The Netherlands Puts 25,000 Civil Service Jobs on a 2035 Clock

The Netherlands plans to cut about 25,000 civil service jobs by 2035, roughly 16% of its civil servants, and State Secretary Eric van der Burg has put that deadline in a letter to parliament. The plan runs on a 9-year clock. A civil servant who is 52 today will be 61 when it reaches its full target.

DutchNews.nl reported the letter on October 26, and Bloomberg filled in the money a day later. The Dutch government is aiming for savings of €900 million a year by 2027, rising to €3.5 billion a year from 2035. Central-government headcount reached a record 160,000 at the end of 2025.

Any 52-year-old civil servant reading those numbers at the kitchen table does the same subtraction. That leaves 9 more working years, and no one can yet say what the job will look like at the end of them. I call that position too old to retrain, too young to retire. In the Netherlands, it now has a date attached.

The Dutch plan names its levers and its dates

The Dutch government deserves plain credit for how much of this it put on the record. Van der Burg's letter gives a date and a savings figure, then names every lever it intends to pull, with AI as one of several: expanding AI and standardized IT, scrapping 500 rules a year, fewer advisory bodies, less external hiring, and doing some things less or not at all. The cabinet says it will present its plans for deploying AI early next year.

Just as important, the reporting attributes no specific share of the 25,000 jobs to AI, and the government has made no claim that software alone delivers the number. That restraint matters. A workforce plan that credits every reduction to technology hides the choices people made, and this one keeps those choices in view.

“We cannot keep doing what we have been doing.” Eric van der Burg, Dutch State Secretary, quoted by Bloomberg

Van der Burg described the goal as a government that is “simpler, smarter, and smaller.” Readers will hold their own views on the cuts. Whatever those views are, a published timetable is useful to anyone who has to plan around it, and the people with the most planning to do are the ones whose work is about to change.

Most workforce decisions run on a quarterly clock

Most of the workforce decisions I see get made on a quarterly clock. A headcount target arrives in one planning cycle and the reductions follow in the next. Retraining gets discussed afterward, if at all. Under that rhythm, the people affected learn the timeline at about the same moment everyone else does, with little time left to prepare.

A 9-year horizon changes what is possible. Rebuilding real skill in a new role takes years, because a person has to learn the new work and then earn trust doing it. That is long enough to retrain someone for a different job and still see them settled in it before 2035. When you read the AI job loss statistics we track, ask how much notice came with each number. The Dutch plan came with 9 years.

Inside companies, the same problem shows up as stalled internal mobility, which I looked at in the Workday internal mobility post. There I argued that an internal move is the cheapest retraining program a company owns, and a long runway is what makes those moves possible.

A long runway helps only when the relearning is scheduled on it

A 9-year runway helps people only when the relearning is booked inside it, with dates as firm as the savings targets. Left unbooked, the time turns into the learning debt: each year retraining waits, the skills a person holds lose value while the skills the new work needs go unlearned. Start in year 7 and the bill comes due with too little time left to pay it.

Cuts can be scheduled in a quarter. Relearning needs years, and this plan has them.

Nations have relearned on a long clock before. Between 1910 and 1940, America took high school graduation from roughly 1 in 10 teenagers to 1 in 2, and built its middle class on that work. It took 30 years of steady building and it served ordinary families.

The 9 years are a window to design the relearning

I come to this from inside institutions. I serve as Innovator-in-Residence at Tulane University. At Cisco, I spent a decade establishing innovation centers across 14 countries and shaped a $1.1 billion innovation portfolio. The lesson from both is consistent: new capability takes hold when the people expected to use it get time and a plan to learn it.

That is why I read the Dutch timeline as an opening. The cabinet's AI plans arrive early next year, which is the natural moment to set a relearning schedule beside each lever, naming which roles change first and what the people in them will learn by when. Work like this moves faster when a government brings in partners at the start, and it is work I would welcome contributing to, in the Netherlands or anywhere a public employer sets a clock this long.

The same question reaches well beyond one country. Across our Future of Work coverage, careers are getting longer while the work inside them changes faster. A published clock gives the 52-year-old at the kitchen table something rare: enough warning to plan. For the managers who will carry the plan out, that warning comes with a responsibility, because the schedule people can see is the one they can prepare for.

Here is a question to take into your next leadership meeting: if your organization published its own 9-year clock, what would the relearning schedule beside it promise your 52-year-olds? I'm easy to find.

Here is what makes Alex a credible voice on this topic: Alex Goryachev is Innovator-in-Residence at Tulane University and spent a decade establishing Cisco's innovation centers across 14 countries. At Cisco he shaped a $1.1 billion innovation portfolio, which gave him a close view of how large institutions bring new capability to the people expected to use it.

Talk with Alex about your workforce strategy →