The other day, I mentioned that Jim Wallis is promoting the “living wage” as major plank in the Democrats’ platform. A living wage law sets a minimum legal salary that’s high enough to allow one full-time wage-earner to support a family of three or four at or above the poverty line.

Unions, religious groups, and others on the left have invested a lot of energy in living wage campaigns–giving this policy more attention than other anti-poverty initiatives, such as the Earned Income Tax Credit or expanded coverage for Medicaid. Yet economic theory says that businesses will respond to an increase in the minimum wage by cutting workers. Empirical evidence from a “natural experiment” (when New Jersey raised its minimum wage but Pennsylvania did not) generated ambiguous or controversial results. Indeed, the effects of increasing the minumum wage are likely to be fairly complex–and different for various subgroups of poor people. Whether the net effect is good or not is an empirical question, not to be settled by economic theory alone.

Scott Adams and David Neumark of the Public Policy Institute of California have published a very well written and persuasive paper entitled “A Decade of Living Wages: What Have We Learned?” (Full disclosure: Neumark was part of the controversy about the Pennsylvania/New Jersey natural experiment–arguing that the minimum wage increase was not beneficial.) Their paper finds that local living wage laws tend to increase wages among the lowest-paid workers but also cause employment to fall among the least skilled. Both effects are bigger when the living wage applies in several neighboring jurisdictions. The two effects work at cross-purposes, but the net result is a small reduction in the poverty rate. Evidence suggests that a few of the lowest skilled people lose their jobs, but others in the same family units see pay increases; and the whole poor population gains income, on balance. But not very much income. Futhermore, the families that benefit are those that start close to the poverty line. Those deep in poverty do not gain income–presumably because they are likely to be unemployed.

The evidence about local living wage campaigns cannot tell us what would happen if the national minimum wage were raised. However, the implications of the existing studies are not very impressive. Therefore, I wonder why living wage campaigns have absorbed so much energy. It could be because …

1. We can envision a living wage at the local level, as a law passed by a city government. Some cities are very liberal, so they are likely to pass these laws. In contrast, the federal government will not do anything very progressive about poverty. And no one is advocating, for example, an earned income tax credit at the local level. But could such a policy work? [See the comment by Nick Beaudrot for a correction; many states do have their own EITC.]

2. Like rent control and environmental protection, the minimum wage is a mandate that government passes for businesses. There is no need to appropriate public funds explicitly for a minimum wage, although the cost of government may rise if the state has to pay more for labor. It’s politically easier to pass a mandate than an appropriation. Proponents predict large benefits and minimize the costs. In any case, they say, “corporations” will pay the price (and who likes corporations?). However, most economists reply that the potential benefits of a minimum wage are small, at best, and the price will be paid by consumers and some low-skilled workers.

3. Proponents of the living wage don’t like economic theory. Indeed, the dismal science rests on some fundamental assumptions that should be questioned. However, I think the limits of economic theory call for more empirical evidence, especially data drawn from experiments or quasi-experiments. My reading of the available evidence suggests that living wage laws may be mildly beneficial, but they are nowhere close to sufficient.

Since November, many Democrats have asked Jim Wallis, the editor of Sojourners magazine, to help them develop a moral message–one that might reduce the Republican advantage among religious voters. Wallis says that he has been telling them to change their policy proposals, not just their rhetoric. He writes:

the minority party has been searching, some would say desperately, for the right ‘narrative’: the best story line, metaphors, even magic words to bring back electoral success. The operative term among Democratic politicians and strategists has become ‘framing.’ How to tell the story has become more important than the story itself. And that could be a bigger mistake for the Democrats than the ones they made during the election.

… What are your best ideas, and what are you for–as opposed to what you’re against in the other party’s message? Only when you answer those questions can you figure out how to present your message to the American people.

This is 100% right, in my opinion. Wallis provides an additional service by sketching out the main points of a liberal agenda that is explicitly moral. He has prompted the right discussion, but his proposals raise questions for me.

For example, Wallis recommends that the Democrats set a target of cutting the number of abortions in half. This would move them past the Clinton-era slogan that abortion should be “safe, legal, and rare.” It could potentially give them very broad support on the abortion issue, because they would safeguard the right to terminate a pregnancy while also telling “pro-life” voters that they (and they alone) had a plan to cut the actual number of abortions.

But would the policy work? Wallis recommends “adoption reform, health care, and child care; combating teenage pregnancy and sexual abuse; improving poor and working women’s incomes; and supporting reasonable restrictions on abortion, like parental notification for minors (with necessary legal protections against parental abuse).” Democrats (including me) like it when the number of abortions falls because women have better welfare and more real choices. We do not like the number to fall because of legal restrictions, even “reasonable” ones. I could imagine embracing a policy that included both sides of the coin, as long as most of the reduction in the abortion rate came from the additional social support, not from the new legal limits. What do we know about the relative impact of those two kinds of proposals? For instance, would “adoption reform” really help?

Wallis also recommends that the Democrats take on poverty. Indeed, it is remarkable how little John Kerry said about the poor and near-poor, given that he was the candidate of the center-left. Even though the median family income of American voters is well above $50,000, I believe that some voters would respond to moral language about poverty, which would pay off politically.

Again, the issue is not what people want–they want less abortion and less poverty–but how to achieve that goal. Wallis is angry about “wartime tax cuts for the wealthy, rising deficits, and the slashing of programs for low-income families and children.” So am I. However, changing the distribution of wealth through the tax code only helps if the government spending is beneficial. Some programs help poor people, but others are wasteful or even counterproductive.

Wallis recommends a national “fair wage”: in other words, an increase in the federal minimum wage. There’s controversy about this proposal, but it appears that “moderate minimum wage increases do not reduce poverty rates,” partly because most of the lowest-paid workers are teenagers who are not poor, and partly because employers cut benefits when the minimum wage goes up. Bigger than “moderate” increases might reduce poverty; then again, they might increase unemployment.

The Center on Budget and Policy Priorities estimates that existing spending programs–mainly, Medicaid, the Earned Income Tax Credit, Food Stamps, and Supplemental Security Income–prevent 27 million Americans from being poor. Nevertheless, about 31 million remain poor, and many more are near-poor. It is quite likely that we could help more people if we expanded the existing programs, although that requires separate evidence. Furthermore, I doubt that many Americans would be inspired by a call to increase spending for Medicaid or the EITC. Who has more innovative and persuasive proposals?