Skip to content
Back to search results

Livelihood Diversification Analysis

Introduced by
Frank Ellis; Christopher Barrett, Thomas Reardon & Patrick Webb
Year
1998
Type
Quantitative and analytical method for studying livelihood portfolios
Family
Livelihoods analysis
Font size

What does it do?

Livelihood diversification analysis studies how rural households spread their activities and income across multiple sources rather than relying on a single occupation or crop.

Developed conceptually by Frank Ellis and refined empirically by Christopher Barrett, Thomas Reardon, and Patrick Webb, it combines the enumeration and classification of household income activities with quantitative measures of diversity — the number of income sources, the share of non-farm income, and concentration indices such as the Herfindahl or Simpson index — to characterise livelihood portfolios and distinguish diversification driven by distress from that driven by opportunity.

The intuition behind the method

Most rural households do not earn their living from one activity. A single family may farm, herd, run a small trade, send a member to work in town, and receive remittances, weaving these strands into a portfolio that spreads risk and smooths income across seasons and shocks. Diversification analysis quantifies that portfolio. It counts the strands, measures how evenly income is spread among them, and tracks the weight of non-farm earnings. A household whose income is concentrated in one source is exposed; one whose income is spread is, in principle, more resilient. But spreading can mean two very different things — a poor household scrambling to survive after a failed harvest ('push'), or a better-off household investing in profitable new opportunities ('pull') — and the analysis aims to tell them apart.

History

The analysis of rural livelihood diversification crystallised in the late 1990s as researchers documented that rural African and Asian households derive large and rising shares of income from non-farm sources, challenging the image of the purely agricultural peasant. Frank Ellis's 1998 article in the Journal of Development Studies offered an influential conceptual treatment of why and how households diversify, linking it to assets, risk, and the sustainable livelihoods framework. Christopher Barrett, Thomas Reardon, and Patrick Webb's 2001 Food Policy paper sharpened the concepts, dynamics, and policy implications of non-farm diversification in rural Africa, and the field grew into a large empirical literature using survey data and diversity indices.

When to use it

Use livelihood diversification analysis when you have household-level income data and want to characterise how rural households combine activities, assess their exposure and resilience, or track the structural transformation of a rural economy through the rising share of non-farm income. It is well suited to poverty, food-security, and rural-development diagnostics and to evaluating how households respond to shocks. It is less useful where income data are unreliable or cover only part of the year, where the farm/non-farm motive cannot be interpreted without complementary qualitative insight, or where a single index is treated as a sufficient summary of a complex, dynamic strategy.

Limitations and considerations

  • Depends on accurate, full-year income data, which is hard to collect and prone to recall and seasonality error.
  • A single diversification index conceals the composition and quality of activities and can equate very different portfolios.
  • The same index value can reflect either distress or accumulation, so motive cannot be read from the number alone.
  • Classifying activities into farm/non-farm and capturing informal, in-kind, and joint activities is often ambiguous.

Source conditions do not replace independent scientific review or validation by an analysis engine.

Strengths

  • Reduces a complex livelihood portfolio to interpretable measures (source counts, non-farm share, concentration indices).
  • Distinguishes distress-driven ('push') from opportunity-driven ('pull') diversification when combined with wealth and shock data.
  • Links household behaviour to resilience, risk management, and the structural transformation of rural economies.
  • Built on standard household survey data, allowing comparison across households, regions, and time.

Applications

  • Diagnosing rural poverty and resilience through the composition of household income.
  • Tracking the non-farm income share as an indicator of rural structural transformation.
  • Distinguishing distress from accumulation diversification to target social protection and enterprise support.
  • Analysing how households diversify in response to drought, price, and market shocks.
  • Comparing livelihood strategies across wealth groups, agro-ecological zones, and over time.

How it works

How it works

1

Using household survey data, the analyst enumerates all the income-generating activities a household engages in over a reference period — crop and livestock production, agricultural wage labour, non-farm self-employment, salaried work, migration and remittances, transfers, and natural-resource harvesting — and records the income or value each contributes. Careful enumeration over a full year is essential because many rural activities are seasonal and easily missed in a single-visit survey.

2

The enumerated activities are grouped into analytically meaningful categories, most commonly the farm versus non-farm and on-farm versus off-farm distinctions, and by sector and location, following Barrett, Reardon, and Webb's conceptual scheme. This classification underpins headline measures such as the share of household income derived from non-farm sources, which is widely used to gauge the structural transformation of rural economies.

3

A diversification or concentration index summarises how income is spread across sources. The Herfindahl-Hirschman index is the sum of squared income shares, and its complement, the Simpson index of diversity D = 1 - sum of s_i squared (where s_i is the share of source i in total income), rises toward one as income becomes more evenly spread across many sources and falls toward zero as it concentrates in a single source. Counts of income sources and entropy measures are common complements.

Formula
D = 1 - \sum_{i=1}^{n} s_i^2
4

The measures are interpreted in light of why households diversify, distinguishing 'push' factors — risk reduction, coping with shocks, fragmented markets, and seasonality that force survival diversification — from 'pull' factors — realising the returns to specialisation and complementarities that attract accumulation-oriented diversification. Cross-tabulating diversification indices with wealth, assets, and shocks reveals whether diversification reflects distress or opportunity, with direct implications for policy.

Common pitfalls

  • Measuring diversification over a single season and missing activities concentrated in other parts of the year.
  • Interpreting higher diversification as automatically positive without checking whether it signals distress.
  • Ignoring intra-household and gendered differences in who undertakes which activities and controls the income.
  • Relying on the index alone without the qualitative context needed to read the underlying strategy.

Frequently asked questions

What is the difference between the Herfindahl and Simpson diversification indices?

Both are built from the squared income shares of each source. The Herfindahl-Hirschman index is the sum of squared shares and measures concentration: it is high (near one) when income is concentrated in few sources and low when income is spread widely. The Simpson index of diversity is its complement, D = 1 minus the sum of squared shares, so it rises toward one as income becomes more evenly diversified across many sources. They convey the same information in mirror-image form; analysts simply choose whether to report concentration or diversity.

What do 'push' and 'pull' factors mean in diversification?

'Push' factors drive defensive or distress diversification: households spread activities to reduce risk, cope with shocks, and survive seasonality or missing markets, often because no single activity provides enough. 'Pull' factors drive accumulation-oriented diversification: households add activities to capture profitable opportunities and the returns to specialisation. The same observed diversification can stem from either, so analysts cross-tabulate diversification measures with wealth, assets, and exposure to shocks to infer the dominant motive, which matters greatly for policy.

Why measure the non-farm income share?

The share of household income from non-farm sources is a headline indicator of how rural economies are transforming. A rising non-farm share signals diversification away from agriculture, the growth of rural labour markets and enterprise, and changing exposure to agricultural risk. Barrett, Reardon, and Webb highlighted how large and variable this share is across rural Africa, with implications for poverty, inequality, and the design of agricultural versus broader rural-development policy.

Related methods

Sources

  1. Ellis, F. (1998). Household strategies and rural livelihood diversification. The Journal of Development Studies, 35(1), 1-38.

    Open DOI reference
    Copied.
  2. Barrett, C. B., Reardon, T., & Webb, P. (2001). Nonfarm income diversification and household livelihood strategies in rural Africa: concepts, dynamics, and policy implications. Food Policy, 26(4), 315-331.

    Open DOI reference
    Copied.