The Retirement Crisis That Isn't?

Half Full
When I type "America's retirement" into Google, the first suggestion to complete the phrase is "crisis." That generates 6 million references. On the first page, Forbes states that this is "the greatest retirement crisis in American history." MarketWatch writes that the "retirement crisis will be America's next big disaster." For its part, bankrate.com doesn't even feel the need to make the case, noting that "America's retirement crisis is well documented."

Who will argue otherwise? Certainly not the financial-services industry. Nor critics of the 401(k) system, who brandish the retirement crisis as their battle flag. Not journalists, either. If crises did not exist, journalists would be forced to invent them.

Last month, though, somebody rose to the challenge. Gaobo Pang and Sylvester J. Schieber of Towers Watson issued "American Workers' Retirement Income Security Prospects: A Critique of Recent Assessments," which takes a relatively rosy view of the matter. I'm not sure what Towers Watson gains from publishing the paper--perhaps nothing, as the disclaimer states that the views are solely the authors'--but it is refreshing to see somebody, somewhere arguing against the orthodoxy.

The authors write:

We do not dispute the value of helping workers understand how much they need to save. But some recent assessments of what workers should save and when they should save it dramatically understate the adequacy of retirement savings for many households.

While some might view overstating the retirement-savings problem as a good thing, on the principle that one can never be too prepared, the authors disagree:

If workers are led to believe that they are so far off a reasonable savings path that an adequate retirement income is beyond reach, it raises a question of whether some workers will not become so discouraged that they give up saving in the future. In other words, some of the recent bleak assessments may actually be exacerbating the retirement risks workers face.

(That passage struck home. In a previous Morningstar existence, I oversaw the development of Morningstar's retirement-advice software. In early testing, almost half the users quit the program after seeing their "gap" page--the difference between the retirement income they sought and the retirement income they were on track to receive. We modified the software to present the news more gently, thereby improving the dropout rate, but that section remained difficult.)

Key to the paper is the life-cycle approach, which models lower savings rates, higher consumption of temporary goods (for example, raising children, accumulating the down payment for a house), and, frequently, a reduced lifestyle during the middle earning years. In later years, the retirement-savings rate can and will increase, as the wage earner no longer faces those temporary costs.

The authors are not the first to adopt the life-cycle approach, of course, but they use it to push hard on the question: What does it mean to maintain a similar lifestyle to the working years during retirement? Surely, they argue, this can't be found by simple comparisons of preretirement and in-retirement earnings. If the worker's later-year wages are similar to those of the middle years, then targeting retirement income that matches the conditions of the later years effectively gives that worker a higher retirement lifestyle than he enjoyed on average during the working years--because, for many years, that level of income was used on costs that no longer exist.

Many articles that claim a retirement crisis cite low retirement-savings rates of midcareer workers and extrapolate those rates into the future. The authors are not buying:

The idea that workers can or will save as much when they are buying their first home, starting a family and educating their children as they will after the children finish school and move out, and the mortgage is paid off is simply not feasible. There are significant differences in appropriate savings rates across households of different ages.

Children, of course, are a significant factor. The authors cite research showing that a couple without children is the cost equivalent of 1.63 adults living alone, while a couple with two children is the equivalent of 2.64 adults. That is, during the long child-rearing phase, the couple with two children must support 1.01 more adults (presumably, Gulliver plus one Lilliputian) than it must support during retirement. Clearly, that couple can have a significantly lower income during retirement than it had during the child-raising years, while maintaining a similar lifestyle.

The authors cite several studies, in addition to Fidelity's retirement software, as establishing a retirement-income target of 85% of preretirement, pretax income. According to the authors, such an amount will likely give the retiree more consumable income during retirement than he had while working. The paper advocates replacement-income ratios of roughly 65% to 75%. Of course, as the authors acknowledge, even this estimate is simplified, as households that raised many children are in a different position from those that raised none.

The paper addresses several other topics: that defined-benefit plans can sometimes over-save, so that between Social Security and the pension benefit the worker has a wealthier retirement than during the working years; that Social Security benefits are not always modeled correctly in other studies; and policy proposals from the New America Foundation and Teresa Ghilarducci's Economic Policy Institute (which the authors pan). Those are beyond the scope of this column.

The authors save their strongest language for the conclusion:

Using measures of preretirement income that significantly exceed workers' real earnings over their careers reduces the perceived income replacement capacity of Social Security and raises the specter that workers will not be able to maintain their preretirement spendable income when in fact their retirement income prospects are much more optimistic than that. If we wish to make our situation look dire, this might make some sense but, if we truly want workers to achieve standards of living in retirement comparable to the way they lived during their working years, it makes no sense at all.

There is no doubt that some portion of the workforce is inadequately preparing for retirement. But poorly conceived standards of how much workers will need as they approach retirement, naive models of retirement savings behavior and underestimates of existing retirement annuity and savings programs cannot help us discern how many workers are at risk, who they are and how best to help them.

I wish I could write that the authors are correct. I have long distrusted the term retirement crisis, as it seems unlikely that things are worse today than in the past (over the past 30 years, income for those over the age of 65 has grown faster than the income for any other age segment). It's the sort of thing that people say because it makes sense to them and they wish to believe it. So, I do appreciate the gauntlet being thrown--and I suspect the paper is on track. However, it's a complex subject, and I do not yet have the expertise to make that call.

Well, perhaps in time. In my next column, I'll review the authors' recommendations for replacement-income ratios. 

John Rekenthaler has been researching the fund industry since 1988. He is now a columnist for Morningstar.com and a member of Morningstar's investment research department. John is quick to point out that while Morningstar typically agrees with the views of the Rekenthaler Report, his views are his own.