Cap and Trade Revenue

This is from the budget outline released by the Obama administration on Thursday:

After enactment of the Budget, the Administration will work expeditiously with key stakeholders and Congress to develop an economy-wide emissions reduction program to reduce greenhouse gas emissions approximately 14 percent below 2005 levels by 2020....

I wonder what their economic assumptions are here?  Here's the revenue timeline, starting in 2012:

At first glance, this strikes me as odd.  With only slight variations, it assumes $80 billion in revenue every year between 2012 and 2019.  But that doesn't really make sense.  What you normally expect with a carbon trading program is that you begin with a high cap (carbon emissions in 2012 will probably start out 10% higher than 2005 emissions) and then ratchet the cap down every year after that.  As the cap goes down, the price of permits goes up.  It's true that the number of permits goes down at the same time, but this shouldn't be enough to make up for the higher permit price.  Overall, until the green technology buildout hits a critical mass, the revenue from the program should go up considerably over time.

But not in this one.  I wonder why?