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

September 30, 2025

Most people know what the word “sweep” means, but I doubt most Americans know what “sweep” means to the federal government. In the words of the beloved Inigo Montoya (if you don’t know this movie reference, I feel sorry for you and urge you to correct your Hollywood illiteracy), “I do not think that word – “sweep” – means what you think it means.”

In DC, “sweep” refers to the year-end budget process (year-end here by the way is September 30) by which all government agencies “sweep up” (or gather) any unused appropriated funds before their ability to use those funds expires with the end of the fiscal year.

Sounds good so far, but then what happens with those unused dollars? Traditionally, there is a mad rush to spend (or “allocate” as we euphemistically say here in DC) those dollars prior to Sept 30. GSA reports show the contract dollar amount awarded in September is roughly double that in all other months.

In fact, the contractor army that dominates a huge portion of federal spending has coined a term for this process of grabbing the cash before it disappears – “sweep-up awards”. It’s easy to understand why, as it’s estimated that the federal government spends about $700 billion a year on contractor services (that’s about 10% of the total federal budget). This amount dwarfs the roughly $250-300 billion the government spends on its own 2+ million civilian employees.

But why is there this mad rush to spend money – whether via contractors or otherwise – at year-end?

Well, you’ve heard me say this before, but it’s another example of how organizational incentives drive behavior. If funds are unused at the end of the year, the agency “loses” them – meaning they would go back to the U.S. Treasury and be used to pay down the $36+ trillion national debt. I would certainly like that – and I hope that some of you would as well – but in DC returning money to pay down debt is anathema. Why?

Well, if I do that as an agency head then I am going to get lambasted by the opposing party in Congress who has oversight responsibility for my agency for not spending the money. After all, they appropriated it in the budget, so heaven forbid I should exercise any independent judgement in trying to deliver on my objectives at lower cost. And then, when I go back to these same representatives for the budget discussion for the new fiscal year, they will make my life difficult if I ask for new funds for important areas of investment because my track record of spending money is so poor! [And don’t get me started on the great debate around the mechanics of impoundment and recission.]

The budget process adds another wrinkle. Agencies submit their budget proposals to the Office of Management and Budget (OMB) about a year before the fiscal year begins. OMB then iterates with the agency, the President signs off, and it then goes to Congress. That means by the time we get to the sweeping process, we are working off a budget that was submitted roughly two years ago. Two years ago, we had a different administration and priorities were very different, yet typical practice is to spend all of those funds. It is no surprise a 2017 study found spending in the last week of the year is lower quality, as measured by CIO assessments, cost, and timeliness.

Even as I type this it all sounds crazy.

So, what if we had a different system of incentives that actually rewarded operational efficiency? What if I were rewarded for being able to do more for less? Or for deploying new technologies that can deliver higher quality services with greater automation? Or for not spending money on non-statutory activities that no longer yield value for taxpayers? After all, one would think we would want agency leaders to be good stewards of taxpayer dollars. If not, we should just appoint an automaton simply to execute on the same gameplan the agency has done for the past 50 years!

With all that being said and in the interest of transparency, let me tell you what we at the Office of Personnel Management did with our year-end sweep.

We underspent our budget this year by about $38 million in appropriated funds – that’s just under 10% of our budget. But, instead of allowing independent groups to spend their budgets, we asked the teams to submit their suggestions on whether it would make sense to use any excess funds to invest in critical areas. Thus, I reviewed an agency-wide view of opportunities and made decisions based not on which group had underspent but rather based on what initiatives overall were in the best interest of the American taxpayer. [In contrast, in most agencies historically, she who has the money left over in her budget gets to spend it independently.]

So, what did we do – in round numbers?

• We made our maximum contribution (~$12 million) to our technology development fund. This enables us to invest in critical, multi-year technology projects – modernizing our retirements services operation; deploying AI projects to improve operational efficiency; improving our ability to better serve our health insurance customers; etc. The funds are available for up to three years, a recognition that technology investments don’t always lend themselves well to fiscal year boundaries. Precisely where and when we deploy those funds will depend upon a more in-depth review and prioritization of each potential project.
• We are investing about $6.5 million in the President’s Merit Hiring proposals – to ensure we can attract the best and the brightest and hire based on demonstrated skills vs self-attestations and unreliable proxies for merit (e.g., college degrees, minimum tenure requirements). Unfortunately, doing so requires we essentially rewrite the classification and qualification standards for 600+ federal job descriptions to properly reflect a skills-based framework; not a small task, but one that we will accelerate massively with these efforts. In the absence of deploying this capital, it would have taken us 10+ years to do the work that we think we will now be able to do in about two years.
• We are reserving (not yet investing as we need to work with our partners at GSA on their long-term office building plans) about five million dollars for our Building Fund. If we determine appropriate, these dollars will enable us to tackle some of the estimated $40 million in deferred maintenance for our headquarters in DC. If we end up determining that’s not the best use of funds, we will simply return the money to the US Treasury.
• We are investing about three million dollars in modernizing our retirement services operations. While there is a lot more investment that will be required to get to job done on this one, these dollars will accelerate our efforts to automate the calculation of retirement benefits and to begin to replace much of the paper-based system with an advanced digital file system.

And finally – my favorite one – we are returning the rest (more than $11 million) to the US Treasury to help paydown the exorbitant federal deficit. Obviously, that’s peanuts in the context of the overall debt, but we are a small agency. Imagine the cumulative effect across the whole federal government if every agency were able to – and incented to – contribute as well. And as I’ve said before, small dollars ultimately add up to big dollars and will help drive the cultural change that has been long needed in government.

Now that I’ve spilled the beans, wish me luck with the Democrats on my oversight committee!