With Pay for Performance, Money Is Not the Problem

Pay for performance is not a compensation problem; it’s a performance management problem. The focus in government is on the money but the acceptance of the policy depends on ratings and metrics that are perceived as fair. Anecdotal evidence suggests strongly that performance ratings across government have little credibility. When salary increases or cash awards are seen as unfair, the practice can trigger dissension and costly disengagement.

To illustrate the magnitude of the problem: In 2013, the most recent year summary data have been reported, 74 percent of employees were rated as “outstanding” or “exceeds” fully successful. The senior executive service is worse. It’s analogous to telling children they did great even when their performance is disappointing.

Performance management practices in business are undergoing a sea change to address essentially the same problems. The biggest change is a significant redefinition of a manager’s role. Simply described, managers are becoming coaches and teachers. The revolution is in its early stages but companies are beginning to emphasize ongoing feedback throughout the year.

A key difference in the business environment is that performance issues are discussed regularly at all levels. Progress in achieving goals is reviewed routinely. Everyone understands they can expect to benefit when the company is successful. High performance is celebrated. It contributes to a strong performance culture.  

Another important difference is that ratings and financial awards are usually confidential. That masks problems.

The problems in government are several but they can be minimized. Strengthening the way performance is managed is the single best way—possibly the only way—to improve performance. Managers will need training to augment essential skills, along with new tools to support their decision making. Many offices will need two or more years before they are ready to support pay for performance.

The Manager’s Role

A starting point is to recognize that managers have three core responsibilities. First, at every level they need to work with their people to develop individual performance plans. Second, all employers need to identify the star performers and those whose performance is unacceptable. Special policies apply to each group. Third, everyone needs constructive feedback to improve performance. That captures a large part of a manager’s responsibility. Their effectiveness is the key to raising performance levels.

Historically, performance ratings served to identify the few poor performers. That continues to pervade the thinking in government. The focus on high performance is relatively recent. Managers and supervisors have been solely responsible for managing day-to-day staff activities. “Experts” worked to develop valid appraisal practices but failed. The state of performance management was summarized in a 1991 National Research Council report, “Pay for Performance: Evaluating Performance Appraisal and Merit Pay.”  In many organizations there has been little change.

But there have been critics at least since the 1980s. The quality management guru W. Edwards Deming observed that one manager might say an employee performed at the 4 level while another might rate the same employee as a 3. He was right, of course, and that means the ratings are not valid measures. That’s essentially the same point made in a recent column, “Why the Typical Performance Review Is Overwhelmingly Biased.” We all see things through our own prism. The problem is compounded by the traditional use of vague performance criteria (e.g., dependability, cooperation, judgment) that require subjective judgments.

When management by objectives or goals gained acceptance for executives, managers and professionals, it represented an important new direction. It’s effectively universal outside of government.  

Goal-Based Management

Goals are used in managing senior executive performance, but the inflated ratings suggest there are problems. The practice needs to be extended to managers, supervisors and most professionals.  Metrics-based goals are not always feasible but managers should be able and expected to state what they plan to accomplish. As used in business, goals are stated in terms that define a commitment to improved results. The SMART acronym sets forth a proven approach to goal setting. Linking results to rewards reinforces individual accountability. That’s a missing piece today.

At year-end an individual’s performance relative to goals can be summarized in empirical terms. Verifiable results, supported with metrics, are a solid basis for determining financial awards.  

There are occupations where performance goals are not feasible (e.g., law enforcement, research). I learned that lesson years ago working with parole officers and again with nurses. In both fields the best performers are assigned to the toughest cases. Where goals are not appropriate, individual competence is a commonsense alternative. Individual capability is recognized and valued in virtually every society.

The use of goals highlights a significant change in appraisal practices. Goals are specific to the individual and what he or she accomplished. It’s very different than older systems where everyone was evaluated on the same criteria and encouraged rating comparisons. A useful analogy is a football team where each position has a unique role. No team would use the same criteria to assess tight ends and linebackers. In the same way it makes no sense to use the same performance criteria for obviously different occupations.

A new Harvard Business Review article says it all, “People Don’t Want to Be Compared with Others in Performance Reviews. They Want to Be Compared with Themselves.” That’s central to improvement.

Occupational Competencies

In the most widely used appraisal model goals are combined with competencies, defined as the knowledge, skills and abilities needed to perform at a high level. Mandating the same competencies across diverse occupations makes no sense. Occupation-specific competencies facilitate discussions of individual strengths and developmental needs. Small teams of high performers can, with guidance, define a profile of key competencies in two or three meetings.

As with goals, the best practice limits the number of competencies to those that differentiate the best performers. The purpose is to reinforce and highlight where employees should focus their development.

The National Geospatial-Intelligence Agency chose to rely on occupational competency profiles when it created its pay-for-performance salary system. The agency bases pay increases on demonstrated “knowledge, skills, abilities, and competencies necessary for successful job performance.”

Making Ratings Defensible

All of this depends on the ability and the commitment of managers and supervisors to take their responsibility seriously. Their evaluation and financial rewards should be based on the achievement of unit goals combined with a profile of competencies that define being an effective manager. Competency profiles defined for managers at each organizational level and specific to the roles they play would also support staffing. For example, a supervisor at the FBI holds a far different position than a supervisor at the Bureau of Prisons.

A four-fold strategy to address the rating inflation problem and institute more consistent and defensible evaluations would include:

1. Inviting co-workers and “customers” of an employee to provide feedback to an employee’s supervisor.
2. Soliciting feedback from employees on their supervisor’s effectiveness.
3. Requiring supervisors to explain and justify high and low ratings to a committee of peer level managers.  That will reduce possible bias and discrimination.
4. Requiring HR to analyze ratings to look for evidence of bias and discrimination.

In addition, agencies should adopt as standard practice NGA’s year-end practice of reviewing operations, where the agency solicits feedback from managers and employees, and commits to addressing problems.

In organizations with an announced policy to reward the better performers, employees need the assurance their performance will be evaluated and rewards granted fairly. It’s certainly possible to create that reality.  There is no more effective strategy to improve performance. Leaders should make it a priority.