How do you know whether your project needs a RAP?

One test decides it: does the project take land, assets or access from anyone. Scale then decides which instrument, not whether.

Olule Solomon7 min read

One test: does the project take land, assets or access from anyone, voluntarily or not. If yes, a resettlement instrument is required. Scale then determines which instrument — a full plan, an abbreviated one, or a policy framework where the footprint is not yet known — but not whether you need one at all.

The triggers

  • Land acquisition of any kind, permanent or temporary, including easements and wayleaves.
  • Restrictions on land use that reduce what a holder may do — a pipeline easement, a buffer zone, a height limit.
  • Loss of access to resources — grazing, water, fishing grounds, forest — even where no land changes hands.[1]
  • Economic displacement — anyone losing income because of the project, including traders, employees and sharecroppers.
  • Physical displacement — anyone having to move.

Note what is not on that list: ownership. Whether the affected people hold title is irrelevant to whether an instrument is needed.[2]

Which instrument

  1. Full RAP — significant displacement, especially physical displacement or households losing most of their productive base.
  2. Abbreviated plan — few people, simple and homogeneous losses, no physical displacement. Same obligations, lighter analysis.[3]
  3. Policy framework — displacement is likely but the footprint is not yet known, as with programmatic lending or subprojects to be identified later.

The screening questions that actually decide it

A headcount threshold is the usual filter and a poor stopping point. Three better questions:

Is anyone physically displaced? Even a handful of households pulls in site selection, replacement housing, host communities and long monitoring — the analytical load of a full plan.

Is any household losing most of its productive base? Severity of loss per household predicts impoverishment better than the total number affected.

Are people without title, or common-property users, affected? That is where the difficult eligibility work is, and shortening the analysis removes exactly the part that matters.

Two mistakes worth avoiding

Assuming a road reserve is exempt. Building within land the state already owns is a statement about title, not about the people living there. If they are displaced, they are in scope.

Assuming a negotiated purchase is exempt. Where a seller could not realistically refuse, the acquisition is treated as involuntary and the full requirements apply.[4]

Screening should happen at the earliest design stage, when the number of people displaced can still influence the alignment — which is the standards' first requirement, and the one skipped most often.

Sources

  1. [1]Performance Standard 5: Land Acquisition and Involuntary Resettlement — International Finance Corporation, 2012.
  2. [2]ESF Guidance Note 5: Land Acquisition, Restrictions on Land Use and Involuntary Resettlement — World Bank, 2018.
  3. [3]Good Practice Handbook: Land Acquisition and Involuntary Resettlement — International Finance Corporation, 2023.
  4. [4]Guidance Note 5: Land Acquisition and Involuntary Resettlement — International Finance Corporation, 2012.

Olule Solomon

Lead Consultant, ValueSpace

Olule Solomon is Lead Consultant at ValueSpace, where he works on land acquisition and resettlement systems for donor-financed infrastructure in East Africa. He writes about the practical gap between what the safeguard standards require and what a project can actually evidence at completion audit.

Related reading

Free entitlement matrix template

15 loss categories, eligibility split by tenure, valuation basis and the PS-5 provision behind every row. CSV, no registration wall.

Get the template →

The software behind this

SmartLARMS keeps the record this article describes

PAP register, replacement-cost valuations, entitlements, recorded payments reconciled against disbursement files, and grievances — every change attributed and time-stamped, so a completion audit is evidenced rather than reconstructed. Offline-first in the field.