In public sphere, setting wages is a tricky business

Few administrative tasks are more difficult than setting efficient and equitable compensation levels for public employees. The Minnesota public employees strike bears this out.

Minnesota’s two largest state employees’ unions went on strike largely because offered increases in pay for a new contract period were mostly offset by increases in health care costs to be borne by employees. The level of dissatisfaction with the state’s offers has varied widely between geographic locations and skill levels.

In an economist’s ideal world, wages and salaries in private business would be driven by market rates for workers with specific sets of skills. If an employer does not pay close to what good workers can get elsewhere, she will not be able to put together a work force capable of doing the job.

If she pays her employees substantially more, relative to productivity, than other firms, her costs will be higher and she will have a hard time selling her products in a competitive environment. The owner faces a difficult budget constraint. If workers don’t produce enough to pay their cost, they get laid off or the business goes bust.

In such an environment, workers with better or more needed skills or higher productivity would earn more. Less skilled, less needed or less productive workers would earn less. These pay differences would provide motivation for people to get training in areas where they would be most needed.

That ideal does not always play out in the real world but is true to a workable degree for many private businesses, especially smaller ones. It is harder to achieve in the public sector.

This is not to say that an economy does not need a public sector to thrive economically and socially. Economists may quibble among themselves about how big a government is necessary, but there is a theoretical consensus that some amount of government plays a vital role.

But it is hard to measure the value of public services. They are not priced, and thus there is no simple test of whether they are “selling” or not, the way one can with bagels or automobiles. It is not easy to tell if a particular set of workers is “worth” what it is paid or not. That is a judgement that has to be made by the legislature.

Furthermore, political culture imposes constraints on personnel management procedures that private sector firms need not bother with. Across history, government officials have been tempted to use their public position for private benefit by giving employment or business to family or friends. This practice was the rule in the U.S. federal government for more than 100 years and still erupts periodically at the state and local level.

Patronage and nepotism can be avoided with a comprehensive civil service system aimed at hiring and promotion based on merit. This can also reduce discrimination against minority groups.

But such systems inevitably seem to bring substantial levels of bureaucracy in hiring, compensation and promotion. This leads to greater job security in the public sector than in the private sector. And it impedes tying public employees’ compensation levels to those of comparable jobs in the private sector.

Public employees’ earnings generally come from some table based on skills or duties and experience. It is hard to include differences in opportunity costs across different locations. In plain English, what that means is that government employees tend to be highly paid compared to local averages in low-wage, low-living cost rural areas, but less well paid compared to private sector employees in higher-cost urban areas.

The fact that health costs, rather than basic pay, have been the point of contention drives home another lesson. Regardless of who writes the check, workers in a market economy inevitably bear most of the cost of noncash compensation. Public employment or union bargaining may delay the process, but ultimately both the supply of and demand for labor are based on total compensation rather than nominal pay.

© 2001 Edward Lotterman
Chanarambie Consulting, Inc.