Which occupations stay hard to fill until 2030 is decided less by job growth than by departures. Where many people leave each year, the market stays tight even when the occupation shrinks. Where few leave, it eases even when the occupation grows. For an employer that changes the first question to ask about a tight role: not how fast it grows, but how fast its people go.
Forecast and measurement, read together
In late 2025 the ROA published its forecast of the Dutch labour market by occupation to 2030, and UWV set that forecast beside its own tightness measure, the Spanningsindicator. For the first time, current tightness and its expected path can be read per occupation group. The starting point is lopsided: at the end of 2024 almost every scored occupation group sat on the tight or very tight side. Only a handful were average or loose.
Departure does the sorting
Within that picture, replacement demand separates the groups that stay tight from those that ease. Butchers are the clearest case. The trade barely grows, yet it has the highest replacement rate of all occupation groups and stays tight to 2030. Bus and tram drivers follow close behind. At the other end sit advisers in marketing, public relations and sales: few of them leave each year, and their market eases slightly despite steady demand for commercial staff.
The strongest evidence comes from occupations that shrink and stay tight anyway. Representatives and buyers lose jobs over the forecast period, yet their market moves only from very tight to slightly less tight. Gardeners and growers also shrink and tighten regardless, because many must be replaced. Primary school teachers face large replacement needs and stay tight. ICT managers grow fast, but so few leave that their market eases on balance.
Where forecast and measurement disagree
In technical occupations the instruments tell different stories. Most technical groups were very tight at the end of 2024. The forecast expects most of them to tighten further or hold level, and a small group to ease, welders and sheet metal workers most of all. UWV notes that this easing was not yet visible in its measurements up to the autumn of 2025.
In ICT the reverse happened. All scored ICT groups, including software developers, the largest occupation group in the country, were very tight at the end of 2024 and are forecast to ease slightly, mainly because few people leave. For ICT user support that easing has already been measured.
Tightness also has a postcode. Engineers are very tight in almost every province. For doctors, therapists and specialised nurses, ROA expects relief in Groningen, Utrecht and Gelderland only.
What this does not show
The projections to 2030 are expectations, not measurements. The Spanningsindicator divides vacancies by people on unemployment benefit for a short period, a narrow measure of supply. For a number of occupations, such as checkout staff where many pupils and students work, UWV computes no representative score at all. Provincial figures average a whole year, national ones a single quarter, so they do not compare directly. If the forecast easing in technical work shows up in the measurements during 2026, the outlook for those roles improves.
What you can do
- For a hard-to-fill role, check the replacement rate before job growth. A high outflow keeps a role tight, growth or no growth.
- Do not read a shrinking occupation as an easy one. Commercial roles such as representatives and buyers stay tight while they shrink.
- In technical recruitment, plan on today's tightness, not on the forecast easing. It has not arrived in the data. See engineering.
- In marketing, public relations and sales advice, expect slightly more room. Use it to raise the bar on selection rather than to shorten the process.