Breaking windows will stimulate the economy, according to a leading public pension advocacy group. Skeptical? The National Institute on Retirement Security (NIRS) has not literally endorsed breaking windows, but a report recently published by the organization relies on the same economic fallacy.

According to NIRS-whose membership consists principally of public employee unions, the pension plans in which they participate, and the actuarial and investment firms that serve them-the best economic stimulus is not tax cuts or unemployment checks, but increased pension benefits to public-sector retirees. Each dollar of pension benefits produces $2.37 in economic output, NIRS says, creating millions of new jobs and billions in additional labor income.

Put simply, this is nonsense.

Many people have heard of the "broken window" fallacy in economics. The argument is that breaking a window is actually good for the economy, since the window owner has to pay a glazier, who uses the money to pay a butcher, who uses the money to pay a cobbler, and so on. (A little bit like the old TV commercials for Faberge Organics shampoo: "I told two friends, and they told two friends, and so on....") One doesn't need to be an economist to see the problem: Had the window not been broken, the owner could have spent or invested the money elsewhere.

NIRS gives its own version of the broken windows story: "A retired firefighter uses his pension money to buy a new lawnmower. As a result that purchase, the owner of the hardware store, a lawnmower salesman, and each of the companies involved in the production of the lawnmower all see an increase in income, and spend that additional income. These companies hire additional employees as a result of this increased business, and those new employees spend their paychecks in the local economy."

According to NIRS, each dollar of pension payments results in an additional $1.37 of income flowing to non-retirees. By this logic, states could spend themselves to prosperity by borrowing money to increase public sector pension benefits. And, some states appear to be giving this approach a try.

"According to NIRS, each dollar of pension payments results in an additional $1.37 of income flowing to non-retirees. By this logic, states could spend themselves to prosperity by borrowing money to increase public sector pension benefits." -- Andrew Biggs Of course, this reasoning is wrong because it ignores the cost to the economy of providing public pension benefits. Each taxpayer dollar flowing into public pensions is a dollar that cannot be spent or saved on something else. To the degree it is not spent, today's economy is smaller as a result. To the degree it is not saved, tomorrow's economy is smaller, since saving helps boost productivity. If the stimulus provided by pension benefits has a multiplier effect, the cost of funding pensions presumably has a divisor effect - and this effect is entirely unaccounted for by the NIRS study.

Indeed, NIRS acknowledges that its study measures only the "gross economic impact" of pension payments, not the net effect once taxpayer and employee contributions are accounted for. But leaving out half of the equation makes the entire exercise meaningless. Unless we assume that public pensions have a magic ability to create money, the net economic impact of pension benefits is roughly zero.

It's even possible that defined-benefit pensions reduce economic output. After all, it is well-known that employees of DB pension leave the workforce earlier than workers with defined contribution 401(k) plans. Researchers at the Center for Retirement Research at Boston College found that workers "covered by a defined benefit plan will retire about one year earlier than those covered by a defined contribution plan." If we assume that the average person works 35 years over their career, DB pensions would lower labor supply and economic output by roughly 3 percent.

Public-sector pensions are not economic stimulus. They are simply a transfer of money from taxpayers to public employees. To the degree that pensions are fair compensation for public service, these transfers are perfectly appropriate. But as my work with Jason Richwine of the Heritage Foundation has shown, public-sector pensions are often several times more generous than the 401(k) plans enjoyed by private sector workers.
In fact, research on "fiscal consolidations" - attempts to balance government budgets and reduce public debt - shows that the most successful efforts in terms of cutting deficits and boosting growth include reductions in public sector compensation. Of which retirement benefits are by far the most generous component.

The faulty NIRS study illustrates a larger problem-namely, the willingness of those who run public pensions to jump into the political fray. Across the country, pension administrators are publishing their own localized versions of the NIRS fallacy, touting the supposed economic benefits of generous pension payments. Administrators also regularly exaggerate the transition costs of moving to a 401(k)-style system, and they insist to critics that even the most generous benefits are really quite modest.

Pension administrators should be apolitical public servants. Instead, too many fight tooth and nail to preserve the existing pension systems, advancing dubious arguments along the way. Since the managers of public retirement plans justify generous benefits with economic claims easily dismissed-not just by trained economists, but by the average person using his common sense-the only reasonable conclusion is that administrators have taken sides in the pension debate. And it's not the taxpayers' side.

Andrew G. Biggs is a resident scholar at the American Enterprise Institute. He previously served as principal deputy commissioner of the Social Security Administration.