Reform Government Pay with Step Decreases
Wednesday, February 6, 2013
In the private sector, workers are often compensated on the basis of performance. In government, workers are compensated primarily on the basis of credentials (the higher the degree, the higher the pay grade) and longevity.
Longevity is rewarded through the mechanism known as a “step increase.” Within a given pay grade, known as a GS level, there are steps. A GS-9, for instance, starts out at step one, then after a year receives an increase to step two, then after another year transitions to step three, and so on.
I think that the step increase is the opposite of what should happen. That is because I think that longevity is the wrong characteristic to reward in the case of government workers.
The IETF approach has been very successful. Internet standards are robust. The cost of governing the Internet is low.
With government in Washington, agencies last forever, and the problems that they are created to address never get solved. With the Internet, the opposite is true.
Part of the problem in Washington is the incentive structure embedded in step increases. In a government agency, each worker is paid more the longer the problem persists. In contrast, no one gets paid to work on an IETF task force. Instead, businesses and universities lend experts to work with the IETF. This represents a cost to the employers and perhaps also to the employees, who are kept from research or other activities that could advance their careers. Everyone involved in the process has an incentive to get the problem solved as expeditiously as possible.
We ought to see what happens if we transfer some functions that currently are performed by a permanent, salaried workforce to temporary task forces of volunteers.
One can imagine Washington adopting the IETF model. That is, instead of creating, say, a permanent Department of Housing and Urban Development, Congress could commission a task force of unpaid volunteers to come up with ways to improve housing. The task force would work as quickly as possible to implement solutions, and then the members would return to their previous occupations.
One concern with the approach of using a volunteer task force is that the firms most willing to donate workers would be those with a special interest in the outcome of the task force. Thus, the task force could become a vehicle for rent-seeking — that is, for gaining private benefits rather than producing social benefits. This is a potential problem with the IETF model as well — in theory, Microsoft or Google could try to pack the IETF working group. However, in practice, the various competing interests appear to check one another.
Note also that having workers paid by agencies does not seem to have solved the problem of rent-seeking. If anything, lobbyists seem to have more influence in Washington than the biggest corporations have over the IETF.
Washington ought to experiment with using the IETF model. We ought to see what happens if we transfer some functions that currently are performed by a permanent, salaried workforce to temporary task forces of volunteers.
More generally, I would recommend replacing the system of automatic step increases with a system of automatic step decreases. That is, a government worker's salary should go down as the worker spends more years at an agency. I believe this would have a number of beneficial effects:
1. It would increase turnover at government agencies. It is unhealthy for lifetime service to be standard in government. It leads to a culture in which government workers are permanently detached from the private sector, and where workers in one agency lack familiarity with other agencies. That creates narrow thinking and a lack of empathy for people in business.
2. It would decrease the extent to which government workers are invested in the programs they work on and the problems they deal with. Instead, they might be more inclined to think in terms of solving a problem, perhaps as a way to get rewarded with a better job later in a different agency or in the private sector.
3. It would allow managers more latitude to compensate workers based on performance. If the baseline is for the typical worker's salary to decrease every year, then agencies will have more salary money available to use for promotions and bonuses.
4. It would help to create a culture in which working for the government is a public service, not an entitlement. It would attract fewer people looking for a government career and instead attract more people who are motivated by a desire to contribute to public services for a few years.
Overall, automatic step decreases for workers would help make government more effective.
Arnold Kling is a member of the Financial Markets Working Group at the Mercatus Center of George Mason University. He blogs here.
FURTHER READING: Kling also writes “16 Tons of Keynesian Economics,” “Skin in the Housing Game,” and “What Would Churchill Do?” Andrew G. Biggs and Jason Richwine contribute “A Primer on Government Pay” and “The Truth about Federal Salary Numbers.” Ivan Osorio says “More Than Good Enough for Government Work.”
Image by Dianna Ingram / Bergman Group