Any one who is genuinely interested in learning about how and why complex systems change catastrophically should read "Limits to Growth" - the classic by the MIT team headed by Donella Meadows.
Yes, let's all jump on the 1970s Club of Rome fad and do exactly what they suggested---controlling the reproductive and development rights of third world countries while amassing all of their resources in the first world. Then we will all bask in the wonders of the United States empire while masturbating to Kissinger porn while the neutered africans can stick to their caves where they belong. All for the sake of Malthusian paradise, of course.
As far as your economics analogy, it is pretty poor. No one agrees that economic law is a natural science or that it's entirely accurate or even measurable. Do you not know follow the ten trillion debates that follow every data collection technique and every data series, nevermind the models upon which the data is based? Do you not wonder why companies spend hundreds of billions a year publishing and researching economics studies? You might have caught a bit of it if you, uh, ever read the newspaper in the past 100 years.
But to suggest that it is possible that the data collection techniques and models of such an exacting "science" such as climate change might be off? Why that's a Limbaugh conspiracy!
You people are funny---for some that always harp about complexity and exponential feed back loops you sure have a hard time thinking that it's remotely possible you've missed any exponential feedback loops in the opposite direction and the resulting massive change that would have on your predictions.
But what am I saying---I'm just a mathematician.