Corrective action after a non-compliance finding: what remedy actually requires

1. How findings arise

A non-compliance finding can originate in several ways, and the origin shapes the response. A lender's supervision mission may identify a departure from an agreed plan. An independent monitor may report a discrepancy between what a project asserts and what field verification shows. A completion audit may find that entitlements described in a resettlement plan were not delivered. [1] A grievance may escalate to an independent accountability mechanism whose investigation reaches a finding the project disputes. [2]

These differ in adversarial temperature. A supervision finding is generally cooperative: the lender and the borrower share an interest in resolving it quietly. An accountability mechanism finding is not, because it arises from a complaint by people who consider themselves harmed and who have already exhausted the project's own processes.

The distinction matters because the corrective action plan produced in a cooperative setting is negotiated between institutions, while the complainants — the people whose situation generated the finding — are usually not party to that negotiation. A plan can therefore be agreed, implemented and closed to the satisfaction of both institutions without the complainants ever accepting that anything was fixed.

2. Why plans drift towards process

Examine a sample of corrective action plans and a pattern is unmistakable. Where a finding concerns compensation paid below replacement cost, the actions are frequently to revise the valuation methodology, retrain the valuation team and strengthen review procedures. Each is sensible. None of them pays the person who was underpaid.

This is not usually evasion. It is what institutions are capable of doing. IFC's handbook describes the same drift toward procedural remedy in its own account of common findings. [3] Revising a procedure is within the authority of the project team, requires no new budget line, produces a documentable deliverable and prevents recurrence — which is precisely what a supervising lender wants to see. Paying additional compensation to a defined set of households requires funds that were not appropriated, a legal basis for a payment outside the approved framework, a defensible method of identifying who qualifies, and an admission that the earlier payment was inadequate. The procedural action is available; the restorative one requires authority the project team does not have.

The result is a systematic bias in what correction means. Future affected people are protected by improved procedure. The people whose treatment generated the finding are not compensated by it. Since findings are generated by harm to identifiable people, this bias is the central problem in the field rather than a marginal one.

3. Correction, remedy and the difference

Correction addresses a deficiency in a system. Remedy addresses a loss suffered by a person. They are related but not substitutable, and a plan that delivers the first while describing itself as delivering the second obscures a gap that ought to be visible.

Remedy in a resettlement setting normally requires one or more of: a payment that closes the shortfall between what was received and what was owed; restoration of an asset or access that was taken; provision of a benefit that was promised and not delivered; or, where the loss cannot be restored, an agreed substitute negotiated with the person affected. Each demands individual determination, because entitlement is individual.

The practical test for whether a plan contains remedy is simple and rarely applied: does the plan name, or provide a method for identifying, the specific people who were harmed, and does it commit to a determination for each? A plan whose actions all operate at the level of the project rather than the household is a correction plan, whatever its title.

4. The evidentiary problem

Remedy requires knowing who was affected and what each received, which is exactly what a project that has been found non-compliant frequently cannot establish. The finding often arises because record-keeping was inadequate; the same inadequacy then obstructs the response. Where a project cannot show what it paid to whom, it cannot compute a shortfall, and the corrective action plan retreats to process almost by necessity.

This gives contemporaneous record-keeping a significance beyond audit convenience. A project with complete, individuated records of entitlement and payment can respond to an adverse finding with an arithmetic exercise. One without them faces a reconstruction exercise conducted years later among people with every reason to state their loss at its highest, and no way to distinguish a genuine claim from an opportunistic one.

The asymmetry compounds. Poor records make remedy harder, which makes procedural correction more attractive, which leaves the original harm unaddressed, which sustains the grievance that produced the finding. Projects that have been through this cycle generally describe records as the thing they would do differently.

5. Timeliness and the value of delayed remedy

Correction plans typically run on institutional timescales — quarterly reporting, annual supervision, closure over one or two years. The harm they respond to operated on a household timescale, and the interval between the two is where remedy loses most of its value.

A household that received inadequate compensation for agricultural land does not simply hold an unpaid balance. It has already adjusted: sold livestock, withdrawn a child from school, taken on debt at informal rates, or migrated for work. A payment arriving three years later arrives to a household that has absorbed the loss through irreversible decisions, and cannot restore the position that would have obtained had the payment been correct at the time.

This argues for provisional or interim measures where a finding is likely but not yet concluded — a position accountability practitioners advance and institutions resist, since interim payment resembles an admission. [4] The resistance is understandable and the cost of it falls entirely on the household, which is the reason the argument keeps being made.

6. Conclusion

Corrective action planning is the point at which a project's stated commitment to accountability is tested against what it is actually willing to do. The pattern is visible across the accountability mechanisms' own published caseload. [5] The test is not whether a plan is produced — one always is — but whether it reaches the people whose treatment generated the finding, and whether it reaches them while it can still matter.

The recurring answer in the documented record is that plans improve systems and rarely restore individuals. Understanding why this happens structurally, rather than attributing it to bad faith, is what makes it addressable: the constraint is authority and evidence, and both can be arranged for in advance by a project that expects, as every large project should, to be found wanting in something.