Livelihood restoration: what the evidence shows about outcomes after compensation
Abstract
Safeguard standards require that displaced people's livelihoods be restored, and increasingly that they be improved. This is an obligation expressed as an outcome, which distinguishes it from every other requirement in a resettlement instrument and makes it the only one that cannot be discharged by a transaction record. This paper examines the mechanisms through which cash compensation fails to produce restoration, considers the evidence on programme approaches that perform better, and argues that the weakness of the evidence base is itself the central finding: restoration is rarely measured well enough to establish whether it occurred.
1. An obligation of a different kind
Every other requirement in a resettlement instrument can be discharged by an act: land was acquired, compensation was paid, a meeting was held, a grievance was logged. Livelihood restoration cannot. It asks whether a household's capacity to earn a living is at least what it was, which is a state of the world some years after the acts have concluded.
This has an immediate implication that projects routinely fail to draw: a livelihood restoration plan consisting of activities is not responsive to the obligation. Training delivered and inputs distributed are inputs. Whether income recovered is the question, and answering it requires a baseline of income and its sources, a defined observation period, and a monitoring instrument capable of measuring the same thing twice.
2. Why cash compensation under-performs
Cash is the default instrument and it performs poorly against restoration for reasons that are well documented and largely predictable. A lump sum arriving in a household with no history of managing one is frequently consumed rather than reinvested, and consumption is not irrational where immediate needs are pressing. Replacement land is often unavailable at any price in the vicinity, so a payment calculated to purchase equivalent land purchases nothing equivalent. Where many households are compensated simultaneously in one locality, the resulting demand raises local land prices — sometimes sharply — and the compensation is eroded by a price movement the project itself produced.
Cash also transfers risk. A household that held productive land held an asset generating income indefinitely; a household holding cash holds a wasting asset unless it can be converted back into something productive, and the conversion is precisely what displacement has made difficult.
3. What performs better, and what it costs
The approaches with better evidence behind them share a feature: they extend beyond the transaction. Land-for-land replacement, where feasible, avoids the conversion problem entirely. Phased payment reduces the lump-sum consumption risk. Sustained agricultural extension, rather than one-off training, addresses the fact that a household farming unfamiliar land needs support across seasons rather than a workshop. Access to financial services determines whether a payment can be stored and deployed at all.
Each of these is more expensive than a payment and, more importantly, extends the project's obligation years past construction. That timing conflict is the practical reason restoration programmes are under-resourced: the budget is set when the project is being justified, and the spending falls when the project has moved on.
4. The evidence problem
The most defensible summary of the literature is that outcomes are poor where they have been measured, and that they have not often been measured well. Studies capable of establishing whether restoration occurred require baseline income data of reasonable quality, a follow-up period long enough for effects to appear, and a means of distinguishing project effects from general economic change. Few resettlement programmes generate data meeting those conditions.
This matters beyond scholarship. A project unable to demonstrate restoration is in the same position at completion audit as a project that failed to achieve it. The evidentiary requirement and the substantive one converge, and the instrument that satisfies both is the same: a baseline that measures income and its sources rather than assets alone, and monitoring that returns to the same households and asks the same questions.
References
- [1]Performance Standard 5: Land Acquisition and Involuntary Resettlement. International Finance Corporation, 2012.
- [2]Guidance Note 5: Land Acquisition and Involuntary Resettlement. International Finance Corporation, 2012.
- [3]Handbook Module 5: Livelihood Restoration and Improvement. International Finance Corporation, 2023.
- [4]Good Practice Handbook: Land Acquisition and Involuntary Resettlement. International Finance Corporation, 2023.
- [5]Evicted and Abandoned: The World Bank's Broken Promise to the Poor. International Consortium of Investigative Journalists, 2015.