Thinking Broadly My dog had no idea that the world was larger than one square mile.
Neither did I when advocating that the United States replace 401(k) plans with a different version of defined-contribution system, dubbed the New American Retirement Plan, or NARP. That proposal, which called for a single national plan to replace the current Babel of individual company plans, struck me as sweeping. However, my colleague Aron Szapiro, Morningstar's director of policy research, published an article yesterday that changed my view. Revamping only defined-contribution plans, he writes, would be a "piecemeal" approach to retirement reform.
He was correct. (I highly recommend his article.) Defined-contribution plans supplement the first tier of retirement safety, Social Security, and therefore must be evaluated within that context, as the foundation helps to determine the upper layers. Also, writes Szapiro, one should consider traditional pensions. Their glory days are behind them, but pensions still cover 35 million workers. Finally, there is the issue of retirement income. Many countries assist their workers in converting lump-sum savings into income streams. Maybe the U.S. should do likewise.
Fine topics, for a later date. Today, however, I will do Szapiro one better, and expand the discussion further: Should the federal government mandate retirement planning? I will not answer that question, as it is for politicians, not investment researchers. However, I can sketch out the possibilities and show how defined-contribution schemes (including my NARP proposal) would fit into whichever framework the politicians adopt.
20% Some background: Szapiro's article mentions that with NARP, the combined contribution rate for workers and employees would exceed 20%, including Social Security payments. That caught my breath, but it shouldn't have. We're already there. Mandated Social Security contributions sum to 12.4%, while the median 401(k) rate, including company match, hovers near 10%. (Vanguard reports 9.8% for its plans.)
As a society, we backed into that number. Before Social Security was enacted, the contribution rate was zero. Then it became 2%, as the first iteration of Social Security required 1% payments from the company and employee. Over time, that figure gradually rose, supplemented initially by pension plans, and then by the 401(k) system. The increases were, to use Szapiro's term, piecemeal. As responses to changing conditions, they were also reactionary, rather than anticipatory.
Whether a retirement savings rate of more than 20% is too high, too low, or Goldilocks' chair cannot be judged without specifying the goal. What should retirements in America look like? Should they be uniformly comfortable? If not that, at least acceptable? Broadly speaking, I see three possible federal paths: 1) avoid imperatives entirely, 2) require a safety net, or 3) mandate investing.
Option 1: Hands Off Washington could decide that introducing Social Security initiated an 85-year mistake, whereby the government overstepped its boundaries. The Constitution expects the federal government to safeguard its citizens from foreign invasion, and from domestic violence should the states require assistance, but nowhere does it state, or even imply, that the government should oversee its citizens' financial decisions. Let the people be free. Let them make their own decisions.
That bird, as they say, has flown. The federal government will not be repealing Social Security, although it certainly will modify the program to make it more sustainable. It's worth pointing out the possibility, though, if for no other reason than this first option has been by far the most historically common, even in the relatively modern United States. This is how things were (almost) always done.
Even with the hands-off approach, the federal government would seem to have a role with retirement planning. It wouldn't mandate any savings, by either worker or employee, but it would establish guidelines to smooth the process. Only the strictest of libertarians would object to a nationally established retirement scheme that people and companies could either use or ignore, as they wished.
Option 2: Safety Net The present policy. The government requires that all workers participate in the insurance scheme of Social Security, but not that they invest. Effectively, policymakers decided that the public good was best served by compulsory service. The loss of liberty from forcing all employees to join the program, they determined, would cause less social damage than would destitute retirees.
But they wished to go no further. Some workers will be ants, diligently drying their grains for the winter, while others will be grasshoppers, dancing through the summer. Such decisions are not for the government to make. It enacted Social Security to safeguard its citizens' basic financial needs. After that, it's their choice.
Being fully voluntary while (allegedly) ubiquitous, defined-contribution plans are a suitable complement to the second option's foundation of insurance. Subject to plan limits, employees may invest as little or as much as they wish. Unfortunately, 401(k) plans are not ubiquitous. They only reach 65% of private-sector employees. NARP would fix that rather large problem by increasing coverage to 100%.
Option 3: Aiming Higher The first option was the past; the second is today; and the third may be the future. Most developed countries have decided that insurance programs alone are not enough. Their citizens aren't enamored with the idea of merely scraping by during retirement. They want something better. They wish to be comfortable.
Satisfying those expectations requires real money, raised either through taxes, ongoing investments, or a combination of the two. Were the U.S. to select this route--which is far from certain, the nation being proud of its "exceptionalism"--it almost certainly would do so by mandating that employees, employers, or a combination of the two participate in defined-contribution programs, rather than further increasing retirement-related taxes.
Once again, defined-contribution plans would play a critical role. (And once again, NARP would improve upon the current version.) They can be configured to be merely around, for those who wish to find them; or persistently present, as with today's auto-enrollment and automated-default routines; or to be unavoidable lifetime companions ("death, taxes, and defined-contribution accounts"). The structure is highly flexible.
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.