GovernmentExecutive.com

In a lab on the coast of Australia last winter, academic researchers conducted a real-life social experiment on the usefulness of pay-for-performance systems.

Not all the findings came as a surprise, but the results were certainly informative for federal employees poised to leave the General Schedule for a workplace where performance ratings dictate pay raises.

Subjects of the study were asked to complete anagram word-creation games and choose to be paid either per word or at a set salary. The results showed that pay-for-performance had two advantages. First, people who were paid for their performance were motivated to do a better job.

The second benefit was self-selection. Subjects who chose to be paid per word were better at completing the anagrams, even when pay was removed as a motivator. So, employees who choose to work for companies, or agencies, with pay-for-performance systems self-selected as higher performers to start.

Those benefits are often touted in government personnel reform circles. To attract the best and the brightest to fill the jobs of retiring baby boomers, and to keep them motivated, government needs pay-for-performance, supporters of such systems argue.

But C. Bram Cadsby and Francis Tapon, economists from the University of Guelph in Guelph, Ontario, and Fei Song, a business professor from Ryerson University in Toronto, whose work will be published in December's issue of the Academy of Management Journal, also came upon some unexpected results.

They measured participants' preference for risk and found that subjects who were risk-averse high-performers chose a fixed-pay system over pay for performance. They opted for a set salary even though, in all likelihood, they would have made more money with pay for performance.

So under the pay-for-performance approach, agencies could be recruiting a new breed of government employees. The so-called golden handcuffs of solid federal pensions and the 40-hour workweek that attract one type of employee could clash with this new pay approach.

And in a balkanized federal system where some agencies have pay-for-performance and others don't, the culture gap could widen.

"It's an important caveat," Song said. "They're not only attracting highly productive people, [but] risk-loving people. May be a blessing, may not be."

Another finding to emerge from the data was that stress can hamper productivity. A quarter of the subjects actually performed worse under pay for performance compared with the fixed salary.

"The message ... is companies do not need to give a contract that's one size fits all," Song said. "It's actually better to customize the compensation contracts according to the employees they deal with."

Unfortunately, 1.8 million customized contracts don't seem to be in the cards for government employees.