Ask any poverty expert, and they'll tell you that the Earned Income Tax Credit is the most effective anti-poverty program in the United States. In 2010, it lifted 6 million people out of poverty, half of whom were children.

But like many government programs, it's hard to take advantage of the EITC if you haven't heard of it, meaning that it has the potential to benefit a great deal more people than it's helping out now. What's more, as with many tax provisions, it takes a fair bit of knowledge to get the maximum benefit from the EITC. So, even those currently receiving it could perhaps be benefiting more than they are today.

A new paper by Raj Chetty and John Friedman of Harvard and Emmanuel Saez of UC Berkeley tries to nail down exactly how big this knowledge effect is. They locate neighborhoods with high levels of knowledge of the EITC by seeing which areas have the most self-employed people reporting the exact amount of income that would maximize their credit. The idea is that if a lot of people in a given area are being that savvy about reporting income, that area has a high level of tax knowledge.

And, interestingly enough, the areas with the most knowledge about the EITC tend to be those with high concentrations of poverty: big cities like New York and Los Angeles, states such as Michigan and those in the South:








The darker the region on the map above, the higher the percentage of people reporting just the right amount of income to max out their rebate. The authors also found that all people--not just self-employed people reporting their own income--living in higher-knowledge areas got more out of the EITC than people in lower-knowledge areas. All people get a larger credit if they have a child, but as the following graph shows, you get a larger bump if you're in a high-knowledge area:









That green line is the size of the credit for people in high-knowledge areas, the red line for medium-knowledge areas, and the blue line for low-knowledge areas. Sure enough, those in high-knowledge areas get the most out of the credit. All told, the authors find that those in high-knowledge areas get an average of $122 more from the credit than those in low- knowledge areas -- a 5.1 percent increase. What's more, those who move between areas change their behavior, with those moving to high-knowledge areas becoming savvier about getting the credit and those moving to low-knowledge areas growing less likely to max it out.