Google “Teresa Ghilarducci” and you’ll find countless references to her as the most dangerous woman in America. That dubious distinction stems from her 2008 book, When I’m Sixty-Four, in which she advocated replacing voluntary 401(k) plans with government-mandated savings accounts.
Her critics, who accuse her of advocating a wealth-redistributing nanny state, may be partly justified, even if their rhetoric is overheated. But Ghilarducci was attempting to address a problem that thus far has eluded solution, so it’s important to consider her arguments, which have drawn praise from some quarters, too.
Ghilarducci’s plan was cited by the New York Times as one of the best ideas of 2008, and she has maintained a high media profile in the four years since her book was published. She is a labor economist and a professor at the New School in New York.
Ghilarducci says that most Americans are ill prepared for retirement, victims of inadequate savings and poor investment results from their 401(k) plans. In a recent op-ed piece for the Times, she claimed that 75% of Americans nearing retirement age in 2010 had less than $30,000 in retirement accounts.
The core problems with 401(k) plans, she says, is two-fold: their voluntary nature, and the high fees and inefficiencies associated with managing them.
Her recommendation is to replace 401(k) plans with government retirement accounts (GRAs), in which workers would be required to save a fixed percentage of their earnings. GRAs would be managed by institutional investors, and the government would guarantee a 3% real annual return.
I’ll explain the problems with her solution, but first let’s look at the broader problem she is attempting to address.
The retirement crisis
Ghilarducci traces the problems in our retirement system to the demise of defined-benefit (DB) plans. Over the last century, worker participation in those plans has declined steadily, a trend that accelerated after the introduction of 401(k) plans 30 years ago.
That’s lamentable, she argues, because DB plans are better for workers and for corporations. The former is indisputable – any retiree would choose a pension over a 401(k) plan, especially if it is partially funded by the employer, inflation-indexed and backed by the government through the Pension Benefit Guaranty Corporation.
Her second claim is problematic. Ghilarducci says that pensions improve productivity (“or else employers wouldn’t offer them”). She reasons that workers will be more committed to their employer if they know that their retirement has been well-funded and properly managed.
In trying to explain the demise of private-sector pension plans, Ghilarducci examines a number of possible causes, but fails to accept what is surely the underlying issue. Pensions are far too expensive and risky for the private sector to finance and manage, when compared to 401(k) plans. The employer costs of 401(k) plans are low and highly transparent – a plan administration fee and whatever matching contributions the company makes. Pensions incur similar administrative fees and employer contributions, but their cost depends on actuarial assumptions, the longevity and demographics of retirees, and – most importantly – the investment results of the pension fund.
The private sector figured this out a long time ago, recognizing that the costs of pension plans far outweigh whatever productivity or other benefits they offer. Ghilarducci documents the steady decline in pension plan participation over the last half century; the watershed event was probably when IBM abandoned its pension plan in the early 1980s. Ford and GM recently became the latest major corporations to sunset their pension plans.
That has left most Americans to plan and save for their own retirements through 401(k) and other defined-contribution plans, and the results are distressing. Ghilarducci states that workers need approximately 20 times their final year’s salary for a comfortable retirement, factoring in Social Security and Medicare benefits. An alarmingly small percentage of soon-to-be-retired workers have saved that much.
Lower-class workers can rely on Social Security, and the wealthy will have sufficient assets to fund their retirement. With the loss of equity that accompanied nosediving home values, however, this emerging retirement model will be a crisis for the middle class. They, Ghilarducci warns, “will find themselves downwardly mobile.”
401(k) plans versus GRAs
Why have 401(k) plans failed to produce adequate results for retirees? And would Ghilarducci’s GRAs be any better?
Ghilarducci says one major problem is that too few workers participate because 401(k) plans are voluntary. “Basing a system on people’s voluntarily saving for 40 years and evaluating the relevant information for sound investment choices is like asking the family pet to dance on two legs,” she said. That charge is indisputable, and behavioral science is working towards fixing this by, for example, making participation the default choice for new enrollees, who would then have to opt out rather than opt in. University of Chicago economist Richard Thaler has advocated for a number of other improvements along these lines.
Her other claim, though, is very troubling. Ghilarducci says that 401(k) plans suffer because investors are forced to choose among retail-oriented investment products (mutual funds). Instead, she says her GRAs could be overseen by the same people who manage public pension funds. Those institutional investors, she claims, have the ability to select investments that will outperform the market and can do so with lower costs than investors incur in 401(k) plans.
Ghilarducci claimed that “professional managed funds get 2-3% more than retail investors, because of fees and economies of scale.”
There’s no data to support those claims and a lot of evidence to the contrary. Look no further than the nation’s largest pension fund, CALPERS. Indeed, Ghilarducci has advocated for a law that would allow California citizens to invest their retirement funds alongside CALPERS. But CALPERS earned a meager 1% in the fiscal year that ended June 30, and it barely broke even over the last five years, with a total return of 0.01%.
Those pension fund managers in whose hands Ghilarducci wants to place our nation’s retirement assets have invested aggressively in high-cost, illiquid assets, including hedge funds, private equity and venture capital. There’s no evidence that pension accounts have outperformed an appropriate risk-adjusted benchmark over the long term.
They’ve certainly incurred higher costs than 401(k) plans, though. The only difference is an accounting technicality: the management costs of 401(k) plans are paid by employees, while employers (and shareholders) pay for the cost of DB plans. Ghilarducci even cites the state of Washington’s pension plan as one that “has done well.” Yet its plan’s long-term performance has been 8.43%, less than that of a passive equity fund, and it has had notoriously high allocations to alternative investment products.
Ghilarducci ignores a critical cost of GRAs. She proposes that GRAs would guarantee an annual real return of 3%, but the market offers no such investing strategy. A risk-free GRA is equivalent to a ladder of TIPS, and TIPS yields range from -1.15% (for short-term maturities) to 0.43% (for 30-year maturities). Thus, the government – which really means taxpayers – would have to provide funds to make up for any shortfalls. In fairness, real rates have declined since her book was written in 2008, but that also underscores the risk in guaranteeing a 3% real return.
One of Ghilarducci’s primary complaints about 401(k) plans is their tax-favored status (contributions are made from pre-tax earnings). Her GRAs would instead offer a flat $600 annual tax credit to participants. She argues that 401(k) tax benefits accrue to the wealthiest individuals, since they contribute significantly more to investment accounts than their middle class counterparts. (Mitt Romney’s IRA was reported to hold $100 million.) But the same could be said of virtually all tax deductions; the wealthier make greater use of home-mortgage and charitable-contribution deductions, because they have bigger mortgages and give more to charity than the less wealthy.
The tax-favored status of 401(k) plans incents workers to save for retirement. There’s no reason to make that incentive more progressive in the tax code than the incentives for home ownership or charitable contributions.
Can 401(k) plans be improved?
GRAs are not the answer, but 401(k) plans can be improved.
Let’s start with requiring that all plans offer very low-cost, passively managed target-date funds as the default investment option. That would ensure that all participants would be guaranteed performance equal to that of the markets, with an asset allocation that would reduce risk as they near retirement.
Target-date funds are currently an acceptable default option for 401(k) plans, but many have unreasonably high costs, with expense ratios approaching 1%. Incidentally, members of Congress and federal employees are able to save for retirement under the Thrift Savings Plan that offers five low-cost index funds and low-cost single-premium immediate annuities (SPIAs).
For workers who want to control their own asset allocation, plans should be required to offer a complete set of equity and fixed-income index funds from which to choose.
As for fees, the Department of Labor recently took a modest step to improve 401(k) fee disclosure, but I would go a step further. Corporations should be required to pay the administrative costs of 401(k) plans; the only fees that employees should pay are the fund management fees (expense ratios). Such a system would incent 401(k) plan providers to lower their administrative fees; when workers bear those costs, that incentive is lessened.
Lastly, Social Security could be expanded to allow increased worker participation. Ghilarducci rejects this idea in her book, because she fears for the solvency of Social Security. Indeed, as she notes, Social Security is more progressive than her GRAs would be, because it provides disability and spousal benefits and a higher replacement ratio for low-income workers.
Our existing retirement system isn’t working. The question is whether it is so flawed that it needs to be replaced, or whether – as I advocate – it can be repaired.
Ghilarducci and I share the desire to improve investor outcomes and to provide for a more predictable and secure retirement, especially for lower- and middle-income Americans. Her solution, however, is based on the vision that institutional investors can achieve above-market returns and, if they fail to do so, then taxpayers will make up the shortfall between their results and her guaranteed 3% annual real return.
Instead, we should maintain the voluntary nature of 401(k) plans, but provide every possible incentive to stimulate participation. That, combined with some simple reforms to lower costs and broaden fund selection, will go a long way toward achieving the goals that Ghilarducci and I share.
Read more articles by Robert Huebscher
