By Ezra Klein
Washington Post Staff Writer
Saturday, November 27, 2010; 2:50 PM
For anyone who would like to retire someday, reading the news can be depressing. There's the Social Security shortfall, of course, and the reports that pensions promised to state employees are terribly underfunded. The financial crisis wiped out plenty of 401(k)s, and many families had to borrow against - or withdraw from - their retirement savings to stay afloat.
Too often, these are three separate conversations, and they're frequently lashed to other issues. Social Security is usually discussed in terms of the deficit, and Washington brims with plans for cutting Social Security benefits. The conversation over state pensions usually is really about the sustainability of state budgets. And when we talk 401(k)s, we tend to be talking about volatility in the stock market.
But really, these are all the same conversation, about the same problem: retirement security. The late 20th century saw a great shift in risk, in which uncertainty that had been borne by employers and the government was shunted onto individuals. And in our efforts to solve our deficit and economic problems, we must be careful not to make our retirement problem worse.
Consider the 401(k): When Congress created the provision in 1978, lawmakers didn't realize they were going to transform the American pension system within a generation. But that's what happened. Previously, employers had defined-benefit systems in which they had to worry about saving enough to pay for the retirement of their workers; the 401(k) - and similar defined-contribution systems - let them push that responsibility onto the workers.
By 1995, there were more 401(k)s plans than traditional pension plans. Now there are about twice as many. And they're not working out that well, Robert Hiltonsmith, a policy analyst at the think tank Demos, shows in his paper "The Failure of the 401(k)."
The failure, experts say, basically, is this: The typical worker approaching retirement needs about $250,000 in a 401(k). Most don't come close. The average is closer to $98,000 - only a bit more than a third of the recommended amount.
That's necessary context for considering the arguments over Social Security and state pensions. In both cases, most of the solutions don't solve the problem so much as switch to a different one. The costs come off the government's balance sheet but land on the backs of individual retirees. And it's not clear how those retirees are supposed to pay for it. On average, baby boomers can expect to subsist on an income of about 77 percent of what they earned in their peak working years. For Gen-Xers, it's 65 percent. And if they have the bad fortune to retire during a market slump, well, it's not clear what they'll do.
The deficit-reduction plans under consideration will accelerate this trend. Under current law, an average -income worker (that's someone making about $43,000 a year) is projected to get the inflation-adjusted equivalent of about $15,000 a year in Social Security benefits in 2050. Under the Simpson-Bowles plan, that would drop to about $12,700. That's not nothing, but it's not much. In fact, neither amount is very much. Social Security is not a generous program. It's actually among the least-generous of any developed nation, and in making it less so, we raise the question of what, exactly, seniors with insufficient retirement savings are supposed to do.
Social Security, after all, is not a minor part of most retirement incomes. For more than 20 percent of couples and 40 percent of individuals, it provides more than 90 percent of their retirement income. We can take some of those costs off the government's books, but then we need at least a theory for how seniors would make up the shortfall.
The best-case outcome is that they'll work longer, contributing more to the economy. But that's not always possible. Age discrimination is rampant. Employers know that younger workers are often more productive, and always cheaper, so they push out older workers or refuse to hire them. And once unemployed, it's particularly hard for older workers to find new jobs. It's something that's become all too clear in the current recession, which has seen a startling rise in long-term unemployment among older workers.
There are policy solutions that might work, but they require looking at retirement security as a whole and not just as a subcategory of deficit reduction. Increasing benefits for low-income seniors even as you reduce benefits for high-income seniors probably makes sense. So, too, does automatic enrollment in 401(k)s, which has been shown to vastly increase participation. Hiltonsmith supports an expanded version of the 401(k) called a "guaranteed retirement account," in which contributions from employees, employers and the government would be mandatory.
And the problem involves more than just pensions. If we want seniors to work longer, we're going to have to think harder about why so many don't. Age discrimination is often part of it, and so, too, is physical exhaustion among those with manual jobs. Many experts suggest that the typical retirement - where you go from working one day to not working the next - could be replaced by partial retirement, where workers move to more flexible, part-time arrangements as they get older.
Shifting risk is not the same as eliminating it, and sometimes can even add to it. With the rise of 401(k)s, fewer employers need to worry about pension costs, but more workers need to worry about retirement security. As Congress and state governments turn to address the solvency of Social Security and state pension systems, they need to do more than just continue moving the costs to retiring workers. We need to solve our budget problems, of course, but in a way that doesn't worsen our retirement problems.