What do you think of our site upgrade?
Hosting an industry conference? Ask us about including it in this ticker?

Exclusive Interview with Alicia Munnell: Auto-Enrollment Without Auto-Escalation Reduces Deferral Rates

Exclusive Interview with Alicia Munnell: Auto-Enrollment Without Auto-Escalation Reduces Deferral Rates
February 17
00:03 2016

We’ve had the opportunity to interview nationally recognized retirement researcher Alicia Munnell on several occasion during the course of our normal reporting. It was an honor for us, then, when she graciously agreed to participate in our monthly Exclusive Interview series. Alicia H. Munnell is the Peter F. Drucker Professor of Management Sciences at Boston College’s Carroll School of Management. She also serves as the Director of the Center for Retirement Research at Boston College. Before joining Boston College in 1997, Professor Munnell was a Member of the President’s Council of Economic Advisers (1995-1997) and Assistant Secretary of the Treasury for Economic Policy (1993-1995). Previously, she spent 20 years at the Federal Reserve Bank of Boston (1973-1993), where she became Senior Vice President and Director of Research in 1984.

FN: Our readers are always interested in the back story of our interview subjects. How did you end up where you are today? What triggering experiences had the most impact in your career choice? If you had the choice, is there anything you would do different?
Munnell: I majored in economics by default; worked for men who had daughters; and really liked the idea of tackling issues involving big bucks, which led me to study retirement income programs like Social Security and employer pensions.

FN: Last fall you co-published a report (“How Has Shift To Defined Contribution Plans Affected Saving?” Center for Retirement Research at Boston College, September 2015, Number 15-16), that dramatically changed your perception regarding the success of the 401k plan. Tell us a little about the “before” and “after” of this story. What were your perceptions about defined contribution plans prior to conducting the research on this paper? What were the specific research points that prompted you to reconsider those earlier thoughts? How can others – primarily practitioners – use this data to help people save for retirement?
Munnell: Our thought going into the study was that the shift from defined benefit to defined contribution plans had reduced the amount of total saving that was occurring in the employer-sponsored retirement system. But we never had seen any data to support or refute this hypothesis. Recent changes to the government’s National Income and Product Account database provided an opportunity to shed some light on the issue. Our main finding was that the shift to defined contribution plans appears to have not resulted in less total saving. But it is quite possible that the shift to defined contribution plans has led to distributional differences in the accumulation of retirement assets – for example, lower income workers could be saving less than higher income workers than under the traditional defined benefit system. This question is one that we are currently examining. One last thing – one clear-cut effect of the switch to a defined contribution system is that individuals, rather than employers, now bear all of the risk of providing for their retirement.

FN: In October of last year I spoke to Teresa Ghilarducci following her keynote at a forum of business writers and editors. The positions she advocated there seemed to be in contradiction with the conclusions from your report. I asked her if she was aware of your (then very recent) publication and she said she wasn’t. Have you had a chance to speak to her about your report? How would you contrast her position with the findings of your report?
Munnell: Teresa and I are in contact on various topics. In fact, we had a lovely lunch together recently in Rome, but we have not had a specific conversation about that study. Note that the study doesn’t change anything about the retirement challenge that today’s workers are facing, an issue that both Teresa and I care deeply about. Individuals will need more retirement income due to improving longevity, growing health care costs, and low interest rates. And our major sources of retirement income will not provide enough. Social Security replacement rates are actually shrinking and 401k balances are far too small to allow many to maintain their pre-retirement standard of living when they retire. In addition, at any given point in time, about half of private sector workers are not participating in any employer-sponsored retirement savings plan.

FN: You recently appeared before the Senate committee hearing on retirement plans and made some very specific suggestions. Can you enumerate those again for our readers?
Munnell: Yes. First, I indicated my general support for the efforts of the Committee’s bipartisan working group that has proposals to help reduce the pension coverage gap and encourage saving among lower income workers. However, much bolder changes are needed to address the retirement saving shortfall, so I offered two specific changes of my own: 1) require 401k plans to adopt auto-enrollment and auto-escalation, retaining the ability for workers to opt-out of these arrangements; and 2) enact auto-IRA legislation at the national level to provide coverage to those workers who currently lack it.

FN: While the idea of a “default” auto-enrollment provision in all plans is sound, not all employees have the same financial needs as others. This being the case, would you advocate the standard “opt-out” choice remains or do you see a need to make opting out only available on a “hardship” basis?
Munnell: Yes, as noted in my previous answer, my proposal would definitely allow individuals to opt out. The idea of auto-enrollment is to nudge individuals toward participating without requiring them to do so.

FN: You’re proposed auto-enrollment number is 6%, much high than the average default rate (and still considerably lower than the target deferral advocated by many financial advisers). What is the percentage deferral that you think employees should target? How does default auto-escalation policy help employees get there and what percentage increase should that be set at?
Munnell: Yes, 6 percent is just my suggested starting point. I believe that it is absolutely necessary to have auto-escalation in the default contribution rate until the combined employee contribution and employer match reaches 12 percent of a worker’s salary. Auto-enrollment without auto-escalation would actually reduce contribution rates, as research has shown.

FN: What are the three best improvements you’ve seen in the defined contribution marketplace in the last 10-15 years?
Munnell: 1) The adoption of auto-401k plans by some firms; but this positive development has not spread far enough and seems to have stalled out in recent years; 2) The Labor Department’s focus on investment fees, along with some moves within the industry to find lower-cost ways to invest retirement assets, have helped to bring fees down, although average fees are still too high; and, 3) The declining share of assets invested in company stock.

FN: What are the two most important improvements you feel need to be implemented in the defined contribution marketplace in the next 10-15 years?
Munnell: 1) Auto-enrollment and auto-escalation in all 401k plans, covering both existing workers and new hires; and, 2) A reduction in leakages, which requires some tightening up by the federal government such as not allowing cash-outs when people change jobs and raising the age for penalty-free withdrawals above 59½

FN: Turning the conversation to defined benefit plans, what do you see as the best way to address the looming Social Security Crises? How do you think a “Child IRA” might best be utilized as a way to fundamentally change the role of government in retirement planning? [Editor’s Note: If you’re not familiar with the Child IRA, you can read about it here: and here:]
Munnell: The best way to address the Social Security challenge is to increase the revenue going into the system, which means some combination of higher payroll tax rates, an increase in the payroll tax cap, and perhaps more unorthodox ideas like transferring the program’s legacy cost burden to the income tax or investing a portion of Social Security’s reserves in equities. The reason for focusing on the revenue side is that benefits are already modest and replacement rates are declining under current law, so further reductions would only make the retirement security challenge worse.

With respect to the “Child IRA,” the idea of starting to save very early for retirement is a fine notion. Realistically, though, very few Americans save for such purposes on their own. My sense is that only a fraction of people would take much advantage of a Child IRA. Now if you were to couple the Child IRA with some sort of government match and auto-enrollment, that could change things.

FN: Can you give us a sneak preview of your latest research?
Munnell: As I mentioned earlier, we are currently looking at the distribution of defined benefit (DB) and defined contribution (DC) wealth by income level to see if DC wealth is more skewed to higher income individuals than DB wealth.

FN: Where can readers go to learn more about your research?
Munnell: Everything is on the website of the Center for Retirement Research at Boston College (

FN: Is there anything else you would like to add?
Munnell: Our nation is really facing a retirement income crisis, so it is necessary to act soon to avert it. 

FN: Professor Munnell, thank you very much for taking the time to share your thoughts and ideas with our readers. We continue to be impressed with your depth of research and encourage our readers to continue to watch for more news on that front by going to your web-site.

About Author

Christopher Carosa, CTFA

Christopher Carosa, CTFA


  1. Jack Towarnicky
    Jack Towarnicky February 18, 11:17

    You quote her as saying: “Auto-enrollment without auto-escalation would actually reduce contribution rates, as research has shown.” Just not true. Auto enrollment compliance requires the individual be given the opportunity to make an election, then where she/he does not make an election, they are defaulted to a specific percentage. So, when people say they would have contributed more, they had the chance and failed to act. It is this failure at execution, at enrolling when first eligible, that auto enrollment was specifically designed to counter. I have yet to see the survey that asks those who assert they would have saved more, well, why didn’t you act when provided the opportunity?

    More importantly, while the contribution rate patterns of plans with auto enrollment and plans without auto enrollment are very different, there is no doubt that saving is greater in plans with automatic enrollment – if you look at savings (and not contribution rates as a percent of pay), because savings includes the 0’s (such as by comparing ADP before and after introduction of automatic enrollment).

    I agree with the professor that the default contribution percentage should be higher – I favor the greater of 6% or the percentage of pay needed to obtain the maximum company match. Unlike most, however, I believe automatic enrollment and automatic escalation should be applied perennially, to all workers who are not saving at a defined target percentage (I favor a total of 15% of pay). We implemented such a program of perennial auto enrollment and auto escalation back in 2007. The results are that average participation has consistently been 95+%, and average deferral percentages of non-highly compensated employees have significantly increased.

    Finally, I also believe employers should adopt Roth 401(k) provisions. That would force them to make a call on defaulting to traditional pre-tax 401(k) or Roth 401(k) – perhaps varying the default based on specific participant circumstance. Remember, many of those who are defaulted are younger, lower paid, shorter service employees – meaning that Roth 401(k) may be the better alternative because:
    (1) Participants may well be in the lowest tax bracket they will experience in their working careers, and
    (2) Participants are likely to change employers a number of times (EBRI reports median tenure relatively unchanged at between 5 – 6 years, based on surveys over the past 40 years), and, where leakage occurs, any early withdrawal penalty tax on Roth 401(k) contributions will likely be less than on regular, pre-tax 401(k).

  2. Dennis Myhre
    Dennis Myhre February 18, 11:51


    Good interview, but the focus is on auto-enrollment/auto-escalation of 401k plans, when the real problem is fiduciary abuse and self dealing. Her comment having the most merit is that “one clear-cut effect of the switch to a defined contribution system is that individuals, rather than employers, now bear all of the risk of providing for their retirement.” This is true, and the government must develop standards that preclude other interested parties from stripping away the hard earned funds owned by those individuals that bear “all of the risk.”

    It is no secret that proprietary funds offered to retirement plans carry a high risk of failure for reasons of self dealing and cross trading by insurance companies, and the new DOL fiduciary standard, if enacted, will be the first step in a long effort to resolve this issue.

    I would ask Ms. Munnell to offer her insight in bringing fraud and corruption in the retirement scheme under control first, then address enrollment issues.

    Dennis Myhre, AIC

  3. Christopher Carosa, CTFA
    Christopher Carosa, CTFA Author February 19, 21:44

    Jack: Here’s the info on the quote you questioned:
    “401(k)/IRA Holdings in 2013: An Update from the SCF”

    “The median contribution rate in 2013 was 9.2 percent – roughly 6 percent by the employee with a 50-percent employer match (see Figure 7). Interestingly, the contribution rate declined slightly after the passage of the PPA, when employers began to auto-enroll workers into the plan, typically with a 3-percent default deferral rate. Since only 40 percent of plans with auto-enrollment have auto-escalation in the default contribution, many of those who are enrolled at low contribution rates remain at those rates.”

  4. Jack Towarnicky
    Jack Towarnicky February 29, 08:28

    Chris, as you state, the data you cite compares the median contribution rate. It doesn’t compare actual savings rates, but contribution rates.

    Show me the comparative savings rates – including the zeroes in your averages, before and after adoption of automatic enrollment – before you (others) embrace a suggestion that auto enrollment is flawed. Show me the survey data that confirms why individuals did not enroll at a higher percentage when approached during the enrollment process.

    When we implemented auto features (enrollment, QDIA, escalation, etc.), we saw significant increases in average contribution/average deferral rates – adding significantly to employer matching contribution costs. I have never seen even one study that remotely suggests that auto enrollment (even applied without escalation) negatively impacts average savings rates. The only circumstance where that could happen is where a plan already had active participation of > 80% or 90%, and an average deferral rate of > 6% of pay, and then introduced an auto enrollment (without escalation) with a 3% default. Is there such a plan out there?

    Again, as I mentioned in my note, I agree with the professor that plan sponsors should implement higher default rates, should include escalation, and unlike most, I also recommend that the auto features be applied perennially – forcing people to make a decision each and every year if they are not saving at an arbitrarily selected target percentage (again, I favor 15%).

Only registered users can comment. Login is sponsored by…

Order Your 401k Fiduciary Solutions book today!

Vote in our Poll


The materials at this web site are maintained for the sole purpose of providing general information about fiduciary law, tax accounting and investments and do not under any circumstances constitute legal, accounting or investment advice. You should not act or refrain from acting based on these materials without first obtaining the advice of an appropriate professional. Please carefully read the terms and conditions for using this site. This website contains links to third-party websites. We are not responsible for, and make no representations or endorsements with respect to, third-party websites, or with respect to any information, products or services that may be provided by or through such websites.