If Time Out or New York magazine rated social programs along with boites and botox clinics, there’s no question the Earned Income Tax Credit (EITC) would be “In.” From Mayor Bloomberg to Tommy Thompson, policy makers agree that this is one program for the poor that really works. Nationwide, it has become the country’s largest source of cash aid to the poor-with little of the controversy that normally dogs such programs.

At nearly $35 billion a year, spending on the EITC dwarfs that of TANF (traditional welfare) and food stamps combined. Thanks in part to an advertising campaign and free tax-filing help, last year 700,000 New Yorkers filed for the credit, receiving an average of $2,000 apiece.

Good news for the working poor, you may say. In both its state and federal forms, the program is essentially a government subsidy for the earnings of the lowest-paid workers. But here’s the puzzle: Why do so many people who hate welfare love the EITC? Is this really a benefit for the working poor-or for their employers?

Introduced during the Nixon administration to stave off then-popular ideas of a negative income tax or a guaranteed minimum income, the EITC has grown steadily ever since, under both Democratic and Republican administrations. Ronald Reagan once pronounced it his favorite program for the poor. The editors of the welfare-averse New Republic have nothing but praise for it, while living-wage opponents like the Employment Policies Institute insist that the EITC is one labor market intervention that works. No doubt many of these surprising supporters are motivated by genuine admiration for the program’s virtues; but advocates for the poor finding themselves in such unexpected company could be excused for asking themselves, Is there a catch?

Maybe there is. Even more ominous than raves from Reagan is the program’s historical precedent: the “Speenhamland system,” a little-known law in force in England from 1795 to 1834. Like the EITC, Speenhamland linked welfare benefits to paid employment-18th century English worthies were no more eager to subsidize idleness than their twenty-first century American equivalents. Also like the EITC, it was an attempt to raise earnings without placing an additional burden on employers: If wages fell below a certain level, local governments were required to make up the difference, so the level of government benefit fell as wages rose.

However laudable its intentions, Speenhamland is now considered one of the great welfare disasters of all time. Employers soon discovered that they could cut wages freely without getting malnourished, resentful workers, or none at all; the county would always have to take up the slack.

“In the long run the result was ghastly,” wrote economic historian Karl Polanyi in his 1944 classic The Great Transformation. “Wages which were subsidized from public funds were bound eventually to be bottomless.” The result was that, as Notre Dame University economist Teresa Ghilorducci puts it, “The government subsidized wages so much they went broke.”

_______

To be sure, there are major differences between Speenhamland and the EITC. Nonetheless, the historical parallel points to complexities and potential difficulties the program’s champions often gloss over.

The most obvious problem with the EITC is that it is a creature of the tax code. Navigating the income-tax labyrinth is a headache even for educated middle-class taxpayers. For many potential EITC recipients-whose incomes are so low that in most cases they would have no need to file were it not for the program-the barriers are insurmountable. According to Amy Brown of the Community Food Resource Center (CFRC), an estimated 230,000 eligible New York workers failed to file for the program this year, even after a massive outreach program. (As Brown points out, with benefits averaging somewhat over $2,000 per person, that is half a billion dollars lost to New York’s economy.)

Another eligibility concern is the policy of capping benefits at a very low level for households without children, effectively excluding many low-wage workers from the program. Punitive family-friendliness cannot have hurt the program’s support among Republicans. On the other hand, the structure of payments can act as a disincentive to marriage: If two single workers with children, both eligible for the credit, marry, their benefit will almost certainly be less than their two former benefits combined.

These are all quibbles, though, compared with the more profound worry about the EITC: its potential for Speenhamland-like effects at the low end of the labor market.

The mechanics of the EITC can be a little tricky. There are three benefit levels, depending on how many children a household has, but the basic structure is the same for all three: wage subsidies up to a certain (subpoverty) level, after which the credit begins to diminish as the workers’ wages rise.

Consider a household with two children: for each dollar earned up to $9,540, the government kicks in an extra 40 cents. Between $9,540 and $12,460, the benefit is the same, $3,816, neither increasing nor decreasing with additional earnings. So if our hypothetical two-child household earned $10,700-the annual salary for a full-time minimum wage job-they would receive an extra $3,816 after filing their income taxes. Once our family hits $12,460, they enter the “phase-out” range, where each additional dollar earned subtracts 21 cents from the benefit, until they reach $30,580-at which point, their credit dwindles to zero.

For our hapless family, this means that after they reach a salary of $12,460-hardly a substantial income-there’s a strong disincentive to find a job with better pay. The data backs this up; economists generally agree that, except for married women, the program increases labor force participation rates, but decreases hours worked. In other words, more people work, but for fewer hours-on balance, the total number of hours worked goes up. Because of this, economists’ worries about the program typically take the form of questions about its effects on recipients.

Less studied, but potentially much more worrisome, is the effect on employers’ incentives. The phase-out range of the EITC runs from approximately $12,000 to $30,000, covering the bulk of the working poor. Add in Social Security and other income taxes, and in some states more than 50 percent of a pay raise goes to taxes-a higher tax rate than even the wealthiest taxpayers face. Employers have been quick to suss out the implications of the government subsidizing minimum-wage jobs, while heavily taxing moderate wages. In Illinois, for instance, McDonalds enthusiastically took part in a campaign to publicize the program-not surprisingly, since the majority of its employees are probably eligible.

More broadly, the danger of the EITC is that, like Speenhamland, it relieves employers of the cost of adequately compensating their workers. No doubt, this is why some businesses and their public representatives applaud the program. While traditional income-support programs, including cash assistance and labor-market interventions like wage and hour laws, tend to increase costs to employers, the EITC tends to decrease them. Over the short run, the EITC is certainly effective in moving people out of poverty. But over the long term, the danger is that wages will shift downward to take advantage of the subsidy, gradually transforming the program into an ever more costly transfer from taxpayers to low-wage employers, rather than workers.

In a 1996 paper for the economic research center the Levy Institute-one of the few to directly address this issue-Ghilarducci and Barry Bluestone warned that to substantially reduce poverty, the EITC requires a higher minimum wage. Otherwise, it runs the risk of becoming a subsidy to employers rather than to workers, and an ever-larger burden on taxpayers.

“Today, most economists believe that the EITC does encourage employers to offer jobs with low wages,” says Ghilarducci. “It infuriates me that we have subsidized, for so many years, the wages of some of the richest employers in the country-McDonalds and Wal Mart!”

None of this is to say that the EITC is a bad program, only that it can’t replace more traditional forms of income support. The EITC, for instance, does nothing for those temporarily out of the workforce. (One of the great overlooked virtues of the old AFDC program was its role as a sort of public maternity and/or sick leave for workers whose employers seldom provided these benefits.) To offset its perverse effects on the labor market, the EITC would require more direct regulation, especially increased minimum wages. “You have to have the minimum wage as well, or you’re going to have employers taking advantage of the government,” says Ghilarducci.

Perhaps the EITC isn’t even best seen as a welfare program, but as a form of economic development spending. After all, it brings public dollars to neighborhoods and local businesses that badly need them. As the CFRC’s Brown puts it, “The city has spent a lot of money bringing the Grammys back to New York. Well, this program brings in a lot more money than the Grammys.”

J.W. Mason studies economics at the University of Massachusetts-Amherst.