Defined contribution plans, as account-based plans, in which nearly all retirement savings risk is borne by the participant, present (broadly) three “adequacy” challenges: (1) adequate savings; (2) adequate investment; and (3) adequate payout. Tools have been developed, over several decades, to address challenges (1) and (2), generally through the use of defaults, savings incentives, and fiduciary practice (e.g., in fund menu design).
There has been less success developing solutions to challenge (3) – translating the DC account into a stream of retirement income that will last over the participant’s (generally uncertain) expected life.
In this article we consider the obstacles to finding a DC “lifetime income solution” and then survey the (limited) policy initiatives that have thus far been adopted or proposed to address the DC/401(k) payout challenge.
Obstacles to lifetime income solutions
There are several obstacles to the development of a DC retirement income solution, including those inherent in DC design, current regulatory policy, and individual biases and (sometimes clearly rational) preferences.
DC plans are generally designed to pay out lump sums
To begin with, as account-based plans, the “intuitive” form of payment of a DC benefit is a lump sum equal to the account balance. Further, payment of a lump sum comports with the 401(k) system’s general bias towards devolving as much decision-making as possible to participants. Instead of locking her into a particular payout form, the participant is simply given her cash, and she can then choose what sort of payout strategy – e.g., annuity or IRA drawdown – she prefers.
Internal Revenue Code rules also encourage this approach: qualified joint and survivor annuity (QJSA) rules generally require that benefits under a tax-qualified retirement plan be paid in the form of a qualified (spousal) joint and survivor annuity, unless the participant elects a different form with the (notarized) consent of the participant’s spouse. This rule does not (with limited exceptions) apply to a DC plan where (1) the participant’s benefit is payable on the participant’s death to the participant’s spouse and (2) the participant does not elect payment in the form of a life annuity.
The result is that most 401(k) plans, other than those marketed by annuity carriers, simply provide for payment on termination of employment in the form of a lump sum with, perhaps, an option to have payment over a period of years (e.g., over the participant’s life expectancy).
There remains widespread concern that sponsor fiduciaries may be held responsible, perhaps for decades, for the financial viability of any annuity carrier selected to offer an annuity under the plan.
In 2008 DOL finalized a regulation providing a safe harbor for the purchase of annuities in a DC plan. While that rule improved on DOL’s “safest available annuity” standard, it still imposed significant duties on plan fiduciaries, including that the fiduciary appropriately conclude that “at the time of the selection, the annuity provider is financially able to make all future payments under the annuity contract and the cost of the annuity contract is reasonable in relation to the benefits and services to be provided under the contract.”
While not widely discussed, nondiscrimination rules require that in-plan annuities be “unisex;” those rules do not apply to outside-the-plan annuities. Given these underwriting policies, it is conceivable that, e.g., for a male, an outside-the-plan annuity may be a “better deal” than an in-plan annuity. The inefficiency of the individual annuity market (relative to the institutional annuity market) has generally obscured this issue. But if that inefficiency is in the future eliminated, or at least significantly reduced, this difference may become an obstacle to in-plan annuities, at least under certain circumstances.
Individual bias against annuities
The consensus among sponsors and providers is that participants are significantly biased against selecting annuities and in favor of taking lump sums. Many different reasons have been given for this, including: (1) the human bias in favor of present consumption over future security (aka hyperbolic discounting); (2) the complexity of annuities as a product; and (3) inefficiencies in the annuity market.
In addition, we would note at least two other factors. First, Social Security, which ispaid as an annuity, represents a significant portion of the retirement benefits for lower paid employees. And second, research shows that uncertainty, e.g., about life expectancy and possible future spending needs, plays a significant role in participants not electing annuity payouts.
The heterogeneity of needs and preferences in retirement
Finally, and perhaps most problematic, is the heterogeneity of individual needs and preferences in retirement – some of which is implicit in the last point we made about participant bias. Some participants may (often quite realistically) see themselves as likely “losers” in the annuity tontine – compromised health (for instance) may make them likely to die before their “average life expectancy” date.
In addition, a participant might also anticipate, or at least be anxious about, significant, unpredictable one-time expenses.
Finally, there is conflicting data as to whether spending in retirement is actually “smooth.” Where, e.g., the participant anticipates significant expenses early in retirement (on, e.g., travel and the “bucket list”), followed by a long period of lowered activity/lowered spending, followed by significant end-of-life medical expenses, an annuity may not look like a good fit.
Policy initiatives to encourage in-plan annuities
In this context, the policy initiatives we have seen to encourage the inclusion of retirement income solutions in DC plans have been modest and limited.
QLACs: On July 1, 2014 IRS released its final regulation on longevity annuities, generally providing a limited exception to the required minimum distribution (RMD) rules for qualifying longevity annuity contracts (QLACs), generally defined as insurance company annuities the premiums for which do not exceed the lesser of $125,000 or 25% of the employee’s account balance.
There are proposals in Congress to expand the QLAC RMD exemption, e.g., the Portman-Cardin proposal to eliminate the 25% of the account balance limit and increase the dollar limit to $200,000 and to allow a “free look period” and certain variable and indexed annuity contracts to qualify as QLACs. The Family Savings Act of 2018(part of House Republicans’ 2018 Tax Reform 2.0) included a $50,000 exemption from RMD rules.
More remotely related to the lifetime income issue, President Trump instructed Treasury to update its mortality tables for purposes of calculating RMD requirements in his August 2018 Executive Order on Strengthening Retirement Security in America.
Rev. Rul. 2012-3 clarifying the application of QJSA rules to 401(k) plans offering annuities as investment options: This revenue ruling holds that the QJSA rules (described above) do not apply until annuitization where, under an annuity investment option in a DC plan, the participant can transfer assets out of the annuity contract at any time before annuitization. It is intended to facilitate participants “getting used to the idea of annuitization” by investing in them during the accumulation phase.
Rev. Rul. 2012-4 providing a “roadmap” for certain DC to DB rollovers: This revenue ruling generally provides a compliance roadmap for DC to DB rollovers. It is intended to make it easier for a DC participant to “annuitize” her DC account by directly rolling it over to an (annuity-providing) DB plan (or at least clarify how DC-to-DB rollovers are to be done).
Lifetime income disclosure: In 2013 DOL released an “Advance notice of proposed rulemaking” (ANPR) describing changes to rules for defined contribution plan benefit statements that would require inclusion of disclosure of an estimated lifetime stream of payments based on (1) the participant’s current account balance and (2) the participant’s account balance projected to retirement. There was considerable push-back against this project; many sponsors have developed their own (differently formatted) lifetime income illustrations, and there was concern that basing illustrations on projected account balances might be misleading.
Annuity portability: There are a number of Congressional proposals to facilitate the portability of lifetime income options (generally, annuities), e.g., when a specific annuity option is eliminated from the plan fund menu, e.g., Congressman Neal’s (D-MA and now Chairman of the House Ways and Means Committee) Automatic Retirement Act of 2017.
ERISA Advisory Council Lifetime Income Recommendations: The 2018 ERISA Advisory Council recommended that DOL modify current QDIA regulations to allow the inclusion of annuities and other lifetime income options in QDIAs and allow outsourcing of fiduciary obligations with respect to the selection and management of lifetime income options. With respect to the latter recommendation, DOL “should publish guidance confirming that a named plan fiduciary may appoint a 3(38) investment manager to select and monitor annuity and other [lifetime income] options for DC plan decumulation, as well as accumulation.”
* * *
As an increasing number of American workers are retiring with DC account balances, concern about the adequacy of DC/401(k) payout options and strategies has significantly increased, among policymakers, sponsors and providers.
We will continue to follow this issue.