Exceedingly High Tax Rates and the Poverty Trap

Another topic I have discussed before - see How Good Intentions Lead to Crushing Marginal Tax Rates on the Working Poor.

FreeExchange wonders if effective marginal tax rates of above 100% really do create a poverty trap:

Here is what I'd like to see CBO, or someone, put together: a statistical investigation into whether or not the trap actually traps. It's not that I don't understand the negative incentive effects at work here, it's just that those aren't the only factors being considered by individuals deciding how much to work.

If the line shown above ended with the dip just below the $40,000 level, the effects of the change in implicit tax rate would be clear. But it doesn't. Notably, it continues to the right, on a steady upward slope. And one assumes that the only way to move from a point to the left of that dip to a point on the right of that dip, which is where you want to be, is to go through the dip. Any worker opting to return to a lower income level would essentially be cutting off the possibility of moving to the fat part of the line with the next raise.

I do not find it particularly plausible that someone earning $10,000 a year would accept above-100% marginal tax rates because it is the only way they will be able to earn $40,000+ levels of income in the future. I suspect that an income level that high would be seen as too far off.

That being said, I think there are reasons to wonder how much the poverty trap actually traps. Because the tax and transfer system is so unbelievably complicated, I suspect that ex-ante people do not realize exactly how high marginal tax rates are. They only discover, ex-post that their 'take' would have been higher had that not attempted to increase their market income.