Are government workers overpaid or underpaid? That question is behind the fight now waging in state capitals as several governors push for cuts in public employees' retirement benefits. But new research suggests that the question is misframed: public workers are neither overpaid nor underpaid, but mispaid. If that's right, it offers a solution to states' retirement benefits problem.

My last column[1] explained that, though states face serious problems with their 2012 budgets, the real fiscal crisis lies in the future. For years, state and local governments have not been setting aside enough money to cover public workers' future pensions and other retirement benefits. What money they have set aside has not earned good investment returns. As a result, promised retirement benefits for government workers are woefully underfunded. When baby-boom-generation public workers begin retiring in the coming years, governments will face a difficult choice: either renege on those promised benefits or cut government spending in other areas and raise taxes to cover the benefits.

Some people justify the benefit cuts by arguing that government workers are paid more than private sector workers.[2] Public employees and their unions respond that they receive lower wages than private-sector workers with similar education and experience.[3] Cut supporters reply that government employees may receive lower wages, but they also have high job security and receive high levels of benefits like health insurance, pensions and retirement health coverage. When the total value of all compensation is taken into account, public workers earn more than similar private sector workers.[4]

Wading through all the analyses of the overpaid/underpaid question leaves readers feeling that they're caught in a never-ending game of "dueling studies." Besides, even if public workers do earn more, that doesn't justify breaking retirement promises to them. Like it or not, and fair or not, taxpayers are responsible for those benefits.

But what if governments could lower the cost of government employees, but do it in a way that pleases those employees? Remarkably, that's a very real possibility, according to research by Stanford University postdoctoral fellow Maria Fitzpatrick.[5]

In 1998, Illinois lawmakers voted to raise teachers' pension accrual rates for coming years, boosting their retirement benefits. Lawmakers also gave teachers the option to increase their accrual rates for prior years, in exchange for the teachers paying a relatively small one-time fee. For most teachers, the option was a tremendous bargain, made even more appealing by Illinois offering generous terms for the fee's payment and allowing teachers to exercise the option at any time. Yet despite strong, favorable publicity for the option, a large number of teachers chose not to take the bargain.

Fitzpatrick used the Illinois data to calculate that, on average, teachers valued the new pension benefit at a mere 17 cents for each dollar it costs Illinois to provide it. Teachers preferred to have a little more money today, rather than a lot more money in retirement.  Other studies of different public pensioners have found similar results.[6]

It's important to understand that Fitzpatrick calculated the pension value "on the margin"- that is, how much the teachers value having a little more (or less) of what they already have. This doesn't mean they would sell off all their retirement benefits at 17 cents on the dollar. Moreover, her analysis involves only one small group of government workers: teachers in Illinois. Still, it suggests that public workers would willingly give up some of their promised pension benefits in exchange for cash today, at a cost much less than what taxpayers would pay for the forgone benefits.

 Policymakers thus have a rare opportunity: a chance to benefit some people at no cost to others. They could create a program in which government workers could voluntarily sell back some of their pension benefits in exchange for cash. If engineered correctly, the overall savings to taxpayers from the buy-back would be large, while participating workers would receive cash that they value more than the benefits they give up-a win-win situation.

There would be a few losers from this deal. It would expose the misdealing of politicians and union chiefs who agreed to higher retirement benefits instead of the better wages that government workers prefer and that cost taxpayers less. And it would expose the falsehood of supposed labor activists' claims that workers who don't receive large retirement benefits are being cheated. Consider all this to be yet another win.

 [1] Thomas A. Firey, "Retirement Woes Fleece Taxpayers, Public Workers," Herald-Mail, March 31, 2011.

 [2] See, e.g., Chris Edwards, "Public Sector Unions and the Rising Costs of Employee Pensions," Cato Journal, Vol. 30, No. 1 (Winter 2010).

 [3] See, e.g., Gregory B. Lewis and Chester S. Galloway, "A National Analysis of Public/Private Wage Differentials at the State and Local Level by Race and Gender," Georgia State University Working Paper No. 11-10, February 2011.

 [4] See, e.g., Jason Richwine and Andrew Biggs, "Are California Public Employees Overpaid?" Heritage Foundation Center for Data Analysis Report No. 11-01, March 17, 2011.

 [5] Maria D. Fitzgerald, "How Much Do Public School Teachers Value Their Benefits?" Stanford Institute for Economic Policy Research, 2011.

 [6] See, e.g., John Warner and Saul Pleeter, "The Personal Discount Rate: Evidence from Military Downsizing Programs," American Economic Review, Vol. 19, No. 1; Jeffrey R. Brown et al., "Who Values the Social Security Annuity? New Evidence on the Annuity Puzzle," NBER Working Paper No. W13800, December 2007.