Federal Employee Viewpoint Survey: The Price of Mistrust

“There’s a world of difference between insisting on someone’s doing something and establishing an atmosphere in which that person can grow into wanting to do it.”  Fred Rogers (of Mister Rogers’ Neighborhood)

Agencies are releasing their 2014 Federal Employee Viewpoint Survey (FEVS) results and many of them are not pretty. Here are results from a few agencies:

Looking at DoD as an example, the results for 2014 are lower than those for 2013 for more than half of the questions. The numbers in other agencies are not much better.  The FEVS questions cover a wide range of issues, but the ones that concern me most this year are those regarding agency leadership. Saying they are not pretty is an understatement. Here is how they look in agencies that represent 65% of the Federal work force:

2014 to 2013 FEVS Result Comparison

These numbers might lead someone to conclude that employees simply have poor opinions of their bosses, but the FEVS also includes questions regarding how employees view their supervisors and those numbers are significantly better. For example, the question “Overall, how good a job do you feel is being done by your immediate supervisor?” got the following positive responses:

  • Department of Defense   –   69.5%
  • DHS   –   63.1%
  • Interior   –   66.7%
  • Veterans Affairs   –   64.5%

Some of the difference can be attributed to familiarity. Employees see and talk with their supervisors every day, but may rarely (or never) have a conversation with higher ranking leaders. Regardless of the reason, the fact is that employees are reporting they do not have great trust and confidence in their leaders and the numbers are heading in the wrong direction. Some of the numbers are shocking. Only DoD was able to muster more than 50% positive responses for “I have a high level of respect for my organization’s senior leaders” and “Managers promote communication among different work units.” None of the agencies broke 60% on “Managers communicate the goals and priorities of the organization.” Think about that. How can an employee work to accomplish the agency’s goals if managers do not communicate what they are? Do we expect them to just stumble blindly into accomplishing them?

Not knowing where the agency wants to go is just one of the consequences of the trust and confidence issues the FEVS reveals. Mister Rogers was right – we have to create an environment where people want to work. A workforce that does not think much of its leaders can be pushed only so far. It will eventually have problems with productivity, recruiting high quality talent, customer service and other factors that are driven by employee engagement. Trust and confidence in leadership can also be a leading indicator that presages drops in other FEVS questions. If this trend continues, it is likely to drag all of the results even farther down. At some point, maintaining morale and productivity will become increasingly difficult. The downward spiral has to be stopped soon. Stopping it will take concerted actions by the key players. Here are a few ideas that should be considered.

  • Link manager and executive performance goals directly to agency strategies and objectives. One reason managers do not adequately communicate goals and priorities is that many of them do not have performance plans that are tied to the overall agency objectives. If manager performance plans always contained that type of linkage, managers would have a more personal stake in communicating those goals to their subordinates. Failure to communicate would decrease the probability that they could accomplish the objectives in their performance plans.
  • Conduct regular reviews of the organization’s progress in meeting goals. When I was HR Director for the Defense Logistics Agency, I implemented a monthly In-Progress Review for all of our critical programs. The first few were rough – they came in barely aware of the status of their programs and unprepared to discuss anything in detail. After a few painful meetings, people came in ready to discuss status, issues they were facing, support they might need, and any problems that needed to be addressed. We went from struggling to implement programs to being able to successfully develop and deploy new initiatives.
  • Create cross-functional and cross-organizational teams as a normal operating process. Too many organizations tolerate stovepipes that operate as though they are islands. Those stovepipes will never go away on their own. They are too comfortable for their inhabitants and give some managers a sense of control and power. Sharing information is viewed as weakening that power. The truth is that sharing information makes the entire organization more effective. Agencies can improve their performance by ensuring that program teams have representation from multiple organizations with diverse skill sets and views.
  • Invest in Leader Development programs for managers and executives. “Supervisory training” is common in Federal agencies – usually taking the form of a 40-hour or less course that is heavily focused on the HR rules and policies managers need to know. That training is typically focused on new supervisors. Less common are programs that focus on improving the skills of managers and executives. I have heard too many people tell me “Senior Executives don’t need training.” Nothing could be further from the truth. The roles of executives and supervisors are different. We often assume that a newly-minted SES has all of the skills he or she needs when the new SES is wondering if s/he is the right person for the job. If we want to see improvements in how senior leaders are perceived, we have to develop those leaders.
  • Do not exempt political appointees from leader development programs. Few agencies invest in training for political appointees. The Obama Administration has a program that provides appointees with an initial orientation, but once they get to their agencies, formal training is rare. Given the lack of government experience most appointees have, agencies should identify critical skills and conduct training to fill skill gaps.
  • Hire better leaders. This may seem a bit too obvious, but 30+ years in Federal HR tell me it does not always happen. The very best leader development strategy is to hire leaders who need less development. Senior manager and executive vacancies should be filled using processes that focus on rigorous assessment of skills. Most selection processes have such questionable validity that they are only a little better than random chance. For example, most agencies use unstructured interviews. The correlation coefficient (a 0.0 – 1.0 measure of the relationship between the assessment and performance) for unstructured interviews is typically reported by researchers as being in the .20 range. Structured interviews and other more valid assessments techniques are more likely to produce selectees who are able to perform.

Unless agencies begin to take steps such as these, it is unlikely we will see significant improvement in FEVS scores. They will continue to fall and the downward spiral will continue until the Federal government becomes a toxic workplace. We cannot afford to let that happen.


Performance Ratings – Why Do We Do Them?

This is the third post in my series on performance ratings.The first two are (1) here and (2) here.

My earlier posts explained why I believe most performance rating processes are inherently flawed and doomed to fail unlikely to produce good results. There is another reason – most performance rating processes were designed without a clear purpose in mind. In the past year I have talked with Chief Human Capital/Resources Officers, line managers, employees, union officials and HR practitioners about the subject of performance. One of the most striking facts is that most can explain the alleged benefits of their performance appraisal process, but few will claim they explicitly designed it with those ends in mind. Even fewer will claim it actually accomplishes those objectives.

Most Federal HR leaders and practitioners will tell you they have the system they do because it is mandated by law (5 USC 4302) to do it. Section 4302 requires agencies to use the results of performance appraisals as a basis for training, rewarding, reassigning, promoting, reducing in grade, retaining, and removing employees. In other words, for almost all significant decisions regarding employees. Doing something solely because it is required by law is not enough to produce a good result. In fact, it tends to produce poor processes that are nothing more than compliance exercises. Current rating processes affect more than 2 million Federal employees, cost a fortune, often harm morale and productivity, and generate few benefits.

Consider the time it takes for those folks to participate in the rating process, then add the time their supervisors and any other reviewers are consuming. If each employee spends only 5 hours annually in the process and each supervisor or manager spends 40 hours (very conservative estimates), the annual cost* of the current processes is over $900 million. That does not include systems and other non-labor costs, nor does it include the cost of grievances, appeals, arbitration and lawsuits. What does our $900+ million buy us? The vast majority of employees receive a rating at or above the mid-point. The number of marginal or less-than-satisfactory ratings is small. The number of unsatisfactory ratings is miniscule. Few managers, employees, union officials (or anyone else) will tell you the rating process serves a useful purpose. Many will say it is harmful. So – the Federal government spends more than $900,000,000 every year on a process that generates little positive return and most likely does harm.

A big part of the problem is lack of clear intent. A process that is not designed with a clear purpose in mind produces goodness only by accident. While we could be charitable and say the government has a clear intent – the legal requirement to use the ratings for training, rewarding, reassigning, promoting, reducing in grade, retaining, and removing employees, saying it doesn’t make it so. Let’s take a look at those requirements one at a time:

  • Training. To be effective in indentifying training requirements for employees, an agency would need to link its training and rating processes. It would identify goals, performance deficiencies, the competencies that are lacking, courses or other training that might provide those skills, and a means of incorporating them into individual development plans. Some agencies have the component parts, but few agencies formal procedures that link performance ratings to the training process. Training that results from a performance rating is likely to be in those rare cases where an employee is placed on a performance improvement period following an unsatisfactory rating.
  • Rewarding. Most agencies link ratings and some awards, while others are attempting to decouple awards/bonuses from the appraisal process. The thinking is sound – ratings are done once each year and recognition for great work should be more proximate to the work being done. The dollar value of awards in the Federal government is so small, that little meaningful difference exists between awards for above-average and top performers. Now that OMB has frozen awards as a result of sequestration, the question is, for the time being, moot.
  • Reassigning. Reassignments based on performance ratings are virtually impossible to quantify, because there is no way to track them in current HR systems. Anecdotal evidence suggests they are uncommon, and when they do occur, they are in lieu of a more severe action, such as a downgrade or removal.
  • Promoting. Most agencies factor the rating into the promotion process, but it is not the primary or even a key driver of promotion eligibility or selection. Some opponents of using performance ratings in the selection process make the argument that allowing the same people to assign ratings, then use the ratings they assigned as the basis for selections they will make is a type of pernicious double-dipping that is not in the public interest. For the most part, it is safe to say agencies do, to some degree, use ratings for promotion purposes.
  • Reducing in grade. Downgrades for performance purposes are uncommon. The regulations governing such actions require an agency to prove the basis for the action, but do not provide for review of the type of action by the Merit Systems Protection Board. MSPB can review whether the agency followed the correct procedures and proved the employee did not meet approved performance standards. They cannot decide a demotion should have been one grade rather than two or more or a removal should have been a demotion.  As a result, an agency that goes through the long process of taking a performance-based action is unlikely to stop at a demotion. Add to that the negative effects on morale that can result from a demoted employee still being in the workplace, and the rationale for downgrading versus removal is weak.
  • Retaining. Retention decisions typically involve probationary periods, such as for new hires or newly promoted supervisors. How does that work out? New supervisors must complete a 1-year probationary period. MSPB’s 2010 report, A Call to Action: Improving First-Level Supervision of Federal Employees, found that less than one half of one percent of supervisors are removed from their supervisory position during probation.  In MSPB’s 2005 report, The Probationary Period: A Critical Assessment Opportunity, MSPB reports that 1.6% of competitive service employees are removed from their jobs in the first year of service. Clearly, little is being done with respect to retention decisions.
  • Removing. Since 2008, between 7,488 and 8,187 full-time, permanent Federal employees have been removed for performance or misconduct reasons each year. The majority of those are for conduct. If less than two tenths of one percent of employees are removed for performance reasons, it is safe to conclude the performance rating process is not working for making removal decisions.

Federal HR professionals who administer these systems do not claim they are great systems. Many will agree they are frustrating, difficult to administer, generate far too many grievances and appeals, too costly for the benefits they return, and generally not satisfactory. They recognize it is difficult to design a process to accomplish significant results with respect to people issues. They know it is next to impossible to design a successful process that accomplishes many conflicting goals all at once. Imagine how difficult it is to design a process that accomplishes outcomes that were not in the process designer’s minds during the design process.

The Swiss Army Knife approach to rating processes is almost certain to fail. Conflicting goals, inconsistent messaging and employee pushback are the result when we claim our rating processes can do everything. The only way to design an effective rating process is to set limited, very clear goals for what it will accomplish. It requires the discipline to throw out the Swiss Army Knife approach, determine what outcomes are most needed by the organization and its people, and establish clear measures to determine whetherthe rating process actually does what it is designed to do.

* Cost estimates are based upon 2 million employees with an average salary of $77,535 and approximately 260,000 managers and supervisors with an average salary of approximately $107,000. Because supervisors and managers also are employees for purposes of performance ratings, they are included in the count of employees as well.