Le crédit d'impôt sur les revenus d'activité (Earned Income Tax Credit, EITC) a été mis en place aux Etats-Unis en 1975 pour réduire les trappes à inactivité en s'assurant que l'écart entre revenu du travail et l'ensemble des revenus de remplacement soit assez grand pour que l'incitation à travailler soit forte. Initialement réservé aux familles pauvres avec enfants ayant des revenus d'activité, il a été étendu à plusieurs reprises et notamment, en 1993, aux familles sans enfant. Le crédit d'impôt est proportionnel à la rémunération (phase ascendante) puis est plafonné à un certain seuil de revenu total (phase stationnaire) avant d'être réduit lorsque le revenu excède ce seuil. Les taux de progressivité-dégressivité et les seuils sont variables selon la composition familiale. La forme de la contrainte budgétaire correspondant à un dispositif de type EITC n'est pas fondamentalement différente de celle de l'ACR (F. Bourguignon et D. Bureau (1999), "L'architecture des prélèvements en France", Conseil d'analyse économique, Paris La Documentation Française, p. 40)

In recent years, the most important change for single mothers in the financial incentive to work bas probably coule from the Eamed Income Tax Credit. EITC credits increased fifteen-fold from $1.6 billion in 1984 to a projected $25.1 billion in 1996. Single mothers received about two thirds of these EITC dollars (1996 Green Book, pp. 808-9). In 1996 a single woman with two children who earned less than $8,890 (the phase-in range) received a 40 percent credit on dollars earned, up to a maximum of $3,556. Because the credit is refundable and a mother of two with those earnings was not subject to any federal income tax (due to the standard deduction and personal exemptions), she would have received a check from the IRS for the credit amount. With additional earnings up to $11,610 the credit amount did not change. Additional earnings beyond $11,610 and up to $28,495 (the phase-out range) resulted in a reduction in the credit by 21.06 percent of the additional earnings, until the credit was reduced to zero at earnings of $28,495: This credit schedule meant that a woman with two children earning between $5,000 and just under $19,000 received at least a $2,000 credit.

Between 1984 and 1996, there were enormous changes in many of the tax and transfer programs that affect single mothers. The Eamed Income Tax Credit was expanded, welfare benefits were cut, welfare time limits were added and cases were terminated, Medicaid for the working poor was expanded, training programs were redirected, and subsidized or free child care was expanded. All of these changes would be expected to encourage single mothers to work.

These changes were followed by large increases in the employment rates of single mothers. The employment of single mothers in a typical week rose six percentage points, while employment at all during the year rose eight and one-half percentage points. These employment increases were not shared by other low-wage groups such as single women without children, married mothers, or black men. This evidence suggests that policy changes specific to single mothers are likely to be responsible for the recent rise in their employment.

We then examine which policies were the likely cause of the employment increases. There were large relative increases in the employment of single mothers with two or more children beginning in the year when there was a substantially higher EITC for those with two or more children. There were also larger increases in employment in states with a low cost of living, where a given dollar EITC would be expected to have a larger effect.