(Not) Managing Performance

By Scott Kupor, Director, U.S. Office of Personnel Management

December 5, 2025

We just wrapped the annual performance review cycle for OPM employees. It was – in a word – frustrating.

Why? Because today the federal government lacks a performance management process that would create a high-performance culture – one in which great people can achieve great things and be rewarded appropriately for their contributions. Instead, we have rampant ratings inflation and a lack of accountability for poor performers that fails to meaningfully differentiate between excellence, successful achievement of one’s objectives, and poor performance. As OPM wrote in a guidance memorandum from June, “[f]or many decades now, performance management across the Federal workforce has fallen short of what the American people should expect.”

To be clear, this is not an indictment of federal employees, nor are they at fault – they are doing exactly what they have been told and what the incentive system rewards. This is a fundamental organizational systems and leadership problem. Addressing it requires a complete overhaul.

Let’s start with some 2025 data from OPM’s performance management process.

Everyone gets rated on a 5-point scale, with 5 being the highest and 1 being the lowest.

Over 98% of employees are rated 3 or above – which means “fully successful” – and virtually nobody (less than 2%) is rated a 1 or 2.

Here are the data broken out by managers versus individual contributors

Interesting to see that the manager performance is more highly skewed towards the top rating relative to individual contributors. More on this later.

There is a modicum of good news – here are the data compared with the prior year.

Below are the annual comparison data when you strip out our largest organization that also happens to be about 90% unionized. Again, a meaningfully better distribution relative to the prior year.

And here’s the comparative year-over-year data for the most senior executives.

This one is notable in its improvement because we have put out formal guidance of a cap on 4/5 ratings for senior executives that goes into effect in FY’26.

So, what do we need to do to address the system failures?

First, ratings inflation is so deeply embedded in the performance management culture that it will likely require a set of formal regulatory changes to force more normalized distributions of ratings. We need to change the rules to drive the desired behavior.

We at OPM have already implemented this for the most senior government executives – starting in FY’26, no more than 30% of executives can be rated above the 3/fully successful rating. We plan to propose similar distribution guidelines to cover the rest of the federal employee base as well. 

Second, good performance management starts with annual objective-setting with measurable metrics that are aligned from the top of the organization and cascade down into individual contributor objectives. As OPM wrote in June, performance plans “should set forth clear performance expectations and goals that align individual employee efforts with organizational goals.” That is currently lacking in many places.

At OPM, for example, we have some organizations where the individuals who, according to the objectives they were given last year, did in fact earn a 5 (the top rating). However, these objectives were not aligned with the organization such that the organization missed its objectives. That makes no sense – if every individual far exceeds expectations, the organization should do the same since it is just the sum of the individuals who comprise the organization!

And, as you see in the data above, we have some disconnects in manager ratings relative to those of individual contributors. It seems strange that you would have many more managers rated a 5 than you have individual contributors in that ratings distribution. After all, manager objectives and individual contributor objectives should be tied to one another, so outperformance in one group is likely to be similarly reflected in the other.

How could this possibly happen and why don’t we just change it by fiat?

In some cases, the answer is simple. Many government organizations do not have a robust annual planning process that sets objectives at the top level and ensures they properly cascade down. This is a behavioral discipline that we will need to drive consistently.

In other cases, although the organizational objectives may have been updated, the individual contributor objectives have not been. At OPM, for example, some individual objectives haven’t been reviewed in as many as 10 years!

Third, we need much greater differentiation between what constitutes “fully successful” (that is a 3 rating) and the higher ratings of 4 and 5. For example, in one of OPM’s organizations, the difference between receiving a 3, 4, or 5 is a function of doing as little as a few percentage points of incremental performance.

Instead, we should have very meaningful step functions of incremental contributions for employees to earn above a 3 – we want to recognize exceptional performance, not minimally incremental performance.

At the same time, we should reward uncapped upside behavior as well. Earning a 5 means you are uniquely among the very best in the organization. We all know who those people are in our respective organizations, so we should start by defining a 5 to be commensurate with those individuals’ contributions and work our way down the rating scale from there. A 5 is a true bar raiser in the organization; ratings should reflect that. OPM is making sure agencies fix this and has already issued guidance to this effect.

We also need to re-visit whether a “2” makes sense as a rating step. A 2 is defined as “minimally successful” but I am not sure anyone could really give you a good definition of what that means. At some point if you are not a 3 (which stands for “fully successful”), do we really need to distinguish between someone who is minimally successful vs wholly unsuccessful?

Fourth, performance management needs to be intimately tied with rewards – compensation and other ways of recognizing achievement. For example, the SES ratings changes that OPM has implemented for FY26 not only cap the 4’s and 5’s at 30%, but also directs agencies to assign 60% of total rewards to this group of exceptional individuals. We need to disproportionately recognize those who truly move their organizations forward through exceptional performance. We have provided the same guidance for non-executives.

At the same time, we need to relook at compensation barriers that may be contributing to ratings inflation. For example, at OPM we have a policy that prohibits those ranked a 3 from receiving annual bonus rewards. It’s not surprising this policy contributes to the artificially high number of 4 and 5 ratings; managers don’t want to deny bonuses to their team members.

But, if we are defining the new “3” as the likely predominant rating in organizations, we will need to make sure we don’t have artificial barriers that prevent this. A “3” is a great score – you are doing everything that we have asked you to do – and for that you should also be eligible for some reasonable bonus-based compensation. (And we have told agencies they should make this clear to their employees: a “3” means you are achieving all expectations for your position and contributing in a meaningful way to the agency’s success in meeting organizational goals.). We can both recognize the vast majority of team members who do what they are asked and still disproportionately recognize the minority of employees who truly drive outperformance.  

Fifth, performance management is not just about set it and forget it. Yes, it starts with setting good, consistent objectives at the beginning of the year, but it requires regular check-ins with employees designed to manage adherence and provide ongoing feedback. That too is lacking in many organizations. And, as a result, when we get to year-end performance reviews, an employee who may not be performing has not received adequate ongoing feedback – and an opportunity to improve – such that it would be unfair to assign that individual a rating of 1 or 2.

Compound that with the very onerous process of putting an individual on a performance improvement plan (PIP) and ultimately trying to remove them from the organization for cause. We have a legal process that results in managers being tied up in paperwork, depositions, and other legal processes that take far too long versus being able to spend their time working on behalf of the American people. Despite the fact that almost very few employees historically get terminated for performance issues, we have more than 1,500 employment litigation attorneys FTEs employed across government to deal with these cases!

We can fix this.

Under President Trump’s direction, OPM is already making changes to the performance management process, as we’ve begun to announce in several guidance memos and rulemakings including one from June of this year.

Among the changes: shorten PIP time periods and streamline the process for managers to remove individuals who truly are not performing their jobs; limit performance improvement plans to no more than 30 days; do not require supervisors to use progressive discipline or a table of penalties before recommending termination; and eliminate the use of suspension as a substitute for fully removing an employee who engages in misconduct.

We are also creating Schedule Policy/Career to make sure key policymaking officials can be removed at-will if they perform poorly. And, thanks to President Trump’s Executive Order from April, we have also cut regulatory red-tape around removing probationary (that is, newly-hired) employees who perform poorly.  We also need to ensure every manager has objectives around performance management – starting with regular check-ins with their employees so they can provide ongoing feedback. We have just rolled out mandatory supervisor training on this very topic.

It’s unfair to terminate any employee for performance whose manager does not provide them feedback and an opportunity to address that feedback. At the same time, it’s unfair to everyone else in the organization who is in fact performing their job responsibilities to be saddled with individuals who fail to carry their weight. None of this is about firing people arbitrarily, and none of it is about firing people for political reasons.

Quite the opposite. It’s about building high-performing organizations that are accountable to the American taxpayer who underwrites their paychecks. Effective performance management is the cornerstone of a merit-based employment system—and the current lack of one in the government shows how far we have strayed from bedrock principles of merit, towards seeing federal jobs as entitlements.

The good news is that all of these challenges are addressable. They require systematic changes to the overall organizational incentive structure, but we do know that the right incentives will drive the right behavior. Turning a 2.1 million employee ship will not be easy, but it is fundamental to our ability to deliver on behalf of all Americans.