Governments are slowly coming to grips with the problems created by running their pension systems with pie-in-the-sky investment returns. Unfortunately, most still haven’t.

Today’s Wall Street Journal gives a quick look at this problem. The good news: At least 19 state or local plans have scaled back unsustainable targets. The bad news? More than 100 (and probably many more than that) have not.

When pension plans make their forecasts, one thing they do is estimate how well their investments will do. Why does that matter? “The assumed rate of return is critical,” says the Journal, “because it determines how much a city or state and its workers must contribute to a pension system.”

Lowballing the amount of contributions sounds like a good idea. But when it’s paired with legally binding pension plans, it’s a disaster. It’s like taking one of those mortgages (the kind that threw the economy into a tailspin) based on the assumption that ¬†your income will, three years from now, be double what it is today. It might pay off. But if it doesn’t?

Overly optimistic assumptions about investment returns encourage fund managers to pursue investments that are too risky.

As for Minnesota, the Journal quotes¬†Laurie Hacking, executive director of the Teachers Retirement Association of Minnesota. She still expects a return of 8.5 percent–at a time when ten-year treasury bills are yielding 2 percent.

It is true, as the Journal says, that “the lower the [projected] rate [of return], the greater the obligations appear.” But the key word is appear. The actual obligation–based on the formulas of a defined benefit plan and the demographics of the government workforce–is not changed.

Making the expected rate of return more realistic may create a temporary panic (“The bill is HOW MUCH?”). But as they say in 12-step programs, you’ve got to realize you’ve got a problem before you can fix it.