Careful. Don't hurt your knee! For some of you, what you are about to read may cause a knee-jerk reaction so sharp and immediate that it might be dangerous.

How about a new government program that pays benefits to laid-off manufacturing workers who accept new jobs at lower pay? We'd call it "wage insurance," and the payments would make up a portion--perhaps 50%--of displaced workers' wage loss. A guy making $60,000 a year at a steel mill who got laid off and took a new job earning $40,000 at the widget factory, for example, would get $10,000 from the Feds during his first year at the new place.

I know. I know. We don't want any more government entitlement programs. We're already paying out $20 billion a year in unemployment insurance, for heaven's sake, and the last thing we need is another big drain for our tax dollars.

But wait. Let's take a closer look.

The wage insurance idea has actually been around for a few decades. It's gaining new credence lately thanks to the work of a couple of Washington, D.C., think-tank economists, Lori Kletzer and Robert Litan.

Kletzer, a visiting fellow at the Institute for International Economics, and Litan, director of economic studies at the Brookings Institution, present the idea as a way to help get U.S. free-trade initiatives back on track.

In a policy brief last February, the pair cited a recent survey in which 78% of respondents said that "protecting the jobs of American workers" should be the top priority in shaping U.S. trade policy. They argue that despite the demonstrated economic benefits of free trade and globalization, no U.S. president can succeed in persuading Congress or the American public to accept further trade liberalization until steps are taken to ease the pain of trade-related worker dislocations.

That's where wage insurance would come in. And it turns out that it could cost less than you might think.

Under the Kletzer/Litan proposal, wage insurance would not be available to all unemployed, only those "displaced" by a plant closing, a plant relocation or job elimination. Only full-time workers on a job for at least two years would be eligible. While benefits could range from 30% to 70% of the total re-employment earnings loss, they could not exceed an annual cap; Kletzer and Litan suggest $10,000 per worker, per year. Further, a worker would not receive benefits until accepting a new job, and payments would end two years after loss of the first job--both incentives for laid-off workers to get back to work quickly, even if it means taking a pay cut.

Based on unemployment statistics, Kletzer and Litan calculate the cost of the program at less than $4 billion a year, assuming 5% unemployment and 50% replacement of displaced workers' earning loss. That's not bad, given the current $20 billion annual tab for unemployment insurance. And it's a tiny fraction of the estimated $500 billion in benefits that the United States would get from unrestricted global free trade.

The wage insurance concept has drawn support from various influential individuals, including Carla Hills, U.S. trade representative for the former Bush administration. It's something to think about.