A recent Freakonomics podcast featured “ideas that must die” — situations where we’d be better off if we stopped clinging to a bad idea, inspired by this book.

Here’s my candidate bad idea: a mistaken argument I heard repeated from several Canadian fiscal policy pundits this week.

I’ll start with an analogy before I get where I’m going:

Say you want to lose weight. To help you achieve this goal you impose a strict dieting rule: no snacking after 10pm.

You stick to this rule for a while. And it’s working. You are losing weight. Hooray!

But then you go on vacation. You’ve spent lots of money, you’re in a new city, you’re out at night and you want to enjoy yourself. It no longer makes sense to stick rigidly to your no-late-night-snacking rule.

So you throw caution to the wind, indulge in some late-night snacking and enjoy yourself. It’s a tough habit to break, so your weight creeps up a bit while you’re away.

Before you know it your vacation is over, and you’re back on the no-snacking regime. On net, even with your indulgence, you’ve still managed to lose weight. Would anyone argue that your dieting rule doesn’t work?

Now let “snacking” = “undisciplined fiscal policy choices”; and “vacation” = “recession”. Analogy over.

My complaint is this: why do discussions about balanced budget laws lead many people — even very smart people — to rely on very bad arguments (and to make exaggerated claims about how clear the evidence is on this or that)?

Why is it acceptable to argue that balanced budget rules are useless just because people mistakenly think that at the first sign of trouble all governments abandon them?

First, there’s some truth in the notion that fiscal rules are not always kept indefinitely. In my study, which covered a relatively benign economic environment of the “Great Moderation” before the “Great Recession”, over 40 percent of the fiscal rules used by Canadian provinces were abandoned or allowed to lapse. I suspect this number would be higher when economic situations were tougher.

But here’s the thing: what does that tell you about whether the rules worked, when they were being used? Not much.

If I’m right about this, and I hope I am, then please do your part for better fiscal policy discussions: don’t use this argument and let this bad idea die.