Research
August 23, 2012
401(k) Tax Break May Be Weak Incentive
Typical American households approaching retirement age had just $42,000 saved in 401(k)s and IRAs in 2010. This raises the question: Does the federal tax incentive designed to spur savings even work?
In what one retirement expert called “landmark” research, a new study has found that employers’ automatic enrollment and other employee mandates are far more effective ways to increase retirement savings than the federal tax exemption granted for retirement-fund contributions.
Harvard University Professors Raj Chetty and John Friedman, together with Soren Leth-Petersen and Torben Nielsen at the University of Copenhagen, tested the impact of both types of incentives on an enormous sample of 4.3 million people in Denmark. Chetty said the findings also hold implications for the United States.
They found that every $1 increase mandated for retirement savings – in this case, by a temporary Danish policy that required workers to contribute 1 percent of their earnings to government pension savings accounts – spurred 86 cents in additional savings by individuals. In contrast, the Danish government’s tax subsidy, which is very much like our own 401(k) tax break, spurred only 20 cents more in savings.
“This is a landmark study,” Dartmouth College professor Jonathan Skinner said about the paper, presented during the Retirement Research Consortium’s conference in Washington in early August.
The savings boost spurred by the Danish savings mandate, in effect from 1998 through 2003, is analogous to U.S. employer programs with automatic features, Chetty said. Such programs might automatically enroll workers in a 401(k) or raise employees’ savings as their earnings increase.
Some 42 percent of U.S. employers now automatically enroll employees in 401(k)s, giving them the option of dropping out. Research has also shown that automatic enrollment is more effective than trying to push employees to actively sign up for these savings programs.
To test the impact of tax subsidies, the researchers examined a different government policy: a 15-percent reduction in the tax break for high-income Danes who contributed to their capital pension retirement savings accounts – what would amount roughly to a partial reversal of 401(k) tax breaks in the United States. They found that contributions by high earners fell sharply when the tax break was reduced but that individuals were simply moving those funds to other types of savings accounts. In other words, saving overall was not very responsive to the tax incentive.
It’s important to note that this focus on “total savings” is critical to the researchers’ findings. They wanted to measure a change in savings in all of an individual’s various accounts, rather than savings that merely shifted among accounts in response to each policy.
“[T]he impacts of this [tax subsidy] reform on total savings are very limited,” the paper said. Chetty cautioned that further research, using data from the United States, is needed before one can draw definitive conclusions for U.S. policy, but the Denmark findings suggest that tax subsidies may not be the most effective way to raise retirement savings in this country.
At the research consortium, Chetty previewed the results of a forthcoming working paper, which is expected to go live on the National Bureau of Economic Research’s website in a few months. The Center for Retirement Research, which sponsors this blog, is a consortium member.
Full disclosure: The research cited in this post was funded by a grant from the U.S. Social Security Administration (SSA) through the Retirement Research Consortium, which also funds this blog. The opinions and conclusions expressed are solely those of the blog’s author and do not represent the opinions or policy of SSA or any agency of the federal government.
I continue to think that the Chetty et al. article is a landmark study, but I don’t think that it tells us anything about the effectiveness of the U.S. 401(k) program, as suggested above. The policy change in Denmark removed one flavor of retirement saving but left the other one intact — hence, no surprise that savvy Danes simply switched over to the other retirement saving plan.
This policy change was roughly equivalent to closing down Roth IRAs in the U.S. We wouldn’t be surprised that, as a consequence, people would move their Roth IRA money into traditional IRAs, with no impact on overall saving or tax revenue. But we can’t conclude from such a policy experiment that 401(k) tax incentives are ineffective.
I really cannot see how this could be considered comparable. High earners are not effected by the [savers] tax break. When referring to a regular 401(k) vs. a Roth 401(k) again, higher earners are not the ones who need to matter as much when considering retirement savings. Sure they do need to save more, but they also have more options than lower earners, and will always find the most tax advantaged ways to save for retirement.
The tax benefit is critical to enticing employers to offer plans, especially for owner-managers of small businesses. The Danish study underscores this. Drop the incentive and decision-makers will move to other venues but in the U.S. that plan termination would accompany that change. Sponsoring a plan in the U.S. takes significant time and imposes fiduciary liability on decision-makers. Many would not take this on without a benefit.
I’ve always said that the most important reason to join a 401(k) is you save before you spend. Automatic features help that.
Tax deferral matters the more you make, so many of the people who find the deferral meaningful are also highly compensated employees who have their contributions capped, and have most of their retirement money someplace else.
Mr. Miller’s comments are important. Even if the tax benefit did not induce people to save any more money for retirement, the money accumulates much more rapidly in a tax deferred account. At 8% earnings for 40 years, the accumulation from tax deferral more than doubles the after tax terminal amount.
Chetty, Friedman, Petersen and Nielsen have asked some important questions in an effort to explain savings behaviors. They took a logical approach, comparing the impact of mandatory contributions versus tax policy.
I would like to suggest that they look more carefully at tax policy and its impact. I would counter their argument by saying that the tax impact on savings behavior is not clearly understood by individuals. As a result, when individuals make decisions about the use of their hard earned income, they do not calculate the tax benefits of saving through a 401(k).
Over the last decade in my work as a financial advisor, it is clear that most middle class and upper middle class income earners do not even think about tax breaks when they decide how much to contribute to their 401(k). While mortgage interest deductions are very salient to individuals, the tax impact of 401(k) contributions are not noted, nor I believe, understood as clearly. In fact, many people overestimate the tax impact of mortgage interest deductions (as a recent researcher found in his contrast of mortgage interest and charitable deductions). So the mortgage interest tax break influences middle and upper class behavior quite clearly (as seen in the high rate of home ownership). By contrast, the lower salience of the retirement savings tax breaks do not seem to do so.
I would suggest that this line of inquiry be considered in formulating subsequent research. In my role as advisor, I have explained the positive tax impact to countless people—from spouses of high-paid lawyers down to single individuals who earn $50,000. They do not believe it. I have to repeatedly point out the dual positives: reduction in their tax payments plus the increase in their Net Worth.
The mortgage industry has clarified the tax impact of mortgage interest deductions. Perhaps retirement vendors should clarify the immediate tax impact of 401(k) contributions. They can demonstrate how an increase in 401(k) contributions can reduce taxes as much if not more than mortgage interest deductions. As a result, they may help increase retirement savings among income earners who are not part of the mandatory savings plans.
One additional area for possible further investigation is automatic enrollment in the U.S. federal government’s Thrift Savings Plan (TSP), which as you may know is the government’s version of a 401(k).
While the TSP celebrated its 25th birthday earlier this year, automatic enrollment in the TSP began two years ago in August 2010, and so far 97% or so of automatic enrollees since then continue to invest in the TSP — the total number of participants since then is approaching 300,000, according to the Federal Retirement Thrift Investment Board (FRTIB) figures at the end of May. Perhaps not quite the 4.3 million cited in the Denmark study, but in total there are about 4.5 million TSP account holders including uniformed service personnel and former government employees who have left service, so there is a sizable sample to compare and contrast among participants, if anything. To contrast the automatic enrollment of FERS civilian employees, for example, uniformed service personnel are not automatically enrolled, and the participation rate is only about 40% since service members were first allowed to participate in the TSP in 2002.
At any rate, please keep up the interesting posts in “Squared Away.” I enjoy reading what you and BC’s Financial Security Project have to say.
The percentage of companies eligible companies for 401(k) style programs is continuing to grow rapidly. The percentage of people who participate is continuing to grow rapidly as well. Add to that the growing percentage of companies that are instituting or increasing company contributions, and there is more than enough evidence to support this retirement savings option. In fact, reliable research indicates increased employee contributions, and satisfaction as time (and balances) grow.
Also, the growing lack of confidence is Social Security as a retirement “hope” makes more and more people want to find other ways to control their own retirement options, with less reliance on government. I would not be surprised to see serious effort to “re-invent” the self-controlled Social Security Investment Account in the near future, particularly if Romney is elected.
There is some thinking that Retirement as a concept has lost value; a “second life” either as a self-fulfillment goal, or as a second career (with much less stress and pressure) is becoming a real possibility, IF saving for that period in time were to lead to economic independence at a much earlier age, say 50 to 55, leaving time to pursue leisure while health still allows, and providing significant incentive for much more savings to achieve the goal.
Some people do not want to have taxes, but taxes might also be favorable. There are retirement plans that charge you through your taxes so that you can get a pension when you are older.