Why should Social Security, Medicare, and Medicaid be untouchable and not other programs? And shouldn’t there be more to the liberal message than,“Don’t touch entitlements”?

To think clearly about “entitlements” and their role in social policy, we need to strip away the moral and emotional barnacles that have attached themselves to a simple word describing a budget category. Entitlements differ from other government spending in only one way: The amount spent is determined by the rules of the program (who is eligible, what benefits are promised) rather than by the amount set by Congress each year. Changing their costs requires changing the rules, and the projected savings may not materialize for years, which creates the need for long-term budget plans.

But the word has long been freighted with more significance than it merits. Remember the welfare reform fights of the mid-1990s, when “ending the entitlement to welfare” was the one non-negotiable position of conservatives? The implication was that individuals had been “entitled” to welfare benefits regardless of their behavior, and that ending the entitlement would force those individuals to adopt “responsibility.”

In reality, individuals never had an automatic entitlement to benefits. States could and did cut off or reduce benefits for almost any reason at all, including having an additional child while on welfare or failing to seek work. It was an “entitlement” only because of the program’s budget structure – states got money from the federal government calculated as a percentage of their own spending on welfare. And after 1996, and continuing today, when the “entitlement” had been ended, states still get money from the federal government, but now as a fixed amount —a scheme that Senator Marco Rubio recently proposed to apply to almost all federal social programs. The alternative to “entitlement” wasn’t “responsibility.” Because entitlements are a budget category, the alternative was a different budgetary approach, one that showed its fatal weakness in the long recession.

Something similar, in reverse, has happened on the left during the period after President Obama  moved toward embracing a budgetary “grand bargain,” in 2011. Entitlements have been elevated from a budget category to a moral cause. Throughout the spring of 2013, I got emails from organizations such as the Campaign for America’s Future, all imploring me to sign petitions demanding that Obama stay away from Social Security, Medicare and Medicaid in any budget negotiation. More than two million people signed. (I wasn’t among them.) Why should the big three entitlements be untouchable, and not all the other programs – K-12 and higher education, nutrition, housing, energy research – that were cut and cut again over the last three years, and, that, in total, are now at the lowest level as a share of GDP since the Eisenhower years? And shouldn’t there be more to the liberal message than just, “Don’t touch entitlements”? While the big entitlements provide economic security, they do very little to address “predistribution” – that is, incomes, opportunities, and working conditions.

The liberal devotion to the big three entitlements has two roots. One is the widely accepted assumption that only programs that are universal, or that “benefit a broad, cross-class constituency,” as Harvard political sociologist Theda Skocpol put it in her 1993 book, The Missing Middle, can attract lasting political support in the U.S. Medicare, Social Security, and to a lesser extent Medicaid (which pays the nursing home bills for millions of middle-class families) meet that criteria. The corollary is the aphorism, “programs for poor people are poor programs,” attributed to the English social researcher Richard Titmuss. Broad support for an active federal government depends almost entirely on these near-universal programs, this argument holds, and without them, voters would perceive most government programs as for poor people or minorities. The recent focus on “the 99%,” which creates a cross-class narrative by suggesting that almost all Americans have suffered while the top 1% has made huge gains, is correct about the takings of the super-rich, but also obscures the enormous difference in living standards between those in the top half of the 99%, and the 40% of households in or near poverty, who gained almost nothing during the decade before the economic crash.

But there’s plenty of evidence, especially from the subsequent twenty years, that casts doubt on Skocpol’s thesis. The State Children’s Health Insurance Program (CHIP), enacted in 1997, serves only low-income working families, but has been extraordinarily popular. Republican efforts to cut the program in 2006 hurt them in that year’s elections, Bush’s vetoes of CHIP expansion were unpopular (polling showed 72% support for the program), and Obama signed a long-delayed CHIP expansion on his 14th day in office. Tax credits for low-income families, education programs, and many other benefits that don’t reach a broad cross-class constituency have been resilient and popular. Perhaps this is because children are involved, perhaps it’s because they are seen as rewarding work or responsibility, or perhaps it’s that the programs are perceived as benefiting low-income whites as well as minorities. Whatever the reason, targeted programs aren’t always political losers. Unconditional cash transfers to poor adults, without regard to work, may still be a tough sell, but many other mechanisms to lift the economic prospects of the poor have proven popular and resilient. The more recent Republican effort to decimate the Food Stamp program, now known as SNAP, which seems to have levels of public support comparable to CHIP, will provide a further test of the proposition that programs that are not universal are uniquely vulnerable.

There’s a second, more tactical, reason that liberals are tempted to treat entitlements as untouchable. In 2005, when George W. Bush, relishing his narrow reelection to the presidency, declared that “I have political capital and I intend to use it” to push through privatization of Social Security, Democrats and liberals panicked, but soon figured out that the only response was no response. Resisting the temptation to offer their own alternatives to Bush’s privatization plan, such as the well thought-out package of fixes assembled a few years earlier by economists Peter Orszag and Peter Diamond, they stopped the process in its tracks. And that was a smart call, because in earlier negotiations on the “No Child Left Behind” education initiative, Medicare coverage for prescription drugs, and an energy bill, progressives had negotiated in good faith and moved the legislation forward, only to be cut out of the final deal.

The lesson to take from that episode is not “never touch Social Security.” It should have been, that on any issue, you should not negotiate with people who operate in bad faith, when you don’t have enough power. Social Security has been changed 30 times since it was created, often by Democrats, including Franklin Roosevelt himself. With a Democratic president and Senate that are committed to the basic concept of Social Security, changes such as a switch to Chained-CPI (a plausible method of calculating cost-of-living increases floated in 2013 that would have a small effect for most recipients, but a larger one for the very old) should be considered on their own terms, and in the context of other changes, not rejected out of hand because every detail of Social Security is sacred.

Each entitlement program is its own story. Social Security can be improved in dozens of ways, some of which would improve the long-term projected trust-fund balance and others of which might not. (Everything regarding entitlements in the budget is a matter of projections, which are a lot shakier than they seem, since they’re subject to assumptions about economic growth, population growth, and inflation. Still, Social Security projections are a lot sounder than projections for Medicare and Medicaid.) Instead of focusing on Social Security in isolation (or, worse, in the context of “entitlements”), we should look at the crisis of retirement savings for most Americans. There are a number of good small solutions, such as automatic enrollment in 401(k) plans, universal 401(k) plans, or a new system of private accounts, perhaps with a match for lower-income workers, grafted on to Social Security. But the more you look at the reality of work and retirement, in which ever fewer workers will have secure pensions, the clearer it becomes that the most efficient and even the most modern way to secure retirement is simply by expanding Social Security. This would require another level of financing as well, and possibly some benefit cuts in some areas, such as for higher earners, making the program slightly more redistributive. Resistance to any cuts in Social Security should not inhibit these initiatives.

Medicare and Medicaid, on the other hand, we should be desperate to cut. We should not aim to cut eligibility or coverage, but total costs. If, as current projections suggest, Medicare spending were actually to reach 5.6% of GDP by 2035, that would be an extraordinary level of spending in a sector that is generally not adding much productive capacity to the economy, and it would be propping up a level of total spending on health care in the U.S. that is 2.5 times the average of developed countries, with little to show for it. It is not a question of Medicare “going bankrupt.” Rather, Medicare and Medicaid, now joined by the Affordable Care Act, can provide invaluable leverage against overall health inflation, and we shouldn’t be afraid to use them. Fortunately, even without a budget bargain, the Obama administration has succeeded in putting measures in place that are likely to reduce Medicare, Medicaid, and overall health spending, against massive Republican resistance. As these dozens of experiments begin to show results, we should be willing to move quickly to expand those that work to reduce public health-care costs, even if the results won’t be fully captured in the projections of the Congressional Budget Office for some time.

Other participants in this forum are likely to argue that my initial premise – that holding entitlements untouchable has come at the expense of other spending and investment – is a false one, that there’s not a fixed federal budget that’s allocated between these two categories. That’s true, and some earlier efforts to mobilize children’s advocates and others against the entitlement programs on the grounds that they were “squeezing out” investment in kids were indeed efforts to split the progressive base. In theory, we can have plenty of both. But after three years of budgetary trench warfare, in which Republicans don’t budge from blind opposition to tax cuts, and Democrats hold entitlements off the table, the results are quite clear – one category remains vulnerable, and it’s no accident that non-defense discretionary spending is now on track to reach just 2.9% of GDP in a few years. That’s the lowest level for that budget category in more than half a century and barely half of the peak, in the early 1980s. Anything progressives can do to break this cycle of cuts to domestic spending and investment, which really does represent theft of economic potential from the future, we should embrace.  And that starts with treating the big entitlements as what they are: a budget category, not a sacrament.