One of the complaints made against economists is that they are no good at predicting the future. This is, of course, true, but it is not the biggest problem. A far bigger problem is that economists make policy recommendations based on arguments that contain a hidden premise: economists can predict the future with certainty!
Let's go back to the example of whether or not I trade in my car for a new one, and how this choice affects American GDP.
Situation A:
Imagine I own an old SUV that gets terrible gas mileage. The nice thing is that it is paid off and the high gas expenses don't bother me because, really, it's for trips to Home Depot where I buy big bulky things that require an SUV. Although... I do drive it for other reasons, too. This goes on for a number of years, all the while there exists Zipcar, a company that will rent me a pickup truck whenever I need it. It turns out that the rental fee for using the Zipcar pickup to go to Home Depot is actually less than I spend on gas and insurance for the same trip with my old SUV. After three years of this situation, the joy of ownership just doesn't seem worth it, so I turn my SUV into the dealer, take out a loan for a brand new tiny hybrid and use the Zipcar pickup when I go to Home Depot.
Thus, I participate in the glorious free market, maximizing my happiness and growing our common GDP.
Situation B:
Instead of me waiting 3 years, in this situation the big bad government forces me to trade in my gas guzzler for a hybrid before I want to. Grumpy, I cast about for a solution after a trip to Home Depot in the little hybrid goes horribly arry. Oh well, I guess I'll just use a Zipcar Pickup when I need one.
Thus, government regulation makes me grumpy while it imposes costs and inefficiencies.
But wait: in both situations I trade in my old SUV for the latest hybrid, taking out a loan to do so, but this happens three years earlier in situation B. Given the time value of money, that means that Situation B actually causes the bigger jump in GDP. There's no "cost" to the regulation! It's all benefit.
The point is this: when economists talk about the efficiency of the market versus the inefficiency of regulation, they are claiming that they know how you will spend your money: they know that you will face a range of choices where choice A will grow the economy a little and choice B will grow the economy a lot and they know which you will choose, already (like God knows you, even when you are but a twinkle in your mother's eye). You will choose B! They know it.
But as I hope my example shows, us stupid humans routinely pick choice A!
As at least one economist has told me, economists are smart people. How do they not know that they are making the ridiculous claim that they can predict the future with certainty? Because all that work is done by three words: "the free market." When economists say that a transaction happens in "the free market" they have defined that to mean, out of the infinite number of possible transactions, the one transaction that maximizes value. It's wrapped up in the first week of Econ 101 and they never question it after that.
Having trouble believing that a market transaction might not be more efficient than a government command? That's because the definition is deeply ingrained in every American brain. In America, we teach our children that the free market is always more efficient than government. And the result is a disaster.