W. Scott Simon
Section 403(b) was first added to the Internal Revenue Code (IRC) in 1958. In 1964,
regulations were first issued that detailed some of the basic statutory provisions of
section 403(b). The Internal Revenue Service is now in the process of finalizing new
regulations to govern 403(b) plans. The best guess at present is that these
regulations, the first in over 40 years, are to become effective Jan. 1, or shortly
A Suggested Model 403(b) Plan
The new 403(b) regulations make clear that officials at school districts offering 403(b)
plans will take on some fiduciary duties to plan participants and their beneficiaries.
Smart (or even purely self-interested) officials should understand that by placing the
interests of 403(b) plan participants first, they will gain the greatest fiduciary
protection against lawsuits filed by the government and the plaintiffs’ bar. (I know, I
know, such officials are protected by governmental immunity laws, contractual
indemnity clauses and insurance policies. Yet I don’t know of anyone, including school
officials, who likes to prepare for, and be subjected to, depositions and the other
niceties of modern litigation.)
The truly wonderful thing about placing the interests of 403(b) plan participants first is
that it also actually produces the best outcome for them. A “win-win” if there ever was
one. To help ensure that outcome, though, school district officials must jettison the
absurdities currently plaguing so many 403(b) plans. What follows, then, are some
suggestions for creating a model 403(b) plan.
1. Get rid of the multi-provider model and adopt the single-provider model.
The multi-provider 403(b) model, which is the most prevalent by far at school districts,
allows teachers to choose their investment options from among more than one
retirement plan services provider such as an insurance company or mutual fund
The multi-provider model can be confusing to teachers (or anyone else with a pulse)
and often generates the “deer in the headlights” phenomenon: teachers have so many
investment options they make no selections at all, or make them by throwing darts.
This model has reached its logical absurdity in my state of California where, as a result
of adoption of insurance code section 770.3, California teachers get to choose from (at
last count) 123 providers – each of which has a glut of investment options.
The multi-provider model also balkanizes the 403(b) assets held by teachers into
separate insurance contract accounts held at separate insurance companies. Schools
districts with no fiduciary duties don’t need to keep track of all that money so they
have no incentive to try to get a low cost deal for teachers.
The single-provider model, on the other hand, allows school districts greater
purchasing power, thereby resulting in lower and more transparent investment costs
for teachers; this enhanced employee benefit gives districts a competitive edge in
recruiting teachers. This model also offers a centralized way to provide plan
documents, monitoring and reporting of participation rates, monitoring and reporting
of participant contributions and withdrawals, a full range of investment options and
customized and consistent communication materials, all of which help reduce the
administrative burden at school districts.
2. Get rid of annuities and offer only mutual funds. Annuities and other insurance products have their place–just not in 403(b) plans (or 401(k) plans or 457(b) plans
for that matter). I have, for the record, purchased lots of insurance products and
believe fully in the idea of insurance. I just don’t think such products should be on the
menu of investment options in retirement plans. They are just too costly. I detailed in
last month’s column the fact that investment products offered by insurance companies
in 403(b) plans cost between 200 and 500 basis points. In his testimony before a
congressional committee in March, noted independent fiduciary Matt Hutcheson was
less charitable in his estimate of 300-500 basis points. Given this, it just seems crazy
for school districts to offer teachers investment options that have an added layer of
fees. It’s not that such options have no value at all; rather, it’s that they do not have
value enough to justify their (oftentimes) grossly higher costs.
Compounding this problem is the fact that these costs are usually hidden–because
they’re high. A cost, under any body of fiduciary law, must be reasonable in relation to
the service or product received in return. If a cost is hidden, how is it possible to
determine whether that cost is reasonable in relation to the service or product received
so that the recipient knows it’s getting full value?
Insurance companies favor this kind of obfuscation because it allows them to get away
with charging higher costs than would otherwise be the case in a cost-transparent
environment. A recent example of this absurdity involved one of the large insurance
companies active in marketing retirement plans to school districts. The Los Angeles
Unified School District (LAUSD), led by a very well-informed advisory committee,
insisted that the insurance company disclose to plan participants a revenue-sharing fee
in the district’s 457(b) plan. An insurance company vice president refused, stating that
“I’m afraid the revenue-sharing would just confuse people.”
Apart from the contempt that statement reveals for those who are educating our next
generation, perhaps this person could (a) actually try to make revenue-sharing
understandable to teachers or (b) better yet, advise his company to do away with its
revenue-sharing fees altogether. No school district official should have to spend time
debating the intricacies of such nonsense with insurance companies or any other
retirement plan services provider. (LAUSD eventually prevailed in this debate.)
Investment products offered by non-insurance retirement plan services providers (i.e.,
brokerage firms and mutual fund companies) in 403(b), 401(k) and 457(b) retirement
plans are generally less expensive but there are still too many of them that are too
costly, especially considering their general resistance towards the idea of accepting
any fiduciary responsibility. Some of these providers will accept fiduciary responsibility
if a 403(b) plan is large enough, but historically they haven’t been too keen to do so.
The exemption that brokerage firms have hidden behind, the so-called “Merrill Lynch
Rule,” was struck down by the United States Circuit Court of Appeals for the District of
Columbia a few months ago, and this may have a positive impact on the 403(b)
3. Be more responsible and offer fewer mutual funds, not irresponsible by
offering more mutual funds. An exhausting array of studies has shown that the
more choices (of anything) people are given, the less confident they are in their
selections – if they even make them. I just ran across the winning bid for a menu of
investment options in a 457(b) plan at a large school district comprised of 20
individual mutual funds and two “portfolios.” Each of the individual funds was
separated into one of 5 or 6 asset classes, resulting in 3-4 funds in each asset class. I
guess the idea is that a plan participant should select a fund from each asset class
since the funds in each are clearly duplicative (more or less) of one another.
I cannot count the times over the years that friends have called for advice because
they haven’t a clue when faced with this kind of selection “process.” Many retirement
plan service providers, as Matt Hutcheson noted in his congressional testimony in
March, “understand and count on participants making imprudent investment
decisions.” “The more fund choice is offered, the more mistakes participants make.
[Participants] tinker with the investments within their accounts, incurring hidden
trading costs that reduce their returns. The current [retirement plan] environment
encourages mistakes, for no good or necessary reason. [Many retirement plan service
providers] rely on [participant] ignorance to generate revenue. This is a substantial
and hidden cost about which participants are almost universally unaware.” It’s this
current environment that leads me to suggest that school districts offering 403(b)
plans should provide perhaps no more than 10-12 individual mutual funds, each of
which is broadly diversified, low cost and actually differs from the other funds offered
within a portfolio context.
4. Be even more responsible and offer portfolios of mutual funds, not
individual mutual funds. Let’s remember that Nobel laureate Harry Markowitz is the
father of modern portfolio theory, not modern investment theory. It was the Employee
Retirement Income Security Act of 1974 (ERISA) which first applied key tenets of
modern portfolio theory to the investment and management of assets by investment
fiduciaries (see Labor Reg. § 2550.404a-1).
Modern portfolio theory also provides the underpinnings of the 1992 Restatement 3rd
of Trusts (Prudent Investor Rule), the 1994 Uniform Prudent Investor Act, the 1997
Uniform Management of Public Employee Retirement Systems Act, the 2006 Uniform
Prudent Management of Institutional Funds Act (and its predecessor, the 1972 Uniform
Management of Institutional Funds Act).
A basic tenet of modern portfolio theory is that properly constructed portfolios of
investments are more conducive to wealth accumulation than individual, stand-alone
investments. The “proper” construction of a portfolio involves assembling mutual funds
that have dissimilar price movements to each other. This produces a portfolio with
reduced risk which not only decreases loss but can also simultaneously increase
return. This kind of portfolio is greater than the sum of its parts.
5. Be super responsible and offer a prudent range of model portfolios
comprised of low cost, passively managed mutual funds. The ideal conditions for
achieving investment success are created by disciplined application of three major
themes found in modern prudent fiduciary investing: broad diversification of risk, low
costs and (for taxable investors) low taxes. Over the years, I have detailed in this
column (some would say in an ad nauseam way) why passively managed mutual funds
(i.e., index funds and asset class funds) are the best way to apply these themes in a
It’s not hard to understand, of course, that every dollar saved in investment costs and
taxes goes straight to the bottom line to increase return. It’s also true (though little
understood) that broad diversification of risk can help increase return because of
reduction of loss. Maximum reduction of loss, through broad diversification (both
“horizontally” across the portfolio itself and “vertically” within each mutual fund held in
the portfolio), occurs as a result of minimizing a portfolio’s exposure to (bad)
uncompensated risk so that the only risk remaining is (good) compensated risk.
The most effective and efficient way for a participant in a 403(b) plan (or any
retirement plan) to reduce loss and thereby increase return is to invest in an
automatically rebalanced model portfolio appropriate to the participant which is
comprised of institutional level asset class mutual funds and index mutual funds,
providing market level returns at market level risk. Offering a range of prudent model
portfolios of passively managed funds also generates reliable returns relative to model
portfolio and asset class performance benchmarks, which greatly simplifies monitoring
and reduces its costs.
Such portfolios are rarely changed but when they are it’s because (a) academic
research has determined the existence of a new asset class and the investment
provider adds a corresponding asset class fund to the portfolio to achieve even broader
diversification of risk (which, again, decreases loss and thereby increases return) or
(b) the provider replaces an existing higher cost fund in the portfolio with an
equivalent lower cost fund (which increases return). Passive investing gives plan
participants the best bang for their buck.
The demonstrably inferior alternative, investing in actively managed mutual funds,
invites big trouble: the higher costs and risks inherent in active investing. There’s also
the constant heartburn caused by managers who haven’t performed up to snuff for
whatever reason (whether it be the current price of oil or that the Giants have been
winless in a World Series since 1954) and the (costly) need to keep replacing them, or
soap operas at mutual fund companies over the (rumored or not) departure of a
superstar fund manager and what that might do to a school district’s 403(b) plan.
6. Be super-duper responsible and offer a menu of prudent model portfolios
comprised of low cost, risk-based passively managed mutual funds. Fiduciaries
of many retirement plans have jumped on the band wagon of age-based “lifecycle”
fund portfolios. These portfolios are invested with an eye towards the target date that
a particular plan participant will retire, say, 2030. As the portfolio advances in time
towards that date, the portfolio will grow more conservative by adding more bonds
and shedding more stocks as the retirement date draws nearer.
The problem with age-based fund portfolios is that two people of the same age (42 in
this example) could have $50,000 and $500,000 in their respective retirement
accounts. Perhaps the participant with the larger account should become more
conservative in its investment strategy as time goes on but is that a reasonable
strategy for the participant with the smaller account? On the contrary, it may be that
this participant should become less conservative by adding a greater stock component
to increase long term return in order to catch up. Most age-based fund portfolios are
actively managed and are therefore subject inherently to higher costs and risks.
Better fiduciary conduct is to offer 403(b) participants a menu of prudent risk-based
“lifestyle” fund portfolios. Each participant determines which portfolio is appropriate for
it based on its age, and risk tolerance determined through a short online exercise. A
participant tends to stay invested in a risk-based fund portfolio because it’s geared
towards the participant’s own comfort zone vis-à-vis risk. The kind of prudent
portfolios that I’ve suggested for a model 403(b) plan (see suggestions nos. 5 and 6)
are focused on what investors can control – management of risk and reduction of
costs–rather than the attempting to corral the random variable of return. This
approach makes risk-based fund portfolios a more rational choice for participants in
7. Don’t allow revenue-sharing. Revenue-sharing is a legacy of the cost structure
that needs to be in place to support a very inefficient system of gathering assets and a
very inefficient system of distributing 403(b) benefits on the basis of individual
contracts with individual participants. The current system is based on a retail model,
not a wholesale one that could efficiently provide a true group benefit. There’s no
institutional aspect about the current system: no institutional provider and no
All my previous suggestions have been made with the intention of simplifying a 403(b)
plan. To further simplify a 403(b) plan, get rid of revenue-sharing; there is no need for
it. Insurance companies and other retirement plan services providers have bamboozled
school districts into thinking that revenue-sharing is a way of life in 403(b) plans; it
isn’t. There’s just no need for school district officials to have to deal with insolent
insurance salesmen such as the one in the Los Angeles Unified School District example
Now, there’s nothing illegal about revenue sharing per se. The problem, rather, lies
squarely with retirement plan services providers that aren’t transparent about
revenue-sharing fees and refuse to be forthcoming about them. This refusal implies
that there’s something rotten in the state of Denmark. That’s why I wonder: isn’t it a
bit ridiculous for any retirement plan services provider in the 21st century to purposely
hide its fees? Even the hint of fee camouflage and a provider saying that it’s
impossible to know what is being shared on a fund basis is simply absurd. Think about
it: why would anyone want to do business with a retirement plan services provider
that claims it has no way of knowing how much money it’s really making? Any
company that spews this kind of nonsense should be immediately removed from
consideration as a retirement plan services provider by school district officials.
8. Get rid of unnecessary costs and fees. Jettisoning annuities as investment
options in 403(b) plans (see suggestion no. 2) gets rid of mortality and expense risk
fees, death benefit charges, administrative fees and surrender charges. There’s also no
need to pay commissions on investment products, any revenue-sharing fees (including
sub-transfer agent fees, shareholder servicing fees and 12(b)-1 fees) or allow
excessive brokerage commissions under SEC rule 28(e) “soft dollars” trading revenue
since the prudent portfolios I’ve suggested for a 403(b) plan (see suggestion nos. 5
and 6) need to pay for little or no “research.” In addition, don’t pay finder’s fees or
9. Don’t allow a self-directed brokerage option (SDBO) or give plan
participants the ability to borrow from their 403(b) accounts. Let the small
minority of plan participants demanding a SDBO in order to trade exotic mutual funds
or annuities do it in their own personal non-retirement plan accounts. A SDBO is a plan
expense that other participants shouldn’t have to subsidize. Sometimes the cost of an
SDBO is actually wrapped into the overall cost of a 403(b) plan; that’s a very big no-no.
Officials at many retirement plans give in to the small but loud minority of participants
that want the ability to borrow from their 403(b) accounts. This capitulation is often a
response to those who insist that participants won’t enroll in a 403(b) plan unless they
can get their money out through a loan. That’s a lot of baloney. Those who scream
loudest for loans are the same ones who should avoid them like the destroyer of
accumulating wealth that they are. This is yet another plan expense that other
participants shouldn’t have to subsidize. Just say no to loans.
10. Enroll, educate and advise plan participants online. I’m a member of the
State bar of California, a Certified Financial Planner® and an Accredited Investment
Fiduciary Analyst®. To continue keeping the acronyms of these designations next to
my name requires that I complete continuing education, which I do – all online. So
why not enroll, educate and advise 403(b) plan participants via the internet? After all,
teachers are among the most internet-savvy folks around.
Studies show, however, that investors don’t utilize, for example, online investment
education very much. This can be overcome by tying the use of online investment
education to mandatory continuing education credits for teachers just as it’s done in
the financial services industry.
This requirement should be augmented by a voice response system, a 24/7 call center
to handle non-investment questions (e.g., what’s my pin number?) and a call center
staffed by salaried financial professionals delivering impartial investment advice
without selling any investment products. Schools districts are generally reluctant to
pay for this kind of education. Instead, the (surprisingly modest) cost for such
education can be priced (in a fully disclosed way) into a 403(b) plan which allows
school districts to avoid paying for the education. All school districts have to do is offer
such educational efforts and get behind them but they don’t have to pay for them.
11. Provide 403(b) plan participants with automatic enrollment and
escalating enrollment via the Internet. Suggestion no. 10 can be supported
further by a centralized online information system that automatically enrolls
participants in their plan (unless participants overcome human inertia and actually
take the time to opt out of the plan) and escalating enrollment by which the deferral
rates for participant accounts are increased automatically (unless participants
overcome human inertia and actually take the time to opt out of such increases).
12. Develop a Request for Proposal (RFP) and Investment Policy Statement
(IPS). The use of a RFP brings rationality to the process of finding the right retirement
plan services provider for a 403(b) plan. A well-developed IPS is the roadmap to a plan
so it should be so clear and complete that a competent stranger could read it and
understand how the plan works.
13. Despite the fact that public school districts are exempt from ERISA,
require the winner of the RFP to accept ERISA section 3(38) fiduciary
responsibility. School district officials should require any winner of an RFP to accept
via contract ERISA section 3(38) fiduciary responsibility for the selection and
monitoring of a 403(b) plan’s investment options. This delegation allows school
officials to unload much of their responsibility and most of any associated liability onto
the winning retirement plan services provider who will then owe fiduciary duties to the
403(b) plan fiduciaries (and participants and beneficiaries by implication), not just to
the funds offered by the plan.
W. Scott Simon is an expert on the Uniform Prudent Investor Act, Restatement (Third) of Trusts and Title I of ERISA. He provides services as a consultant and expert witness on fiduciary investment issues in depositions, arbitrations and trials as well as in written opinions. Simon is the author of two books including The Prudent Investor Act: A Guide to Understanding. He is a member of the State Bar of California, a Certified Financial Planner® and an Accredited Investment Fiduciary Analyst™. The author’s views expressed in this article do not necessarily reflect the views of Morningstar.