An economist's heart would be more gladdened at the sight of the federal and Alberta governments pulling out of the Lloydminster heavy-oil upgrader if those involved gave any sign of having learned from the experience. Unfortunately, in support of doing the right thing, all the wrong reasons have been cited. People who don't know why they're doing the right thing today are as likely to do the wrong thing tomorrow.

According to a federal official, the project was "losing money, and in the current fiscal environment we don't want to continue pouring good money after bad." Said another, "In this era of fiscal responsibility, we are not in the business of owning business." You can see a pattern developing. As yet another unnamed adviser put it, more grumpily, "the deficit has constrained us from doing anything innovative as a federal government."

If that is the case, then roll on deficits forever. As an investment in keeping governments out of mischief, it might well be worth the annual interest cost. But of course, the deficit is, or ought to be, beside the point. Ask yourself: Would this use of public funds make sense even if we didn't have a $700-billion national debt? Subsidizing heavy-oil upgraders is not wrong "in the current fiscal environment"; it's wrong at any time - not because we can't afford it, but because it's a hideous waste of money.

This has nothing to do with whether the upgrader is profitable, now or ever. As far as state funding is concerned, the economics of Lloydminster - what the price of heavy crude is, or how much it costs to process - are irrelevant. All you need to know is this: Either the project is economic, or it isn't. If it is economic, it doesn't need public money. If it isn't, it shouldn't get it.

This is what separates the economist from the politician, or, if you prefer, the cynic from the sentimentalist. Economists are relativists. They place no inherent value on anything. They believe just two things: (a) People should get what they want, so long as they understand that (b) more of one thing means less of another.

Subsidy's enduring appeal is essentially sentimental: We are invited to look only at the value of the thing subsidized, ignoring its price in terms of things unsubsidized. It is to replace what the people want, measured by the choices they make in the market, with what the politicians want.

If we assume that people are the best judges of their own welfare, and that prices are good signals of the relative costs of things, the case against subsidy becomes watertight. Given accurate information on the comparative scarcity, now and in future, of different goods and services, people exchange one for another, as consumers, workers and investors, in the way that makes them happiest. Any departure from that set of arrangements must inevitably make them less happy. In jargon, subsidy distorts the allocation of resources.

These assumptions don't always hold. But it is hard to see any reason why they should not in the market for heavy crude, in a way that would justify subsidy. If anything, there is an argument for taxing oil more heavily, to account for the costs its consumption might impose on the environment that are not fully captured in its price - the so-called carbon tax. But to be subsidizing its production, even as we are taxing its consumption, is simply bizarre. It is the work of neither cynics nor sentimentalists, but fools.

The most encouraging thing about the Lloydminster decision is the tantalizing suggestion that at least some politicians may have begun to understand the fallacy of sunk-cost thinking. In walking away from the upgrader, the two governments left behind nearly $1-billion of public funds spent on its construction. This is hard to do. It's only natural to say, "We're in so deep, we can't turn back now." But that money has been spent already, and nothing will get it back. It can have no bearing on whether to spend still more. Perhaps now someone will get the same idea about Hibernia.