The slogan is simple: remove the migrant, free the job for a local. The labour market is not. Between the vacancy created by a departure and the local worker who fills it lie four frictions — skill, location, wage and willingness — and each of them takes time that a protest does not allow for.
South Africa’s unemployment is genuine and severe; the official rate rose to 32.7% in the first quarter, and the pressure behind the slogan is real. But the Cabinet statement of 6 May 2026 and the government’s broader posture reflect a harder truth than the placard: substitution is a process, not a switch, and assuming it is instantaneous is how job-creation promises quietly fail.
The Four Frictions: Why a Vacancy Is Not a Placement
A migrant-held job is a bundle of specifics. It sits in a particular place, demands a particular skill, pays a particular wage and asks for a particular tolerance of conditions. A local jobseeker fills it only when all four line up.
Skill mismatch is the first gap: a spaza operator’s supplier network, a welder’s certification or a farm supervisor’s experience is not conferred by nationality. Location is the second: the unemployed worker in one province does not automatically move to a vacancy in another, and the cost of moving is real. Wage is the third: many of the roles pay at a level that assumes hardship acceptance, not aspiration. Willingness is the fourth: some work is arduous, informal or insecure enough that it goes unfilled at the offered wage regardless of who is available.
A vacancy is not a placement; it is the start of a search that can end in nobody.
The Sectors Where the Gap Bites
Migrant labour concentrates where these frictions are sharpest: commercial agriculture, hospitality, construction, domestic work and the informal retail economy. These are not accidents of preference. They are sectors with seasonal peaks, dispersed worksites, thin margins and demanding conditions — precisely the profile that makes rapid local substitution hardest.
Remove the labour faster than it can be replaced and the immediate result is not a transfer of jobs but a loss of output. A farm that cannot harvest on schedule, a building site that stalls, a kitchen that cannot open: each is a business that shrinks, and a shrinking business hires fewer people, not more. The promised gain to local workers can arrive, if it arrives, only after a contraction that costs some of them their jobs too.
Where substitution lags removal, the economy loses twice before it gains once.
Measuring Before Asserting
The corrective is empirical. Before anyone predicts a net gain in local employment from displacement, the mismatches should be measured: how many of the vacated roles match the skills of the locally unemployed, how many sit within reach of where those workers live, how many pay a wage that a South African jobseeker will accept, and how many are wanted at all at that wage.
That audit will not always comfort either side of the argument. But it replaces a slogan with a number, and a number is what a serious labour-market policy is built on.
Until the mismatch is measured, the job gain is a forecast without arithmetic.
So What: Price the Adjustment Lag Into Every Plan
For employers, investors and policymakers, the intelligence angle is to model the adjustment lag explicitly. Map migrant-labour dependency by sector and site, quantify the skills, location, wage and willingness gaps, and treat the transition period as a cost to be managed rather than a benefit to be assumed. Workforce continuity plans should assume replacement measured in seasons, not slogans.
Labour substitution is slower than the chant that demands it. The operators who accept that arithmetic will keep producing while others discover it the hard way.
