If You Pay a Neighbour to Guard Your Border, Why Does It Become Their Leverage?

When a border is “outsourced”, what is actually being outsourced?

What gets outsourced is not the border itself but the act of interception. The line, its legal status and its sovereignty do not move. What moves is who stands on the shore, who patrols, and who reaches the traveller first.

The typical form: one side supplies money, equipment and training; the other uses its own forces, on its own territory, to stop people heading for the first. The EU and Morocco, the EU and Turkey, the United States and Mexico all share this structure of border externalisation.

For the payer this has a very practical benefit. People stopped that way never enter your jurisdiction, so they never trigger your asylum procedures, never create legal obligations for you, and never appear in your statistics. The money buys not just manpower but distance from legal responsibility.

That is also why these arrangements are almost always written as technical cooperation or development assistance rather than security agreements. The naming is deliberate: development assistance does not have to explain why one country is doing another country’s law enforcement.

If it leaks leverage, why does the payer do it anyway?

Because in the short run it works, and it is cheap. Doing the interception with your own coastguard and navy is expensive, legally exposed and politically ugly. Paying someone to do the same thing on their own coast reduces all three at once.

There is a less-stated reason too: it moves the problem out of electoral view. Interceptions happen abroad, the footage does not reach domestic news, the numbers do not enter domestic statistics, and the political pressure falls accordingly.

The difficulty is that this effectiveness has a shelf life. Once the arrangement has run for a few years, the payer steadily loses its own capacity to intercept — staffing, equipment and procedures all atrophy. The gap is discovered at the moment it has to act alone.

It is the familiar problem with outsourcing, relocated to a border: what you contract out is not only the work but the ability to do the work. And once the ability is gone, so is the bargaining position.

What is the difference between lighter enforcement and deliberate pressure?

From the outside you usually cannot tell — which is exactly what makes the instrument useful.

The keeper holds a continuous dial, running from “publicly suspend cooperation” all the way down to “change nothing, simply stop warning in advance”. The further down, the smaller and less identifiable the action — but not necessarily the smaller the effect. One patrol fewer shows up in the numbers on the far shore within hours.

The crux is deniability. Rosters change anyway, notifications are sometimes slow anyway, personnel have to be allocated anyway. Every individual action is explainable as routine operation, and stringing them into an intention requires communications records or policy instructions — precisely what outsiders cannot obtain.

So these episodes almost always end on the same sentence: “no conclusive evidence”. That sentence is true, but it means cannot be shown, not did not happen. Treating the two as identical is the most common error on this subject — and it runs both ways: it can excuse real pressure, and it can turn a simple capacity shortfall into an accusation of conspiracy.

EU law calls it the instrumentalisation of migration; the whole approach is a textbook gray zone action: clear effect, below the threshold, blurred responsibility.

The money has been paid — so why is the payer still the one being led?

Because leverage does not follow the money. It follows who can adjust their own end at will, and how quickly the other side notices.

For the payer, withdrawing resources means budgetary, parliamentary and diplomatic process — months, every step of it public. For the keeper, adjusting enforcement takes a sentence, works in hours, and requires no explanation. The same agreement, two entirely different sets of things you can move.

Alternatives widen the gap. The keeper has other partners beyond the EU, on worse terms but real. The payer cannot find a third party to hold the same border — geography is not renegotiable.

This is why such arrangements so often become a point of application for economic coercion: not because the recipient is unusually forceful, but because the thing it holds can be adjusted daily, and the thing the payer holds cannot.

One qualification: none of this makes the keeper free. Funding and diplomatic ties remain real constraints — they simply bind far more slowly. The difference in speed is itself the leverage.

Which signals show that such an arrangement is loosening?

First, watch whether a shelved dispute is being raised again. Arrangements like these almost always rest on an agreement not to discuss something — sovereignty, historical status, the position of some third party. When an official reopens it in public, that is rarely a slip; it is usually a test of the other side’s reaction.

Second, watch coincidences in timing — but do not convert them into causation. Events following one another is a signal worth noting, not evidence; evidence means records and instructions. Keeping the two apart is what keeps the judgement stable.

Third, watch how much capacity the payer still has to act alone. Are the vessels, the personnel and the contingency procedures still there? If they have atrophied, the bargaining position is weaker than the paperwork suggests, however well the agreement is drafted.

Fourth, watch where the money goes after a crisis. If the payer responds to a failure by enlarging the same arrangement, it has looked for alternatives and found none — a more reliable reading of dependence than any statement.

For companies the portable judgement is this: when a cooperation rests on a dispute both sides have agreed not to raise, its stability is set not by the terms of cooperation but by the political temperature of that dispute. Contract clauses cannot protect against that risk, because the risk is not in the clauses.