The Garment Factory That Cannot Hire

Framework. Treat the factory as facing a labour supply decision by a potential worker: accept the job if the net utility of the job exceeds the reservation option (rural underemployment plus home-based work). Net utility is wage minus the costs of taking the job. The vacancy coexists with underemployment, so the market is not clearing - something is driving a wedge between the wage offered and the net value the worker perceives.

What the evidence says.

• Volume up 9 percent, unit price down 4 percent means margins are thin and the industry's willingness to pay a higher wage is limited. A wage subsidy would raise the offered wage, but the wedge here is not mainly the wage level.
• 30,000 vacancies concentrated in machine operators signals a specific skill, not general labour scarcity.
• 12,000 trained, 4,100 hired, most leaving within six months is the decisive clue. If the problem were simply skills, trainees would be hired. If it were simply wages, the 4,100 would stay. High attrition points to non-wage costs and job quality: safety on night transport, childcare, hours, and supervision.
• Tenure under two years confirms that retention, not recruitment alone, is the leak.

Binding constraints.

1. Mobility and safety: night-shift transport risk raises the perceived cost of the job for women, who are 78 percent of the workforce. This is a supply-side wedge independent of the wage.
2. Reproduction costs: absent childcare, the worker bears the cost of care, which can exceed the wage gain.
3. Matching and retention: the training programme produces skills that either do not match the machine types in use, or are not paired with placement and post-placement support.

So the vacancy is a job-quality and matching problem more than a wage-level problem.

Wage subsidy. Attractive to owners because it is fast and flexible. But:

• It subsidises a job that workers are already unwilling to take at the going wage because of non-wage costs. Paying more may not fix the night-transport and childcare wedge; it may simply raise the reservation wage and be capitalised into owner margins.
• It is fiscally open-ended and hard to withdraw without a worker exodus.
• It treats the symptom (too few applicants) while attrition stays high, so the subsidy must be paid repeatedly.

Expanded training. Addresses the operator-skill gap, which is real. But the existing programme already trained 12,000 with a 34 percent placement rate and heavy early exit - expanding a programme with that placement record multiplies the failure. Training without addressing why hires leave will produce more trained leavers. Training also does nothing for the 30,000 immediate vacancies.

Third option: a bundled 'job-quality' package. Fund safe night transport (factory-shared buses with a subsidy), on-site or zone-level childcare, and a placement-linked training contract where the training provider is paid partly on six-month retention. Rationale:

• It attacks the binding constraints identified - the non-wage wedge and attrition - rather than the wage level or the training headcount.
• It is self-targeting: factories that provide safe transport and childcare attract and keep workers, so the subsidy rewards the behaviour that solves the problem.
• It improves the information problem: retention-linked payment forces training providers to match skills to actual machine types and to support placement.

Caveat and judgement. Bundled measures take longer to show results than a wage subsidy and require coordination between factory, local government, and training providers. But the wage subsidy's speed is illusory if workers still leave within six months. On the evidence given - high vacancies, high attrition, low placement - the bundled package dominates on cost-effectiveness per retained worker, which is the outcome that matters for export capacity.