I’m sympathetic to the basic thrust of Jody Heymann’s argument. I would, however, rephrase the problem as follows: As a nation, we’ve done a reasonably good job of providing a public safety net for poor and near-poor families—roughly speaking, families with incomes below 150 percent of the federal poverty line—while families at or above the median income—roughly 300 percent of the poverty line—are typically able to meet basic needs through a combination of employer-provided benefits and services purchased out of earned income. But families in-between—let’s call them working class—all too often lack access either to public programs or to employer benefits and cannot afford to plug the gap out of their earnings. As Heymann argues, this gap amounts to a denial of equal opportunity and has negative consequences for families and especially children.

From a policy perspective, I would characterize Heymann’s argument as moral and qualitative. She does not spend much time grappling with questions of program design and cost, or with issues of political coalition-building and sustainability. Within these self-defined limits, her case is strong. Considerations of both social morality and long-term collective self-interest should lead us to respond aggressively. But not only should we do the right thing; we should do it in the right way. Below, I submit some guidelines for a sound strategy:

(1) Programs to assist working-class families must be compatible with vigorous long-term noninflationary economic growth. The second half of the 1990s witnessed a rapid increase in incomes across the board, coupled with a significant reduction in poverty. There is no substitute for sustainable tight labor markets, which under contemporary circumstances, require sustainable fiscal discipline, as former Treasury Secretary Bob Rubin has rightly insisted.

(2) Programs to assist working-class families must also be compatible with rapid private sector job creation. Consistent with this principle, we should avoid payroll taxes and employer mandates in favor of social provision based on general revenues. This strategy has the additional merit of emphasizing that society as a whole, not just a single sector, is responsible for fostering equal opportunity. For example, the federal government should enter into a fiscal partnership with the states to create a pool of funds for paid family leaves.

(3) Considerations of both equity and affordability should direct us to means-tested programs as our basic model of provision for the working class. Means-tested programs are not inherently feeble and endangered, as many social policy advocates have wrongly argued. Many means-tested programs enjoy broad bipartisan support, and the most significant assault on working-class poverty in the past generation—President Clinton’s dramatic expansion of the Earned Income Tax Credit—was pegged to family income. The best and fairest means-tested programs use sliding scales, with the largest subsidies for the worst-off families and phased-out benefits above a suitable income threshold.

(4) We should use measures of need as a basis of eligibility for government-provided services as well as income. For example, research suggests that low-income and minority children stand to gain the most from educational reforms such as smaller class sizes and summer instruction. It would not be wise to rush toward universal programs in these areas, even if they were affordable (which they aren’t). We should start where both the need and the return on social investment are the greatest.

(5) We should distinguish between the public sector as a provider of opportunities and as a provider of services, and emphasize the former whenever possible. For example, in order to create universal access to preschool for all four-year-olds, Georgia provides sliding-scale, income-based subsidies redeemable at all pre-K providers—public, private, and non-profit—who comply with basic state requirements. (During the 2000 election, then Vice-President Gore proposed a federal-state partnership, backed by $5 billion in annual federal subsidies, to spread the Georgia model nationwide.)

A similar model could be used to ensure universal access to health insurance for working-class families: The federal government should define a percentage of income that each family is expected to contribute to its health care expenses and provide a subsidy to close the gap between that sum and the cost of a standard package of insurance benefits. The state would define basic requirements for private program eligibility to redeem these subsidies and would also allow families to buy into public programs such as Medicaid and CHIP.

(6) In designing programs to equalize opportunity, we must pay attention to research on quantifiable success and cost-effectiveness. I do not read the evidence in quite the same way as Heymann does. For example, my years of teaching education policy have convinced me that the research basis for many popular reforms is weak at best. While certain “model programs” (Heymann’s phrase) work very well, it’s hard to know whether these programs would be as effective if scaled up. Much of the research on the long-term effectiveness of Head Start, for instance, revolves around the famous Perry Preschool experiment, which is far from the typical Head Start program.

In a political environment defined by shrinking federal budget surpluses and a return to deficits in the majority of the states, we must ask some hard questions about costs and benefits. In a perfect world, for example, the school year might well be forty days longer. But each additional day could cost as much as a billion dollars. If the measure of benefit is educational attainment, the evidence suggests that we could do better by investing scarce dollars in teacher training, recruitment, and retention.

• • •

I offer these comments as friendly policy amendments within Heymann’s broad normative framework. I agree with her basic premise: justice in liberal democracies is the promise of equal opportunity, and we cannot say that working-class families in the United States enjoy that opportunity. If we are patient and smart, we can construct a coalition for fairness across party lines to implement an equal-opportunity program that public budgets and the private sector can sustain.