Tax Incentives for Retirement Plans: Lessons from Denmark?

Notes.Jan13.FinalFlow.TxIncns.Pg1A recent study found that tax incentives for retirement savings in Denmark had virtually no impact on increasing total savings But are those findings relevant to the United States?

Maybe not, according to a new report by EBRI: The two retirement systems have some similarities but also major differences—mainly that, unlike in the United States, in Denmark the availability of employment-based, tax-deferred retirement plans is not tied to the tax-deferred status of the accounts.

At issue are so-called “tax expenditures” in the United States—preferential tax treatment for public policy goals such as retirement, health insurance, home ownership, and a variety of other issues—that currently are under heavy scrutiny in the debate over the federal debt, taxes, and spending.

The authors of the study on Danish savings behaviors offered statistical evidence that changes in tax preferences for Danish work place retirement savings plans had virtually no effect on total savings of those affected by the change. This has drawn the attention of those interested in considering a modification of the long-standing tax preferences for employment-based retirement savings plans in this country.

However, aside from the differences in incentive structures between the two countries, the EBRI report notes that study of Danish workers examined only the impact that changes in tax incentives for work place retirement plans might have on worker savings behaviors—but did not address how employers might react to changes in retirement savings tax incentives.

The EBRI report notes recent surveys have found many American private-sector plan sponsors have expressed a desire to offer no plans at all in the absence of tax incentives for workers. If this happened, low-wage workers—who are generally less prepared for retirement—would suffer on several counts, said Sudipto Banerjee, EBRI research associate and co-author of the report.

“The Danish study provided insight into the savings behavior of Danes, conditioned by the culture and influences of public policies and programs of Denmark,” Banerjee said. “But the ‘success’ of work place retirement plans in the United States depends on the behavior of two parties: workers who voluntarily elect to defer compensation, and employers that sponsor and, in many cases, contribute to them.”

“While the study of Danish savings behaviors presented the impact of tax-incentives and the ‘nudges’ of automatic mandatory savings as an ‘either/or’ solution, the optimal solution—certainly for a voluntary system such as the one currently in place in the U.S.—may well be a combination of the two,” noted Nevin Adams, co-director of the EBRI Center for Research on Retirement Income, and co-author of the report.

The full report is published in the January 2013 EBRI Notes, “Tax Preferences and Mandates: Is the Danish Savings Experience Relevant to America?” online at

Some Rare Good News: Retiree Health Savings Needs Slip

Projections for how much elderly Americans need to save for out-of-pocket health care in retirement have edged lower, due to a provision the federal health reform law that will cover more of their prescription drug costs, according to a new report by EBRI.

The Patient Protection and Affordable Care Act (PPACA) reduces cost sharing in the Medicare Part D “donut hole” to 25 percent by 2020. This year-to-year reduction in coinsurance will continue to reduce savings needed for health care expenses in retirement, all else equal, for individuals with the highest prescription drug use, EBRI reports.

The Medicare Prescription Drug, Improvement, and Modernization Act of 2003 (MMA) added outpatient prescription drugs (Part D) as an optional benefit. When the program was originally enacted, it included a controversial feature: a coverage gap, more commonly known as the “donut hole.” PPACA included provisions to reduce (but not eliminate) this coverage gap.

Medicare generally covers only about 60 percent of the cost of health care services (not including long-term care) for Medicare beneficiaries ages 65 and older, while out-of-pocket spending accounts for 13 percent (see figure, below).

The EBRI analysis finds 1–2 percent reductions in needed savings among individuals with median (mid-point, half above and half below) drug use and 4–5 percent reductions in needed savings among individuals at the 90th percentile in drug use since its last analysis in 2011.

The full report, “Savings Needed for Health Expenses for People Eligible for Medicare: Some Rare Good News,” is published in the October EBRI Notes, online at

The press release is online here.

Means, Tested

By Nevin Adams, EBRI


Recently the Center for Retirement Research at Boston College released a paper titled “Can Retirees Base Wealth Withdrawals On the IRS’ Required Minimum Distributions?”¹ The answer to that question, according to CRR, is “yes.” A more complete response might be, “yes, or any number of other random withdrawal methodologies.”

There are some advantages to a drawdown strategy based on the schedule provided by the Internal Revenue Service (IRS) for required minimum distributions, or RMDs.² First off, and as the CRR paper notes, it’s relatively straightforward. Secondly, it effectively defers initiating withdrawals until age 70-½, which also provides some additional accumulation opportunity. Perhaps most importantly, it helps avoid the stiff penalties the IRS imposes on those who don’t withdraw funds from these accounts at least as rapidly as the RMD schedule provides. The CRR paper cites as an advantage the reality that those drawdowns are based on the portfolio’s current market value, though surely some people remember how the impact of the 2008 financial crisis on those accounts triggered a more aggressive withdrawal schedule than many found optimal or necessary.

A previous post dealt with another popular drawdown method: the so-called 4 percent rule (see “Withdrawal ‘Symptoms”). As with that 4 percent “rule,” once you stipulate certain assumptions about the length of retirement, portfolio mix/returns, and inflation, those type guidelines are really “just” a mathematical exercise that involves stretching a finite pool of resources over an estimated period of time.

The CRR paper outlines a series of reasons as to why the RMD approach might be superior to alternatives such as the 4 percent rule, but ultimately the biggest shortcoming of the RMD schedule as a basis for withdrawal may be that it fails to take into account how much income is needed, much less when it is needed—and it’s based on a series of assumptions that may or may not apply to an individual’s real-life circumstance.

Of course, in a very real sense, relying on any arbitrary systematic calculation to determine how much, and how fast, to drawdown savings can be seen as a way of living within your means—an approach that can work just fine if you have first made preparations to have adequate means upon which to live.


¹ The CRR report is online here. 

² Information about required minimum distributions is available at the IRS website, online here. IRS worksheets to calculate the amount of the RMD are online here. 

For some interesting data on actual withdrawal rates from individual retirement accounts, see “Withdrawal Symptoms.”

Higher Starting Point Would Significantly Increase Retirement Security “Success”

Setting a higher starting point for 401(k) contributions would make a significant difference in improving workers’ likelihood of a financially viable retirement, according to new research by EBRI.

Most private-sector employers that automatically enroll their 401(k) participants do so at a default rate of 3 percent of pay, a level consistent with the starting rate set out in the Pension Protection Act of 2006, but a rate that many financial experts say is far too low to generate sufficient assets for a comfortable retirement. Raising the default saving rate to 6 percent would significantly increase the chances for achieving retirement adequacy “success” for both low- and high-income workers, EBRI found.

Using its proprietary Retirement Security Projection Model® (RSPM), EBRI evaluated the impact of raising the default contribution rate on younger workers (with 31–40 years of simulated 401(k) eligibility) to see how many would be likely to achieve a to achieve a total income real replacement rate of 80 percent at retirement—within the typical range of replacement rates suggested by many financial consultants.

A key variable in considering the impact of auto-enrollment in 401(k) plans with automatic escalation of contributions is whether workers who change jobs continue to save at their previous (and typically higher) contribution rate, or whether they “start over” in the new job at the typical lower automatic deferral rate of 3 percent.

Jack VanDerhei

Under the EBRI modeling, more than a quarter (25.6 percent) of those in the lowest-income quartile who had previously NOT been modeled to have a financially successful retirement (under the actual default contribution rates) would be successful as a result of the increase in starting deferral rate to 6 percent of compensation. Even workers in the highest-income quartile would benefit, although not as much: Just over 18 percent who would not be successful under the actual default contribution rate would be successful due to the higher 6 percent default rate.

“This study shows that substantial increases in success rates were found for both low- and high-income employees if employers raised the default 401(k) contribution rate to 6 percent of pay,” said Jack VanDerhei, EBRI research director and author of the report.

Full results are published in the September EBRI Notes, “Increasing Default Deferral Rates in Automatic Enrollment 401(k) Plans: The Impact on Retirement Savings Success in Plans With Automatic Escalation,” available online at

Average IRA Balances a Third Higher When Multiple Accounts are Considered

The average IRA balance is about a third higher and the median (mid-point) balance is almost 42 percent larger when multiple individual retirement accounts (IRAs) owned by an individual are taken into account, according to a new report by the nonpartisan Employee Benefit Research Institute (EBRI).

EBRI’s new analysis, based on its unique EBRI IRA Database,™ shows that in 2010 the average IRA individual balance (all accounts from the same person combined) was $91,864, while the median balance was $25,296. By comparison, the average and median account balance of all IRAs was $67,438 and $17,863, respectively. Compared with 2008, the average and median individual balances are up 32 and 26 percent, respectively.

“The results show the importance of being able to look at an aggregation of an individual’s combined account balances to determine the potential total retirement savings he or she has,” said Craig Copeland, EBRI research associate and author of the report. The report provides results for the second year of data available from the EBRI IRA Database.™

The full report is published in the May 2012 EBRI Issue Brief, “Individual Retirement Account Balances, Contributions, and Rollovers, 2010: The EBRI IRA Database,™” online at It analyzes 2010 data from the more than 11 million individuals with more than $1 trillion in the EBRI IRA Database™ and highlights the distribution of IRA owners by IRA types, account balances, rollovers, and contributions to IRAs. A unique aspect of the EBRI IRA Database™ is the ability to link the balances of individuals with more than one account in the database, providing a more complete picture of their IRA-based retirement savings.

The press release is online here. The full report is online here.

Titanic Proportions

By Nevin Adams, EBRI


This weekend marks the 100th anniversary of the sinking of the now iconic RMS Titanic, at the time the world’s largest ocean liner. Its passengers included some of the wealthiest people in the world, as well as a large number of emigrants seeking a new life in North America. On the ocean liner’s maiden—and only—voyage, it carried 2,244 people, 1,514 of whom would perish as a result of a decision to carry only enough lifeboats to accommodate about half those on board. Despite that, the tale of Titanic’s closing hours is generally one of an orderly, “women and children first” evacuation, with the band playing on while passengers stood in line and waited their turn.

In an NPR report this week titled “Why Didn’t Passengers Panic on the Titanic?” David Savage, an economist and Queensland University in Australia, compared the behavior of the passengers on the Titanic with those on the Lusitania, another ship that also sunk at about the same time. Both involved luxury liners, both had a similar number of passengers and a similar number of survivors. But on the Lusitania, passengers panicked—and the survivors were mostly those who were able to swim and get into the lifeboats. The biggest difference in the reactions in these two similar circumstances, Savage concludes in the report, was time: The Lusitania, struck by a U-Boat torpedo, sank in less than 20 minutes, while the Titanic took approximately two and a half hours. Time enough, in the case of Titanic, according to Savage, for social order to prevail over “instinct.”

I’ve not yet seen anyone link the nation’s retirement prospects to the Titanic, though the importance of starting to save early, establishing a plan, and having a goal all fit within the moral of that example: Act while you have time, as you may not always have all the time you need, or the resources to take advantage of it. But what kind of time do Americans have?

The concept of measuring retirement security—or retirement income adequacy—is an extremely important topic, and EBRI has been working with this type of measurement since the late 1990s. When we modeled the Baby Boomers and Gen Xers in 2012, approximately 44 percent of those households were projected to have inadequate retirement income for basic retirement expenses plus uninsured health care costs; so the glass is more than half full, but just a bit. That’s a big percentage, but it’s still an improvement of 5‐8 percentage points over what we found in 2003, when the glass was, indeed, half empty. The improvement occurred despite the impact of the 2008 financial crisis, and the subsequent “great recession.”

In part, that is because the passage of time allowed more funds to be saved, but that improvement is largely due to the fact that in 2003 very few 401(k) sponsors had automatic enrollment (AE) provisions in place, and participation rates among the lower-income workers (those most likely to be at risk) was quite low. However, with the uptick in adoption of AE the past few years after passage of the Pension Protection Act (PPA), participation rates have often increased to the high 80 or low 90 percent—and that stands to make a big difference in shoring up the retirement financial security of many of those who need it the most. In fact, EBRI simulation results show that approximately 60 percent of the AE-eligible workers would immediately be better off in an AE plan than in a plan design relying on voluntary enrollment, and that over time (as automatic escalation provisions took effect for some of the workers) that number would increase to 85 percent.

The tragic loss of life on the Titanic is generally attributed to a shortage of lifeboats, while on the Lusitania, it was the lack of time to get people into the lifeboats available. For many of those looking ahead to retirement, plan design changes such as automatic enrollment and contribution acceleration look to be a real retirement life-saver, as they encourage both early action and the effective use of time.


While the lack of retirement income adequacy for the lowest-income households is a matter of concern, so is the rate at which they will run “short” of money during retirement. Indeed, as documented in our July 2010 Issue Brief (“The EBRI Retirement Readiness Rating:™ Retirement Income Preparation and Future Prospects”), 41 percent of early boomers in the lowest income quartile could run short of money within 10 years after they retire.

More information is included in recent testimony by Jack VanDerhei, EBRI research director, before the Senate Banking Committee, on “Retirement (In)security: Examining the Retirement Savings Deficit” (T-171), available online here.

“Good” Vibrations

By Nevin Adams, EBRI


Last week the Employee Benefit Research Institute (EBRI) and Mathew Greenwald & Associates, Inc., unveiled the 22nd annual Retirement Confidence Survey (RCS).

Among the things we have learned after doing this survey for more than two decades: People’s confidence about retirement frequently seems out of line with the financial resources they indicate they have on hand to fund it. Of course, most (56%) of this year’s respondents admit neither they nor their spouse have made even a single attempt to determine how much they need to achieve that comfortable retirement—so it shouldn’t be too surprising that, asked how much they think they need to have saved in order to provide for a comfortable retirement, many hold forth a number that seems lower than some might expect.

While we spent a fair amount of time this week discussing the results with reporters, one question that came up repeatedly was “Why do you do this survey?  What do you hope people take from it?”

The survey itself is meaningful both for the kinds of issues it deals with and the trends it measures: Questions that, as in this year’s RCS, deal not just with confidence as a “feeling” but also the criteria that underlie and influence that sentiment. It looks at the perspective both of those already in retirement, as well as those still working and heading toward that milestone. It also (with a perspective based on two decades of conducting this particular survey) offers insights on how those feelings and factors have changed over time.

Those good reasons notwithstanding, this past week EBRI reminded reporters that the RCS has found that people who have taken the time to do a retirement needs assessment are generally more confident than those who haven’t done so, and not necessarily because they find that they are in better shape than they’d thought. In fact, most report that they set higher savings goals AFTER they had done the assessment—and were THEN more confident in their situation. That is why EBRI joined many others in 1995 to establish the American Savings Education Council (ASEC), and then the ChoosetoSave® program and the BallparkE$timate.®  Millions of Americans have used the BallparkE$timate® at to help them climb the hill to savings and greater financial security, and—according to the RCS—a more realistic view of the future.

There’s something to be said for knowing the size and extent of what was previously unknown, particularly when it comes to setting a financial goal as complex as planning for retirement can seem.

If the annual publication of the RCS does no more than remind individuals of the importance of taking the time to do so, then it’s not only good information—it’s information that does some good.

Full results of the 2012 Retirement Confidence Survey (RCS), along with the press release and seven related RCS Fact Sheets,  are now available online here. 

“Managing” Expectations

By Nevin Adams, EBRI


On March 13,(1) the Employee Benefit Research Institute (EBRI) and Mathew Greenwald & Associates, Inc., will unveil the 22nd annual Retirement Confidence Survey (RCS)—the longest-running annual retirement survey of its kind in the nation. Indeed, the RCS is unique in offering a perspective on retirement that is now as long as the retirement at age 65 of those living to average life expectancy.

Consider that back in 1996, the sixth annual RCS found that 24 percent of retirees were not confident that they would have enough money to live comfortably throughout their retirement years, and more than 1 in 5 said their lifestyles were worse than when they first retired (with nearly 1 in 10 calling them “a lot worse”). The report noted that “[t]wo-thirds of those working then predicted they would work after they ‘retire,’ and nearly 40 percent of those say they think they’ll need to for financial reasons, to pay the bills and make ends meet.”

Five years later, the 2001 RCS  noted that the percentage of individuals who say they have personally saved for retirement decreased from 75 percent in 2000 to 71 percent, though that was still better than the 59 percent cited in 1998. At the time, the RCS noted that the “changes in individual behavior regarding retirement savings may in part be attributed to recent declines in consumer confidence, employment, the economy, and the equity markets.”

While a press release about the 2006 RCS stated that “RCS data over the past 12 years continue to show that retirement confidence overall among workers does not seem to be affected by either stock market performance or varying economic conditions,” subsequent events—notably the 2008 financial crisis—did seem to undermine confidence levels.

The 2010 RCS acknowledged that Americans’ confidence in their ability to afford a comfortable retirement had plunged to a new low at the same time that the recent declines in other retirement confidence indicators appeared to be “stabilizing.” And yet, just one year later, the 2011 RCS cautioned, “Instead of making fundamental adjustments to their spending and saving patterns in response to the decline in confidence, workers continue to change their expectations.”

How we view—and anticipate—retirement can have a dramatic impact on that reality, and the RCS provides valuable insights into the perspectives of those heading toward, and those already dealing with, the realities of retirement.

What matters, of course, isn’t one’s confidence about having a financially secure retirement. What matters is having taken the time and energy to actually do something about it.


(1) Full results of the 2012 RCS will be available online at the morning of Tuesday, March 13.

“Difference” Strokes

By Nevin Adams, EBRI


In response to concerns that tomorrow’s retirees will run short of money, we are often told to save more, to work longer, or—as often as not these days—to work longer AND save more. Certainly working and saving longer can do wonders in terms of stretching your retirement nest egg.

However, the timing of the retirement decision is often not within an individual’s control. In fact, the Retirement Confidence Survey has consistently found that a large percentage of retirees leave the work force earlier than planned. In fact, nearly half (45 percent) of retirees reported that they were in this situation in 2011 (see EBRI Issue Brief No. 355, March 2011, “The 2011 Retirement Confidence Survey: Confidence Drops to Record Lows, Reflecting ’the New Normal.’”

Real-world data have shown (and shown for some time now) that the median retirement age for Americans is not even as old as 65 (it’s been 62). Still, EBRI research has shown that working longer—even working past the age of 65—is no guarantee of a financially comfortable retirement.

In fact, a June 2011 EBRI Issue Brief titled “The Impact of Deferring Retirement Age on Retirement Income Adequacy” notes that, even if a worker delays his or her retirement until age 80, just 61.7% of the lowest preretirement income quartile households would have a 50 percent probability of not running out of money in retirement.

What Matters

The research notes that how workers fare financially after retirement is directly tied to three factors: their salary level at retirement, how long they work beyond 65, and whether they save in a defined contribution retirement plan during their working lifetime. In fact, the report notes that “a major factor that makes a difference” in their ability to meet basic and uninsured health-care costs in retirement is “whether they are still participating in a defined contribution plan after the age of 65.” How much difference? At least a 10 percentage point difference in the majority of the retirement age/income combinations.

Ultimately, the research should remind us of a couple of things: First, that the assumption that we’ll be able to work past “normal” retirement age is just that—an assumption. Second, and more important, even if that assumption pans out, it cannot be assumed that it will, in and of itself, prove to be sufficient.

But finally, and significantly, there is at least one thing individuals can exercise some control over in the “here and now”: their current—and continued—participation in a defined contribution plan.

And that’s something that can directly—and significantly—make a difference in building a financially viable retirement.


(1) Admittedly, except for those in the lowest- income quartile, this would be a small percentage of the population, depending on your expectations of success (see “Short” Comings). Still, according to the EBRI Retirement Security Projection Model (RSPM) baseline results, the lowest preretirement income quartile would need to defer retirement age to 84 before 90 percent of the households would have a 50 percent probability of success.

You can read more about how these factors impact retirement income adequacy online here.

Goals Tending

By Nevin Adams, EBRI


The bad news is that times are still tough for many Americans—and surveys suggest that even those with jobs are nervous about their prospects for the future. The good news is that the current level of economic uncertainty seems to have brought about—at least for some—a heightened awareness of the need to set money aside for a rainy day, perhaps even those rainy days in retirement.

That said, the weak economy has certainly constrained the ability of many to save. In fact, a recent national survey found that an increasing number of Americans are having difficulty saving to meet goals ranging from meeting emergencies to affording retirement. The survey—released as part of America Saves Week(1)—noted that over the past three years, there has been a decline in the number of people who spend less than their income and save the difference, are building home equity, have adequate emergency savings, and think they are saving enough for retirement.

However, the survey also revealed that having a savings plan has beneficial financial effects, even for lower-income families. Consider that 85% of those who had a savings plan spent less than their income, (86% of this group felt they had sufficient emergency savings), and 3 in 4 said they were saving enough for retirement.

On the other hand, just 44% of those without a savings plan claimed to be spending less than their income, while 43% said they had sufficient emergency savings. Fewer than 1 in 4 of those with no savings plan said they were saving enough for retirement.

And yet, while those with a plan for savings had significantly better savings behaviors, fewer than half of survey respondents said they had “a savings plan with specific goals.”

It is admittedly simplistic to chasten those truly unable to save because of challenging economic circumstances. On the other hand, there are surely some today better able to weather those economic storms because they chose to set money aside during less tumultuous times.

That first step on that path, and it’s a critical one, is to Choose to Save.®  As America Saves Week reminds us—and those trying to help others save—there’s no better time to start on that path to Save For Your Future® than today.

Note: Organizations interested in building/reinforcing a workplace savings campaign can find free resources—and a handy schedule of events around which to construct a program—courtesy of the American Savings Education Council (ASEC).


(1) America Saves Week is an annual event where hundreds of national and local organizations promote good savings behavior and individuals are encouraged to assess their own saving status. America Saves Week is managed by the Consumer Federation of America (CFA), and the American Savings Education Council (ASEC). ASEC is a program of the Employee Benefit Research Institute (EBRI), which commissioned the survey, undertaken by Opinion Research Corp. You can find out more about America Saves Week.

(2) Choose to Save® is sponsored by the nonprofit, nonpartisan Employee Benefit Research Institute Education and Research Fund (EBRI-ERF) and one of its programs, the American Savings Education Council (ASEC). The website and materials development have been underwritten through generous grants and additional support from EBRI members and ASEC Partner institutions.