Can economic analysis reliably predict anything though?
In a science if what is observed does not correlate to theory then the theory is ripped down and started again, instead of getting the basics right economists build their 'complex' models on top of foundations that are meaningless ideological constructs. A market comprised of millions of 'rational' people who consider every single opportunity cost based on a hypothetical measurement of 'utility' when making every decision, sounds about right.. Every speculative bubble in history is a testament to the failure of the 'market', too much government intervention in the 'market? The GFC, they say not enough government intervention in the 'market'!
Not to mention the idea of equilibrium is built on a world operating under fixed currency systems based on taxation and distribution, where in reality sovereign currency governments have no constraint on expenditure.. Socially optimal? More like ideologically socially optimal based on theoretical economies that existed 40 years ago.
You've clearly never studied economics.
There are essentially two branches of economics: mathematical and political. The only difference between the two is where they start. Mathematical starts with assumptions about the world, builds a model and extrapolates upon this with tweaks to the the underlying assumptions, creating an ever more complex model that better represents reality. Political economics, meanwhile, starts with an observation of the world and comes up with an theory to explain it. More observations are added and the model refined until some basic maths can be wrung from it or the model can help predict future behaviour. Both are entirely valid to the world around you, if for no other reason than everyone else is listening to the results and acting upon them, so you'll fuck yourself over by not listening as well. The fundemental problem is that you can't create a model that could perfectly map and predict behaviour by humans, because a model is supposed to be a simplification of the real world. Humans are so complex that any model that did them justice would be more complex than the world it is supposed to be modelling, and thus no longer a model.
Also, biology and medicine are commonly labelled as branches of science, yet they can not predict the future path of evolution nor disease, respectively, any better than economics can predict the behaviour of humans and the economy.