Ok, so this multiplier effect that is defined as 1/(1-c) for government spending and X c for tax cuts... Of course like all keynesian economics it's really abstract and the C can not be known. Anyway I just don't get it how it would work even if you knew c (and mind you I am talking about increasing aggregate demand, not if it's good or not). This is because sure you can calculate the multiplier, but the multiplier effect should be calculated against what would have other wise happened. The model considers just crowding out effect... So for example if you increase government spending, you can't just calculate the multiplier, you also would have to take the multiplier that otherwise would have happened had you not taxed the money away, same goes for decreasing taxes. I just don't get it, how does it make any sense to say moving money from one pocket to other, and then calculating only the multiplier of the 2nd pocket and not calculating the multiplier that would have happened in the first pocket, make sense? So shouldn't the multiplier be calculated as: A (the demand after tax breaks or government spending) - B (the demand before it). Any explanation for this? I do understand that if you print money then you can get a real multiplier... But it makes sense because printing money means more money and higher inflation expectations, which means more demand for stuff and less demand for moneys.