It's likely the working poor, as indicated by Arnold Kling in Low Income, High Marginal Tax Rate :
This does lead me to ask a behavioral economics question.
We all know that high marginal tax rates discourage labor effort. But for this effect to take place, would the taxpayer not need to know what their marginal tax rate is, or at least know that it is high? Since economists often have difficulty estimating the effective marginal tax rate of our hypothetical single mother, how realistic is it to believe that she has any idea of what her marginal tax rate is? If she does not recognize that it is high, then will it really discourage her from earning more income?
I expect that it still will, because I suspect the person in this example will thing "if I earn any more money, I'll start losing benefits, so there's no real point", even if they don't have even a rough estimate of the size of the trade-off involved.
More on Taxation
Jeff Frankel quotes Jeff Liebman:In 2000 I attended a half-day mini-conference regarding a soon-to-be-announced Canadian federal budget. I recall one of the economists stating that the highest effective marginal tax rates were not borne by the rich, but were on single-mothers with 2 or more children, earning between $20,000-$40,000 a year. The effective marginal tax rate for these women, I was told, was well over 80%. I wish I could find a reference to this in the literature, but a quick search on JStor is not coming up with anything.
"There are some excellent papers that carefully model how the cumulative effects of the welfare system create a poverty trap. But I don’t think either of these papers includes all of the factors facing the woman above — so they would probably indicate that she faced a 60 percent marginal tax rate rather than the 130% (or whatever it really is) rate that she actually faces.”
The problem is that when a poor person starts earning income, he or she loses eligibility to a variety of in-kind benefits, especially Medicaid.
In my view, the key to changing this is to change the default measurement of income from "only count cash income" to "include in-kind benefits as income." A cynic could say that those who prefer to ignore in-kind benefits do so in order to overstate poverty.
But the consequences are real. Because so many subsidies and benefits are phased out or cut off on the basis of cash income, the effective marginal tax rates on poor people are enormous.
This does lead me to ask a behavioral economics question.
We all know that high marginal tax rates discourage labor effort. But for this effect to take place, would the taxpayer not need to know what their marginal tax rate is, or at least know that it is high? Since economists often have difficulty estimating the effective marginal tax rate of our hypothetical single mother, how realistic is it to believe that she has any idea of what her marginal tax rate is? If she does not recognize that it is high, then will it really discourage her from earning more income?
I expect that it still will, because I suspect the person in this example will thing "if I earn any more money, I'll start losing benefits, so there's no real point", even if they don't have even a rough estimate of the size of the trade-off involved.
More on Taxation

Comments
But for this effect to take place, would the taxpayer not need to know what their marginal tax rate is, or at least know that it is high? Since economists often have difficulty estimating the effective marginal tax rate of our hypothetical single mother, how realistic is it to believe that she has any idea of what her marginal tax rate is? If she does not recognize that it is high, then will it really discourage her from earning more income?
It may take a PhD to calculate the marginal rate +/- 1%, but an engineer (myself) could probably nail it +/- 7 or 8 pct, and even someone with any degree can probably ‘calculate’ it +/-15 pct. If the number is between 65 and 95% does it matter exactly where it is?