center for Retirement Research at Boston College.

]]>
https://crr.bc.edu/caregivers-share-their-stresses-and-joys/feed/ 0
Can Employer Demand Support Older Workers Today…And Tomorrow? https://crr.bc.edu/can-employer-demand-support-older-workers-today-and-tomorrow/ https://crr.bc.edu/can-employer-demand-support-older-workers-today-and-tomorrow/#respond Tue, 12 Mar 2024 13:03:04 +0000 https://crr.bc.edu/?p=46738

The brief’s key findings are:

  • Many older workers are inclined to work longer, but will employers hire and retain them – today and in the future?
  • A series of CRR studies on this topic provide a case for tempered optimism.
  • First, hard data suggest that older workers are at least as productive as younger ones, though they do cost more.
  • Second, survey data show that employers’ views are largely in line with these hard data, and job postings confirm a willingness to hire.
  • Finally, while the jobs that older workers do today may be less prevalent in the future, jobs that they have the skills for should be available.

Introduction 

A common refrain among retirement researchers is that longer careers are the best way to ensure an adequate retirement.  This refrain is often directed at the workers themselves – “you need to work longer!”  And that message seems to have been getting through.  Since the 1990s, the labor force participation rates of older individuals have increased.  But, workers are just one side of the market: the supply side.  Their ability to work longer also depends on whether employers are willing to hire and retain them.  The question is, what does employer demand for older workers look like today and in the future?  

To answer the question, this brief synthesizes the results of several recent Center studies.  These studies examine employer demand from three different angles: 1) the reality of older workers’ value today; 2) employers’ perceptions of that productivity; and 3) whether older workers’ skills will be a good fit for the jobs of the future.  

The discussion proceeds as follows.  The first section focuses on how today’s workers affect the bottom line in terms of productivity and profitability.  The second section considers how employers perceive the value of older workers using two measures: what employers say to a survey-taker and what they do with respect to postings on a job board.  The third section looks ahead to assess how well older workers’ abilities match projections of employer demand in 2030.  

Overall, room for tempered optimism exists.  First, older workers may be just as good as younger workers for a firm’s bottom line.  Second, employer perceptions appear mixed – they say older workers are at least as productive but relatively expensive, which may explain why their job listings specifically targeting older workers are mainly for lower paying positions with limited benefits.  Third, while the jobs older workers do today may be less prevalent in the future, other jobs that older workers have the capacity to do should be plentiful.

Are Today’s Older Workers Good for Business?

In assessing labor demand for any group, the hope is that workers’ contributions to firm profitability are the primary consideration.  But, while measuring these contributions may sound straightforward, it is not.  The major challenge is the availability of data that contain both information on workers’ characteristics – like age – and objective measures of employer profitability.

Fortunately, Center researchers obtained access to restricted Census data and were able to combine three databases: 1) the Longitudinal Employer-Household Dynamics; 2) the Longitudinal Business Database; and 3) the Census’ Business Register.1  These datasets contain information on employee characteristics and earnings, payroll and revenue, and location and industry.  So, the data allow researchers to track businesses and establishments over time, while observing their revenues, payroll, and the age composition of their workforces.

With these data in hand, the study estimated how two measures of firm performance would change if younger workers were replaced by older ones.  The first measure was worker productivity, defined as revenue divided by the number of employees.  The second measure was profitability, defined as revenue per dollar of payroll.  The effect of exchanging younger workers for older ones was estimated using regression analysis to compare firms with workforces that are otherwise similar in racial, ethnic, and educational makeups, as well as geographic location.  This estimation was done in two ways: 1) across many industries in a manner that yielded non-causal (i.e., correlational) results; and 2) in manufacturing only, exploiting special features of this industry to obtain causal results.

Table 1 shows the correlational results.  On the productivity side, no clear evidence supports the notion that older workers are less productive.  Three of the five significant results are positive, with management, manufacturing, and retail all showing productivity gains from older workers.  Just two industries – in particular finance – show a significant reduction in productivity.  The other results are not significantly different from zero. 

Table showing the estimated effect of increasing the share of workers ages 55 and over on productivity and profitability

On profitability, the picture is more lopsided, with the estimates generally indicating that a larger share of older workers is associated with lower profits, which is consistent with a large body of research showing that wage growth for older workers continues even after their productivity flattens out.  Still, in the majority of industries the results were not significantly different from zero, indicating no clear difference in profitability between older and younger workers.  And, the more sophisticated estimation method – which sought to obtain causal results – found no evidence of lower profitability in manufacturing for firms with older workforces.  

So, while estimates vary by industry, the evidence suggests that older workers are at least as productive as younger ones.  And, while older workers’ higher costs may eat into profitability in some industries, in most cases older and younger workers are indistinguishable on this front.  Indeed, the highest-quality evidence, relevant to the limited but important manufacturing industry, finds no evidence of reduced profitability from older workers.  However, if these findings are to be reflected in actual demand for older workers, then employers must have a perception that matches this reality.  

How Do Today’s Employers Perceive Older Workers?

To understand current employer perceptions of older workers, the Center used two approaches.  The first was to simply ask employers through a survey.  The second was to explore what employers actually do by analyzing job postings.

What Do Today’s Employers Say about Older Workers?

In 2019 – before the pandemic – the Center commissioned a telephone survey by Greenwald and Associates.2  The survey queried employers on their views of workers’ productivity and costs – two issues that came up in the study above.  Figure 1 shows how employers viewed the productivity and costs of workers ages 55 and over versus those under 55, separately for professional and support staff.

Bar graph showing Employer Evaluations of the Relative Productivity and Cost of Older Workers

The results indicated that employers’ stated views of older workers’ relative productivity and costs are roughly consistent with the objective measures of workers’ value discussed earlier.  On the productivity front, very few employers view older workers as less productive.  The majority say that older workers are equally productive, with a large fraction seeing them as more productive.  On the cost side, the news is less positive.  While the majority of employers see older workers as equally costly, a sizable minority see them as more costly than younger ones.3

OK, so the majority of today’s employers say that older workers are at least as productive as younger ones, and many also view them as no costlier.  But, do employers act this way, recruiting older workers for opportunities on par with younger ones?

Do Today’s Employers Seek Out Older Workers?

To explore whether employers actively recruit older workers, the researchers turned to RetirementJobs.com, the only major job posting site targeted to individuals ages 50 and over.  Specifically, the Center researchers were given access to the website’s postings as of November 2019, which captures the picture before the disruptions of the pandemic era.4  Although RetirementJobs.com is considerably smaller than Monster.com or Indeed.com, it is the only one of these websites that can provide data on jobs targeted to older workers.  

The analysis divided the jobs on the website into two types.  First, it used openings directly posted to RetirementJobs.com, which represented 20 percent of all listings.  These “direct” postings capture employers aiming explicitly at older workers.  The second type of postings are those fed to the site from CareerBuilder.com.  These “indirect” postings suggest employer willingness to hire older, as well as younger, workers.  Any difference between direct versus indirect postings would provide some insight into the types of jobs that target older workers specifically as compared to those that are not intended solely for a specific age range.   

Center researchers next turned to a comparison between RetirementJobs.com and one of the largest general job boards in the country.  Since this general job board contains millions of listings, a random sample of 15 listings from each state was selected for a total of 765 listings.5  The analysis compares these jobs to both all jobs posted on RetirementJobs.com and the jobs directly targeted at older workers.

When focusing on all jobs posted on RetirementJobs.com, the results contain some positive news for older job seekers, with an important caveat (see Table 2).  The first positive point is that the salaries for both the part- and full-time jobs on RetirementJobs.com are significantly higher than those on the general jobs board.  Another piece of good news is that the jobs are more likely to be full-time positions.  The main caveat is that these jobs seem less likely to mention either health or retirement benefits.  So, older workers may have good opportunities for bridge jobs to retirement, but fewer chances to obtain the full benefits often associated with “career” jobs.

Table showing a comparison of job postings between retirementjobs.com (all and direct) and a general jobs board

When restricting to jobs posted directly to RetirementJobs.com, however, the positive takeaway of higher salaries does not hold.  The direct postings offer significantly lower wages than the general jobs board.  And, directly posted jobs have less full-time work, more temporary work, and fewer benefits.  So, the job postings most specifically targeting older workers seem to be of lower quality than the postings that are not age specific, either those appearing on RetirementJobs.com or the general jobs board.

Summing Up Demand for Older Workers Today

Based on the findings of the three Center studies discussed above, the situation seems relatively positive for older workers today.  In most industries, they are at least as productive as younger ones.  And, while their higher wages eat into profitability, in many industries older and younger workers are indistinguishable with respect to this metric.  In a survey, today’s employers indicate that they recognize these objective measures, viewing older workers to be at least as productive as younger ones, albeit sometimes with higher costs.  And, to an extent, employers appear to act this way.  They post relatively high-salary jobs to a website targeting older workers, although jobs that most directly aim for these workers pay less and have fewer benefits.  All-in-all, things seem OK today.  But, what about tomorrow?

Will Demand for Older Workers Hold up Tomorrow? 

To assess whether older workers are likely to have good access to jobs in 2030, Center researchers tried two different approaches.6  The first one was simply to look at the jobs older workers are doing today and compare them to Bureau of Labor Statistics’ projections on the level of those jobs in 2030.  This analysis answers the question: are older workers currently doing jobs that are expected to be plentiful at the end of the decade?

The second approach addressed a slightly different question: are older workers able to do the jobs that will be plentiful in 2030, even if they are not doing them now?  This approach required the creation of a “Suitability Index” that measured how well older workers are likely to do certain jobs.  The Index is composed of three indicators for each occupation: 1) the extent to which the skills needed erode with age; 2) the disability application rates; and 3) the average retirement age.  The Index provides a convenient summary measure of which occupations are most congenial to older workers, which can then be compared to the growth rates of various occupations projected to be available in 2030.7

The first approach yields a discouraging result.  Figure 2 shows that a one-percentage point increase in the share of older workers in an occupation today is associated with fewer jobs in 2030.  This negative association exists whether the availability of future jobs is measured absolutely using the projected level in 2030, or as a rate of change between 2020 and 2030.

Bar graph showing the Relationship of Older Workers’ Share of Current Occupations to Employment Outlook in 2030 (Thousands of Jobs)

But what about jobs older workers can do, but may not currently be doing?  Here, using the Suitability Index, the Center study found no statistically significant relationship – neither a positive one nor (importantly) a negative one – between the suitability of occupations for older workers and the projected number of jobs in occupations in 2030 or the growth from 2020-2030.  This finding is somewhat encouraging, as it suggests that there may not be a mismatch between the jobs older workers are capable of doing and those available in the future.  In other words, those jobs older workers can do are apparently not going away, even if the ones they are currently doing appear to be becoming less common.  

Conclusion

For decades, researchers have been encouraging workers nearing retirement age to keep working.  But this prescription won’t work unless employers are also interested in hiring and retaining these workers.  On the employer-demand side, the overall picture from the Center’s research seems fairly encouraging.  Today, the hard data suggest that older workers are at least as productive as younger ones and largely indistinguishable on profitability.  Furthermore, employers’ stated perceptions on a survey are largely in line with these data.  So are employer actions; while employers seem to target lower-paying jobs specifically at older workers, they also show a willingness to post high-paying jobs to RetirementJobs.com.

Finally, little reason exists to doubt that the future will look much different.  While the specific jobs older workers do today may be less prevalent in the future, our analysis indicates that jobs in occupations that are suitable for older workers are likely to grow at a similar pace as other jobs.  So, while older workers may need to change with the times and enter some new occupations, their skills should enable them to do so.  Taken together, it seems that if older workers are willing to supply their labor then demand should likely be there, today and into the future. 

References

Munnell, Alicia H. and Gal Wettstein. 2020. “Employer Perceptions of Older Workers – Surveys from 2019 and 2006.” Working Paper 2020-8. Chestnut Hill, MA: Center for Retirement Research at Boston College.

Munnell, Alicia H., Gal Wettstein, and Abigail N. Walters. 2020. “What Jobs Do Employers Want Older Workers to Do?” Working Paper 2020-11. Chestnut Hill, MA: Center for Retirement Research at Boston College.

Munnell, Alicia H., Steven A. Sass, and Mauricio Soto. 2006. “Employers Attitudes towards Older Workers: Survey Results.” Issue in Brief 6-3. Chestnut Hill, MA: Center for Retirement Research at Boston College. 

Quinby, Laura D., Gal Wettstein, and James Giles. 2023. “Are Older Workers Good for Business?” Working Paper 2023-19. Chestnut Hill, MA: Center for Retirement Research at Boston College.

Siliciano, Robert L. and Gal Wettstein. 2022. “Will the Jobs of the Future Support an Older Workforce?” Working Paper 2022-2. Chestnut Hill, MA: Center for Retirement Research at Boston College.

Endnotes

1    Quinby, Wettstein, and Giles (2023).
2    Munnell and Wettstein (2020).
3    It is worth noting that the 2019 survey was meant to replicate an earlier study, conducted in 2006.  The 2019 results are more favorable to older workers – specifically older support/production workers – than the 2006 edition.  See Munnell, Sass, and Soto (2006).
4    For each listing, RetirementJobs.com provided the employer’s name, region of operation, job title, and whether the job was full- or part-time.  These structured fields were supplemented by the job descriptions, which were manually coded to identify information on salary and benefits.  For more detail, see Munnell, Wettstein, and Walters (2020).
5    Because the total number of listings varies by state, the sample for each state was weighted to reflect its total underlying listings.
6    Siliciano and Wettstein (2022).
7    For more details, see Siliciano and Wettstein (2022).
]]>
https://crr.bc.edu/can-employer-demand-support-older-workers-today-and-tomorrow/feed/ 0
Shrinking Social Security’s Racial Gap – but Only a Little https://crr.bc.edu/shrinking-social-securitys-racial-gap-but-only-a-little/ https://crr.bc.edu/shrinking-social-securitys-racial-gap-but-only-a-little/#respond Thu, 07 Mar 2024 14:09:50 +0000 https://crr.bc.edu/?p=46852 Consider the myriad reasons Black Americans get 19 percent less from Social Security during their retirement years than White retirees.

Inequities in the labor market limit their job opportunities and earnings, which are the basis for determining how much they will get in their Social Security checks. Poorer health or the demands of caring for children or ill family members shortens careers and cuts into benefits.

Smaller benefit checks are also a problem because Blacks are less likely to marry. The strain on single workers’ finances spills over into old age if they lacked a higher-earning spouse who would get a bigger Social Security check.

Hispanics also get less – 14 percent less – for some of the same reasons, including lower pay and caregiving duties.

This very large racial gap in Social Security’s retirement benefits exists despite the progressive formula the government uses to try to even the playing field for workers with lower earnings.

A new Urban Institute study finds that making the formula even more favorable to low-income workers would be the most effective way to close the gap in benefits between Black and Hispanic retirees and their White counterparts.

However, the researchers concluded that even a more generous benefit formula would not do as much to shrink the racial gap as an increase in Black and Hispanic workers’ earnings. 

The in-depth analysis compared the future impact of numerous reforms that have been proposed by Congress or policymakers. The estimates apply to the benefits that would kick in when members of Generation Z are retiring.

One change in the formula analyzed in this study would disproportionately increase Blacks’ and Hispanics’ benefits by about 4 percent to 5 percent. Black retirees would get $23,800 more during their retirement years and Hispanics $20,500 more. Whites would get only $800 more.

The reform would close the gap by increasing future retirement benefits for lower-income workers and trimming them at the top. Social Security uses a tiered formula. Currently, under the first tier, Social Security benefits equal 90 percent of a worker’s average monthly earnings under $1,174 if he retires at the program’s full retirement age. The first tier would become more generous, increasing to 95 percent of earnings under $1,503.

To reduce higher-earning workers’ future benefits, the formula would drop from 15 percent currently to 5 percent of the monthly earnings above $7,078 that Social Security would replace.

Two other reforms would also reduce the gap but by smaller amounts. One proposal, which mostly helps women, gives working parents credit in their Social Security earnings records for taking care of a dependent child more than 80 hours per month. A minimum benefit pegged to the federal poverty level is geared toward helping workers with very low incomes or spotty employment histories.

On the tax side, eliminating federal income taxes on all Social Security benefits is a popular idea that appeared in a 2016 bill in Congress. But eliminating taxes would disproportionately help White retirees, the researchers found, because the taxes apply only to people with middle and high incomes.

Their takeaway: “Changing Social Security alone seems unlikely to narrow existing racial and ethnic gaps substantially.”

Achieving equity for Black and Hispanic retirees, they said, would have to start with expanding opportunity for workers and increasing pay equity.

To read this study by Richard Johnson and Karen Smith, see “How Can Changes to Social Security Improve Benefits for Black and Hispanic Beneficiaries?”

The research reported herein was derived in whole or in part from research activities performed pursuant to a grant from the U.S. Social Security Administration (SSA) funded as part of the Retirement and Disability Research Consortium.  The opinions and conclusions expressed are solely those of the authors and do not represent the opinions or policy of SSA, any agency of the federal government, or Boston College.  Neither the United States Government nor any agency thereof, nor any of their employees, make any warranty, express or implied, or assumes any legal liability or responsibility for the accuracy, completeness, or usefulness of the contents of this report.  Reference herein to any specific commercial product, process or service by trade name, trademark, manufacturer, or otherwise does not necessarily constitute or imply endorsement, recommendation or favoring by the United States Government or any agency thereof.

]]>
https://crr.bc.edu/shrinking-social-securitys-racial-gap-but-only-a-little/feed/ 0
Saving for Retirement Can Mean Adding Some Debt Too https://crr.bc.edu/saving-for-retirement-can-mean-adding-some-debt-too/ https://crr.bc.edu/saving-for-retirement-can-mean-adding-some-debt-too/#respond Tue, 05 Mar 2024 14:07:33 +0000 https://crr.bc.edu/?p=46864 In today’s world, workers need to save if they want to be comfortable in retirement. But there are also limits to what many people can afford.

A new study finds that when U.K. workers were automatically enrolled and started contributing to a retirement savings plan, their household debt – credit cards, bank overdrafts, and other unsecured loans – increased. For every 32 to 38 pounds (or $40-48) in combined monthly contributions by the employer and employee, their debt rose by just over 7 pounds (about $9).

Stepping back to look at the big picture, this research also confirms the benefit of auto-enrollment: it encourages workers who might not otherwise have saved to get started. And the increase in unsecured debt, with its higher interest rates, is much less than the amount that goes into retirement plans. On the other hand, saving is not cost-free.

The researchers admit they’re not sure why the savers borrow more but they had a few ideas. One reason might be that workers spend more because they feel more optimistic about their future finances after they start saving. The employer’s contributions could have that effect, but it still seems a stretch that someone would change their thinking in the 41 months they were being tracked for this study.  

Another reason to add debt makes more sense: people who are on a tight budget and aren’t paying close attention to the change in their finances “may be failing to reduce their spending to sufficiently finance their pension contributions,” the researchers said.

One piece of evidence to back this up is that the increases in debt are larger for workers with lower incomes, who have serious budget limitations, and for younger adults, who may still be finding their financial footing.

The people included in this research had small employers with up to 29 employees and were automatically enrolled in the National Employment Savings Trust, or NEST, which is a U.K. government-sponsored plan that includes various options for investing the account. Like U.S. auto-enrollment plans, workers have the option of withdrawing from NEST.

The researchers also revealed another connection between saving for retirement and debt. Workers who started contributing to the savings plan were more likely to take out a mortgage than the people who opted out of the plan.

In contrast to credit card debt, however, adding mortgage debt can be a good thing. It probably is either tied to investing in an asset – a new house – or is borrowing against an existing property that’s rising in value. Other financial indications of the savers’ finances were also positive, including a slight improvement in their credit scores and a drop in bankruptcies. 

The workers’ household mortgage debt increased by 118 pounds (or about $149) for every 32 to 38 pounds of new savings.

The reasons for workers’ decisions aren’t clear here either. But auto-enrollment “has complex effects across different facets of the household balance sheet,” the researchers concluded.

Squared Away writer Kim Blanton invites you to follow us @SquaredAwayBC on X, formerly known as Twitter. To stay current on ourblog, join our free email list. You’ll receive just one email each week – with links to the two new posts for that week – when you sign up here.  This blog is supported by the Center for Retirement Research at Boston College

]]>
https://crr.bc.edu/saving-for-retirement-can-mean-adding-some-debt-too/feed/ 0
More Households Are Prepared for Retirement – But This Good News Might Not Last https://crr.bc.edu/more-households-are-prepared-for-retirement-but-this-good-news-might-last/ https://crr.bc.edu/more-households-are-prepared-for-retirement-but-this-good-news-might-last/#respond Thu, 29 Feb 2024 15:59:48 +0000 https://crr.bc.edu/?p=46963

Even with the improvement, two out of five households are still at risk, according to the National Retirement Risk Index.

The release of the Federal Reserve’s 2022 Survey of Consumer Finances (SCF) offers an opportunity to reassess Americans’ retirement preparedness as measured by the National Retirement Risk Index (NRRI).  The NRRI estimates the share of American households that are at risk of being unable to maintain their pre-retirement standard of living in retirement. 

Constructing the NRRI involves three steps: 1) projecting a replacement rate – retirement income as a share of pre-retirement income – for a nationally representative sample of working-age households; 2) constructing a target replacement rate consistent with maintaining a pre-retirement standard of living in retirement; and 3) comparing the projected and target replacement rates to find the percentage of households “at risk.”

Since the last SCF was conducted in 2019, the nation experienced a global pandemic and economic disruption, and 2022 was a very bad year for stock and bond returns.  These factors would have reduced households’ retirement preparedness.  At the same time, the government provided unprecedented fiscal support, employment remained strong, home values rose substantially, and the stock market – even with the drop in 2022 – ended up significantly higher than in 2019. 

The 2022 NRRI shows that the gains in asset values more than offset the economic disruption to produce the lowest level of households at risk since the NRRI first started.  Specifically, between 2019 and 2022, the share at risk dropped from 47 percent to 39 percent (see Figure 1).

Bar graph showing the National Retirement Risk Index, 2004-2022

Figure 2 breaks down the reasons for the big reduction in the NRRI.  The increase in home prices leads the list, followed by new savings during the pandemic, and stock market gains.  Rising interest rates had small offsetting effects by reducing how much home equity that households can tap through reverse mortgages.

Bar graph showing the decrease in percentage "at risk" from 2019 to 2022, by contributing component

What do the 2022 results imply for the future?  Two major contributors to the stunning improvement in the NRRI seem unlikely to persist.  First, housing prices are about 14 percent above their long-run trend for the last 30 years, and may well revert to trend over time.  Second, “new saving” is almost certainly a one-shot COVID phenomenon.  Indeed, personal saving rates have returned to pre-pandemic levels and so has credit card borrowing.  Thus, the spectacular decline in the share of households at risk may not hold for the future.     

But assume the good news is permanent, and future NRRIs hover around 40 percent.  That finding means about two-fifths of today’s working-age households will not have enough retirement income to maintain their pre-retirement standard of living.  This analysis continues to confirm that we need to fix our retirement system so that Social Security is financially sound and employer plan coverage is universal.

]]>
https://crr.bc.edu/more-households-are-prepared-for-retirement-but-this-good-news-might-last/feed/ 0
Surging U.S. Centenarian Population Requires Action https://crr.bc.edu/surging-u-s-centenarian-population-requires-action/ https://crr.bc.edu/surging-u-s-centenarian-population-requires-action/#comments Thu, 29 Feb 2024 14:12:01 +0000 https://crr.bc.edu/?p=46790 The growth in the over-100 population is gob-smacking. In 30 years, Pew reports that the number of U.S. centenarians will quadruple to more than 400,000. To get a sense of what this will mean, picture every resident in Tampa, Fla., being over 100. In 1950, only 2,300 people in the entire country were.

Most centenarians today are women and will be in the future but men are also living longer. Men will make up about a third of them in 2054, up from a fifth currently, according to Pew.

The rapid growth in the over-100 population is part of a larger trend of an aging nation, which creates three pressing policy concerns.

First, all these old people are going to need a lot more caregivers. Daughters usually take care of their aging parents. But someone who lives to 100 might either outlive their children or their children might be too old to care for them.

The systems currently in place are already under strain as baby boomers increasingly need care. Caregiving is difficult, low-paid work. Job openings for in-home aides are going unfilled, and staff turnover at nursing homes and assisted living facilities is a chronic problem.

Another issue is the financial drain on individuals of an extremely long retirement. Already, about four in 10 workers today workers today are not on track with their 401(k) savings and may experience a drop in their standard of living when they retire.

Figure showing male centenarians

Amid rising longevity, many older workers are taking steps to improve their retirement outlook by working longer to save more and hold out for a larger monthly Social Security benefit. But even if someone delays retiring until 70 and then lives to 90 or 100, they will need enough savings to cover two or three decades.

Finally, unless policymakers act, the Social Security trust fund is expected to be depleted in the 2030s, and the payroll taxes paid by workers will provide only about three-fourths of the revenues required to cover retirees’ benefits.

Social Security’s fiscal problems will have to be addressed at the same time that the Medicare program is being increasingly burdened by the cost of medical care for the growing elderly population.

These fiscal problems with be compounded by fertility declines. As more people live to 90, 100 or beyond, there will be fewer and fewer workers funding each retiree. This is a problem throughout developed countries and not just in the United States.

Meeting the growing need for caregivers, doing more to encourage retirement saving, and repairing Social Security’s and Medicare’s finances – these are pressing issues that Congress needs to address. The clock is ticking.

Squared Away writer Kim Blanton invites you to follow us @SquaredAwayBC on X, formerly known as Twitter. To stay current on our blog, join our free email list. You’ll receive just one email each week – with links to the two new posts for that week – when you sign up here.  This blog is supported by the Center for Retirement Research at Boston College.

]]>
https://crr.bc.edu/surging-u-s-centenarian-population-requires-action/feed/ 6
ERISA Killed Defined Benefit Pension Plans, Yale Law School Expert Argues https://crr.bc.edu/erisa-killed-defined-benefit-pension-plans-yale-law-school-expert-argues/ https://crr.bc.edu/erisa-killed-defined-benefit-pension-plans-yale-law-school-expert-argues/#respond Wed, 28 Feb 2024 13:57:30 +0000 https://crr.bc.edu/?p=46760 How important was the 1974 legislation compared to other factors?

This year is the 50th anniversary of the Employee Retirement Income Security Act of 1974 (ERISA), so Yale Law School professor John Langbein’s recent article “ERISA’s Role in the Demise of Defined Benefit Pension Plans” seemed timely.  My view is that a lot of factors contributed to the demise of defined benefit (DB) plans, and ERISA was just one more on the list.  Langbein gives ERISA more prominence, arguing that the 1974 legislation “ultimately sealed the fate of the American DB pension system.” 

First, a brief history.  Retirement plans date from the last quarter of the 19th century when a large prosperous transportation industry and trade unions established benefit arrangements for their workers.  By the 1920s, 40 percent of union members had coverage, but the Great Depression devastated union plans and saw the enactment of Social Security.  So, in the 1940s, business and labor had to recreate the pension system.  Pension coverage expanded dramatically during the 1950s primarily through the establishment of new plans, and grew in the 1960s and 1970s primarily due to the expansion of employment in firms that already had pension coverage.  By the 1980s, roughly half of private sector workers were covered by a retirement plan, and these retirement plans were primarily defined benefit. 

Under defined benefit plans, the employer makes the contributions, pays benefits in the form of a lifetime annuity, and bears the investment and mortality risk.  From the employer’s perspective, defined benefit plans help manage the workforce by encouraging longer tenure and efficient retirement.

But beginning in 1980 things started to change, and by 2000 the majority of workers with coverage relied on a 401(k) plan.  Here’s my usual list of reasons:

  • Globalization increased competition and undermined the financial stability of large corporate employers, making long-term pension obligations much riskier.
  • Employment declined in large hierarchic firms and unionized industries, which typically offered defined benefit plans, and grew in high-tech firms and small, non-unionized companies, which typically did not.  
  • Plans became more expensive, as workers were living longer, and declining inflation raised the cost of unindexed lifetime benefits.  
  • The nature of the labor force changed – becoming more female, educated, and young – which reduced the appeal of lifelong careers.
  • And importantly, 401(k) plans became available, just as the stock market began a two-decade climb.

Langbein doesn’t disagree with these underlying factors, but rather puts considerable weight on the damage done by ERISA.  The legislation was designed to save the rights of defined benefit plan participants so that a greater share of them would receive their benefits.  It introduced participation and vesting standards to make it easier for workers to establish legal claims to benefits, and funding and fiduciary standards to ensure that money was available to pay the legal claims.  And, should a plan terminate with inadequate assets, ERISA created the Pension Benefit Guaranty Corporation (PBGC).

While all well intended, Langbein concludes that ERISA made defined benefit plans too burdensome.  At the top of his list is PBGC premiums, which he characterizes as a tax on healthy plans to subsidize sick plans, creating a major incentive for firms to get out of the defined benefit business.  Second on his list is accelerated vesting rules – five years since 1986 – which are restrictive and costly and prevent sponsors from using the pension to reduce turnover.  Third is forbidding employers from limiting their liabilities to the plan’s assets, which led to the Financial Accounting Standards Board requiring the display of the plan’s impact (with assets measured at market) on the firm’s income statement and balance sheet.  The final item on Langbein’s list is compliance and litigation costs, especially considering that during the 1980s Congress passed significant legislation affecting defined benefit plans every few years.

It is certainly true that legislation designed to make pensions fairer also made the pension system more complex and costlier to administer.  The only question is how much ERISA contributed to the decline in defined benefit plans compared to developments in the global economy, industry shifts, decline in unions, and changing composition of the workforce.

]]>
https://crr.bc.edu/erisa-killed-defined-benefit-pension-plans-yale-law-school-expert-argues/feed/ 0
The National Retirement Risk Index: An Update from the 2022 SCF https://crr.bc.edu/the-national-retirement-risk-index-an-update-from-the-2022-scf/ https://crr.bc.edu/the-national-retirement-risk-index-an-update-from-the-2022-scf/#respond Tue, 27 Feb 2024 13:59:08 +0000 https://crr.bc.edu/?p=46525

The brief’s key findings are:

  • From 2019 to 2022, the National Retirement Risk Index improved substantially, dropping from 47 percent to 39 percent.
  • Weathering the pandemic turmoil, households were buoyed by government stimulus, strong employment, and rising asset markets.
  • The single biggest factor driving the improvement was soaring home values.
  • Though these results are encouraging, most people do not tap their home equity in retirement and prices may not stay at such high levels.
  • Importantly, even with what may be a temporary improvement, the NRRI shows that 2 in 5 of today’s working households could fall short.

Introduction 

The release of the Federal Reserve’s 2022 Survey of Consumer Finances offers an opportunity to reassess Americans’ retirement preparedness as measured by the National Retirement Risk Index (NRRI).  The NRRI estimates the share of American households that are at risk of being unable to maintain their pre-retirement standard of living in retirement.  

The NRRI was originally constructed using the Federal Reserve’s 2004 Survey of Consumer Finances (SCF) and has been updated every three years with the release of this triennial survey.  Since the last SCF was conducted in 2019, the nation experienced a global pandemic and economic disruption, and 2022 was a very bad year for stock and bond returns.  These factors would have reduced households’ retirement preparedness.  At the same time, the government provided unprecedented fiscal support, employment remained strong, home values rose substantially, and the stock market – even with the drop in 2022 – ended up significantly higher than in 2019.  The 2022 survey allows us to see how these disparate factors have affected Americans’ readiness for retirement.

The discussion proceeds as follows.  The first section describes the nuts and bolts of the NRRI.  The second section updates the NRRI using 2022 SCF data and shows that the share of households at risk dropped from 47 percent to 39 percent, the lowest level since we started the Index in 2004, largely due to rising home values.  The third section relates the various reasons for the decline to relevant subgroups.  The fourth assesses the implications of the big drop in the NRRI for the future.  The final section concludes that, even with the immediate impact of the strong housing market, retirement readiness remains a major challenge for many of today’s workers. 

Nuts and Bolts of the NRRI 

Constructing the NRRI involves three steps: 1) projecting a replacement rate – retirement income as a share of pre-retirement income – for a nationally representative sample of U.S. households; 2) constructing a target replacement rate consistent with maintaining a pre-retirement standard of living in retirement; and 3) comparing the projected and target replacement rates to find the percentage of households “at risk.”1

Projecting Household Replacement Rates

The exercise starts with projecting retirement income for households at three retirement ages – 62 for low income, 66 for middle income, and 67 for high income.2  Retirement income is defined broadly to include income from Social Security and defined benefit (DB) plans; income from financial assets in both defined contribution (DC) plans and saved directly; and income from housing. 

The values for financial and housing assets at retirement are calculated from the ratio of wealth to income by age from the 1983-2022 SCFs.  As shown in Figure 1, the aggregate wealth-to-income by age has been stable over time, in the sense that the lines rested roughly on top of each other.  Similar stability holds with regard to the various components of wealth.  Using the stable relationship between each component and income, the analysis estimates DC assets, non-DC financial assets, and housing equity separately, at various retirement ages. 

Line graph showing the Ratio of Wealth-to-Income by Age from the Survey of Consumer Finances, 1983-2022

The aggregate wealth-to-income ratios also provide some hint of the NRRI for 2022.  Historically, wealth-to-income lines from each survey have been bracketed by 2007 values on the high side and 2013 values on the low side.  The 2022 line (red) looks similar to wealth-to-income ratios in 2007, suggesting strong improvements in the percentage of households at risk between 2019 and 2022. 

The NRRI calculations assume that households convert all their assets, including financial assets, 401(k)/ IRA balances, and proceeds from a reverse mortgage, into a stream of income by purchasing an inflation-indexed annuity.3

Sources of retirement income that are not derived from reported wealth in the SCF are estimated directly.  Specifically, Social Security benefits are calculated based on estimated earnings histories for each member of the household, indexed to national average wage growth.  DB pension income is based on the amount reported by survey respondents. 

The remaining step is to calculate pre-retirement income at the three retirement ages.  For homeowners, this measure includes earnings (again, indexed to average wage growth) and imputed rent from housing.  Average pre-retirement income then serves as the denominator for each household’s replacement rate.  This measure excludes income from assets.

Estimating Target Replacement Rates 

Determining the estimated share of the population that is at risk requires comparing projected replacement rates with a benchmark rate.  A commonly used benchmark is the replacement rate needed to allow households to maintain their pre-retirement standard of living in retirement.  People typically need less than their full pre-retirement income to maintain this standard once they stop working since they generally pay less in taxes, no longer need to save for retirement, and often have paid off their mortgage.  Thus, a greater share of their income is available for spending.4  The Index estimates the target replacement rates for different types of households using a consumption-smoothing model, which is based on the assumption that households want the same level of consumption in retirement as they had before they retired. 

Calculating the Index 

The final step in creating the NRRI is to compare each household’s projected replacement rate with the appropriate target.  Households whose projected replacement rates fall more than 10 percent below their targets are deemed to be at risk of having insufficient income to maintain their pre-retirement standard of living.  The NRRI is simply the percentage of all households that fall more than 10 percent short of their target.

The NRRI in 2022

The strong ratio of wealth to income shown in Figure 1 suggested that the NRRI would improve in 2022 – that is, fewer households would be at risk.  And indeed, the share of households at risk dropped from 47 percent to 39 percent – the lowest level since the Index started in 2004 (see Figure 2).

Bar graph showing The National Retirement Risk Index, 2004-2022

Figure 3 breaks down the reasons for the big reduction in the NRRI.  The increase in home prices leads the list, followed by new savings during the pandemic, and stock market gains.  Rising interest rates had small offsetting effects.  The following takes a closer look at each component.

Bar graph showing the Decease in Percentage “At Risk” from 2019 to 2022, by Contributing Component

Housing.  Between 2019 and 2022, U.S. home prices increased by about 22 percent in real terms, (see Figure 4).  For context, home prices grew by 27 percent between 2001 and 2004.  In the NRRI, home prices have a significant impact because households are assumed to access their home equity at retirement by taking out a reverse mortgage.  The higher the home value, the more equity households can extract through a reverse mortgage.

Line graph showing the Index of Average U.S. Home Prices (Real), January 1990 to August 2023

Increased Savings.  A unique factor that improved projected retirement finances for many households was a dramatic spike in personal saving during the pandemic, fueled by the federal stimulus spending.  Personal saving rates jumped from roughly 7 percent of disposable income to over 30 percent during the pandemic (see Figure 5).  While the saving rate returned to pre-pandemic levels by 2022, the build-up in savings in 2020 and 2021 temporarily made many households’ balance sheets much stronger.

Line graph showing the Personal Saving Rate, January 1990-September 2023

Stock Market. Despite the initial dip early in the pandemic and a pullback in 2022, equity prices increased by more than 20 percent after adjusting for inflation since the end of 2019 (see Figure 6).  These gains, however, have been concentrated in the top third of the income distribution, which holds about 87 percent of all equities.5  This pattern means that much of the gains went to households that were already not at risk.

Line graph showing the Dow Jones Wilshire 5000 (Real), January 1990-October 2023

Interest Rates.  While real interest rates had been declining for the last 30 years, during the pandemic they spiked to the highest level since the early-2000s (see Figure 7).  Higher interest rates have two opposing effects on households’ financial security in retirement.  On one hand, higher rates mean that households get more income from annuitizing their assets.  On the other hand, the high rates reduce the amount households can get from a reverse mortgage.  On net, higher rates increased the percentage of households at risk, but the effect is small.6

Line graph showing the Real 10-Year Interest Rate, 1990-2023

Impact of 2022 Changes for NRRI Subgroups

Given the importance of the various factors reducing the percentage at risk, it is interesting to examine the patterns in the NRRI by homeownership status, retirement plan coverage, income level, and age group. 

It is not surprising that fewer homeowners are at risk relative to renters, but the gap between the two groups increased.  The strong growth in the housing market reduced the share of homeowners at risk by 10 percentage points, while the share of renters at risk stayed fairly steady (see Table 1).

Table showing the Percentage of Households “At Risk” at Retirement Age by Homeownership Status, 2019 and 2022

At the same time, the gains in the stock market between 2019 and 2022 reduced the percentage at risk of households with a DC plan, while the percentage at risk with no retirement plan dropped only slightly (see Table 2).  The declining NRRI for households with a DB plan reflects the fact that an increasing share of these plans are in the public sector, where pensions are a major component of total compensation.

Table showing the Percentage of Households “At Risk” at Retirement Age by Plan Coverage, 2019 and 2022

When viewed by income and age, the largest gains in relative terms were enjoyed by high-income households (see Table 3), who are mostly likely to own homes and who hold most of the equities.  The gains for the low- and middle-income groups most likely reflect some homeownership and perhaps some of the new savings.  In terms of age, the oldest group, who hold most of the assets, saw the largest reduction in the NRRI (see Table 4).

Table showing the Percentage of Households “At Risk” at Retirement Age by Income Group, 2019 and 2022
Table showing the Percentage of Households “At Risk” at Retirement Age by Age Group, 2019 and 2022

Implications of 2022 NRRI for the Future

The 2022 NRRI represents the biggest drop in the Index since its creation, largely driven by growth in the asset markets, and in particular the housing market.  Current housing prices are about 14 percent above their long-run trend for the last 30 years.7  It would seem reasonable that, over time, housing prices may revert to trend.  Moreover, it is important to remember that the NRRI has retirees purchasing a reverse mortgage – something that few actually do.  Therefore, the share of households deemed at risk by the NRRI can be viewed as a lower-bound estimate.  In fact, a recent study by researchers at Vanguard found that at least 70 percent of households would fall short of their pre-retirement standard of living in retirement when housing equity is excluded.8 

The other major factor contributing to the decline in the NRRI from 2019 to 2022 that may not persist in the future is “new saving.”  Personal saving rates have returned to pre-pandemic levels and so has credit card borrowing.  Thus, it appears that households are spending some of the additional savings they built up since 2019.9  Thus, one would not expect new savings to be a major contributor to improved retirement readiness over the next three years.

In contrast to these transitory effects, the positive impact of the stock market may be more permanent.  As noted, most households saw a rise in net financial wealth and 401(k)/IRA balances due to higher stock prices and the one-time accumulation in new savings.10  That means, even ignoring growth in the housing market, households across the income and age distribution are more prepared for retirement than they were in 2019.  Although always uncertain, stock market gains may well contribute to improving the financial security of middle- and high-income households in the future.

Conclusion

Between 2019 and 2022, the NRRI dropped substantially – from 47 to 39 percent.  Despite a global pandemic and economic disruption, household finances were buoyed by unprecedented fiscal support, a strong labor market, and considerable growth in the housing and stock markets.  The biggest factor driving the improvement in retirement finances is growth in the housing market.  But most households do not tap their home equity in retirement and home prices may not remain at historically high levels.  The bottom line is that even a conservative estimate shows a substantial portion – about two-fifths – of today’s households will not have enough retirement income to maintain their pre-retirement standard of living.  This analysis continues to confirm that we need to fix our retirement system so that Social Security is financially sound and employer plan coverage is universal.

References 

Barbiero, Omar and Dhiren Patki. 2023. “Have US Households Depleted All the Excess Savings They Accumulated during the Pandemic?” Current Policy Perspectives Paper 97263. Boston, MA: Federal Reserve Bank of Boston.

Federal Reserve Bank of Cleveland. 2023. “Ten-Year Expected Inflation and Real and Inflation Risk Premia.” Cleveland, OH.

Haubrich, Joseph, George Pennacchi, and Peter Ritchken. 2012. “Inflation Expectations, Real Rates, and Risk Premia: Evidence from Inflation Swaps.” The Review of Financial Studies 25(5): 1588-1629.

S&P Global. 2023. “S&P CoreLogic Case-Shiller U.S. National Home Price NSA Index.” New York, NY: S&P Dow Jones Indices.

Tan, Fu, Fiona Greig, Andrew S. Clarke, Kevin Khang, Kate McKinnon, and Victoria Zhang. 2023. “The Vanguard Retirement Outlook: A National Perspective on Retirement Readiness.” Valley Forge, PA: Vanguard.    

U.S. Board of Governors of the Federal Reserve System. 2023. “Selected Interest Rates: Historical Data.” Washington, DC.

U.S. Board of Governors of the Federal Reserve System. Survey of Consumer Finances, 1983-2022. Washington, DC.

U.S. Bureau of Economic Analysis. 2023. Personal Saving Rate Data (accessed from FRED database). Washington, DC. 

U.S. Bureau of Labor Statistics. 2023. Consumer Price Index. Washington, DC. 

Wheat, Chris and Erica Deadman. 2023. “Household Pulse: Balances through March 2023.” Washington, DC: JPMorgan Chase Institute.

Wilshire Associates. 2023. “Dow Jones Wilshire 5000 (Full Cap) Price Levels Since Inception.” Santa Monica, CA.

Yin, Yimeng, Anqi Chen, and Alicia H. Munnell. 2023a. “The National Retirement Risk Index with Varying Claiming Ages.” Issue in Brief 23-23. Chestnut Hill, MA: Center for Retirement Research at Boston College.

Yin, Yimeng, Anqi Chen, and Alicia H. Munnell. 2023b. “The National Retirement Risk Index: Version 2.0.” Issue in Brief 23-10. Chestnut Hill, MA: Center for Retirement Research at Boston College.   

Appendix: Methodological Improvements in NRRI since 2019

Since its inception, the Center has periodically made modest changes to the NRRI.  Recently, however, we undertook a major overhaul to incorporate new research findings and methodological advances.11  Although the overall modeling framework described above remains unchanged, the updated NRRI includes the following major improvements:

Accounts for differential claiming.  One of the biggest and most recent updates in the NRRI is allowing households in different income groups to claim Social Security (and stop working) at different ages.  Differential claiming better reflects the real-world choices households make around retirement. 

Captures growing wealth inequality.  Another major update in the NRRI is shifting from average wealth-to-income ratios to median ratios when projecting wealth.  Using medians for our projections allows the model to better capture real-world wealth distributions, particularly given growing wealth inequality.12  

Better reflects the shift from DB to DC plans.  The new method projects DC assets for different cohorts to account for the growing share of younger workers covered by DC plans.13

Models financial debt separately.  Financial assets and non-mortgage debt are now projected separately instead of in a single net financial asset variable.14

Refines the target replacement rate model.  Projected replacement rates are matched to a wide range of characteristics, allowing for hundreds of targets, instead of just 12 household types in the previous model.15 

Endnotes

1    Target rates are calculated using a lifecycle model rather than a household’s actual, but unobserved, consumption history.  For details, see Yin, Chen, and Munnell (2023a,b). 
2    The original NRRI assumed that all households stopped working and claimed Social Security benefits at the same age – namely, 65.  SCF data on the claiming age for the household head, however, suggest that more reasonable claiming ages for the three income groups would be 62, 66, and 67 for low, middle, and high income respectively.  Varying claiming ages by income group incorporates about two-thirds of the total earnings of workers 62-75.  While allowing for differential claiming across income groups increases the share of low-income households at risk and decreases the share of high-income households at risk, it does not change the overall percentage of households at risk.  For an overview of other methodological improvements to the NRRI, see the Appendix.
3    While inflation-indexed annuities are not widely used by – or even available to – consumers, they provide a convenient metric for calculating the lifetime income that can be obtained from a lump sum.  And while inflation-indexed annuities provide a smaller initial benefit than nominal annuities, they protect a household’s purchasing power over time against the erosive effects of inflation. 
4    The level of replacement required for smoothing consumption before and after retirement will vary by type of household.  For example, low-income households get most of their retirement income from Social Security and therefore need little saving before retirement.  The result is that they get little break from no longer having to save in retirement.  Similarly, low-income households pay little in taxes, so they receive little in the way of tax saving in retirement.  Thus, low-income households need a higher replacement rate in retirement.
5    This metric includes equities in retirement accounts.  If only equities in brokerage and individual investment accounts were considered, the number would be much higher.
6    The NRRI tapers the effect of interest rate fluctuations because it assumes that rates return to their long-run average over time.  In other words, households approaching retirement will experience all or part of the change in interest rates while households under age 50 remain at long-run levels.
7    At the end of 2022, the inflation-adjusted S&P CoreLogic Case-Shiller U.S. National Home Price Index was at 307.7, 14 percent higher than the predicted index of 270.2 based on a linear trend starting in 1990.
8    Tan et al. (2023).
9    Although households have spent down some of the new savings accumulated during the pandemic, researchers have found that households across the income distribution have more savings now than they did in 2019 (Wheat and Deadman 2023; Barbiero and Patki 2023).  The extent to which households have spent down new savings, however, is not clear (Barbiero and Patki 2023).
10    The exception is households ages 40-49, who saw a decrease in their net financial assets and retirement assets between 2019 and 2022.  One explanation could be that these households shifted assets from their financial and retirement accounts to help buy a house.
11    See Yin, Chen, and Munnell (2023a,b) for details about all the updates.
12    The previous projection method was based on the mean wealth growth paths estimated by a linear regression approach.  The previous methodology imposes restrictions on the resulting projected distributions, which can partially reduce the bias.  However, the rise in wealth inequality over time makes it increasingly difficult to rely on this approach.
13    The growing share of workers covered by DC plans since the 1980s means that the level and pattern of DC asset accumulation differs across birth cohorts.  To account for differences in access to DC plans and accumulation, the improved method projects DC assets separately for three broad cohorts: 1) workers born before 1945, who were at least halfway into their careers when coverage under DC plans began to expand in 1980; 2) workers born from 1945-1955, who were early in their careers during the transition to DC plans; and 3) workers born after 1955, whose careers mostly fall in the years when DC plans were already prevalent. 
14    Analyses of previous NRRI results suggest that the dynamics of financial debt can be of interest on their own.  For example, middle-age and middle-income households saw very limited improvements in retirement preparedness in 2016 partly due to increased non-mortgage borrowing.  The new method now projects financial assets and non-mortgage debt separately, allowing for more in-depth analysis as well as counterfactual analysis focusing on borrowing. 
15    Under the new method, much richer household characteristics are used for calculating target rates, allowing the projected replacement rates to be matched to hundreds of targets, which yields more accurate estimates.  Specifically, matching is now based on much more fine-grained income groups and households’ actual DB coverage and homeownership status.
]]>
https://crr.bc.edu/the-national-retirement-risk-index-an-update-from-the-2022-scf/feed/ 0
Crack-era Incarcerations Added to the Disability Rolls https://crr.bc.edu/crack-era-incarcerations-added-to-the-disability-rolls/ https://crr.bc.edu/crack-era-incarcerations-added-to-the-disability-rolls/#respond Tue, 27 Feb 2024 13:58:26 +0000 https://crr.bc.edu/?p=46831 Some 300,000 men in their 50s and early 60s were receiving federal disability benefits in 2016 as a direct result of a past surge in incarceration, according to new research.

The number of American men in federal or state prisons peaked at 1.5 million in 2010. That surge, which continues to affect the disability program today, was rooted in the crack epidemic of the 1980s that entangled many baby boomers and in the three-strikes laws in the 1990s that put more men in prison for drug offenses.

The increase in the prison population and its subsequent fall roughly follows the rise and fall in the number of men on the disability rolls. A new study confirms a link between men’s past incarceration and resulting employment problems and the upward march of the disability rolls before and during the Great Recession years. 

Many of the aging boomers who wound up on disability may have developed their impairments in prison and had difficulty finding or maintaining a job after their release.

Syracuse University economist Gary Engelhardt found very large increases in the likelihoods that the older men applied for and received monthly disability benefits, as well as the cash assistance provided under the Supplemental Security Income program.

The early surge in incarceration also significantly increased the men’s poverty rate, resulting in an additional 375,000 men ages 50 to 61 being impoverished in 2016.

The analysis was based on two different surveys that included questions about whether the participants had spent time in prison. Crack first entered the country through Florida, California and New York in 1981. After identifying the incarcerated people in the surveys, Engelhardt used the spread of the drug to 36 states, which occurred at various times, to distinguish how much of the rise in disability benefits among their residents resulted from the crack epidemic rather than from the aging population or economic conditions that also affect benefits.

Paradoxically, the formerly incarcerated were more likely to get disability benefits even though this group was less likely to meet the U.S. Social Security Administration’s required years of employment. The requirements vary by age. Someone who currently earns at least $6,920 annually receives the maximum four credits per year. Baby boomers need more credits than young adults, which is more difficult to achieve if someone has spent years or decades in prison.

Despite the difficulty of meeting the standard, the big jump in the probability of applying among those who would qualify resulted in the 300,000 additional beneficiaries in 2016.

After the Great Recession, the disability rolls dropped sharply. But the past surge in incarceration has affected the strength of that post-recession trend. The recent decline in people receiving disability “was partially blunted,” Engelhardt said, by the increasing number of aging men being released from prison and re-entering the civilian population.

To read this study by Gary Engelhardt, see “The Impact of Past Incarceration on Later-Life DI and SSI Receipt.”

The research reported herein was derived in whole or in part from research activities performed pursuant to a grant from the U.S. Social Security Administration (SSA) funded as part of the Retirement and Disability Research Consortium.  The opinions and conclusions expressed are solely those of the authors and do not represent the opinions or policy of SSA, any agency of the federal government, or Boston College.  Neither the United States Government nor any agency thereof, nor any of their employees, make any warranty, express or implied, or assumes any legal liability or responsibility for the accuracy, completeness, or usefulness of the contents of this report.  Reference herein to any specific commercial product, process or service by trade name, trademark, manufacturer, or otherwise does not necessarily constitute or imply endorsement, recommendation or favoring by the United States Government or any agency thereof.

]]>
https://crr.bc.edu/crack-era-incarcerations-added-to-the-disability-rolls/feed/ 0
Connecting Black, Hispanic Layoffs to Retirement Wealth https://crr.bc.edu/connecting-black-hispanic-layoffs-to-retirement-wealth/ https://crr.bc.edu/connecting-black-hispanic-layoffs-to-retirement-wealth/#respond Thu, 22 Feb 2024 14:11:24 +0000 https://crr.bc.edu/?p=46693 Homeowners nearing retirement have 40 percent of their wealth tied up in their homes. But to what extent do racial disparities in employment affect workers’ ability to hold on to a home and build up that wealth?

This question is at the heart of an ambitious study of U.S. homeowners that digs into whether stable homeownership – or, rather, a lack of it – contributes to the longstanding gaps in retirement wealth between Blacks and Hispanics and wealthier White retirees.

The researchers find that the racial disparities in homeowners’ finances while they are working continue after they retire and start collecting Social Security. And Black and Hispanic workers’ employment histories, and specifically their higher chances of having been laid off, impact the ability to accumulate home equity. This has implications for their finances years later when they retire.

To connect the two required the researchers to link numerous sources of information on individual homeowners in eight states, including workers’ and retirees’ mortgage and sales data, employment histories, the demographic, income, and property data in the U.S. Census, and even the voter rolls that indicate whether a retiree has moved.

Start with their findings on job instability: Black and Hispanic homeowners quit or were laid off or fired from their jobs at much higher rates than Whites. Their jobs also were more often casualties of the subset of job losses that result from mass corporate layoffs, which mainly tend to occur during steep economic downturns.

Job instability during one’s working years can affect retirement wealth by making it more difficult for working-aged people to keep paying the mortgage and accumulating home equity than for the homeowners who worked continuously. And the researchers did, in fact, find that Black and Hispanic homeowners, in the first months after losing a job, were more likely to experience distressed sales through a foreclosure that wipes out their equity.

“The racial/ethnic disparities in wealth at retirement age are a direct result of racial/ethnic disparities in labor market experiences during working years,” the researchers concluded.

“An underappreciated aspect” of those wealth disparities, they said, “is job instability.”

To read this study by Francis Wong, Kate Pennington, and Amir Kermani, see “The Impacts of Racial Differences in Economic Challenges on Housing, Wealth, and Economic Security Among OASI Beneficiaries.”

The research reported herein was derived in whole or in part from research activities performed pursuant to a grant from the U.S. Social Security Administration (SSA) funded as part of the Retirement and Disability Research Consortium.  The opinions and conclusions expressed are solely those of the authors and do not represent the opinions or policy of SSA, any agency of the federal government, or Boston College.  Neither the United States Government nor any agency thereof, nor any of their employees, make any warranty, express or implied, or assumes any legal liability or responsibility for the accuracy, completeness, or usefulness of the contents of this report.  Reference herein to any specific commercial product, process or service by trade name, trademark, manufacturer, or otherwise does not necessarily constitute or imply endorsement, recommendation or favoring by the United States Government or any agency thereof.

]]>
https://crr.bc.edu/connecting-black-hispanic-layoffs-to-retirement-wealth/feed/ 0