The Bond and the Border: Retaining Nurses

The frame. Nursing education is publicly financed human capital formation. When a trained nurse emigrates, the investing society loses the stream of services and taxes it expected. That is the classic brain drain account, and the 2.4 times agency premium makes the loss concrete and recurring: 210 vacancies staffed at a premium is a continuous drain that exists whether or not anyone migrates.

But the frame needs qualifying in three ways:

1. Brain drain versus brain circulation. Migrants remit, pay some tax at home, acquire skills and sometimes return. The loss is real but not equal to the training cost - see the next task.
2. Push as well as pull. The exit is a rational response to chronic understaffing, unsafe ratios, real-wage decline and security concerns. Much of the 'loss' is caused by management and funding failures, not by foreign recruiters.
3. Sunk cost. The JMD 3.5 million is spent whether the nurse stays or leaves. The marginal loss is the replacement cost.

Individuals or system? Neither purely. Exit is individually rational and no nurse owes the state her career. But there is a genuine collective action problem: if exit is unconstrained and widespread, the training pipeline and the ward rota collapse for the patients and colleagues who cannot leave - including nurses trained at the same expense who stayed. So the defensible question is not whether people may leave but on what terms - notice, cost-sharing, and above all whether the system gives them a reason to stay.

Blame the voluntary code for what it is: unenforceable, and the hospital's own numbers show it changed nothing.

Criteria: does it reduce vacancies; is it fair to individuals; can it be enforced; what does it do to future recruitment of trainees?

Proposal A - the bond. Strengths: targets those who benefit; keeps some nurses; a sliding scale rewards service. Weaknesses: hard to enforce once someone is abroad; regressive, since a nurse from a poor family is most deterred; if ward conditions remain bad it functions as a tax on leaving rather than a reason to stay; and a UK salary may absorb the repayment easily, so the bond deters the wrong people.

Proposal B - withholding the certificate. Strengths: high leverage, because overseas registration usually requires the certificate, so it actually bites. Weaknesses: it converts a professional document into an instrument of staffing policy and restricts a livelihood - the union's objection is not merely rhetorical. It may also fail: the nurse can wait two years and leave anyway, having worked those years resentfully; it may conflict with regional free-movement commitments; and the ward is still short when she goes.

Alternative - retention plus funded service. Use the agency premium as the budget: filling one vacancy at staff rates saves roughly 1.4 times a staff salary, so 210 vacancies represent a large, recurring, recoverable sum. Spend it on safe staffing ratios, pay progression, transport or housing support, and guaranteed study leave. Add a government-to-government agreement with guaranteed return and a defined notice period, and task-shifting to enrolled assistants to relieve registered-nurse workload.

Recommendation: an A-lite bond attached to funded postgraduate training rather than a penalty for leaving, combined with the retention package and a bilateral agreement; reject B as a licence restriction. Effectiveness first, fairness second, enforceability third - and B fails the third test while damaging the first.

Four reasons the figure overstates the loss:

1. Sunk cost. The JMD 3.5 million has already been spent. Whether the nurse stays or goes does not change the expenditure; what changes is the return on it. For a policy decision the relevant figure is the replacement cost - the 2.4 times agency premium and the service lost while the post is vacant.
2. Double counting. If the board counts both the training cost and the agency premium, it counts the same vacancy twice.
3. Remittances and return. Nurses abroad send money home - often in foreign currency, which matters for a current-account-constrained economy - pay some taxes, and a minority return with advanced skills. The net loss is the difference, not the gross.
4. The counterfactual. If the nurse had stayed, would the post have been filled and funded, at what staffing ratio, and with what equipment? A trained nurse working in a ward that cannot function may produce far less than the training cost implies.

How to present it instead:

• A recurring vacancy cost: 210 posts staffed at a premium, expressed annually.
• A service cost: rota gaps, cancelled training, overtime, adverse events.
• A pipeline cost: the deterrent effect on new applicants, which is where the 3.5 million becomes genuinely relevant - it is the cost of training someone who never practises here.

Why it matters: the inflated figure invites a restrictive policy response, and it cannot survive a journalist's arithmetic. The agency premium is smaller, unarguable, and points to the actual remedy.