Catherine Rampell provides the simple, nickel version of why the Romney/Ryan tax plan is mathematically impossible:
As the Tax Policy Center demonstrated, cutting individual income tax rates by 20 percent from today’s levels would reduce tax burdens by $251 billion per year (in 2015) among households with income above $200,000.
If you leave preferential tax rates for savings and investing (e.g., long-term capital gains and dividends) untouched, as Mr. Romney has said he would do, that leaves only $165 billion of available tax expenditures that can be eliminated from this same group of high-income earners once their marginal tax rates fall.
In other words, even if you completely eliminated all tax deductions for high earners — the mortgage interest deduction, the charitable deduction, the exclusion of healthcare benefits, etc. — it still wouldn't make up for the 20% rate cut Romney wants to give them. Their total tax bill would go down. However, Romney has also said that his total tax plan is revenue neutral, which means that someone else's tax bill has to go up to make up for the tax cuts he's giving to the rich. But Romney says that won't happen either. Middle-class taxes, on net, will stay the same. In other words:
-$251 + $165 + 0 = 0
In my 7th-grade pre-algebra class, this bit of arithmetic wouldn't have passed muster. Maybe they taught math differently at Cranbrook. In any case, all I'd really like to see from Romney is a proof of concept. It doesn't have to be his final plan or anything like that. Just any combination of a 20% rate cut and the closing of tax deductions that produces no net tax decrease for the rich. Anything at all that proves it can be done.
But he can't do it, and he knows it. Not without invoking the dynamic scoring fairy, anyway. But at this point, more than a decade after George Bush's tax-cutting extravaganza, if you still believe that tax cuts for the rich will hypercharge the economy and produce huge pots of free revenue, then you deserve whatever you get. For the rest of us, we just want to see the math. It really shouldn't take too long.