Fermi approximations and estimating political commitments

I like journalists who use the usefully practical tool of a Fermi approximation to estimate whether a proposal is practical or feasible.  In this instance, Alan Cole in How Biden’s $400,000 tax pledge hobbles progressive ambitions.  

In his 2020 presidential campaign, Joe Biden pledged not to raise taxes on anyone earning less than $400,000 per year. Biden’s promise was more stringent than that of his old boss, Barack Obama, who promised not to tax anyone making less than $250,000.

In an email last week, reader Andrew K asked me to explore how much revenue can actually be generated by raising taxes on the rich. The result⁠ is bad news for those who want to dramatically expand the welfare state⁠—though not necessarily for more moderate liberals.

He then does a series of back-of-the-envelope estimations, all of which are actually pretty precise, to arrive at an answer.  If you want additional income tax, there is only about $1.3 trillion untaxed income above that $400,000 pledge.  Even if confiscate the entire amount, 100% taxable, you still won't cover the proposed spending. 

He concludes:

This is the kind of point that gets you pushback on social media, especially if you characterize a low-six-figure income as “rich” or worthy of more taxation. Richer Americans are more prevalent on Twitter than poorer Americans, and they⁠—like anyone else⁠—are wont to engage in motivated reasoning. They are likely to sincerely believe they aren’t really rich, and that even the more lofty Democratic ambitions can be funded without them.

From the perspective of an earnest, tax-math-oriented social Democrat, the answer is clear: tax six-figure incomes heavily. But from the perspective of a constituent-focused political-math-oriented member of Congress who picks up a wealthy suburban seat for Democrats in 2012 or 2018, you want to run in the opposite direction.

Arguably, the failure of the Build Back Better plan came from this conflict: Biden made a very Gottheimerist pledge, rhetorically cutting off most possible tax revenue, while simultaneously pushing a large spending package and promising to pay for the whole thing. Something had to give.

This was a non-starter from the beginning.  If we want European levels of state service, we have to bear European levels of taxation.  The top rates would have to reach deep into the middle and lower middle class tax payers.  Nobody in the US has confidence that the quality services would be delivered and no one, rich or poor, wants to bear that tax burden. 

Committing to not raise taxes on those making below $400,0000 was a bad decision.  Just like so many others made in the past year.  Back to the difference between governing (trade-off decisions, consent of the governed, etc.) and ruling (coercion and deception).  

For every political promise, what does the Fermi approximation tell you?