Similarly we reach a point in coding where you don't really need to know how every API or language you use operates beneath the hood, you just need to be able to see where its broken.

What it shows is that 10% means nothing to them. It's just another number that a calculator will spit out. It's not "one out of ten" or "shift everything to the right." They have no ability to evaluate what the number means on the fly. They just plug it into the calculator and check if they have that much. They can't have a reasonable discussion about the number without stopping and calculating every interest rate they can think of, writing them down on a piece of people, and writing the budget next to them.

A calculator here, of course, leads to more work, not less. I bet the vast majority of the time that somebody too lazy or anxiety-ridden to learn how to calculate 20% in their head is also going to be too lazy and bothered to take out a calculator every time they need to know, rather than just trying to bluff their way through conversations until they can get to a calculator.

Calculators are of no benefit to students who have no clear concept of the calculation. Just learn your times tables, they're one of the first things we teach kids.

The pragmatic pro-calculator "nobody needs to calculate those numbers in real life" school I think gets most of its support because constructivists often fail entirely to teach things that have been until now best learned by rote, hoping that an instinct for multiplication and division will naturally and precisely arise from children's mathematical souls. The fact that it doesn't is proof that doing arithmetic is bad and a waste of time actually. They love abstractions, but only vague ones that can't be reliably tested.

> Either way, they probably would not likely have learned the mental math required to do this regardless of if it were being taught at school or not

That's the law of averages. It is not a law.