Public Inaction—and Private Retreat—in the Face of Rising Economic Risks
What can be said with certainty is that existing public and private policies
are not adequately protecting families against economic instability in a new
era of work and family. One revealing piece of evidence is provided by the PSID
data discussed earlier: whereas family income drops were substantially
cushioned by government taxes and transfers in the 1970s, the cushioning effect
of these interventions has declined dramatically. The median drop in family
income in the early 1970s was roughly a third smaller than it would have been
without taxes and transfers. By the early 2000s, taxes and transfers hardly
made a dent in the median drop in family income. To be sure, this may partially
reflect the growing role of in-kind benefits, such as health care, which are
not included in the PSID. But it seems clear that, at least when it comes to
income protection, government is not doing as much as it once did to help
families that experience economic shocks.
In part, this may be because U.S. social programs are increasingly focused on workers’ retirement years, rather than on their youth or working lives. In comparative perspective, only Greece and Japan have a greater skew in their public social policies toward cash and services for the aged (Lynch 2000). Though protections for retirees are vital, the growing economic insecurity of younger Americans suggests that there are major gaps in protection for economic risks that affect families during the working years.
At the same time, public programs of social insurance have come under increasing strain. Most notably, unemployment insurance has contracted in reach and generosity during a period when job insecurity has risen. Between 1947 and 1995, the share of workers in covered employment who actually received benefits fell from 80 percent to less than 40 percent (Graetz and Mashaw 1999, 76). The GAO recently reported, in a brief entitled “Unemployment Insurance: Role as a Safety Net for Low-Wage Workers is Limited,” that low-wage workers are particularly unlikely to receive unemployment benefits. In 1995, only about 18 percent of unemployed low-wage workers were collecting benefits (GAO 2000).
Not only is government less involved in risk protection; employers have also cut back many of the generous benefits they once provided as a matter of course. The United States is unique in the extent to which workers rely on employers for basic benefits, like health care and retirement pensions—benefits that in other advanced industrial democracies are provided mostly by government. Indeed, I have shown elsewhere (Hacker 2002) that the American welfare state, with its heavy reliance on tax-subsidized private benefits, is as large as many European welfare states when such benefits and comparative tax burdens are taken into account.
Yet these private forms of risk pooling are in broad decline. Between 1979 and 1998, the share of workers with health insurance coverage from their own employers fell from 66 to 54 percent; among the lowest-paid fifth of workers, the proportion fell from 46 to 26 percent. Retirement pensions, the other major workplace benefit, have likewise declined in reach among lower-paid workers. Yet the major change in pensions is the dramatic shift away from defined-benefit pensions that promise a fixed payment in retirement toward defined-contribution retirement accounts. As recently as twenty years ago, nearly half of workers were covered by a defined-benefit plan. Today, only 20 percent are, and the share continues to fall as the generation covered by such plans retires or dies. Meanwhile, 401(k)s and other defined-contribution plans have gone from nothing to a national obsession. By 2001, nearly 60 percent of families nearing retirement had some money in a defined-contribution pension plan in 1998, up from 11 percent in 1983.
Because defined-contribution plans are essentially private investment accounts sponsored by employers, they greatly increase the degree of risk and responsibility placed on individual workers in retirement planning. Traditional defined-benefit plans are generally mandatory and paid for largely by employers (in lieu of cash wages). They thus represented a form of forced savings. Defined-benefit plans are also insured by the federal government and heavily regulated to protect participants against mismanagement. Perhaps most important, their fixed benefits protect workers against the risk of stock market downturns and the possibility of living longer than expected.
None of this is true of defined-contribution plans. Participation is voluntary, and due to the lack of generous employer contributions, many workers choose not to participate or contribute inadequate sums. Plans are not aggressively regulated to protect against poor asset allocations or corporate or personal mismanagement. The federal government does not insure defined-contribution plans. And defined-contribution accounts provide no inherent protection against asset or longevity risks. Indeed, some features of defined-contribution plans—namely, the ability to borrow against their assets, and the distribution of their accumulated savings as lump-sum payments that must be rolled over into new accounts when workers change jobs—exacerbate the risk that workers will prematurely use retirement savings, leaving inadequate income upon retirement. And, perversely, this risk falls most heavily on younger and less highly paid workers, the very workers most in need of secure retirement protection.
Recent research by Edward Wolff suggests just how serious these shortcomings are. Examining Federal Reserve Board data, Wolff finds that median retirement savings actually declined between 1983 and 1998, despite the massive run-up of the stock market during this period. One cause of the decline was the Social Security reforms of 1983, which lowered guaranteed benefits under the government system. Additionally, however, the value of defined-benefit plans also declined dramatically, as defined-contribution plans became more common. The net effect was not only to lower net retirement wealth, but also to make the distribution of retirement wealth, always relatively unequal, even more skewed in favor of the well off. In 1983, a family with enough wealth to place it at the 99th percentile of the wealth distribution held just over four times as much pension wealth as families in the middle of the wealth distribution. By 1998, they held almost eleven times as much. These trends are likely to be exacerbated as defined-benefit plans, like sediment from a previous era, slowly disappear from the pension landscape.
The problem is not just that older forms of guaranteed insurance in the public and private sectors are slipping away. Perhaps more important, it is that few of America’s strained social programs have been retooled to deal with the new and newly intensified risks to income of the post-industrial, two-earner economy. The current framework of social insurance was constructed in an era in which the key economic risks were seen as a temporary interruption of male wages and inadequate income in retirement. Today, even well-educated workers face a heightened risk of being displaced from employment without prospects for rapid reemployment at comparable levels of earnings, and women are much more likely to be breadwinners than to stay home to care for children. The distinctive risks to family finances created by these shifts are not well insured by present policies.