What a difference a year can make. If the President’s Pay Agent had released a report responding to the Federal Salary Council’s recommendations last year, it would have reflected the Obama administration’s point of view. The delay of a year meant the newly constituted Pay Agent under President Trump voiced a very different opinion: “We believe in the need for fundamental reforms of the white-collar federal pay system.”
The Pay Agent, a body composed of the directors of the Office of Management and Budget, the Office of Personnel Management and the Labor Secretary, went on to state:
“We believe it is imperative to develop performance-sensitive compensation systems that will contribute to a government that is more citizen-centered, results-oriented, and market-based. We need to empower federal agencies to better manage, develop, and reward employees in order to better serve the American people.”
This is not a new idea, of course. There have been reports arguing that the General Schedule needs to be replaced as far back as a 1991, beginning with a report published by the National Academy of Public Administration. Presidents Clinton, Bush and Obama each released similar statements early in their administrations.
This should not be surprising. Employees working in government at all levels know switching to pay for performance is a slowly moving tidal wave. Federal employees may be reluctant to admit it but the program outlined by the Pay Agent would create a more satisfying work environment than many experience now. The GS system is badly broken and is an impediment to talent management and improved performance.
The problem is the failure of the National Security Personnel System and the headlines it engendered. The unions that had been ignored by the Bush administration were very vocal in their opposition. Several reports confirmed the dissatisfaction. Problems also undermined the pay for performance systems in the Government Accountability Office and the Securities and Exchange Commission. All of that and more contributes to the notion that employees have a lot to lose
But we know how to minimize the transition problems. There are success stories. That includes the demos starting with China Lake almost 40 years ago. It also includes a growing number of states. A couple of which go back more than two decades. Unfortunately, there is no registry of pay for performance programs in government. It would be valuable to revisit the failures to understand the problems.
The Pay Agent report highlights two key points:
- The balance of supply and demand for specific skills differs from area to area. That is true within the current locality areas. New York is an obvious example. Even within Manhattan, midtown employers pay somewhat differently than Wall Street employers. An early step in the planning should focus on identifying the job families and areas where current salaries impede recruiting.
- The current gap analysis ignores the value of benefits. That, however, is not a simple problem. A so-called market analysis starts by identifying other employers competing for talent. For federal agencies, the employers differ in each area. In the Washington DC area, federal contractors dominate the talent market and they maintain the full complement of benefits. Government benefits need to be competitive.
The report sets the stage for true transformative change.
As one writer recently argued, this will require “robust public policy debate.” That is what Connie Newman, then the Director of the Office of Personnel Management, organized in 1990 to discuss the locality pay idea. She led a series of meetings with at times 20 individuals representing the interests of stakeholders. Newman also arranged for me to meet individually with representatives of key employee groups (e.g., the American Nurses Association) plus selected congressional staff.
Newman understood that changes affecting employee careers cannot be dictated. To quote again from a recent McKinsey article, “change is not a top-down exercise.” That is a way of stating the “strategy- execution gap” discussed in a recent column. That, no doubt, contributed to the NSPS failure.
This will be a much more complex and sensitive project than anything government has undertaken in decades. The key is not the pay policy; it’s the effectiveness of managers in managing performance. For success, managers and employees need to understand the intent is to create a performance culture.
This is not an HR responsibility, although chief human resource officers will need to provide leadership and add staff to provide support. It’s also not a problem where unions should be involved in either the planning or the ongoing management. The suggestion is that they play a quality control role to assure fairness.
It’s widely understood that performance ratings are inflated. Linking pay increases to current ratings would guarantee a new program’s failure. Credible performance systems will first need to be developed and tested.
“Results-oriented” has to refer to goal-based management. That’s effectively the universal practice for all executives, managers and most professionals in other sectors. Goals are best defined with objective or verifiable performance measures. But there are jobs where individual goals make no sense (e.g., nurses, scientists); but for others, the practice is an accepted basis for rewards.
The best practice for evaluating performance is based on goal achievement and job-specific competencies. High performers can identify and define key competencies for a job family in two or three meetings. Defining competencies will facilitate coaching conversations that help employees improve—informal, frequent coaching supports continued development. It’s the same as coaching in sports.
This will change the role of managers and their relationships with staff. Performance management skills become much more important. Training to master the new skills will be essential. Equally important is rewarding those managers and supervisors who are most effective. Those who are ineffective should be moved to non-supervisory roles.
The planning for a new salary system is far more straightforward. Multiple salary surveys are available for virtually every job family and metro area. The model for a new system has been discussed in multiple reports and proven in several demonstration projects. Pay for performance policies are simple—increases are defined at each rating level.
New performance systems should be tested for at least a year before salary increases are linked to ratings. To state what may be obvious, a new pay program will need to be fine-tuned. The National Geospatial-Intelligence Agency has the best answer: At year end, employees are asked where improvement is needed; employee teams then address problems. If pay for performance is successful, government will be a better place to work.