The Living Wage Myth
The heat under the "living wage" debate has been turned up a notch. The Public Policy Institute of California has just published an extensive study by Michigan State economist David Neumark, the main conclusion of which is that, despite causing increased unemployment among the lowest-wage workers, "living wage laws raise the wages of low-wage workers."
It has been seized upon by proponents as a "proof text" against critics. The success claimed for such policies by that study, however, is, in fact, far less than implied by the innumerable "Living Wage Laws Reduce Poverty" stories it has spawned.
The study concludes that a living wage one and a half times the minimum wage would raise the average wages of the lowest tenth of income earners by 3.5 percent. That is, it would increase total earnings for the group and, as a result, slightly decrease the likelihood of families having an income below the poverty line.
That is hardly a ringing success. After all, it simply means that the unintended consequences of living wage laws, such as increased unemployment of low-wage workers, are not so large that they totally undermine the intent of the policy, so that there are actually some gains, on average, to low-income earners.
If a policy designed to help low-wage workers actually hurt them, on average, that would certainly be a failure. In fact, that is what Professor Neumark's research indicates occurs by raising the minimum wage--it actually increases the chances of a family being below the poverty line (though groups touting Neumark's living wage results, which also push for higher minimum wages in the name of helping the poor, have not abandoned that campaign as a result). But to achieve any good at all for the intended beneficiaries, on average, is surely too low a standard to call a program a success.
Neumark's conclusion that the average low-income workers gain from living wage laws does not establish that they meet the tests of either efficiency or equity, reflected in the fact that he does not endorse such laws as a result of his study.
His finding of significantly decreased employment with higher average wages means that living wage laws harm many of the poorest workers we are trying to help. And we cannot assume away these inequitable results among the poor simply because the winners gain more in total income than the losers lose; the harm does not disappear because one's gains somehow cancel out another's losses, as is implied when the focus is solely on the "average" effect.
In addition, living wage laws are poorly targeted to help low-income families. They only apply to a small fraction of the low-wage labor force (for which Neumark has said he "can imagine no rationale," if the point is to help low-income families generally). Further, they are paid whether the workers support families or not, and even when total income from multiple workers in the family already puts them well above the poverty line.
Another virtually unreported problem with the study's headlined conclusion that living wages raise low-wage workers' incomes is that his results hold only for some types of living wage laws. The most common type (including those in Baltimore, Boston, Chicago, Denver, Milwaukee and San Francisco), which only apply to contractors supplying goods or services to the government in question but not to all those receiving substantial government assistance or to government employees, does not.
As Neumark says, "the poverty reducing effect of living wage laws stems solely from business assistance laws...the estimated effect of contractor-only laws is small and insignificant." When the most common type of living wage law has no measurable positive effect on poverty, it cannot be justified as an anti-poverty tool.
Beyond even these issues, Neumark's study warns that "many important questions remain to be addressed before policy analysts should feel confident that they have a well-established set of findings on which to draw strong conclusions regarding living wages." These questions remain because the study only addressed the "first-order," or direct, effects of living wage laws, and ignored indirect effects which do not show up in income data but could more than offset the first-order effects in helping the poor.
Neumark's study ignores the increased municipal budgets from higher wages, which will force them to reduce services, many of which are provided to low-income families. Alternatively, it will require higher taxes, including those borne by the poor. To the extent living wage laws apply to nonprofits serving the poor, those organizations will also have to cut back on the services they provide the poor. And to the extent that suppliers' higher costs get passed on in higher prices to users beyond the government, the poor are also harmed (particularly because they spend a larger fraction of their income on consumption purchases than higher-income families).
While living wage laws cannot be justified as reducing poverty, Neumark found that they did significantly boost the wages of unionized municipal workers. And in a significant understatement, he wrote, "Because contractor-only living wage laws do not appear to be associated with benefits to low-wage workers or low income families, evidence that these types of living wages benefit unionized city workers would tend to cast living wage laws limited to restrictions on wages paid by city contractors in a negative light."
How do living wage laws benefit unions? Living wages directly increase wages for lower-skill union workers who previously negotiated below living wage contracts. Further, by forcing producers to pay higher wages even if they are nonunion, they reduce competition from nonunion companies, whose costs are forced up (the mechanism Neumark emphasizes). Because of the increased costs, however, such laws also undermine municipalities' attempts to save their taxpayers money by privatizing public services or by putting welfare recipients to work, either of which threatens union jobs.
By forcing affected contractors to pay employees more, living wages allow unions to get pay raises and job security they couldn't get through traditional means. Because they apply to government contracts, they use governments' ability to force taxpayers to bear increased costs, which unions couldn't use in direct negotiations with private employers (much like the Davis-Bacon prevailing wage law for federally funded construction).
Living wage laws also provide other props for unions. They substitute for attempts to impose higher local minimum wages, which would intrude on state authority. By dealing only with municipalities where supporters' power is strong, they divide and conquer opposition that would prevent passage at the state level. And living wage laws are intended as wedges for further wage and coverage expansions in the future. In any event, as local governments expand their reach, involving themselves in ever more aspects of the local economy, these effects will increase over time.
The main conclusion of Neumark's study--that living wages reduce poverty--is not as strong as the headlines make it appear. The study actually shows that some low-income workers may gain more than other low-income workers lose from some types of living wage laws. That is, it only helps "on average," while directly harming low-income workers who lose their jobs as a result. And the most common type of living wage law produces no detectable progress against poverty, even on average.
Further, those underwhelming conclusions hold only when additional indirect effects which will harm low-income families are ignored. But those laws give unions substantial raises, ultimately funded from higher taxes and reduced services from municipal governments. As a result, a careful reading of his study leaves us a long way from the headlines that "Living Wage Laws Reduce Poverty," unless it only refers to union members.