That's the Keynesian view where you can move a lever (cause) and measure the changes (effect).

Unfortunately, there are way too many variables, interactions, etc, etc to have any real confidence in those measures. It's further complicated by the idea that the normal tools (primarily interest rate & money supply delivered via subsidies, etc) are beyond their "normal" operating ranges.

When the Fed rate was 5% and loans were 8%, lowering rates encouraged borrowing. When the Fed rate is 0.25% and loans are 3-5%, qualified people can get all the money they want.. now what? Do they give money to people who can't pay it back (mortgage crash) or spend it on "shovel ready projects" which take 12-18 months to get started?

Alternatively, when consumer spending makes up 70%+ of the economy, consumer confidence is probably the single most important metric.

This is less engineering and more psychology at scale.

That’s why I say it’s mostly accounting—you have to convert reality into quantities before you can engineer it, and accounting is the set of rules for measuring and manipulating those quantities. The difficulty lies in the fact that unlike physics, where units are essentially hard-coded into the fabric of reality, the unit in accounting is money which is a social construct.