Geoff Colvin’s latest Fortune column warns that the federal government can address Medicare’s long-term unsustainability in one of two ways:

One way to fix it is the Brute Force approach. That’s the concept on which Medicare was built. The federal government dictates which services are covered and how much will be paid to doctors, hospitals, and others for everything they do. To keep costs under control, Washington restricts what it covers or dials down what it pays.

How well has the Brute Force approach worked? “It never works,” says Mark McClellan, former head of the Centers for Medicare and Medicaid Services. The House Ways and Means Committee predicted in 1967 that the new Medicare program would cost $12 billion in 1990. Actual 1990 cost: $110 billion. (2010 cost: $523 billion.) The problem is that eternal irritant to grand Washington plans, the market. Turns out that if you unilaterally cut prices, some providers will quit providing services and some patients won’t get care, so you can’t cut too much. And if you pay providers barely profitable rates when they perform a given service, they will overperform those services, grossly inflating the government’s costs. That’s what has happened.

The other way to fix Medicare is the People Aren’t Dummies approach. It’s the concept on which most markets operate. Let people spend their own money — even if it’s given to them — and let providers compete for it. Providers aren’t dummies, so they’ll innovate in ways that bureaucrats would never think of. Consumers aren’t dummies, so they’ll choose what works for them. Quality rises, and costs stay reasonable.