Imagine for a moment you're talking to a financial adviser about refinancing your mortgage to get a lower interest rate and cut your total finance charges. "You shouldn't do it," he says. "Think of all those points you'll have to pay to get that mortgage. No, stick with what you have now." Would you take his advice?

Probably not. Like the millions of other Americans who have refinanced, you know that what you save over the long term with your new mortgage more than compensates for anything you pay upfront.

So why take the same flawed advice when it comes to Social Security reform?

Critics of personal accounts often point to "transition costs," arguing that there will be a huge taxpayer cost associated with "privatizing" the system. Put that way, it doesn't sound like a good deal. But this cost, like the points you pay to get your mortgage, is certainly worth paying in the long run.

To see why, think of Social Security's unfunded long-term obligations (the gap between promised benefits and projected payroll taxes) like a giant mortgage. It's not the kind of mortgage you or I would want because, under current law, it's never paid off. It's just passed on to our kids and grandkids, and the annual financing charges will soon start to rise rapidly as the baby boomers retire. The total charges over just the next 75 years are projected to be a staggering $27 trillion in inflation-adjusted dollars, or a $100,000 liability for every working household over that period.

Part of the idea of reform is to begin to dig out from this hole, just as you would want to do if you had a mortgage like that.

Now turn to Social Security reform. Advocates of personal accounts often point to the better financial returns one can expect from such accounts. But the other advantage is that this kind of reform, like refinancing a mortgage, will reduce those huge unfunded liabilities that will burden our children. And just like refinancing, there are points involved.

Under most reform proposals, younger Americans would be able to take some of the taxes they now pay to Social Security and put them instead into a government-regulated personal retirement account. They'd be opting to get part of their retirement income from this account instead of relying entirely on traditional Social Security benefits. Over time, as these workers age, more and more people likely would have part of their retirement financed through these accounts.

This reform would have two implications for taxpayers now facing huge unfunded liabilities in Social Security.

First, as more workers opted to take part of their retirement as income from personal accounts rather than from Social Security, the future benefits Social Security have to pay would fall, and so the unfunded liabilities (i.e., the total finance charges) would decline quite sharply over time. In fact, if workers could put half their Social Security payroll taxes into a personal account in return for halving their traditional Social Security benefits, some analysts put the drop in those mortgage-style financing charges at a staggering $20 trillion over 75 years.

Second, there's a cost. When younger workers put some of their payroll taxes into personal accounts, those taxes don't go into Social Security, where they would be used to pay other people's benefits. So workers would have to pay to honor those benefits until the lower financing charges begin to kick in when today's younger workers retire. That short-term tab for reducing Social Security's liabilities is called the "transition cost," and it's just like the upfront points you pay to get your mortgage charges down.

Just as many Americans do, of course, we could pay that transition cost over time, using Treasury bonds that mature over the next 30 years or so.

Like any refinancing opportunity, you would compare the points with the reduction in financing charges to see how far ahead you'll end up. In the case of Social Security reform, the transition cost in the example we've used would be about $7 trillion -- to cut financing charges by $20 trillion. A pretty good deal.

The numbers in Social Security reform are far larger that anyone could imagine in a mortgage refinance, but the principle is exactly the same, and the savings would be huge.

So the next time someone tells you that we shouldn't "privatize" Social Security because there are transition costs, ask him if he's ever refinanced a home.