Stackable Credentials for an AI-Disrupted Workforce: A Policy Framework for Aligning Technical and Vocational Education with Labor-Market Demand in the United States and South Africa

Abstract

Artificial intelligence (AI) is changing the tasks that make up middle-skill work faster than technical and vocational education and training (TVET) systems can revise their qualifications. This article argues that stackable credentials can close this gap: short, quality-assured awards that accumulate into larger qualifications. It also argues that they will do so only if they are governed as a system, not issued as isolated products. Using a comparative policy analysis and an integrative review of labour-economics research and applied research on technical occupations, the article examines the United States and South Africa. The two systems have complementary deficits. The United States has a responsive, employer-led credential market and, since 2025, federal Workforce Pell funding that requires short programmes to articulate into credit-bearing pathways, but it lacks a coherent competency architecture. South Africa has a well-developed national qualifications framework with part-qualifications and a levy-funded skills system, but it lacks strong demand signals, workplace absorption and learner progression, and official unemployment stood at 33.6 per cent in mid-2026. The article proposes the SCALE framework, whose five pillars are Signal (AI-aware labour-market intelligence), Compose (a modular competency architecture with an AI-fluency spine), Anchor (work-integrated learning and employer co-investment), Ledger (outcome-based quality assurance and verifiable learner records) and Equity (financing the whole stack and building educator capacity). For each pillar the article sets out instruments and indicators for both countries, and it closes with a phased implementation roadmap and a risk analysis.