Leaders across the country are proposing restrictions on public employees' pay and benefits in order to put their budgets on a more sustainable path. The political left's counterattack is that government workers aren't overpaid compared to those in the private economy. Who's right?

Consider a study released last October by the Center on Wage and Employment Dynamics at the University of California, Berkeley, which concluded that Golden State public employees "are neither overpaid nor overcompensated." The Economic Policy Institute has generated reports arguing that government workers are underpaid.

These studies are misleading. Public-private pay comparisons vary from state to state, but a full accounting shows clearly that large, union-dominated states tend to overpay their workers. California is a good example.

The Berkeley study begins by studying salaries, where its methods are solid. Using individual-level data from the Census Bureau's Current Population Survey, it compares public and private wages while controlling for differences in age, education and other earnings-related characteristics. Using essentially the same methods, we found that California state and local government employees receive wages about 4% lower than those received by similarly skilled workers in large private firms, which offer the most generous pay and benefits. But if we compare public employees to all private workers, the 4% penalty becomes statistically zero.

Public employees really pull ahead in non-wage benefits. The Berkeley study concludes that counting benefits means that public workers' total hourly compensation is about 2% higher than that of private workers. But our research shows that the study underestimates what public workers receive from pensions and retiree health programs. It also doesn't account for the value of job security in government employment. Once these are noted, the balance tilts clearly in favor of public workers.

The first error in the Berkeley study concerns defined-benefit pension plans. The study erroneously conflated what governments pay into defined-benefit plans with what workers will eventually receive in retirement. So if governments contribute 10% of employee pay to defined-benefit pensions while private employers contribute 10% to 401(k)-type pensions, these studies conclude that pension compensation is equal.

But here's the problem: State and local pensions effectively guarantee employees an 8% return on both their contributions and those made by their employer. By contrast, a private-sector employee with a 401(k) can achieve a guaranteed return of only around 4% by investing in U.S. Treasury securities. Most economists believe governments are foolish to base their funding decisions on the assumption of high investment returns, but the benefits for public employees are guaranteed in any case.

Over a career, the difference between a 4% and 8% return is significant. Using data from California's major pension funds, we calculate that the higher implicit return on public pensions increases the compensation of California's government workers by around 4%.

The Berkeley study's second error is the omission of retiree health benefits. Private workers retire later and relatively few receive retiree health coverage. For those who do, eligibility has been tightened and premiums increased. But almost 90% of state and local governments offer retiree health benefits to employees. They generally retire in their 50s, at which point the government often pays most of their costs, including Medicare premiums and deductibles.

State actuarial reports show the annual cost of California retiree health benefits could top 8% of total compensation. Thus an accurate accounting of pension and retiree health benefits shows that public employees in California are paid about 15% more than individuals working for large private firms (accounting for age, education, etc.).

Another major benefit of public employment is job security. The Bureau of Labor Statistics reports that, on average, a private worker has about a 20% chance of being fired or laid off in a given year. In state and local government, the discharge rate is only about 6%—and several studies have found that public employees are more risk-averse than other workers, meaning they place particular value on job security. We estimate that government job security is equivalent to about a 15% increase in compensation.

Overall, our research suggests that government workers in California are compensated up to 30% more generously than are similar employees in large private firms. And the California experience is similar to that of other large states with powerful public unions. Elected officials are right to reassess public worker compensation as they try to close their budget deficits.

Mr. Biggs is a resident scholar at the American Enterprise Institute. Mr. Richwine is a senior policy analyst at the Heritage Foundation.