Your Degree Didn't Make You Safe.
Your Skills Will.

Singapore degree holders are now being retrenched at the highest rate of any education group. The Ministry of Manpower's Q1 2026 Labour Market Report records the retrenchment incidence at 3.1 per 1,000 resident employees, up from 2.6 the previous quarter. That is the highest figure across all qualification levels. The same report shows that diploma and professional certificate holders find new work faster after retrenchment: 69.4 per cent within six months, against 58.3 per cent for degree holders.

Steven Neo spent 25 years building a career backed by a bachelor's degree in information systems and an MBA. When he was retrenched from a senior programme manager role at a global tech firm in mid-2025, he applied for close to 180 jobs. Almost a year later, he had no offers. His experience, reported by Channel NewsAsia on 23 June 2026, reflects what the retrenchment data has been showing across several quarters: formal credentials are not protecting workers the way they used to.

What the MOM data shows

The MOM Labour Market Report Q1 2026 covers January to March. Total retrenchments were 3,830, up from 3,690 in Q4 2025. The overall incidence, at 1.6 per 1,000 employees, remains below the pre-pandemic quarterly average of 1.7. The aggregate number is not alarming. The distribution within it is.

Degree holders reached 3.1 per 1,000 — the highest across all qualification groups. Workers in their 50s hit the same figure, up from 2.8 the previous quarter. MOM attributes the concentration to “ongoing restructuring in professional and knowledge-intensive sectors.”

This pattern has held across multiple quarters. Associate Professor Walter Theseira of the Singapore University of Social Sciences told CNA that “retrenchment data has actually been consistent over time, in that more educated, higher-skilled workers and older workers have generally been at higher risk of retrenchment in recent years.”

His explanation: retrenchment happens when organisations have more staff than their current business requires. Lower-skilled roles, where hiring is already difficult, do not generate the same surplus. “A job that people want is going to have higher retrenchments, compared to jobs that people do not want,” he said.

The re-entry rate problem

Among retrenched residents in Q1 2026, diploma and professional certificate holders re-enter employment within six months at 69.4 per cent. Degree holders are at 58.3 per cent. That sits below the overall average of 60.7 per cent and below the secondary school graduate rate of 60.4 per cent.

The group with the most formal education finds new work more slowly than those with diplomas, professional certificates, or secondary qualifications. Only workers educated below secondary level show a lower rate.

A role that has deep but narrow applicability becomes high-risk when the sector it belongs to restructures broadly. Comparable positions at the same level may not exist. The credential does not transfer to adjacent fields the way more generalised, applied skills do.

Tan Weng Han, head of policy and partnerships at Vertical Institute, described the contrasting situation for diploma and ITE graduates in a CNA report: “They were never given the luxury of an ultimate credential. They learned on the job, proved capability through applied output, and built a habit of iterative skill acquisition from the start. That habit is now having a structural advantage.”

What employers are actually asking now

The shift in the re-entry rates reflects a broader change in how Singapore employers evaluate candidates. Tan Weng Han put it directly: employers are now asking “can I train this person to use a new tool?” rather than “what did this person graduate with?”

ManpowerGroup's 2026 Global Talent Shortage Surveyidentifies AI Literacy (25%) and AI Model and Application Development (26%) as Singapore's two hardest-to-fill capability gaps, overtaking IT and Data, which led in 2025. Sales and marketing roles are simultaneously among the hardest to fill at 19 per cent.

A practitioner who has spent a career demonstrating applied output is better positioned for this hiring environment than a specialist whose credentials signal depth in a category that is actively restructuring. That is what the re-entry rate data is showing.

Why chasing AI trends is not a strategy

The advice to upskill in AI, as most professionals have encountered it, specifies nothing about which skills, in what context, or toward what result. Without those answers, it is not actionable.

Steven Neo identified the specific version of this problem in his own job search. He noticed that highly experienced AI-related engineering roles are in demand, but require advanced skills he estimated would take years to hone, by which time those skills might already be obsolete.

He is right about the rate of change. AI tools, models, and frameworks are cycling faster than most professional development programmes can accommodate. Workers who organised their upskilling around the dominant AI platform of any given year may find themselves updating again before the learning has produced demonstrable results.

The pattern Tan Weng Han describes, “iterative skill acquisition” built through applied output, is precisely what the AI hype cycle undermines. It encourages recency over demonstrated result and breadth over working competence. Neither produces the transferable capability the re-entry data suggests employers now value.

Results-first skill building

The more useful question than “what should I learn about AI?” is narrower: which AI skill, built to a working level now, would make this specific work measurably more effective?

Building a company website with AI is a concrete example of that question answered well. The skill set covers developing and sharpening copy with AI assistance, structuring a site for search and answer engine visibility, and integrating the tracking that tells you what is working. A business with a well-built AI website communicates more clearly with its customers and attracts better-qualified traffic. The measurement baseline is set from the start. None of those outcomes depend on knowing what the underlying models are or keeping pace with model releases.

The skill transfers forward because the capability being built is not model-specific. It is the ability to use AI as a production tool in the context of business communication and marketing. That understanding carries across whatever generation of tools comes next.

AI-assisted campaign management, content production, and customer communications follow the same logic. In each case, the question is which skill, built to what depth, produces what result in the current operating environment. Breadth across many tools is a different exercise. It is also the one that the MOM data suggests does not protect a career the way demonstrated, transferable competence does.

For marketing professionals working out which applied AI skills are worth the time, our Training programme is built around demonstrable outcomes in marketing and business contexts, not certifications. For SME leaders thinking through what this means for their teams, the Advisory engagement starts with exactly that conversation.

Common questions