A FIRM MAKES MONEY BY CREATING WEALTH. IF IT LOSES MONEY, IT IS DESTROYING WEALTH. IF A GOVERNMENT SUBSIDIZES THOSE LOSSES, IT IS DESTROYING WEALTH. SO LET’S NOT KEEP DOING THAT, WRITES JOHN ROBSON.

So it was money down the drain after all. GM closing North American plants including in Oshawa is hard on affected people and communities. But it also makes short work of the argument that markets are great in theory but massive government bailouts and subsidies work in practice.

Yes, I said so at the time. And for my pains the Harper Tories gave me patronizing lectures about principles being for chumps and ideologues. Did I say “gave?” Actually it’s “give” because Harper’s new book claims the essence of conservatism is to say one thing and do another.

Phooey. As John Stuart Mill explained, if something works in theory but not practice the theory is wrong so get a new one. But here it wasn’t.

On Nov. 21, 2008, I ridiculed a Globe and Mail editorial claim that “While there are ample reasons for Ottawa to tell carmakers they don’t deserve taxpayer bailouts, there are also compelling reasons to provide help for weathering the current storm. The trick will be to provide the right help to keep these critical companies afloat without getting stuck in a corporate welfare quagmire.”

Good luck with that, mate. Especially as, I said, “this argument drops any pretence that GM, Ford and Chrysler are winners but urges backing them anyway” without any explanation of how subsidizing failure suddenly went from “unaffordable folly” to “reputable wisdom.” It just somehow became a wise investment because everyone panicked. Do not try this approach with your own portfolio.

Weirdly, the pre-politics Harper knew exactly why bailouts are bad. He could have recited in his sleep, and possibly yours, too, the “Austrian school” explanation that the insurmountable problem with intervention is that it requires government to possess more knowledge than it conceivably can.

Markets are amazingly effective at reducing the dazzling, baffling complexity of economic interaction to a few simple numbers: sales revenue and input prices. As Leonard Read’s “I, Pencil” explained back when there were pencils (that didn’t break in the sharpener, I mean), even that humble object required the efficient co-operation of everyone from loggers who cut trees to those who make gasoline for their chainsaws and coffee for their cups (and the cups and chainsaws) to producers of drab yellow paint and the little “ferrule” holding the eraser.

If everyone involved had to know what everyone else was doing and why, you’d never see another pencil. But they don’t. They just need to know what things cost. The central planner, by contrast, must know everything about everything to order the right inputs at the right time not just for the pencil factory but for everyone in the whole complex interlocking supply chain in which miners also need pencils and coffee, accountants need desks and so on.

Now wait, you may cry. That theory explains why socialism fails in practice. But to bail out one car plant you only have to know how much cash they want and where the taxpayer keeps it. And it’s amazing how often people stop there, assuming the immediate effect of a policy is the end of the story.

We “lend” GM umptysquillion bucks, the factory stays open, people have jobs, sing Hallelujah. As for taxpayers, well, there’s more where that came from. But to know whether bailing out the plant beats letting it fail, you must know everything that will happen if it gets the money and, crucially, everything that therefore won’t happen because someone else doesn’t. And of course you can’t.

You can’t know everything taxpayers would have spent the money on instead, or who would have borrowed it and why, let alone what would have happened next, and next, and next. You aren’t God. But you can know one simple thing.

A firm makes money by creating wealth. The things it sells are worth more than the things (including labour) used up creating them. That’s what profit is. Literally. Hence, necessarily, a firm that loses money destroys wealth. It makes things worth less than the inputs consumed. So the more you subsidize it, the more wealth gets destroyed. Which is bad.

There’s the first bitter fruit of government intervention. The second is that disguising destruction as creation lets underlying problems get worse, like excessive labour costs, until the hole gets too deep to fill even with other people’s money. Predictably.

When I did predict it, I got patted on the head and insulted. Then our governments gave GM and Chrysler $13.7 billion, got back $10.2 billion (maybe, given characteristically opaque government bookkeeping), and the plant shut anyway. Practice, meet theory.

So can we have our $3.5 billion back? Or have they got some new hare-brained scheme to lose money on every sale but make it up on volume? Yup. Maple Leaf Foods this week.

I was afraid they might.

DISGUISING DESTRUCTION AS CREATION LETS UNDERLYING PROBLEMS GET WORSE.