Why practically all macroeconomics is wrong
Because the discipline's two main statistical models are total crap.
GDP and CPI: Broken beyond repair.
All models are wrong, some models are useful. Two highly cited statistical models in economics – Real Gross Domestic Product and the Consumer Price Index – are so broken so as to not be useful. The models are a hodgepodge of dubious assumptions and subjective judgements that have been munged together and massaged until the results simply mirror the intuition of those constructing the model. By trying to be all things, the numbers end up meaning nothing. The GDP statistic tells us neither about well-being, nor about actual productive output of raw goods. For every purpose that GDP may be used, a better measure exists. I’ll start this essay by deconstructing GDP and the CPI, and then I’ll present the alternatives...
Continued at the link. For the most part, the purpose of macroeconomics is to justify the notion that governments can do what individuals, households and businesses cannot. Actually, governments can do two things that everybody else cannot: kill people with impunity and print money without being prosecuted. But no government can do either forever.