Simplifying Living Income

By Ruerd Ruben

The idea of a “living income” is increasingly considered as an important strategy to guarantee that smallholder farmers’ revenues are sufficient to meet their and their families’ basic needs, as well as to put aside some savings, thus being more likely to find their way out of poverty. There is growing acceptance of an international standard for estimating living income benchmarks and an active community of practice to support its implementation. However  measurements are cumbersome and require a lot of resources.

The last few years, living income benchmarks were defined in a broad range of countries and for different commodities, rural and (peri)urban locations and types of farms. This information is conveniently bundled in the ALIGN living-wage-and-income-dataset that includes more than 120 detailed field studies that measure living income benchmarks, although there are still considerable differences between estimates for the same location.

Notwithstanding these valuable efforts, there is increasing awareness that these living income benchmarks could be assessed in a less cumbersome manner, and regularly updated based on open data sources. Poverty line estimates from the World Bank might offer an acceptable proxy for the living income benchmark. Poverty line data is widely available and consistently considers differences in purchasing power between countries. These poverty estimates are computed from household survey data collected from nationally representative samples of households.  

From poverty line to living income benchmark

It is worthwhile to take a closer look at the relationship between poverty lines and living income estimates.  Based on available data it can be analysed whether such a systematic and meaningful relationship exists. This can considerably reduce time investments in field measurement and create more space for developing of serious proposals for reducing the current living income gap.

We therefore assessed the relationship between (family per month) living income with the poverty line (per family/month) for rural areas in 15 sub-Saharan countries for the year 2021 (see Figure 1). On average, living incomes are 50% higher than poverty lines. In some countries (Uganda, South Africa, Nigeria and Ghana) the living income benchmark is twice as high as the World Bank poverty line, whereas in other countries like Kenya and Tanzania differences are small to negligible

The (dotted) regression takes living income as the outcome (dependent variable), and the poverty line as the independent variable. Living income becomes a function of a fixed intercept – reflecting country-level development as determined by resource availability, social and physical infrastructure, urbanization and education, as well as a correlation coefficient that indicates how the poverty line relates to living income – reflecting the return to resources. The regression line shows that observations are fairly equally distributed around the estimated values, and that 61% of the difference are explained by the regression function. Moreover, the explanatory variable (poverty line) is very significant.

The dotted regression line has the following function: Living Income = 82 + 0.95*(Poverty Line).  The graph indicates that the average relationship between poverty line and living income is fairly robust.  But some countries (i.e. Nigeria, Ghana, Uganda) register higher living incomes benchmarks. We would need an upward shift of the intercept with 35-45% to ensure that households in these countries reach their minimum living income benchmark. The downside is that some other countries that have a very low poverty line (i.e. Ethiopia, Rwanda, Mozambique, Madagascar) would then receive 20-30% too much living incomes (compared to their benchmarks). This indicates that more disaggregation is required between different types of countries for reaching an acceptable average living income estimate.

As can be noted from Figure 1, countries with relatively high living income gaps compared to the poverty line (> 25%) are characterized by stronger growth dynamics and their economies also tend to be more market-oriented. These ‘higher-performing’ countries face different challenges to guarantee that domestic incomes satisfy minimum living conditions, since risky investments and access to innovations are required for improving factor productivity of land and labour.  On the other hand, several other ‘lower performing’ countries in Sub-Saharan Africa (SSA) have a smaller gap between poverty and living income, basically because there is still a lot of subsistence production and better access to land and capital markets are critical to poverty reduction.

Different strategies towards living income

In order to reduce heterogeneity – and thus avoid too much variation between countries that respond to the same poverty-income equation – the sample can be divided in two groups that maintain low standard deviations on the key parameters and thus have on average less than 7% differences between what the equation delivers as living income estimate and what the field data effectively register. We identify two different types of country dynamics (see Figure 2): 

a)            Higher performing countries : Living Income = 87 + 1.02*(Poverty Line)  (N=8, R2 = 85%)

Countries with higher rates of economic development, better infrastructure facilities, higher enrolment in education and more coverage of mobile phone networks have more favourable resource endowments (reflected by the higher intercept) and are therefore better able to achieve reductions in poverty with improvements in living incomes (illustrated by the higher coefficient and the steeper curve). Typical countries in this category are Nigeria, Ghana, Ethiopia and Rwanda.

•         Lower performing countries: Living Income = 41.7 + 0.90*(Poverty Line)  (N=7, R2 = 79%)

Countries with lower levels of economic development, infrastructure limitations, less educational performance and lower mobile phone coverage start their development path at a lower income level (shown by the smaller intercept), and need more reduction in poverty to reach a subsequent improvement in living income (reflected in the lower coefficient of the flatter curve). Typical countries in this category are DR Congo, Burkina Faso, Cameroon and Mozambique.

These categories of countries are far more homogeneous with respect to the Poverty Line – Living Income relationship, and therefore the estimated regression function can be used as a suitable approximation for identifying the living income benchmark. This may not only save a lot of time and money that is nowadays invested in living income measurement, but can also be used to better focus on key constraints for reducing the income gap in practice.

How to support living incomes?

Given the structural differences between SSA countries, two specific strategic priorities can be identified:

a) for higher performing countries where initial resource conditions are better guaranteed, reaching living wages will depend especially on the strategies for increasing the responsiveness to poverty reduction by improving the return to resources. This is mostly related to innovations in crop yields, improvements in crop mix or higher wages in (off-farm) work, reflected in a steeper regression function.

 b) for lower performing countries that start with less physical and social infrastructure, opportunities for reaching and improving living income levels are likely to be based on strategies for enhancing access to land and/or opportunities for engagement in off-farm employment that lead to an upward shift (or rise of the intercept) of the regression function.

A better understanding of the relationship between poverty line and living income is an attractive shortcut to identify structural causes of living income gaps and to identify strategic opportunities to support poverty reduction.

Ruerd Ruben is (emeritus) professor impact assessment for food systems at Wageningen University, the Netherlands. He works on smallholders in tropical value chains, the effectiveness of cooperatives and the impact of certification. He was director of the Director Policy & Operations Evaluation at the Netherlands Ministry of Foreign Affairs.

Image: World Bank Photo Collection under a creative commons licence on Flickr

Note: This article gives the views of the author, not the position of the EADI Debating Development Blog or the European Association of Development Research and Training Institutes.