The chart above is based on data from the Pew Research Center's Global Attitudes project, which surveyed people in 40 countries to ask whether belief in God was necessary for someone to be a "moral person."
The results aren't just a measure of people's own religious beliefs, but also of the character of the place they're in and the exposure they have to people who aren't like them. If you've always been taught that the nature of right and wrong and the enforcement of those rules comes from the church, and virtually everyone you've ever known believes in God, those who don't would seem like something of an alien species...Commentary at The American Prospect asks why China and the United States "fall off the curve" and attributes the U.S. relatively high religiosity to the fact that "we've always had a dynamic, competitive religious marketplace."
At the other end, if you live in a place where most people don't believe in God, even if you do, you probably know many perfectly nice people who don't, so it would be harder to sustain the belief that they're all inherently amoral...
I'll offer a different viewpoint - that the high GDP of the United States is distorted by the concentration of wealth in a relatively small group of individuals, so that the per capita GDP should more properly be shifted to the left, placing the U.S. closer to the curve. (I have not checked to see whether the Pew study used mean or median GDP, which would address my postulate).
Whatever the reason, it's an interesting curve.