The Environmental and Social Action Plan: how lender conditions become dated obligations

1. Where the plan comes from

When a lender appraises a project against its standards, the review produces a set of findings: places where the project as designed or as currently operating does not yet meet the requirement. Some are documentary, such as an assessment that has not covered a required topic. Some are procedural, such as a grievance mechanism that exists on paper but has no staff. Some are substantive, such as land acquired before the lender's involvement in a manner that would not have satisfied the standard.

The lender does not generally refuse the transaction over these findings, nor does it treat them as satisfied by a promise to improve. It converts them into an action plan: a numbered list of actions, each with a deliverable, a responsible party and a deadline, agreed with the borrower before financial close [7] — the operational counterpart to PS-1's management-system requirement. [1][2] That document is the Environmental and Social Action Plan, and the borrower's execution of it is what the lender supervises thereafter — the World Bank's equivalent is the environmental and social commitment plan, agreed under the same logic. [3][4]

The generative logic matters because it explains the plan's shape. An ESAP is not a plan for managing the project's impacts — the management plans do that. It is a plan for closing the distance between the project's current state and the standard, which is why its items can look arbitrary to someone reading it without the due diligence findings alongside.

2. Why drafting determines closability

An ESAP item has to be closed eventually, and closure means a lender's environmental and social specialist accepting that the action is complete. Whether that is possible depends almost entirely on how the item was drafted, and the drafting happens under commercial time pressure at a stage when both parties want the plan agreed rather than perfected.

Items drafted as outcomes rather than deliverables are the recurring problem. 'Improve stakeholder engagement' and 'ensure livelihood restoration is effective' state objectives against which no specific act can be judged sufficient. The borrower does work, submits it, and is told it is not yet enough; the item remains open across successive supervision cycles; and the disagreement is not about the work but about a completion condition that was never defined. Experienced counterparties therefore negotiate the deliverable, not merely the obligation — a named document, a demonstrated capability, a dated event — precisely so that closure has a test.

The second drafting failure is the item whose deadline is fixed to a calendar date rather than a project milestone. Construction schedules move. An item due 'within six months of financial close' may fall due before the activity it governs has begun, producing either a technical breach on a plan that is otherwise being honoured, or an informal understanding between the parties that the deadline is not real — which corrodes the plan generally.

3. Conditions precedent and the disbursement link

Not all ESAP items carry the same weight. A subset is typically elevated into conditions precedent: matters that must be satisfied before financial close or before a particular disbursement. Elevation is a risk judgement — the lender identifies which failures would be unacceptable to discover after money has moved and makes those items structurally, rather than merely contractually, binding.

In land acquisition this most often attaches to compensation. A condition that compensation be paid, and civil works not commence on a given section, before that section's affected households have received their entitlements is the mechanism by which the standards' sequencing requirement acquires financial force. [5] The standard itself states the sequencing; the disbursement condition is what makes ignoring it expensive.

The practical consequence is that ESAP items become schedule-critical in a way that surprises project teams accustomed to treating safeguard work as parallel to engineering. An unclosed condition does not delay a report; it delays money, and therefore the works. This is generally the point at which environmental and social performance acquires the attention of people who do not otherwise attend to it.

4. The relationship to the commitment register

The ESAP and the commitment register overlap and are frequently confused. The distinction is one of origin and audience. The register consolidates every obligation the project has assumed from any source, and its audience is the project itself. The ESAP contains only those actions the lender required as a condition of financing, and its audience is the lender.

Properly handled, ESAP items appear in the register as a subset, flagged by source, so that the project manages one list rather than two. Where they are managed separately — the safeguards team maintaining a register while the finance team tracks the ESAP — the two drift, and the project ends up asserting different completion positions to different audiences without anyone intending to.

The subset relationship also runs the other way in an important respect. Closing every ESAP item does not mean the project is compliant; it means the specific gaps found at due diligence have been addressed. Projects sometimes read a fully closed ESAP as a clean bill of health and are then surprised by findings at completion audit on matters the plan never covered, because due diligence sampled rather than examined everything.

5. Supervision and the reopening problem

ESAP items are reviewed at supervision missions, and a status agreed with a lender's specialist at one mission is not necessarily durable. Staff change, and a successor may read an item's deliverable differently. A closed item may be reopened where subsequent events suggest the closure was premature — a grievance mechanism accepted as operational that then handles no cases, a livelihood programme accepted as established that shows no participation.

Borrowers experience reopening as bad faith, and it is better understood as the consequence of closing items against deliverables that evidenced establishment rather than function. An item closed on the existence of a document is vulnerable in a way that one closed on demonstrated operation is not, which is a further argument for negotiating deliverables carefully at the outset.

The durable protection is the record. A borrower who retains the submission, the specialist's written acceptance and the evidence behind it can defend a closure years later. One who relies on a mission aide-mémoire noting an item as closed, without the underlying material, cannot reconstruct why the lender was satisfied, and is effectively obliged to satisfy it again.

6. Conclusion

The Environmental and Social Action Plan is the shortest document in a project's safeguard architecture and the one most directly connected to money. Its items are negotiated once, under time pressure, by people whose principal objective is to reach financial close, and then govern conduct for the life of the financing. Whether they can be closed at all is determined at that negotiation, not afterwards.

IFC's handbook makes the same point about defining completion at the point of drafting rather than at the point of dispute. [6] The instrument rewards a discipline that is easy to state and rarely applied: agree what completion looks like before agreeing to the item. An action without a completion test is not a lighter obligation than one with a test. It is a heavier one, because it cannot be discharged.