We know that, according to the Tax Policy Center, that the corporate cut alone will cost the country $2 trillion over 15 years.
Most importantly we know that if it has any hope of survival, the plan's architects must engage in a massive generational theft, and they'll to use a classic budget trick to pull it off.
The trick is a method is called dynamic scoring, which in reality is just a fancy way of justifying massive increases in the national debt.
Dynamic scoring has to do with the "growth" part of Mnuchin's explanation. In order to make tax cuts look revenue neutral, budget "wonks" what a future government might do
Now, one might think that so-called fiscally conservative Republicans would be opposed to things like this, and that Trump might face opposition from his own party.
But he won't. And that's because there is a way to make Washington's budgets sound more sensible than they actually are. That's where dynamic scoring, much beloved by deficit hawks like House Speaker Paul Ryan, comes in.
The Republican-controlled House adopted dynamic scoring last year, but it's still up for debate in the Senate, where opponents like Sen. Bernie Sanders of Vermont have been critical of the practice. They say it politicizes the budgeting process.
That's in part because there's no exact way to dynamically score anything. This is not a science. There's no set process, and there are no set rules on the assumptions made. For example, Mnuchin said during the press conference that his office is playing with a bunch of different models (reassuring).
So back when GOP lawmakers put pressure on the nonpartisan Joint Committee on Taxation to use dynamic scoring, it was unclear to Tom Barthold, the economist who heads the group, exactly what that means.
What we do know, though, that both the Reagan and Bush administrations argued that tax cuts, especially for the wealthy, would pay for themselves. In both instances, this got us in trouble.
More from the Tax Policy Center:
"If 'dynamic scoring' means that Congress can use any macroeconomic model it wants, then we are thrown back 100 or 150 years in terms of the rigor of our thinking. There are too many models with a very wide variety of assumptions and implications. It is not exactly true that you can find a model that will support any claims, but this is sometimes uncomfortably close to the truth."
So all Trump has to do is zoom in on the model that shows that cutting taxes for the rich while spending tons of money will be great for the economy, and this plan is a go.
How hard do you think it will be to find that in Washington?
An earlier version of this piece appeared on Business Insider in November of 2016.