Credit where taxes are due

IN RECENT years tax credits for the poorly paid have become a big part of the welfare systems of America and Britain. Nearly a quarter of Americans are eligible for the Earned Income Tax Credit (EITC), which can boost the income of workers with big families by over $6,000 a year by reducing their taxes. In Britain similar credits account for 14% of all welfare spending.

The credits have critics. Left-wing activists have long complained that although the tax credits are paid to the poor, it is really “corporate welfare queens” such as McDonald’s and Walmart that benefit, since the credits allow them to pay lower wages. Now politicians on the right are joining in. David Cameron, Britain’s prime minister, has hinted that tax credits might be next for the chop in his government’s austerity drive.

Wage subsidies had been politically popular because they help the poor while still encouraging work. But politicians like Mr Cameron wonder whether the same objective could be achieved at less cost to taxpayers by raising the minimum wage that firms must pay workers. The hitch, of course, is that a higher minimum wage might dent employment. Tax credits, in contrast, are supposed to encourage hiring.

In practice, their design means that tax credits often have a more nuanced effect on employment. They do seem to pull single mothers into the workforce, but have the opposite effect on married women who are already working. That is because credits paid to men allow their wives to leave work without reducing the family’s income. Moreover, credits are gradually withdrawn beyond a certain level of income, which diminishes the incentive to work once the threshold has been reached.

A review of the evidence from Britain by Mike Brewer and James Browne of the Institute for Fiscal Studies, a think-tank, concludes that the extra jobs for single mothers outweigh the reduced work for married ones. A recent paper by Austin Nichols of the DeBruce Foundation, an American charity, and Jesse Rothstein of the University of California, Berkeley, reaches a similar conclusion about the EITC. In both cases, the impact on married couples diminishes the number of jobs created. What is more, there is little indication that the credits draw many men into work, though this is unsurprising given that few childless men are eligible and most married fathers work no matter what.

The evidence that tax credits reduce poverty seems much stronger. A report last year from the Census Bureau in America found that the rate of poverty among children would rise from 16% to 23% if the EITC were eliminated. Moreover, the fear that they benefit employers more than employees seems misplaced.

Tax credits create a wedge between the wages firms pay and the income workers receive. In a free market, you would expect firms and workers to respond to that difference, the former by hiring more workers, the latter by wanting to work more. Economists know that how much they respond determines who really benefits—the “incidence” of the subsidy, in the jargon. Counter-intuitively, whoever responds least gains most.

Suppose an ice-cream parlour has room for only two employees, meaning however much wages fall it will not take on any more. It employs the cheapest available workers, who cost $10 an hour to hire. If the government introduces a tax credit worth $4 an hour, it need only pay the workers $6 an hour. All of the benefit of the tax credit flows to the parlour and none to the employees.

Instead suppose the ice-cream parlour could take advantage of cheaper workers and expand. The next-cheapest worker available wants $12 an hour. Assuming the parlour must pay all employees the same wage, it now pays its three workers $8 an hour before the tax credit ($12 an hour after it). The parlour’s existing employees are $2 an hour better off, but the parlour is paying them $2 an hour less. The benefit of the tax credit is split (and a new job is created too).

An analysis of the EITC published in 2010 by Andrew Leigh of the Australian National University found that most of the benefit of the credit went to workers. Not all of it did though: a 10% increase in the credit was associated with a 5% dip in wages of high-school dropouts. By the same token, a study conducted the following year by Mr Rothstein found that for each dollar spent on tax credits, existing workers’ income rose by $0.73 (although $0.09 of this was because they chose to work more). Employers gained $0.36, as they spent less on wages.

Economists at Britain’s National Institute of Economic and Social Research are conducting a similar study of the British system of tax credits. Childless workers become eligible for the credits at the age of 25. By comparing wages either side of this threshold, they have been able to estimate how much the credits are depressing wages. Their preliminary (and unpublished) results suggest that, of the 76p an hour the government forks out in tax credits for someone on the minimum wage, 72-79% goes to workers.

Cross-subsidy

A higher minimum wage, meanwhile, might have unintended costs for workers and the government. Imagine a supermarket choosing between employing an extra cashier or investing in a self-service checkout machine. A tax credit would encourage hiring by reducing the wages paid by businesses, whereas a higher minimum wage encourages automation by increasing them. Messrs Rothstein and Nichols point out that a minimum wage and tax credits work best in tandem: a wage floor can restrict the amount by which wages sink due to a tax credit. Mr Cameron should bear their results in mind when his government presents its ideas for reducing welfare bills in the budget on July 8th.

^{Visit our Free exchange economics blog}