Borrowed capability

Across a number of jurisdictions we have watched the same sequence run. A security problem becomes acute. Partners respond, and the response works. Figures improve, sometimes sharply, and the improvement is real rather than presentational. Then, twelve to twenty-four months later, the position has slipped back towards where it started, and nobody is entirely sure why.

The explanation is usually the same, and it is rarely a failure of will or of competence. The improvement rested on capability the institution was using rather than capability it owned.

Borrowed capability is not disguised. It is simply not counted, because most assessments are built to measure what an institution has and what it is currently achieving, and borrowed capability shows up correctly in both.

The four forms it takes

In our experience it arrives in four shapes, and an institution under pressure will usually have all four at once.

• Seconded or contracted personnel. Officers or specialists on loan from a partner force or an external provider. The number is always published; the instrument almost never is. Whether they are attested, appointed under a specific provision or serving under a commercial contract determines what they may lawfully do, who is accountable for them, and what happens to their powers when the arrangement ends.
• In-kind assets. Donated vessels, vehicles, sensors and communications equipment. The gift is announced. The through-life cost is not, and it arrives eighteen months later as a maintenance liability nobody budgeted for, in a currency the donor does not pay in.
• Non-recurrent funding. Support announced package by package, usually following an incident. It is genuinely helpful and it is structurally unreliable, because the thing that triggers it is the thing everyone is trying to prevent.
• A thin trained cadre. The improvement is credited to an institution but delivered by a small number of individuals who happen to be good at it. This is the least visible of the four and the most fragile. It is a personnel dependency wearing an institutional badge.

Why assessments miss it

Because the instruments most reviews use are the wrong shape. An establishment table shows posts. It does not show which of those posts are filled by someone who will be gone in eleven months. A crime statistic shows an outcome. It does not show whether the outcome was produced by a system or by three people. A budget shows what was spent. It does not show what was given.

Each of those instruments is accurate. Together they produce a picture that is true today and misleading about next year.

A review that measures only what an institution has and what it is achieving will describe borrowed capability as strength, because on the day of the assessment that is exactly what it looks like.

There is also an incentive problem, and it is worth naming. Nobody involved is well served by raising it. The partner providing the support would rather report success. The institution receiving it would rather not appear dependent. The reviewer, if there is one, was usually commissioned to assess capability rather than to model its decay. The question goes unasked not because it is difficult but because it is nobody’s job.

The questions that surface it

They are not complicated and they do not require special access. They require somebody to ask them and to write down the answers.

• Under what instrument does each external person serve, and when does it expire? If nobody can produce the document, that is the finding.
• Which functions would stop entirely if the external contribution ended tomorrow? Not degrade. Stop.
• What is the through-life cost of every donated asset, and whose budget line carries it? If the answer is that it has not been allocated, the asset is a liability with a delay on it.
• How many individuals does the current result actually depend on? Name them. If the list is short enough to write on one hand, the capability is personal rather than institutional.
• What is the time from recruitment to competence in each critical role, and does the training pipeline produce replacements faster than attrition removes them? This single ratio predicts more than any equipment inventory.
• Which of the recommendations in the last review were implemented, and which were quietly abandoned? The pattern of abandonment tells you what the institution can actually absorb.

What it means for how capability is bought

The practical consequence is that the sequence matters more than the list. An institution operating under resource constraint cannot do everything at once, and the order in which it does things determines whether any of it holds.

It also means that some of the most valuable recommendations are the ones that cost nothing to implement and produce no procurement. Writing down an authorisation regime. Recording a tasking decision. Establishing who owns a maintenance budget. Naming a successor. None of these appear on a capability shopping list and all of them determine whether the items on that list survive contact with the next twelve months.

And it means being honest about what an external adviser is for. A team that has to be there for the capability to work has not delivered a capability. It has delivered a subscription. The test of any assistance programme is what remains when the people who built it have gone home, and that is a test most programmes are never designed to pass because nobody set the exam.

A note on where this comes from

We have both spent time on the giving and the receiving side of this problem. Between us that has included strategic defence review work, border security advisory roles across West and Central Africa and the Sahel, mentoring of senior and junior officers in the Levant, maritime and intelligence advisory work in Europe and the Caribbean, and, earlier, service as officers on United Kingdom Strategic Defence and Security Reviews. In none of those places was borrowed capability a secret. It was simply not on the form.

We do not name the institutions we have worked for, and the examples above are drawn from a pattern rather than from any one of them. The pattern is the point. It is remarkably consistent, and the twelve-to-twenty-four-month window in which it becomes visible is remarkably consistent too.

Read next · Border policy

The wall and the watchman

A case for fencing Nigeria’s northern border is circulating again, built on a real and worsening threat. The threat is real. The wall does not follow from it.