[Joint House and Senate Hearing, 108 Congress]
[From the U.S. Government Publishing Office]



                                                        S. Hrg. 108-532



                         HELPING AMERICANS SAVE

=======================================================================

                                HEARING

                               BEFORE THE

                        JOINT ECONOMIC COMMITTEE

                     CONGRESS OF THE UNITED STATES

                      ONE HUNDRED EIGHTH CONGRESS

                             SECOND SESSION

                               __________

                             MARCH 10, 2004

                               __________

          Printed for the use of the Joint Economic Committee


                    U.S. GOVERNMENT PRINTING OFFICE
93-761                      WASHINGTON : DC
____________________________________________________________________________
For Sale by the Superintendent of Documents, U.S. Government Printing Office
Internet: bookstore.gpo.gov  Phone: toll free (866) 512-1800; (202) 512�091800  
Fax: (202) 512�092250 Mail: Stop SSOP, Washington, DC 20402�090001

                        JOINT ECONOMIC COMMITTEE


    [Created pursuant to Sec. 5(a) of Public Law 304, 79th Congress]


SENATE                               HOUSE OF REPRESENTATIVES
Robert F. Bennett, Utah, Chairman    Jim Saxton, New Jersey, Vice 
Sam Brownback, Kansas                    Chairman
Jeff Sessions, Alabama               Paul Ryan, Wisconsin
John Sununu, New Hampshire           Jennifer Dunn, Washington
Lamar Alexander, Tennessee           Phil English, Pennsylvania
Susan Collins, Maine                 Adam H. Putnam, Florida
Jack Reed, Rhode Island              Ron Paul, Texas
Edward M. Kennedy, Massachusetts     Pete Stark, California
Paul S. Sarbanes, Maryland           Carolyn B. Maloney, New York
Jeff Bingaman, New Mexico            Melvin L. Watt, North Carolina
                                     Baron P. Hill, Indiana



        Donald B. Marron, Executive Director and Chief Economist
                Wendell Primus, Minority Staff Director


                            C O N T E N T S

                              ----------                              


                      Opening Statement of Members

                                                                   Page

Senator Robert F. Bennett, Chairman..............................     1
Representative Jim Saxton, Vice Chairman.........................     3

                               Witnesses

Statement of Dr. Richard H. Thaler, Robert P. Gwinn Professor of 
  Behavioral Science and Economics, Graduate School of Business, 
  University of Chicago; Research Associate, National Bureau of 
  Economic Research..............................................     4
Statement of Robert C. Pozen, Non-Executive Chairman, MFS 
  Investment Management, John M. Olin Visiting Professor of Law, 
  Harvard Law School.............................................     8
Statement of Ric Edelman, Chairman, Edelman Financial Services...    11
Statement of Dr. Peter R. Orszag, Joseph A. Pechman Senior Fellow 
  in 
  Economic Studies, The Brookings Institution....................    14

                       Submissions for the Record

Prepared statement of Senator Robert F. Bennett, Chairman........    31
Prepared statement of Representative Jim Saxton, Vice Chairman...    33
Prepared statement of Representative Pete Stark, Ranking Minority 
  Member.........................................................    34
Prepared statement of Dr. Richard H. Thaler, Robert P. Gwinn 
  Professor of Behavioral Science and Economics, Graduate School 
  of Business, University of Chicago; Research Associate, 
  National Bureau of Economic Research...........................    36
Prepared statement of Robert C. Pozen, Non-Executive Chairman, 
  MFS 
  Investment Management, John M. Olin Visiting Professor of Law, 
  Harvard Law School.............................................    41
Prepared statement of Ric Edelman, Chairman, Edelman Financial 
  Services.......................................................    49
Prepared Statement of Dr. Peter R. Orszag, Joseph A. Pechman 
  Senior Fellow in Economic Studies, The Brookings Institution...    55

 
                         HELPING AMERICANS SAVE

                              ----------                              


                       WEDNESDAY, MARCH 10, 2004

                    United States Congress,
                          Joint Economic Committee,
                                                     Washington, DC
    The Committee met, pursuant to notice, at 10:00 a.m., in 
room SD-628 of the Dirksen Senate Office Building, the 
Honorable 
Robert F. Bennett, Chairman of the Committee, presiding.
    Members Present: Senator Bennett, Representative Saxton.
    Staff Present: Donald Marron, Ike Brannon, Brian Jenn, Mike 
Ashton, Colleen J. Healy, Chris Frenze, Robert Keleher, Jason 
Fichtner, Wendell Primus, Chad Stone.

        OPENING STATEMENT OF SENATOR ROBERT F. BENNETT, 
                            CHAIRMAN

    Chairman Bennett. The hearing will come to order. I'm told 
that there is a vote on the Floor of the House in 15 minutes, 
and so we will get started right on time, even though the crowd 
is a little slow in gathering.
    But we want to accommodate the Members of the House who are 
here, and I appreciate Vice Chairman Saxton coming over, and we 
will hear from him prior to the time when he has to leave for 
the House, and then we'll hear from our witnesses.
    Good morning and welcome to today's hearing on helping 
Americans save. We politicians have been bemoaning our nation's 
low savings rate since well before I came to the Senate.
    Two years ago, American households saved only 1.5 percent 
of their income, an all-time low. Just over a decade ago, 
households saved 8 percent of their income, and in the 1970s 
and early 1980s, the savings rate was regularly over 10 
percent.
    Personal saving is low, not only by historical standards, 
but by international standards. Nearly every other westernized 
economy saves more than the U.S., as do many developing 
countries.
    There are a number of reasons, I think, why we save so 
little. Many households experienced a large gain in wealth in 
the 1990s from impressive increases in the stock market.
    Over one-half of Americans are involved in the stock market 
in one way or another. People fortunate enough to own property 
on one of the coasts or in certain areas like Chicago saw the 
value of their homes increase, as well. A family that's gained 
significant wealth from stocks or housing might safely assume 
that they can reduce their saving and still have enough to 
provide for retirement in an emergency.
    But, of course, not every family spent the 1990s 
calculating their capital gains. For the typical household, 
capital gains only modestly increased their wealth, yet, while 
middle- and lower-income households experienced sharp increases 
in income in the latter years of the previous economic 
expansion, there is little evidence that it led to higher 
saving.
    The U.S. tax system does not encourage savings. Economists 
of all stripes have noted that our treatment of investment 
income is counterproductive.
    The U.S. corporate income tax system and the treatment of 
dividends, capital gains, and interest income lets the 
governments tax the return from savings, 2 or even 3 times 
before it reaches the worker's pocket. It's no wonder that many 
choose to simply spend their money before it is taxed again.
    Congress has tried to alleviate the pernicious taxation of 
savings by offering a plethora of tax-preferred savings 
accounts. However, the patchwork approach of tax breaks makes 
navigating these programs exceedingly complicated, even for the 
most financially savvy person.
    We have three different types of individual retirement 
accounts, medical savings accounts, educational savings 
accounts, all of which are separate from any employee-sponsored 
retirement plan. Each account has different contribution 
limits, tax treatments, income cutoffs, and allocation rules.
    Professor Richard Thaler's research has shown that people 
often make poor decisions when offered too many choices. I 
suggest that that's precisely what's happening with IRAs.
    In discussing the low U.S. savings rate, it's important to 
recall why saving is important to individual households and 
society, generally. In the first place, households should have 
enough wealth at their disposal to be able to retire 
comfortably and not have to rely on the government.
    The present value of the total unfunded debt associated 
with Social Security is calculated to reach trillions of 
dollars, and as longevity increases and our obligations to 
entitlement programs balloon, it is not realistic to expect the 
Federal Government to pick up the entire tab for everybody's 
retirement.
    Second, savings finances the investment necessary to spur 
future economic growth. The American economy is driven by 
ingenuity and entrepreneurship, but even the most ambitious 
genius with a business plan can do little without ready access 
to capital.
    The innovators of Silicon Valley, from which flowed much of 
the U.S. technological and economic innovation over the last 15 
years, created enormous wealth for themselves and society 
through the combination of creativity, talent, hard work, and 
ready access to financial capital. Each was an essential 
ingredient.
    So, it makes sense to look at institutional factors that 
inhibit savings, and consider what kind of low-cost, common-
sense reforms can be adopted to make it easier for individuals 
to set aside a sufficient portion of their income each year to 
finance retirement, college education, or other significant 
financial obligations, and I think we've assembled an 
outstanding panel to help us deal with that.
    Mr. Saxton, we appreciate your being here, and look forward 
now to your opening comments.
    [The prepared statement of Senator Robert F. Bennett can be 
found in the Submissions for the Record on page 31.]

        OPENING STATEMENT OF REPRESENTATIVE JIM SAXTON, 
                         VICE CHAIRMAN

    Representative Saxton. Mr. Chairman, thank you very much. 
It's a pleasure to be here to welcome our witnesses this 
morning, and I thank you for having the foresight to call this 
hearing, because personal savings is vital for the financial 
security of households, and it also finances the investment and 
capital formation necessary for long-term economic growth.
    Unfortunately, and despite recent changes in the U.S. 
Income Tax, the U.S. Income Tax still retails a systematic bias 
against savings and investment. Under an income tax, a dollar 
saved is taxed and its return is taxed yet again, yet each 
dollar of consumption is taxed only once.
    Some of this bias has been reduced through the expansion of 
IRAs, 401(k)s and similar vehicles. The longstanding anti-
savings bias in the income tax is the reason that I have 
supported higher IRA deductions and 401(k) ceilings over the 
past several years.
    Another problem is that the current tax treatment of mutual 
fund shareholders regarding capital gains distribution is 
illogical, and, I think, very unfair. Under current law, mutual 
fund shareholders must pay taxes on capital gains realized by 
mutual funds, even though they have not sold one mutual fund 
share.
    Furthermore, they pay such taxes, even when the value of 
their shares has plummeted, after it did after the collapse of 
the stock market that began in the first quarter of 2000. In 
other words, when mutual funds generate huge capital gains, the 
shareholders get hammered, even when their own unsold shares 
have declined in value.
    This is something that should be changed, and when the 
mutual funds incur huge capital losses as they did after the 
bubble burst, most of these losses cannot be immediately passed 
on to shareholders. This is a ``heads-I-win/tails-you-lose'' 
situation for the government.
    In addition, given the complexity of the relevant tax 
provisions, it is very easy for confused taxpayers to pay 
capital gains taxes on mutual funds twice. I have offered 
legislation, H.R. 496, which would remedy this inequity by 
providing a tax deferral on capital gains distributions, large 
enough to cover all distributions of over 90-percent of 
shareholders.
    Mutual funds are an important savings and investment 
vehicle for middle-income Americans, and the punitive tax 
treatment of these taxpayers is unnecessary and 
counterproductive.
    Mr. Chairman, as you mentioned, we're going to have a vote 
on the House side sometime between now and 10:30, and I'm going 
to miss, therefore, the question period, so if I may just read 
into the record the question that I wanted to ask, relative to 
this mutual fund tax situation, I wanted to ask the panelists, 
and particularly Mr. Edelman, several questions.
    And if I may just read them now, I would appreciate that.
    Chairman Bennett. We'd be happy to have you do that, or, if 
you prefer, you could leave them with me and I will ask them on 
your behalf.
    Representative Saxton. Okay, that will be fine, and I have 
a copy of them here, and if you would do that, I would 
appreciate it.
    [The prepared statement of Representative Jim Saxton can be 
found in the Submissions for the Record on page 33.]
    Chairman Bennett. I'll do my best to follow up with the 
brilliance and incisiveness that you always display.
    Representative Saxton. That should be very easy.
    [Laughter.]
    Chairman Bennett. Thank you, thank you very much. We'll 
look to see the House Members join us a little later when they 
finish saving the Republic.
    Our witnesses are all international experts on savings, and 
Dr. Thaler, I apologize for mispronouncing your name the first 
time around. I have been appropriately admonished, and will be 
accurate from here on in.
    Dr. Richard Thaler is a University of Chicago Professor of 
Economics and he's been at the forefront of developing 
innovative ways to increase savings through employer-sponsored 
retirement plans.
    Robert Pozen, who is the Non-Executive Chairman of MFS 
Investment Management, a law professor at Harvard Law School, 
author and former Vice Chairman of the Board at Fidelity 
Investments, which is the hat he wore when we first met. He has 
a unique perspective on how institutions can affect savings and 
what types of reforms would compliment the work of financial 
institutions.
    And Ric Edelman, who is founder of Edelman Financial 
Services in Fairfax, Virginia, is the author of three New York 
Times number one best sellers, an award-winning host of radio 
and television shows, and has taught personal finance at 
Georgetown University for 9 years.
    Peter Orszag is a Senior Fellow at the Brookings 
Institution who has also published widely on tax policy, Social 
Security, and pensions. So, gentlemen, we thank you all for 
your willingness to be with us this morning. I think we will go 
in the order in which I introduced you, which means we start 
here at my right with Dr. Thaler.

      STATEMENT OF DR. RICHARD H. THALER, ROBERT P. GWINN 
PROFESSOR OF BEHAVIORAL SCIENCE AND ECONOMICS, GRADUATE SCHOOL 
                  OF BUSINESS, UNIVERSITY OF 
   CHICAGO; RESEARCH ASSOCIATE, NATIONAL BUREAU OF ECONOMIC 
                            RESEARCH

    Dr. Thaler. Thank you very much, Chairman Bennett and other 
Members of the Committee. Strictly speaking, my name should be 
pronounced ``tall-er,'' which you might be interested to know--
--
    Chairman Bennett. Does that go back to the Dutch and is the 
word from which dollar came?
    Dr. Thaler. Correct.
    Chairman Bennett. So, being in the economics business, we 
should call you Dr. Dollar?
    [Laughter.]
    Dr. Thaler. That's right. I'll answer to that.
    Chairman Bennett. Very good.
    Dr. Thaler. So, thank you for inviting me to participate in 
this panel on helping Americans save. I'm Richard H. Thaler, a 
Professor of Behavioral Science and Economics at the University 
of Chicago's Graduate School of Business.
    I'm an economist by training, but for the last 25 years, 
I've been exploring ways to incorporate the findings of modern 
psychology into economic analysis. As you all know, America's 
personal savings rate is hovering near zero, as the Chairman 
indicated earlier.
    Furthermore, as the population ages, there will be growing 
difficulty in financing Social Security, and future generations 
face a very likely prospect of having to finance a large 
fraction of their retirement on their own.
    I thus applaud the attention you're drawing to the 
important question of how to help Americans save, and I come 
bearing good news. By incorporating simple lessons of 
psychology and a little common sense about human nature, it can 
be quite easy to help Americans save.
    By tradition, governments are advised by economists on 
policy matters such as saving. Unfortunately, the traditional 
economic models that economists rely upon for their advice are 
not very helpful in two main respects:
    First, they assume that households are capable of making 
the complex calculations necessary to determine how much to 
consume and how much to save, and as important, that the 
households have the requisite willpower to delay consumption.
    Since the time of Adam and Eve, real humans, as opposed to 
the imaginary creatures populating economics textbooks, have 
had difficulty resisting temptation. There's a news story today 
about obesity that sort of underlines that point.
    Another problem with the standard economic model is that it 
does not give policymakers any guidance on how to increase 
savings. The primary variable under the control of policymakers 
is the after-tax interest rate.
    But the theory does not tell us whether raising this rate, 
say, by making some savings tax-free, will increase or decrease 
saving rates. The problem is that increasing the return to 
saving has offsetting effects. It makes saving more rewarding, 
and, thus, more attractive, but at the same time, the higher 
return implies that households do not have to save as much to 
achieve any particular savings goal.
    In contrast, by studying actual humans, we discover new 
tools that policymakers can use to increase savings. For 
example, here are some useful findings to consider:
    One, many Americans realize that they should be saving 
more. One survey finds that two-thirds of the participants in 
401(k) plans think they are saving too little.
    Two, most people find it easier to accept self-control 
restrictions that do not begin immediately. Many of us here in 
this room today are planning to begin diets next month, not 
today at lunch.
    Three, money that is put into designated retirement 
accounts tends to stay there, compared, say, to money in 
ordinary savings accounts.
    Four, people are loss-averse. Losses hurt more than gains 
feel good.
    Finally, fifth, there's enormous inertia in retirement 
plans and elsewhere. For the vast majority of participants, 
once they join the 401(k) plan, they rarely make changes, 
either to their contribution rate, or to the asset allocation.
    So, although participants agree that they should save more, 
many never get around to doing it. We can think about this list 
two ways: First, it can be considered a diagnosis, an 
explanation for why the savings rate is so low, and, second, 
and, more helpfully, it can provide the ingredients for the 
cure.
    So, what can we do to help Americans save more? One simple 
step that has been adopted by some organizations is called 
automatic enrollment. The idea is simple.
    In the usual 401(k) plan, when an employee first becomes 
eligible to join the plan, he or she receives a form that says 
``if you want to join the plan, please fill out this form.'' 
Under automatic enrollment, the employee receives a similar 
form, but it says, ``you are now eligible for the plan, and 
unless you return this form, we're going to enroll you 
automatically.''
    Notice that under the standard economic analysis, these two 
setups are considered virtually identical. The cost of filling 
in a form is small, relative to the long-term benefits of 
joining the plan, especially when the firm provides a match.
    Nevertheless, automatic enrollment can have huge effects. 
In one company studied by Madrian and Shea, when automatic 
enrollment was adopted, the enrollment rate by new workers 
jumped from 49 percent to 86 percent. That's the good news.
    The bad news is that under automatic enrollment, companies 
must select some default savings rate, an asset allocation, and 
employees tend to adopt and stick with these default choices.
    So, in the company mentioned above, where the plan 
administrator selected a default saving rate of 3 percent and 
100 percent of the money was invested in a money market 
account, most employees passively accepted these choices. This 
is unfortunate, because virtually every expert who has studied 
the problem concurs that 3 percent is not a high enough 
contribution rate, and a money market account is not a suitable 
long-term investment vehicle for 100-percent of one's 
retirement income.
    My collaborator, Shlomo Benartzi from UCLA, and I have 
discovered a better plan that can be adopted in conjunction 
with or separately from automatic enrollment. We call our plan 
Save More Tomorrow, also known as the SMART Plan.
    Under SMART, participants are contacted a few months before 
their next pay increase with the following offer: They can 
commit themselves now to increasing their savings rates later, 
specifically when they get their next raise, say, by 2 or 3 
percentage points.
    Also, their contribution rates will continue to go up 
whenever they get a pay increase, until they either reach some 
specified maximum or opt out of further increases. Notice that 
this plan incorporates the psychological principles I mentioned 
above.
    People are asked to start saving more in a few months, not 
now, and by linking the savings increase to pay increases, 
participants never have to experience a cut in their take-home 
pay. We have now implemented this plan in several companies, 
but let me report on the results from the first company to 
adopt the idea, a mid-sized manufacturing firm in the Chicago 
area.
    The company was concerned that their employees were not 
saving enough for retirement, so they hired a financial 
consultant and made him available to meet one-on-one with every 
worker. The consultant had a computer with software that could 
help calculate how much the employee should be saving.
    Because the employees were not good savers, the software 
typically recommended that the employee immediately increase 
his or her saving rate to the maximum allowed. However, few 
employees were willing to accept this advice, so the consultant 
typically suggested an increase of 5 percentage points, say, 
from 3 percent to 8 percent.
    This advice was also rejected by most employees, so the 
consultant would offer these reluctant savers the SMART plan. 
Specifically, their savings rate would increase by 3 percentage 
points at the time of every raise.
    This plan proved to be popular with the employees. Over 80 
percent of those offered the plan, signed up, and the effect on 
their saving rates was dramatic, as shown in my Table 1.
    In just 14 months from the time the consultant spoke to the 
employees, the participants who enrolled in the SMART Plan 
increased their saving rates from 3.5 percent to 9.4 percent, 
and after 2 more years, they were saving 13.6 percent of their 
salary.
    Their saving rates have nearly quadrupled, and this is a 
group that had been very reluctant savers. Remember that they 
wouldn't accept this advice to increase by 5 percentage points.
    The SMART Plan has implemented by several other employers 
and companies that administer 401(k) plans, such as Vanguard, 
are offering the idea to their employer-customers. We're 
optimistic that hundreds of thousands of employees will be 
enrolled in SMART Plans within the next few years, and within a 
decade, the plan could reach most employees in the U.S.
    At the moment, the idea does not need government 
intervention, but two steps are worth considering: First, adopt 
some version of the SMART Plan for government employees through 
the Thrift Savings Plan.
    Second, give some consideration to firms that adopt the 
best practice combination of automatic enrollment and SMART, 
perhaps exempting these firms from cumbersome nondiscrimination 
testing.
    Such action would simply recognize that in implementing a 
SMART Plan, firms have already met the spirit of the 
Congressional intent that retirement plans should not 
disproportionately benefit high income earners. There are other 
lessons for government to take away from our experience.
    First, we've found a winning recipe for helping people 
save. The key ingredients are: Make it easy to join the plan--
the easier the better--and automatic enrollment is the easiest.
    Take the contributions directly from the paycheck. If you 
don't see it, you don't spend it. Once you get people saving, 
make it a default option to keep saving, or even better, keep 
increasing their saving rate.
    Four, put the money into an account where people are not 
overly tempted to dip in. These basic principles could be 
adopted in many existing and proposed tax-saving vehicles.
    I would like to make one other behaviorally-motivated 
suggestion. One way many Americans do manage to save, albeit, 
temporarily, is through tax refunds. Most Americans receive a 
refund when they file their tax returns.
    Unfortunately, that money is often spent when the refund 
check arrives, or even quicker, via tax refund loan. One way to 
get more of that money into long-term savings would be to allow 
refunds to be deposited directly into an IRA, and still qualify 
for a tax credit for the previous year.
    In other words, people who are now, in March, 2004, filing 
their 2003 tax returns and claiming a typical $1,500 refund, 
could send those funds directly into an IRA account. For 
traditional deductible, not Roth IRAs, the actual amount 
deposited would be increased by the taxpayer's marginal tax 
rate, so a taxpayer in the 25-percent tax bracket would be 
given a choice of getting a $1,500 refund or making a $2,000 
IRA contribution. That could be an attractive inducement to 
save.
    My principal conclusion is simple and optimistic. We can 
succeed at helping Americans save more by employing a 
combination of basic psychology and common sense.
    If I could add one personal note, helping to save is a 
cause I believe in. We're offering this SMART Plan to any 
organization at no cost, as long as they agree to provide us 
with outcome data. This is not a profit-seeking venture. Thank 
you.
    [The prepared statement of Dr. Richard H. Thaler can be 
found in the Submissions for the Record on page 36.]
    Chairman Bennett. Thank you very much.
    Mr. Pozen.

          STATEMENT OF ROBERT C. POZEN, NON-EXECUTIVE 
  CHAIRMAN, MFS INVESTMENT MANAGEMENT, JOHN M. OLIN VISITING 
              PROFESSOR OF LAW, HARVARD LAW SCHOOL

    Mr. Pozen. Thank you. I'd like to begin by supporting Vice 
Chairman Saxton's suggestion that we change the taxation of 
mutual funds. As he pointed out, right now, even if mutual fund 
shareholders do not sell their own shares of a fund, they still 
are taxed on the capital gains realized by the fund, and his 
proposal for a deferral of those capital gains until 
shareholders sell their fund shares is one that I personally 
support.
    I know that this is a little outside the purview of this 
hearing, but I couldn't help but support the very sound 
suggestion of your Vice Chairman.
    Let me say that my general theme today is that we ought to 
consider private retirement plans, together with Social 
Security. From the point of view of the retiree, they are 
obviously considered together, because a retiree has two 
sources of income.
    But in the past, in a lot of legislative sessions, they 
have been viewed as entirely separate, so I want to put them 
together, and I have suggestions for each area.
    In the private retirement area, I think we know that the 
participation rate of people under $50,000 a year is quite low, 
and under $25,000 a year in income the participation rate is 
actually minimal. And it's probably not possible to increase 
the amount significantly that people can put in a 401(k) plan 
under $25,000 a year. If they are married and have two children 
on $25,000 a year, they just don't have enough to live on.
    But I believe that we can do a lot for people with incomes 
between $25,000 and $50,000 per year. We now have a low-income 
tax credit which Congress adopted a few years ago. 
Unfortunately, it is a non-refundable tax credit. I ran the 
numbers on a family of four with two children, who had income 
of $40,000 a year, and it turns out that if you just apply the 
four exemptions, the standard deductions, and the child credit 
of $2,000, their total Federal income tax is $49. If we want 
this incentive for IRAs and 401(k)s to work, where the low-
income tax credit is essentially a form of government matching 
for these programs, we need to make the low-income tax credit 
refundable.
    Now, that involves money. I had a preliminary estimate run, 
and it was roughly a 10-year estimate of about $10 billion or 
$1 billion per year (assuming the existing low-income tax 
credit is already permanent). This is a relatively modest 
amount of money that would make this tax credit viable for that 
group of people with incomes between $25,000 and $50,000, who 
would like to save if given appropriate incentives.
    The other proposal I have builds on one of Dr. Thaler's 
ideas, and takes it one step further. We know that employers 
with fewer than 100 employees are the ones who have the lowest 
percentage of private retirement plans.
    Congress has tried to address this problem by introducing 
the SIMPLE Plan, and what I'm proposing is the ULTRA-SIMPLE 
Plan, going one step further.
    I believe that financial institutions, if the program is 
simple enough, would offer this program to every employer. The 
small employer, at the end of the year, would take all 
employees that earn $25,000 or more, and would put 1 percent of 
their wages into an IRA. This would be invested automatically 
in a specified account, whether it be a money market account or 
a balanced account, unless the employee chose a different type 
of account.
    And the employee could opt out of the whole program. But if 
the employee didn't opt out, the 1 percent would go into this 
ULTRA-SIMPLE Plan for retirement.
    I am establishing a minimum of $25,000 to reduce paperwork. 
Otherwise small employers will ask: what about part-time 
workers, what about temporary workers? So we're eliminating 
those issues. I'm also limiting the contribution to once at the 
end of the year, so we know how much money the person has 
earned. I'm putting the minimum at $25,000 so that the account 
size will be $250, which will be enough to get financial 
institutions interested. They are not interested if the number 
is only $79 or something like that.
    So this is an ULTRA-SIMPLE PLAN. It involves no change in 
the tax law; it just is a way to create a plan that can be 
offered to all employers who now do not have any type of 401(k) 
or other retirement plan. It doesn't really have a cost to 
employers, and if the employees don't want to participate in 
the plan, they can opt out.
    Now, the second subject I want to address is the 
relationship between Social Security and IRAs. I have to 
digress a little to explain something that Peter Orszag is very 
familiar with, but most people are not. And that is the 
difference between wage-indexing and price-indexing.
    In the Social Security area, we usually think of price-
indexing because COLAs after retirement are all price-indexed 
to protect against inflation. But sometimes people don't know 
that the initial Social Security benefit is based on wage-
indexing. That is, we compute your average career earnings, and 
then we adjust this average up by wage-indexing. This is much 
more expensive than price-indexing. Indeed, if we only moved 
from wage to price-indexing of initial Social Security 
benefits, this would have a huge positive effect on Social 
Security's long-term deficit.
    Now, what I would like to do is divide workers into three 
categories: We have all those under $25,000 a year in income. 
As I have said, I think it's very hard for them to save. They 
have very, very low levels of participation in IRAs and 401(k) 
plans.
    It's going to be very hard to significantly increase that 
participation rate, so they are going to be almost exclusively 
dependent on Social Security. Therefore, I would continue to 
let those people be on wage-indexing.
    On the other hand, look at all the people who have $100,000 
or more in career earnings. I would move them all on to price-
indexing, because almost all those people have IRAs and 
401(k)s. The combination of slower-growing Social Security 
benefit with an IRA and 401(k) will give them more than they 
would have received under the original schedule for their 
Social Security benefits.
    And the middle ground, say, someone at $50,000 a year, we'd 
work out some proportional formula whereby their initial Social 
Security benefits would be part wage-indexed and part price-
indexed. So this is a way to treat everybody fairly.
    We would have three different groups of Social Security 
beneficiaries, and their Social Security benefits would grow at 
different rates. So we would be counting on people to put that 
1 percent in IRAs to make up the difference.
    We've run the numbers: if the middle and the higher income 
people would just put 1 percent a year of their wages more into 
an IRA, they would come out more than whole. And the beauty of 
this proposal is that it cuts Social Security's long-term 
deficits by over two-thirds. Thus, if we went to what I call 
progressive indexing, moving partly from wage-indexing to 
price-indexing, with the lowest income people staying at wage-
indexing and the highest moving to price-indexing, we would cut 
roughly 68 to 70 percent of Social Security's long-term 
deficit, as computed by Steve Goss, the Chief Actuary of Social 
Security, the keeper of the numbers, as we all know.
    In short, I propose two rather modest changes in IRAs that 
would help increase the savings rate, moving from a non-
refundable to a refundable credit, and introducing the ULTRA-
SIMPLE IRA Plan based on Dr. Thaler's research. Then if we 
could increase the participation rate in IRAs, we could develop 
an approach that combines more IRA benefits with slower growing 
Social Security benefits for higher earners. As a result of 
this dual approach, we could keep our retirees at roughly the 
same total retirement income, while making a big dent on Social 
Security's long-term deficit. Thank you very much.
    [The prepared statement of Robert C. Pozen can be found in 
the Submissions for the Record on page 41.)
    Chairman Bennett. Thank you. That was a very provocative 
suggestion.
    Mr. Edelman.

              STATEMENT OF RIC EDELMAN, CHAIRMAN, 
                   EDELMAN FINANCIAL SERVICES

    Mr. Edelman. Thank you, Mr. Chairman. It's a pleasure to be 
here this morning. In addition to my work as an author in the 
field of personal finance and doing radio and television work 
in the field, my firm, Edelman Financial Services, is the sixth 
largest financial planning firm in the nation, according to 
Bloomberg. We manage about $2 billion in assets for 7,000 
clients around the country, all of it in mutual funds.
    I was also named by Research Magazine in November of 2003 
as the nation's number one financial advisor for focus on the 
individual client, and it's that perspective that I think sets 
me apart from most folks who come here to speak with the 
Committee, in that I'm in the trenches. I deal one-on-one with 
individual consumers on a regular basis, and that perspective 
is perhaps a little bit different from the typical ivory tower 
environment that we sometimes find ourselves in.
    And I can tell you, from having studied Dr. Thaler's 
research for many years, that one of the most effective things 
this Committee can do is basically whatever he says.
    [Laughter.]
    Mr. Edelman. Because the behavioral side of economics is 
one that has been too largely ignored, and I can tell you, from 
working one-on-one with consumers, that what he describes in 
his research and in his work is exactly what we discover in the 
trenches, dealing with individual clients.
    And to that extent, the number one way to help Americans 
save more is financial literacy. Our American education system 
is widely regarded as the best in the world. We're unmatched in 
our ability to teach skills to America's school children, so 
that they have the ability to get jobs and earn a living.
    What we don't teach them, however, is to make money with 
the money that they've made. We don't teach them how to manage 
their wealth. We don't teach personal financial literacy in our 
school systems on a regular basis.
    And yet all the statistics show that money has an 
incredible impact on everyone's life, including at very young 
ages. According to several studies, the first assisted 
purchases occur at age 3, and they are not in toy stores, but 
in the supermarkets. Think carefully as to who is really 
choosing the cereal that mom and dad buy. It's not mom and dad.
    Ask a typical 6-year old where does money come from and the 
most likely answer is the ATM. After all, they just watch mom 
or dad punch buttons on a box and money comes out. Ask the 
typical consumer. Ask yourself. What's the price of a big-
screen TV? It's not $4,000. It's $39.95 a month.
    So what we have to recognize is that the legal tax and 
financial complexities have never been greater. But we have 
found ourselves letting children graduate school without any 
information about how money works.
    Today life is very complex. Americans are changing jobs, on 
average, every 4 years. Fifteen percent of Americans move every 
year. Almost half of marriages end in divorce and more than 
half of those will remarry. We're also having children later in 
life than ever. That does not even begin to introduce the 
concepts of terrorism, technology and new social issues. 
Everything from gay marriage to euthanasia.
    With all these complexities, it's more important than ever 
that our school children graduate understanding the basics of 
money management. They are graduating high school without 
knowing how to balance a checkbook. They don't understand how 
loans work. They don't understand compound interest.
    But, at the same time, they have access to credit cards and 
will rapidly go into credit card debt. The average indebtedness 
in college of today's undergrads is over $4,000. High school 
seniors are now being offered credit cards. We have to 
recognize that the access to debt is pervasive.
    We're finding an increase in bankruptcies. We're finding an 
increase in indebtedness across all age groups, including 
seniors. And what we have to do is give them the education they 
need to be able to survive and thrive financially.
    So my primary recommendation is to require the nation's 
schools to include financial literacy in their curriculum. This 
is easier than it may at first appear, because it doesn't 
require, necessarily, a whole new classroom. It simply says 
that when you teach history, talk about the economic 
implications of the reasons why nations go to war. Talk about 
what things cost 200 years ago and what they cost today.
    In math classes, talk about compound interest. It's a 
question of geometry and algebra. Let school children, when 
they're learning how to count, count coins instead of marbles.
    Sir, we just need to incorporate money as a routine facet 
of everyday life. That means in the classroom as well as in the 
workplace.
    I also think we need to delay Social Security eligibility 
to age 70 for Americans who are currently under the age of 50. 
One of the shocking statistics is that 30 percent of working 
Americans say that Social Security will be providing most of 
their retirement benefits. There is an over-dependence, an 
over-reliance on Social Security today.
    We need to get the message across to Americans that their 
future is more up to them than it is up to the government. By 
letting American workers who are younger, those who are in 
their 30s and 40s today, by letting them know that they can't 
rely on Social Security until age 70, they will have a strong 
incentive to save.
    Simultaneously, we should delay mandatory withdrawals from 
retirement accounts. Currently, Americans are forced to 
withdraw their money starting at age 70\1/2\, the exact 
opposite of encouraging them to save. We're demanding that they 
start spending. We should delay until age 80 the mandatory 
distribution from IRAs and retirement accounts. We should also 
replace the current retirement plan contribution regulations 
with one universal program.
    Currently, where you work determines what plan you're 
offered, whether or not you have a plan, how much money you can 
contribute and what those choices are. We should not penalize 
Americans who work at small companies simply because they work 
at a small company. We should allow them to have the same type 
of program that those who work for large corporations enjoy.
    Fifth, we should permit people to save for retirement even 
if they are not currently earning an income. There are millions 
of stay-at-home spouses. They'll tell you they work. They 
simply don't earn an income. We should encourage people to save 
by permitting them to establish retirement accounts even if 
they don't currently have an income.
    We should also permit people to save for retirement 
regardless of their age. One of my inventions is something 
called a retirement income for everyone trust, which is a 
retirement planning tool for children. It is a vehicle where 
the parents and grandparents can set aside money.
    Because I'm in the private sector, we had to realistically 
put the number at $5,000. If we can introduce this in the 
public sector, we could knock that figure way down. But a 
$5,000 contribution for a newborn child, where they cannot 
touch it until age 65, that $5,000, at historical market 
averages, would grow to $2.4 million by age 65. Effectively, 
potentially eliminating the need for Social Security income. 
The key is to allow people to save money at incredibly young 
ages, regardless of the source, in a mandatory environment 
where they cannot touch it until retirement.
    Number seven is to eliminate the ability for workers to 
borrow from their retirement plans. Today we allow workers to 
borrow from their 401(k). As a result, an awful lot do that. 
Seventeen percent of all American workers have borrowed from 
their retirement. The average loan is 16 percent of the account 
balance.
    The irony is that most of those who have taken loans earn 
between $40,000 and $100,000. It is not the lowest income 
spectrum that have borrowed against their accounts. By 
borrowing against the retirement plan, they're effectively 
ensuring that they will not have money at retirement.
    Number eight, we should eliminate the ability for employers 
to distribute funds to terminated employees. Under current 
retirement plan rules, if you leave the company and there's 
less than $3,500 in the account, the employer has the right to 
send you the cash. When that happens you're going to pay taxes, 
plus a 10 percent penalty, and you'll probably spend the money 
on something frivolous. We should eliminate the ability of 
employers to do that and, instead, require that the money 
remain in a retirement account.
    Finally, number nine, we should extend to IRAs the same 
creditor protection currently available to qualified retirement 
plans. If you have money in a 401(k) and you are sued, a 
creditor cannot go after that money. But they can go after 
money in an IRA. It's an unlevel playing field and serves as a 
disincentive to move money to IRAs. That should be eliminated. 
Thank you very much, Mr. Chairman, for this opportunity.
    [The prepared statement of Ric Edelman can be found in the 
Submissions for the Record on page 49.]
    Chairman Bennett. Thank you, sir.
    Dr. Orszag, you get to bat cleanup here.

  STATEMENT OF DR. PETER R. ORSZAG, JOSEPH A. PECHMAN SENIOR 
     FELLOW IN ECONOMIC STUDIES, THE BROOKINGS INSTITUTE, 
                         WASHINGTON, DC

    Dr. Orszag. Thank you very much, Mr. Chairman. My testimony 
this morning addresses two issues.
    First, why it is critical to preserve Social Security's 
core insurance features even while reforming the program to 
eliminate its long-term deficit. Secondly, how we can expand 
retirement savings on top of Social Security.
    First, on Social Security, the program provides a well-
defined basic income that is protected against inflation, the 
risk of outliving one's assets, and financial market 
fluctuations. Benefits are progressive, providing a higher 
replacement rate for low earners than for high earners. The 
program provides families with important insurance against the 
disability or death of a wage earner in addition to retirement 
benefits.
    Although we can and should boost retirement savings on top 
of Social Security, we must not forget that for the majority of 
the population, the program provides the key layer of financial 
security during particular times of need.
    For example, one-fifth of elderly beneficiaries receive all 
of their income from Social Security and two-thirds receive the 
majority of their income from Social Security.
    The average Social Security benefit is only a little more 
than $10,000 a year, underscoring the role of the program in 
providing a core layer of financial security above which people 
can build retirement wealth in other forms.
    The program does face a long-term deficit and restoring 
long-term balance to Social Security is necessary. But it's not 
necessary to undermine the program's important protections in 
order to save it.
    In particular, in my view, replacing part of Social 
Security with individual accounts is likely to undermine the 
program's retirement security features because a real-world 
system of individual accounts is unlikely to require that 
benefits keep pace with inflation; unlikely to require 
participants to fully annuitize their account balances when 
they retire (that is, transform the account balances into 
something that lasts as long as they are alive); and unlikely 
to protect surviving spouses as well as the current system 
does.
    Furthermore, and perhaps most importantly, there is likely 
to be very substantial pressure for early withdrawals under a 
system of individual accounts, just like we see with 401(k)s 
and IRAs. Someone with $10-, $20-, $30,000 in an account, a 
sick kid, and the need for a new car, is going to exert a lot 
of pressure to get the money before retirement. In which case, 
the money is not there for retirement.
    Professor Peter Diamond of MIT and I have put forward a 
progressive reform plan that restores long-term balance to 
Social Security without any accounting gimmicks and while 
actually boosting some of the social insurance features that 
the program currently offers.
    The second part of my testimony addresses how to boost 
savings on top of Social Security. Various types of savings 
incentives already exist for households to supplement Social 
Security in building up retirement wealth. But most of these 
savings incentives are upside down. They provide the strongest 
incentives to participate to higher income households who, on 
average, are already better prepared for retirement and for 
whom a dollar going into a retirement account is less likely to 
represent new savings rather than just asset shifting from a 
different kind of account, while providing very little 
incentive to participate for lower or moderate income 
households who are more likely to need additional funds in 
retirement and for whom additional dollars going into the 
accounts are more likely to represent new savings.
    In part reflecting this sort of upside sort of incentives, 
the nation's broader pension system betrays several serious 
shortcomings. First, participation is low, especially among 
lower earners. Only about one-fifth of workers in households 
with income of less than $20,000 participate in any given year.
    Even those who do participate rarely make the maximum 
allowable contributions. Only about 5 percent of 401(k) 
participants, and only about 5 percent of those eligible for 
IRAs, make the maximum allowable contribution.
    Finally, despite the shift from defined benefit to defined 
contribution plans over the past few decades, many households 
approach retirement with meager defined contribution balances.
    The median defined contribution balance among all 
households age 55 to 59 in 2001 was only about $10,000. The 
median was higher for those with accounts. But, among all 
households in that age range the median was only $10,000, which 
does not provide a very large retirement annuity.
    The bulk of the policy changes that have been enacted in 
recent years, moreover, move the system further in the wrong 
direction. They provide disproportionate tax breaks to higher 
income households who would have saved the money anyway and for 
whom the contributions made do not represent net additions to 
savings, while doing little to boost incentives for 
participation among moderate and lower income households.
    In my view, a better strategy would encourage expanded 
pension coverage among those lower and moderate income 
households, through the following steps. First, one could 
expand the income eligibility range for the saver's credit that 
Mr. Pozen mentioned and make the credit refundable. The credit 
is currently scheduled to expire or sunset in 2006. It should 
be extended, expanded and made refundable, in my opinion.
    Second, we could reduce the implicit taxes on saving that 
exist under various government programs. For example, food 
stamps and supplemental security income have asset tests that 
impose significant, implicit taxes on the savings that lower 
and moderate income households do accumulate.
    Financial education is obviously critical. I would join in 
the call to be providing financial education as part of the 
core primary and secondary school education, not waiting until 
people are older before even trying to get to them.
    Finally, as the other witnesses have emphasized, I think, 
perhaps the best thing that we can do is use the force of 
inertia to boost savings rather than forcing people to overcome 
inertia in order to save. That would include changing the 
default choices in 401(k) plans as has already been explained 
and also include a little-noticed part of the 
Administration's--well, it wasn't technically the budget, but 
it was in their Blue Book Treasury proposals--a proposal to 
allow individuals to split their refunds.
    Many people appear to be reluctant to put their entire 
refund into an IRA. The IRS has thus far been reluctant to 
allow individuals to split their refunds into two components. 
Included in the Administration's Treasury proposal this year, 
it doesn't require a statutory change, but they should be urged 
to follow-up on this, is a proposal that would allow people to 
just check a box on the tax return and put part of the refund 
into an IRA and part into a checking account or some other more 
liquid asset. I think that does make sense.
    Finally, I think we also need to be paying more attention, 
as the Baby Boomers near retirement, to the withdrawal stage. 
We've focused a lot on the accumulation stage and trying to 
build up account balances, but we really do need to worry about 
how individuals are going to take the money out of their 
401(k)s and IRAs.
    I have some slight differences of opinion about whether 
loosening the minimum distribution rules, in general, makes 
sense. I think some targeted reforms there would be beneficial. 
But, more broadly, we do need to be worrying a lot about how 
people are going to be taking their money out of these 
accounts, not just about how people get the money in them in 
the first place.
    Thank you very much, Mr. Chairman.
    [The prepared statement of Dr. Peter R. Orszag can be found 
in the Submissions for the Record on page 55.]
    Chairman Bennett. Thank you, sir.
    Let me ask Mr. Saxton's questions so that we're true to the 
pledge we made to him before he left.
    Mr. Edelman, they're all directed to you.
    [Laughter.]
    Chairman Bennett. You get to be Lucky Pierre on this one. 
So he says: ``I would like to ask you a few questions about the 
current tax treatment of capital gain distributions made by 
mutual funds to draw on your experience as a financial advisor.
    ``Based on your experience, is it possible for ordinary 
taxpayers to get confused and effectively pay capital gains 
taxes twice, once on distributions and again when they actually 
sell their mutual fund shares?''
    Mr. Edelman. It's not only possible, Mr. Chairman, for them 
to get confused, it's almost impossible for them not to get 
confused. The rules are extraordinarily complex. I wrote about 
this a lot in my first book The Truth About Money. It's one of 
the biggest tax traps facing mutual fund investors.
    When you make an investment, say, you put $10,000 into a 
mutual fund, you typically reinvest the dividends in capital 
gains. The mutual fund pays out those distributions and they 
are reinvested and then at the end of the year you get a 1099. 
You hand the 1099 to your accountant, who has you pay taxes on 
your tax return.
    Later, when you sell the fund, you get a check in the mail 
from the fund company and the 1099 referring to the gross 
distribution. You then give that to your accountant and your 
accountant says, ``How much did you invest?'' And you tell him 
that one day, way-back-when, you invested $10,000.
    Your answer is not adjusted for the fact that you've earned 
dividends and capital gains over time, and you paid taxes on 
them annually. Thus you pay taxes again because you didn't 
adjust for all the reinvestments, which should have increased 
your cost basis.
    We have a simple solution. And that is to do away with the 
annual 1099 distribution.
    Mr. Pozen. I ought to add that most fund companies now, as 
to more recent investments, give you your average cost basis 
which includes dividend reinvestments.
    Chairman Bennett. So they show the increase in basis?
    Mr. Pozen. That's built into the system over the last few 
years. But it wouldn't be possible for somebody who made an 
investment 20 years ago easily to ascertain his or her basis.
    Chairman Bennett. There's been a lot of controversy 
recently about problems in certain sectors of the mutual fund 
industry. Isn't the current tax treatment of mutual funds, 
capital gains distributions, one of the largest costs imposed 
on owners of mutual fund shares?
    Mr. Edelman. There's no question, Mr. Chairman, that the 
annual cost of taxation on your mutual fund profits is a very 
significant impediment to future wealth. It's forcing you to 
pay taxes on money that otherwise could have remained invested.
    This does not apply to stocks or bonds. If you own a stock 
for 20 years, it is, in essence, tax deferred for 20 years. You 
pay no taxes on the growth until you sell. That same simple 
rule does not apply to mutual funds and creates a substantial 
tax and, therefore, decreased wealth for millions of investors.
    Chairman Bennett. Thus, isn't the current tax treatment of 
mutual fund capital gains distributions one of the biggest 
drawbacks to investing in mutual funds?
    Mr. Edelman. It is a very significant drawback, yes. Once 
investors begin to understand the tax complexity associated 
with mutual funds, it often serves as an impediment to their 
long-term savings.
    Chairman Bennett. When the stock market is on an upswing 
and mutual funds generate capital gains, this results in 
immediate tax liability for shareholders. But when the stock 
market falls and the funds incur losses, these can not 
generally be passed through to shareholders. Isn't this tax 
treatment inconsistent, even if the fund losses are eventually 
permitted to offset gains over time?
    Mr. Edelman. Yes, Mr. Chairman. It is inconsistent. There's 
a significant disparity between having to pay taxes when there 
are distributions and not being able to take a tax deduction 
when there are losses.
    It gets worse than that. Many times, mutual funds are 
selling securities that they've owned for less than a year. So 
not only are they issuing you a distribution, they're offering 
short-term distributions which are taxes at the top tax bracket 
of the taxpayer as opposed to the 15 percent long-term gains 
rate. It's even worse than the question would suggest.
    Chairman Bennett. Others of you can jump in here. Mr. 
Saxton said this should all be directed to Mr. Edelman, but I 
don't want to shut anybody else off if they want to comment.
    His last comment, he says: ``Another bill I've introduced 
would end mandatory distributions from IRAs at 70\1/2\ and also 
end mandatory distributions from 401(k) accounts.
    ``Essentially, this is another form of tax deferral that 
would reduce the negative tax consequences for personal 
savings. Isn't it desirable to end, at least, delay these 
mandatory distributions?''
    Mr. Edelman. I'm not an economist, so I can't really 
address the basis or the need to have a date that the money 
must begin to come out.
    It does seem to be illogical to me, from a financial 
planning perspective and working with individual clients, 
because the money in an IRA or other retirement plan is 
eventually going to be withdrawn and add tax if only at the 
individual's death. It's nothing you escape.
    Since it can't be escaped, and we're not worried about a 
step-up-in-basis at death, for example, which happens on stocks 
or real estate. Since there's no step-up-in-basis and we are 
guaranteed to have the money eventually taxed, I'm not quite 
sure why the government cares when it's taxed. Why does it need 
to be 70\1/2\? Why not let it be at the taxpayer's discretion, 
either death or another age of their choosing?
    Chairman Bennett. Does anyone else want to comment on that? 
I am just turning 70\1/2\, so I have a very strong interest in 
this.
    Dr. Orszag. I have a somewhat different view. And I think 
it's important to distinguish the 401(k) rules from the IRAs. 
Under the 401(k) rules, the minimum distribution requirements 
are that you must begin distributing the funds at 70\1/2\ or 
when you retire, whichever is later. So the rules are different 
between 401(k)s and IRAs.
    Chairman Bennett. As long as I stay in the Senate, I can 
hang on to them?
    Dr. Orszag. A lot of people are worried about the 
incentives provided for older workers to stay in the workforce. 
The partial effect, in least in terms of labor incentives from 
raising that age could be to weaken instead of strengthen 
incentives to continue working. It'll depend on individual 
circumstances, but there is, at least, an important distinction 
between the IRA and 401(k) rules.
    Here's also, I think, an important question, which is, why 
are we providing the incentives for IRAs and 401(k)s in 
general?
    In my view, the reason is to provide an incentive for 
retirement saving. The minimum distribution rules, and I don't 
want to defend them in their entirety because I think they are 
too complicated and there are some reforms that are worthwhile, 
but it is important to remember the goal. The goal is to make 
sure that that saving is actually used during retirement and 
not as an estate-planning device or for some other purpose.
    If there are other ways, and I think there are, of ensuring 
that the bulk of the saving is serving the public policy 
purpose for which the tax incentives were provided, that should 
be explored. I do worry about either eliminating or 
substantially raising the age because I think it then 
undermines that public policy objective for putting the rules 
in, in the first place.
    I'd also note one other point. We face, as you know, a very 
substantial long-term fiscal deficit. That long-term deficit 
already incorporates, into the projections, trillions of 
dollars in present value in taxes on withdrawals from 401(k)s 
and IRAs, which had a tax break up front and tax-free build-up 
and then are taxed on the way out. I think you--you meaning 
policymakers in general--are going to come under increasing 
pressure in the years ahead to reduce those taxes.
    A lot of people don't realize that when they've got 
$15,000, $20,000, $30,000 in their 401(k), they will be taxed 
on those funds. If we start to reduce those taxes, we make an 
already really bad fiscal situation that much worse. So 
anything that crosses the line of either reducing or 
eliminating the taxes on withdrawals, I think, is risking a 
very severe loss of revenue. Because I don't know how you stop 
the ultimate effect from adding up to trillions of dollars in 
an already bad fiscal outlook.
    There is such a danger with regard to the minimum 
distribution rules. Because as was noted, relaxing those rules 
is effectively reducing the implicit tax on the withdrawal that 
you start moving in that direction. That's another 
consideration to take into account.
    Mr. Pozen. Mr. Chairman, I have a slightly different view, 
though it's not inconsistent. A lot of the current rules allow 
people to take money out of IRAs and 401(k)s for purposes other 
than retirement. And I know there's a lot of pressure on 
Congress to keep expanding the range of permissible 
withdrawals. Buying a home is a worthwhile objective, but the 
problem is that these withdrawals undermine the retirement 
objective, which should be the primary objective. I think we 
should do more to keep that money in plans for retirement.
    Another subject that we struggled with when we were on the 
President's commission was the form of distribution from a 
retirement plan. For a lot of people, it would really make 
sense not to take a lump sum out of their IRA, but rather to 
buy an annuity so they would be assured of having money for the 
rest of their life.
    But you then meet with somebody who has $200,000 a year in 
income and they say, ``why should I buy an annuity?'' For that 
reason, it probably doesn't make any sense. But I would follow 
Professor Thaler's idea hereby making the presumptive choice of 
annuity. And then if people didn't want an annuity, they could 
opt out.
    I think that lifetime annuities, joint and several 
annuities, are quite important for a lot of families. And if 
people just take a lump sum out at retirement to buy a car, or 
they take money out of their plans before retirement, for 
example, to buy a house, they are really defeating a lot of the 
purpose of having a retirement plan.
    So I would try to keep a choice available. But I'd like to 
use the force of inertia to push people toward retirement 
objectives rather than other objectives.
    Mr. Edelman. One other point I would like to make about the 
mandatory distribution rules and I would suspect that I won't 
get any objection on this point because it's free. It's not 
going to cost anybody any money or alter policy in a 
significant way and I know you love to hear things like that.
    You had mentioned that you're turning 70\1/2\ this year, 
Mr. Chairman.
    Chairman Bennett. This month.
    Mr. Edelman. That does not mean happy birthday.
    [Laughter.]
    Mr. Edelman. Happy day, but not birth. That does not mean 
you have to withdraw money this year from your retirement 
account. It means you have to do it by April 1st of the year 
following the year you turned 70\1/2\. That rule is absurd. To 
tell ordinary consumers that they have to begin taking the 
money out by April 1st of the year following the year they 
turned 70\1/2\ is silly.
    It's far too complex and it should be simply changed to say 
you have to take out money by December 31 of the year you turn 
70 or 71, pick one. But that whole phrase of April 1 following 
the year in which you turned 70\1/2\ is a mathematical jumble 
that is just pointless and it's a good example of how the tax 
code is unnecessarily complex.
    Chairman Bennett. I have exhausted Mr. Saxton's questions. 
Let me ask a few of my own.
    In 1986, Congress imposed income limits on IRAs so that 
workers above a certain income level were no longer permitted 
to receive the tax benefit of an IRA. Following the imposition 
of those limits, IRA participation levels fell, even among 
those low-income workers who were still eligible to 
participate. Any ideas as to why?
    Dr. Thaler, you're the psychologist here.
    Dr. Thaler. I think it's no big mystery, which is that the 
financial services industry no longer viewed this as an 
activity that was worth spending a lot of money advertising.
    In the early 1980s, around this time of year, it was hard 
not to see ads asking people to invest in IRAs. And when you 
cut that market down, they're going to spend less money 
advertising.
    Mr. Pozen. I would tend to agree. The limits then were that 
you couldn't get an IRA deduction for a family unless you were 
earning less than $40,000 a year. And, as we've seen, those 
deductions are not that valuable for people at those income 
levels. Many of them have a hard time saving. The financial 
institutions are interested in the larger accounts and, of 
course, higher participation rates. So if you're advertising in 
a market where only a small percentage are effectively 
eligible, and even a smaller percentage are actually likely to 
contribute to IRAs, it's not really worth spending large 
dollars on IRA advertising.
    One of the benefits of having a more universal IRA is, 
without the government paying anything, you do get the 
advertising dollars of the financial institutions.
    I agree with what Mr. Edelman says about financial 
literacy. In designing savings programs, you ought to try to do 
something that implicitly gets the financial services industry 
behind you and let them carry the ball to a large degree on 
financial literacy. They will if they see a reasonable return 
on their advertising investment.
    Mr. Edelman. For many in our field, Mr. Chairman, the IRA 
is a loss leader. The amount of revenue we can earn from 
opening an IRA account is minimal. But if it enables us to grab 
additional assets from a client who has other assets, then 
we'll go through that marketing effort. When you limit us to 
dealing with individuals whose sole investment is going to 
potentially be that IRA, then as these two gentlemen have said, 
we'll shrug or shoulders and not bother.
    Dr. Orszag. Senator, I think while agreeing, in part, with 
the other panelists, there are a couple of other perspectives 
that are important.
    First of all, there was a decline for those below the 
income limits post-1986. In many cases, it was a pretty modest 
one. For example, if you look at tax filers with adjusted gross 
income of less than $20,000 in 1984, 5 percent of those made an 
IRA contribution. In 1988, 2.4 percent did. So, yes, there's a 
decline. But it's not like you're going from 90 percent to 15 
percent.
    Mr. Pozen. I think you're making my case about why you 
wouldn't advertise for 2.4 percent of those eligible.
    Dr. Orszag. There you go.
    Chairman Bennett. You're dealing in a political world. On 
the floor of the Senate, that would be trumpeted as it's cut in 
half.
    [Laughter.]
    Dr. Orszag. That is right.
    The second thing that is important to realize is the type 
of advertising matters, also. One advertisement in The New York 
Times in 1984, for example. I'm just going to quote it for a 
minute: ``If you were to shift $2,000 from the right pants 
pocket to the left pants pocket, you wouldn't make a nickel on 
the transaction. However, if those different pockets were 
accounts at''--and I'll leave out the name of the financial 
services firm--``you'd profit by hundreds of dollars. Setting 
up an IRA is a means of giving money to yourself.''
    That is an advertisement for asset shifting. That is not an 
advertisement for new saving. The type of advertisements that 
are likely to attract higher-end clientele are not necessarily 
the type of advertising that will attract where I think the 
focus should be. I'm getting new saving from the lower and 
moderate households or middle income households who most need 
to save more.
    I agree that advertising is important. I just want to 
caution us both to put the magnitudes in perspective and, also, 
obviously, the type of advertising will affect the type of 
response that one gets.
    Chairman Bennett. Of course, every investment is a form of 
asset shifting. I have $2,000 in my pocket. I can spend it on a 
fancy trip or I can invest it in this situation.
    Dr. Orszag. Senator, it matters a lot whether the 
contribution is coming from reduced consumption, i.e., forgoing 
that trip, or shifting assets from your mutual fund, for 
example.
    Shifting assets from the mutual fund doesn't generate any 
new savings. It's just moving existing assets from one pocket 
to another. Forgoing the trip is a form of raising overall net 
saving. That's really where we need to be focusing.
    In my view, the empirical evidence strongly suggests that 
as you go up the income ladder, more and more of the dollars 
that are deposited into the tax deferred accounts do not come 
from reducing consumption, but rather from shifting assets from 
other accounts.
    Chairman Bennett. We can argue the psychology of that. But 
let me, Dr. Orszag, get into an area you made reference to. You 
talked about Social Security. I'd like you and Dr. Pozen, if I 
could, to get into dialogue with this because he talks about 
matching the savings with Social Security in an overall plan. 
And you talk about, perhaps, keeping the two separate so that 
they don't get matched in the saver's mind.
    One of the major problems with Social Security that has 
come in the last 30 years, and is accelerating, which nobody 
wants to talk about because it sounds kind of callous; but, in 
fact, one of the biggest problems with Social Security is 
people are living longer.
    If they just had the courtesy to die on the same schedule 
that was anticipated when Otto von Bismarck invented it, when 
the life expectancy for males was in the low 60s, so if you 
could live long enough to beat the odds, and live until you 
were 65, then the government would take care of you for the 
rest of your life with payments from other people who were 
betting that they, too, would live. But the odds were that most 
of them would not.
    So the lucky retiree who beat the odds and stayed alive 
past 65 was getting the benefit of what everybody else was 
paying in. That sounds a little like a Ponzi scheme if you want 
to put that particular face to it.
    Now we're all living pretty much past 65 and we're all 
getting the benefits. But the guy who now doesn't live to 65, 
pays in his entire life, dies and doesn't get a dime and his 
heirs don't get a dime. So there is no wealth creation.
    I realize the advantage of what you're talking about when 
you say if they get a lump sum that can disappear. But if they 
get an annuity, which Social Security is, they get paid for as 
long as they live.
    And we've got people who are now over 100 who have been 
drawing Social Security for over 35 years and the amount 
they've drawn out is staggeringly higher than anything they 
paid in. They're being paid by the people who are coming in at 
the bottom.
    At some point in this whole retirement situation, 
particularly in the African American community, which 
demographically does tend to die more rapidly than the white 
community, they're not accumulating wealth in the way their 
white counterparts are, or their healthier white counterparts 
are.
    They have children they would like to leave estates. They 
would like to have wealth accumulated in their own name that 
they could do something with in their retirement years. But if 
their whole focus is on a program that just pays them as long 
as they stay alive, and allows nothing to be accumulated in 
their own name in the form of assets, they are missing out on 
something fairly significant.
    Mr. Pozen, you talk about tying the two together so that 
they move forward and the advantages that Dr. Orszag talks 
about of having the payment made as long as you live. If you 
live to 101, it's still going to be there. That's an enormous, 
enormous advantage, an important thing.
    But the flip side of it is, you turn 65 or 70--right now 
it's 67 and it's going to be for the Baby Boomers by the time 
they retire. If it comes to be 70, as you're proposing, Mr. 
Edelman--and Chairman Greenspan is talking about that, too, 
whatever the date is, it would be nice to have a mixture of 
both.
    That you have the certainty of the payouts, but you also, 
from the all the money you have been paying in, some of it has 
been going in under your name and you have a significant chunk 
of wealth that at retirement you can do something with because 
maybe you do need to make a down payment on a house and a 
little bit of wealth will allow you to do that. Then you can 
live in the house on the income.
    Discuss that whole mix of where we are in these benefits, 
if you would, the two of you. Because I caught different vibes 
from the both of you and I'd like to have that conversation.
    Dr. Orszag. I can go first. On life expectancy, I just want 
to emphasize one thing that I think is very important for 
policymakers and that I was struck by in writing the book with 
Professor Diamond. While life expectancy is increasing, as 
you've noted, it is increasing at a particularly rapid rate for 
higher earners and higher educated workers. In other words, the 
gap in mortality rates or in life expectancy over the past 
three or four decades has increased markedly.
    Many demographers expect that to continue, both because 
life-preserving health behaviors are more likely at the upper 
end of the income distribution and because a lot of the life-
extending technologies are really expensive and they won't 
filter all the way down the income distribution.
    The partial effect of that trend is to make Social Security 
less progressive on a lifetime basis because the higher earners 
are, on average, getting their benefits for an increasingly 
longer number of years. And I think it has broader implications 
for a variety of policies. I just wanted to mention that.
    On the mix between the two, my view, it really comes down 
to how much saving one thinks households will do, even as you 
move up to the 50th percentile, 60th percentile on top of 
Social Security.
    Again, just to put some numbers in our head, the average 
Social Security benefit may now be up to $11,000. But it's only 
a little higher than $10,000. We're not talking about $50-, 
$60-, $70,000 a year. It's a very modest benefit for the vast 
majority of the population.
    When we talk about substantially scaling that back and 
replacing it with some non-annuitized form of income, I worry 
that, basically, you wind up with too little being provided, 
especially, to the widow. We have to remember that widow 
poverty rates are already much higher than married women's 
poverty rates, 3 times as high.
    When we don't annuitize wealth, and, particularly, when we 
spend it up front, the person who's really getting hurt is the 
widow. That is the reason I'm concerned about substantially 
scaling back annuitized income. But I think you have identified 
a fundamental tension in individual account plans, which is, do 
you force annuitization of the whole thing or not? The trade-
offs there are very difficult. It becomes even more difficult 
when you think about if the annuitization age were 65 and you 
had the elderly woman who has cancer and knows she's going to 
die in 2 or 3 years.
    If you're forcing her to annuitize also, that's a 
substantial loss of wealth that you are imposing on that person 
in order to protect the widows who live a long time.
    Chairman Bennett. Mr. Pozen.
    Mr. Pozen. I agree with your basic idea. It's a positive 
social goal for Congress to allow families, including moderate 
income families, to have some form of retirement account that 
they can call their own, and that they can bestow on other 
members of the family.
    But the challenge for many of the proposals for personal 
retirement accounts is that they are of what I call a carve-out 
nature. That is, they're taking some portion of the 12.4 
percent in payroll taxes, let's say 2 or 3 percent, and putting 
that portion into a personal account.
    As you point out, we already have this big overhanging 
deficit. The problem with carving out is that you wind up in an 
even deeper hole for 30 or 40 years until you can get those 
people who carve out the 2 or 3 percent to take lower Social 
Security benefits many years later. So you wind up with, 
essentially, a requirement for a large loan from the government 
to the Social Security trust fund.
    What I was trying to do with my proposal is to reach your 
goal, but avoid the budgetary problems of carve-outs from 
Social Security. Also, my proposal avoids the administrative 
cost of creating a whole new set of personal accounts.
    Why don't we declare victory with the IRA as the personal 
account? The IRA is already there. It has an administrative 
structure. And if we can encourage people and incent people to 
build up their contributions to their IRA, that's where they 
will have their wealth.
    But the question remains: how will you deal with Social 
Security's deficit? That's why I've made a proposal for 
progressive indexing; it allows you to recognize what Peter 
says. The life expectancy of the higher-income people is 
diverging more and more from that of lower-income people. 
Therefore, we should have a slower-growing Social Security 
benefit for high-income workers because they're going to live 
longer.
    If we bring in progressive indexing on Social Security, we 
help the financing of Social Security. We offset, to some 
degree, the greater life expectancy of the higher income 
groups. And if we combine progressive indexing with more 
incentives for IRAs, we achieve your goal of building wealth 
without having the problem of impairing the financial solvency 
of Social Security.
    I'm trying to say to those people who want personal 
retirement accounts--use the IRA. We have the administrative 
structure. Let's use that, instead of taking money out of 
Social Security, because then you're raising a whole set of 
other budgetary issues that are very difficult to resolve.
    Chairman Bennett. I think what you're saying makes sense, 
but, having said that, you are cutting into one of the 
fundamental principles, political principles of Social 
Security, which is that it cannot be means-tested in any way.
    Mr. Pozen. I don't think I actually am. Social Security, at 
the moment, is progressive, but it's currently not means-
tested; rather it is tilted in a progressive manner. As higher-
income people live longer, however, they receive larger total 
Social Security benefits over their lifetime.
    Thus, the progressivity that's currently in the system 
gradually is undermined and reduced, so I'm not means-testing 
anything. I'm just bringing in a little more tilt to take into 
account that middle- and higher-income people are living longer 
than lower-income people. Therefore, on a lifetime basis, the 
total value of Social Security benefits are changing on a 
relative basis.
    Chairman Bennett. You're not means-testing directly.
    Mr. Pozen. I'm tilting the system, but I'm tilting the 
system to take into account the point Peter is making. Right 
now, if you have a person who makes $100,000 a year, his or her 
chances of living to 90 are much greater than a person who's 
making $30,000 a year.
    So I'm just tilting the system to make the average lifetime 
Social Security benefit be roughly the same for all income 
groups.
    Chairman Bennett. I happen to agree with that, but it's 
interesting that people who keep saying, well, the tax system 
is unfair because we should make the people at the upper end 
pay most of the taxes, well, actually, they do. Nonetheless, we 
should make them pay even more and more and more, but when you 
get to Social Security, you've got to send Ross Perot exactly 
the same size check you send to one of the clerks that works at 
EDS, if they have the same earning pattern during their years.
    You can't say, but Ross Perot doesn't need it.
    Mr. Pozen. Under my system, if Ross Perot and the clerk 
both made $50,000 a year as a career average, they would both 
get the same Social Security check. However, the person who now 
makes $100,000 a year, rather than $50,000, doesn't get twice 
the earnings check of the person with $50,000. Let's assume the 
higher earner receives 1.6 times the benefit of the lower 
earner, so there is a progressivity tilt already built into 
Social Security.
    If we say that we want to maintain the current relationship 
between the person at $100,000 who gets 1.6 times the check 
that a person receives at $50,000, we must consider that the 
person at $100,000 is now going to live 5 years longer than the 
person at $50,000. We should adjust that tilt a little to 
reflect this difference.
    Chairman Bennett. Dr. Thaler, get into this.
    [Laughter.]
    Chairman Bennett. Talk about automatic enrollment plans and 
how they might impact this and how the government could 
encourage automatic enrollment plans. What would the default 
investment be and where would you be in this whole question of 
trying to get something other than just the Social Security 
into people's deductibility situation, all the way through?
    Dr. Thaler. The fortunate thing is that a lot of the things 
that I'm talking about don't really need the government to get 
involved. So, the private sector is taking initiatives.
    Some companies are doing automatic enrollment, some are 
adopting our Save More Tomorrow Plan. Many of the companies 
that administer 401(k) plans, such as Vanguard and Fidelity, 
TIAA-CREF, ADP, are offering these escalating savings plans, 
offering to do the administrative work to allow employers to 
offer these options.
    I will say that we sometimes talk to employers who are 
reluctant to do this, because they are afraid of getting sued, 
and probably the most important thing the government can do to 
facilitate these things is to create safe harbor rules, and the 
Treasury has been good about issuing rulings, making it 
explicit that it's fine to automatically enroll people into 
plans, and automatically.
    You want to make it clear that it's fine to automatically 
enroll them into something other than a money market, like a 
balanced account. As I said in my testimony, another good thing 
the government could do would be to set up a showcase by doing 
this for their own employees through the Thrift Program.
    Chairman Bennett. I'm never amazed at the inventiveness of 
the plaintiff's bar.
    [Laughter.]
    Chairman Bennett. I can't conceive of the basis on which a 
suit would be brought here.
    Mr. Pozen. Unfortunately, I can.
    [Laughter.]
    Mr. Pozen. The complaint would say that you're 
automatically enrolled, unless you opt out, and that the worker 
wasn't sufficiently appraised of the opt-out, so that would be 
one basis of the complaint. Then, second of all, if the money 
did go into a balanced account or in a stock account, which in 
a particular set of years did less well than a Treasury bond, 
then the worker might complain that you automatically put me in 
an investment with a lower return.
    Chairman Bennett. Okay.
    Dr. Orszag. I think it's important to realize that it's 
actually not even primarily--my understanding is that it's not 
Federal law here that's the problem. There are state labor laws 
that could cause problems here in terms of suits. Frankly, we 
should just have Federal preemption and narrow-targeted Federal 
preemption of state labor laws to allow these kinds of plans, 
and I agree that's one step that the Federal Government could 
take to try to remove impediments to their adoption.
    Dr. Thaler. Let me give you one trivial example of the sort 
of things that we run up against at my own University. Every 
November we have to log onto the Web and enroll in open 
enrollment to choose health plans and so forth and so on.
    With our retirement plan, we have a generous retirement 
plan, and faculty can put an additional amount of money, up to 
$9,000, into some kind of supplemental retirement account. You 
have to go onto the Web every year and do that.
    Now, I'm trying to convince them to make it so that if you 
do nothing and you saved last year, you save the same amount 
again this year, and there is some lawyer who is worried that 
if we do that, that will violate some rule. I don't think there 
is any rule it violates, but people are worried that there is 
such a rule, and we can, by simply making it explicit, that 
it's perfectly fine to assume that you want to save the same 
amount as you did last year, that would help a lot.
    Chairman Bennett. Of course, the individual participants in 
the plan would get no benefit from the lawsuit, but the lawyer 
who brought it would get rich, presumably run for the Senate--
--
    [Laughter.]
    Chairman Bennett. [continuing.]----and go on from there. 
Let's talk about matching. All of this conversation has been 
about what the employee would put in. Let's talk about 
matching, incentives to match. Are there lawsuit possibilities 
on matching?
    Mr. Pozen. I don't think there are a lot of lawsuit 
possibilities for matching, but when we're talking about this 
low-income credit, that's essential to increase IRA 
contributions. We know from the 401(k) experience that matching 
brings in higher contributions from workers.
    People love matching. It's the equivalent to going to the 
store and seeing a bargain sale. People really like matching. 
It has a significant effect on participation rates.
    However, if you're talking about people who are not in 
employer-based programs, or people with lower incomes in IRAs, 
then the question is: how do we create a match? The low-income 
tax credit is designed to create a government match, a partial 
match to be precise.
    The problem is that the credit is constructed in a way that 
doesn't really work for a lot of people. In order for the 
credit to be effective, you have to be paying a significant 
amount of income tax. The good news is that people with lower 
incomes aren't paying a significant amount of tax; the bad news 
is, to the extent you're giving these people the match in the 
form of a non-refundable credit, the match is not effective.
    So, I think the match idea makes a lot of sense, but 
because income tax rates have been brought down and other 
credits and deductions have been increased, like the standard 
deduction, if we're going to have a government match to 
encourage IRA contributions by lower-income workers, we need to 
alter the design of the match.
    Chairman Bennett. We're running probably later than we 
should, but let me ask about just one more issue. One of the 
first public policy questions I got involved in back in the 
1960s when I first came to Washington had to do with President 
Kennedy's proposals with respect to pensions.
    It was one of the first recognitions of the fact that 
Americans were shifting from the old paradigm where your 
pension and your healthcare and any other form of ``benefit,'' 
was provided by your employer. We shift employers every 4 
years, did I hear that statistic around or something along that 
line?
    Let's talk about portability. The Kennedy proposal was the 
first towards creating portable pensions.
    I did an analysis for myself when I turned 50, just to kind 
of quantify it, and from the time I turned 20 until the time I 
turned 50, I changed jobs or situations--this included going to 
school, going to the Army, what have you--17 times. I had 17 
different situations in that 30-year period.
    I've gotten a little more stable.
    [Laughter.]
    Chairman Bennett. Since I turned 50, the 11 years I've put 
into the United States Senate is the longest period of time I 
have drawn a paycheck from the same signatory in my life. So, I 
have a whole series of little buckets that were accumulated 
along that way, some of which, quite frankly, I have emptied to 
pay for the new car, to send my kids to school, to do whatever, 
but some of which I have clung to, and----
    Mr. Edelman. I would argue, Mr. Chairman, did you cling to 
the ones which were later or the ones earlier?
    Chairman Bennett. I clung to the ones that were smaller.
    [Laughter.]
    Mr. Edelman. It wasn't worth bothering.
    Chairman Bennett. And my wife now looks at these statements 
that come in every month and says, ``What is this? Why don't we 
dump this all into a single vehicle, so that when you die, I 
don't have to go through all of this and find out what's 
there?''
    Let's talk about this whole portability question.
    Dr. Thaler. Portability is great. One of the advantages of 
the whole 401(k) style systems is portability. The idea of 
unifying all of these tax-favored savings accounts into one 
type also has some elegance and appeal.
    The thing I think that's vital to remember is that for the 
relevant income group--and I would say it's somewhere between 
$25,000 and $125,000, relevant meaning they have enough that 
they could possibly save, and they're not so wealthy that 
they're kind of saving automatically, for that group, the only 
way people successfully save is if the money is taken out of 
their paycheck.
    If they have to go write a check, that's a huge problem to 
overcome.
    Chairman Bennett. Mr. Pozen runs the ad, so they would.
    Mr. Pozen. On the portability issue, I think you're making 
a very good point. We now have the rollover IRA. Theoretically, 
you should be able to consolidate all these small retirement 
accounts into one rollover IRA.
    The problem is, since many of these accounts come from 
different retirement programs, it is an accounting nightmare to 
try to figure out the consolidation. For instance, if you paid 
no tax on all contributions to these accounts and they all had 
the same distribution rules, then you could put them all into 
the same IRA and you would know that the tax basis was zero. 
Then you could make distributions and know the tax 
implications.
    The problem is that some accounts may be partially 
contributory, some of them may not. Some of these accounts are 
subject to 401(k) rules and some of them not.
    I think the key problem is that we have different rules for 
457s and 403(b)s and 401(k)s. As Mr. Edelman says, this is 
crazy. We ought to have one uniform set of rules called by any 
name you want.
    Another thing we ought to do is try to provide some relief 
for people who pour this money into rollover IRAs. There should 
be some sort of mechanical rule about what the basis is in 
these situations. I've tried it myself sometimes with a 
rollover IRA. I had a rollover IRA from a 401(k) plan, then I 
had an IRA that was a nondeductible IRA, and I tried to put 
them together. It was a nightmare.
    So, we need to make those rules simpler. Maybe we can't do 
it all into one rollover IRA, but at least we ought to be able 
to do two--one for all retirement plans where there was no tax 
on the way in, and second of all, everything else, where you 
might have had some tax basis.
    I think that simplification is the key for consolidation: 
we need to simplify the rollover; the rollover is the vehicle 
that should be available to you. You ought to be able to roll 
over all of those accounts to one account, then be able to 
invest that account any way you want, but the accounting is a 
nightmare.
    Mr. Edelman. The remaining issue on that is the rule that 
allows employers to distribute. Seventy-two percent of 
retirement accounts that are between $5,000 and $10,000, 72 
percent of them are distributed prior to retirement, because 
workers leave; they go somewhere else, and although the 
portability rules are in place, effectively enough, to a 
degree, what happens when you leave work? On your final day of 
work, your HR person or the business owner says, ``So what do 
you want me to do with your retirement plan? Do you want to 
roll it over, or do you want me to send you a check?''
    And, 72 percent of the time, the worker screams, ``Sure, 
send me a check,'' especially if they are laid off and they 
don't have another job to go to. They take the check.
    We need to prohibit that. We need to require that the money 
either stay in the account or roll to an IRA.
    Mr. Pozen. We do have a rule that says that if you don't 
roll to an IRA, you are subject to withholding, and we penalize 
you in that sense. I think when that rule came in, you saw a 
much greater number of people who rolled over.
    To have a rule prohibiting withdrawals doesn't recognize 
that there are some people who are laid off and they actually 
need the money. I don't know whether we might increase the 
penalties in terms of not rolling over----
    Mr. Edelman. Or create the default environment.
    Mr. Pozen. We do have the default environment in the sense 
that the money presumptively either stays in the plan now or 
transfers to a financial institution. But we could make the 
default a lot stronger by increasing the taxes on withdrawals.
    Chairman Bennett. Let me thank you all for your 
participation. I think it's been a stimulating discussion, and 
I appreciate the interchange between witnesses. I'm sorry that 
our associates from the House weren't able to come back and 
join us, but I think we've created a very significant record 
here, and I appreciate your willingness to help us do that. The 
hearing is adjourned.
    [Whereupon, at 11:35 a.m., the hearing was adjourned.]


                       Submissions for the Record

=======================================================================

[GRAPHIC] [TIFF OMITTED] T3761.001

[GRAPHIC] [TIFF OMITTED] T3761.002

[GRAPHIC] [TIFF OMITTED] T3761.003

[GRAPHIC] [TIFF OMITTED] T3761.004

[GRAPHIC] [TIFF OMITTED] T3761.005

[GRAPHIC] [TIFF OMITTED] T3761.006

[GRAPHIC] [TIFF OMITTED] T3761.007

[GRAPHIC] [TIFF OMITTED] T3761.008

[GRAPHIC] [TIFF OMITTED] T3761.009

[GRAPHIC] [TIFF OMITTED] T3761.010

[GRAPHIC] [TIFF OMITTED] T3761.011

[GRAPHIC] [TIFF OMITTED] T3761.012

[GRAPHIC] [TIFF OMITTED] T3761.013

[GRAPHIC] [TIFF OMITTED] T3761.014

[GRAPHIC] [TIFF OMITTED] T3761.015

[GRAPHIC] [TIFF OMITTED] T3761.016

[GRAPHIC] [TIFF OMITTED] T3761.017

[GRAPHIC] [TIFF OMITTED] T3761.018

[GRAPHIC] [TIFF OMITTED] T3761.019

[GRAPHIC] [TIFF OMITTED] T3761.020

[GRAPHIC] [TIFF OMITTED] T3761.021

[GRAPHIC] [TIFF OMITTED] T3761.022

[GRAPHIC] [TIFF OMITTED] T3761.023

[GRAPHIC] [TIFF OMITTED] T3761.024

[GRAPHIC] [TIFF OMITTED] T3761.025

[GRAPHIC] [TIFF OMITTED] T3761.026

[GRAPHIC] [TIFF OMITTED] T3761.027

[GRAPHIC] [TIFF OMITTED] T3761.028

[GRAPHIC] [TIFF OMITTED] T3761.029

[GRAPHIC] [TIFF OMITTED] T3761.030

[GRAPHIC] [TIFF OMITTED] T3761.031

[GRAPHIC] [TIFF OMITTED] T3761.032

[GRAPHIC] [TIFF OMITTED] T3761.033

[GRAPHIC] [TIFF OMITTED] T3761.034

[GRAPHIC] [TIFF OMITTED] T3761.035

[GRAPHIC] [TIFF OMITTED] T3761.036

[GRAPHIC] [TIFF OMITTED] T3761.037

[GRAPHIC] [TIFF OMITTED] T3761.038

[GRAPHIC] [TIFF OMITTED] T3761.039

[GRAPHIC] [TIFF OMITTED] T3761.040

[GRAPHIC] [TIFF OMITTED] T3761.041