Working Poverty in the WAEMU*. An Analysis Based on the Harmonized Household Living Conditions Surveys (HHLCS)#

This article analyzes the labor-intensive poverty in the West African Economic and Monetary Union (WAEMU) area using data from the Harmonized Household Living Conditions Surveys (HHLCS). Despite high levels of employment, a significant proportion of workers remain in poverty, revealing the structural limitations of labor markets in this economic and monetary area. Working poverty appears to be closely linked to rural areas, informal employment, low human capital and the predominance of subsistence agriculture. Women, young people and workers from large households are particularly at risk. Conversely, education, formal employment and integration into industry or services significantly reduce the risk of poverty in employment. These results underline that the sustainable reduction of worker poverty requires integrated policies combining educational investment, formalization of the economy, productive diversification and strengthening of social protection.

• Working Poverty
• Informal Employment
• Labor Market
• Human Capital
• WAEMU

1. Introduction

Despite economic progress in several countries in Sub-Saharan Africa over the past two decades, with average growth rates exceeding 5% per year, poverty remains a persistent and multifaceted reality. The income poverty rate, measured at a threshold of $1.90/day PPP, declined from 58% in 2000 to 36% in 2019. However, the number of poor people continued to rise from 433 million people living on less than $2.15 a day in 2018 to 464 million in 2024 ([3]). Among the most worrying expressions of this poverty is working poverty, which refers to the paradoxical condition of individuals engaged in economic activity but unable to meet their basic needs or rise above the monetary poverty threshold: according to the Harmonized Household Living Conditions Surveys (HHLCS), working poverty remains a prominent feature of the region, with rates reaching 40% to 60% of the working population in some countries.

Indeed, while participation in the labor market, at 71.3% in 2024 according to the World Bank, is important in sub-Saharan Africa, the persistence of a high level of poverty questions the effectiveness of available jobs to lift people out of economic insecurity on a sustainable basis. In a region of sub-Saharan Africa where employment does not necessarily guarantee sufficient income or minimum social security, how can the high prevalence of working poverty be explained despite apparent high economic activity?

What are the individual, familial and structural factors that condition this form of poverty, and how do they vary by country, gender, geographical area or sector of activity?

The issue remains relevant in a context where informal employment dominates the economy and productivity remains low, making working poverty a key indicator for assessing the performance of public policies on employment, social protection and human development. The analysis of this phenomenon therefore requires reliable and comparable data over time and between countries. With this in mind, the Harmonized Household Living Conditions Surveys (HLSCS), implemented in West African Economic and Monetary Union (WAEMU) countries with the support of the World Bank and other partners, provide a valuable analytical framework. They document the reality of households in terms of income, employment, access to basic social services, and other essential dimensions of well-being, as well as identify avenues for reflection for more inclusive and effective policies.

The central hypothesis of this research is that working poverty in the WAEMU area is mainly fueled by the prevalence of low-productivity informal employment, combined with limited access to public services and the absence of inclusive social protection policies. The study also posits that certain population groups—including women, youth and rural people—are disproportionately exposed to this economic vulnerability, due to structural barriers to access to decent work.

The objective of the study is, therefore, to analyze the determinants, distribution and characteristics of working poverty in the WAEMU area from data from the Harmonized Household Living Conditions Surveys (HHLCS). The article measures and profiles, first of all, the working poverty in the WAEMU member countries, then examines its determinants before formulating public policy recommendations.

2. Methodological Framework

2.1. Theoretical and Conceptual Framework

Working poverty is a concept that challenges the idea that employment is automatically a vector out of poverty. This concept builds on the work of [13] and [26], who expanded the concept of poverty by integrating not only income but also the capacities, living conditions and dignity of workers. In this context, working poverty is understood not only as a monetary deficit, but also as a decent work deficit, as defined by [16]: productive employment, with social protection, rights at work and the existence of social dialogue. This phenomenon of working poverty is particularly prevalent in developing economies, where a large proportion of the working population works in low-productivity sectors or in precarious employment ([17]). In this sense, it goes beyond the simple measurement of income, as shown by [27] work on abilities and opportunities and Nussbaum’s work on abilities. Indeed, the authors’ work emphasizes that poverty is not just a lack of money, but also the inability to lead a satisfactory life, access essential services, and participate fully in society.

[27] introduces the idea that poverty is linked to a deprivation of capacities and opportunities that prevent individuals from leading a life that they have reasons to value. This multidimensional approach to poverty includes three dimensions: (i) an economic dimension¹ first, because low incomes are the main characteristic of the working poor; (ii) a social dimension, then, in view of precarious working conditions, lack of social protection and vulnerability to economic shocks ([25]); and (iii) institutional, finally, because of the lack of access to quality services such as health, education, and social security.

From a theoretical perspective, working poverty in the WAEMU can be analyzed through several currents of economic and sociological interpretation of the labor market. The labor market segmentation theory ([10]) provides a relevant framework for understanding the structural duality that pits a primary segment—characterized by stability, protection and decent wages—against a secondary segment, dominated by insecurity, informality and vulnerability. This dichotomy is starkly apparent in Sub-Saharan Africa, where the majority of workers are confined to informal employment. The informal employment model ([23]), an empirical extension of this segmentation, highlights the logics of exclusion from social protection mechanisms and access to capital, which limit the productivity of the working poor ([17]). Moreover, feminist approaches to development, including those of Kabeer (cited by [6]) and [12], provide insight into how social norms and gender inequalities increase women’s exposure to working poverty. Finally, a reading inspired by the theory of capabilities ([27]) sheds light on the deficit of real choices suffered by these workers, prisoners of structural conditions that reduce their freedom to access decent employment. Thus, hard-working poverty results not only from a lack of work, but from a deficit of qualitative opportunities, embedded in unequal economic, social and institutional structures.

2.2. Determinants of Working Poverty

The conceptual framework of this study is based on the hypothesis that several factors determine working poverty in Sub-Saharan Africa, particularly in the WAEMU area. These factors are grouped into three main dimensions: individual, professional, and contextual.

In terms of individual characteristics, education is a protective factor against hard-working poverty, in that a higher level of education is associated with a lower probability of being poor, as educated individuals have access to more stable and better-paid jobs ([19]). In the same vein, age and work experience play a central role. In fact, young workers, often with less experience and qualifications, are more likely to find themselves in precarious jobs and, as a result, in working poverty. However, with age and experience, workers can access higher-paying jobs ([9]). This situation is more marked by sex. Women who are more likely to work in low-paid informal sectors are more exposed to working poverty: according to [21], gender inequalities in the labor market largely explain the disparities in working poverty between men and women.

Occupational factors, highlighted by working conditions and type of employment, are also key determinants of working poverty. Thus, the sector in which a person works directly influences their level of remuneration and their chances of escaping poverty. The agricultural sector, for example, is often associated with low and irregular incomes, increasing the risk of working poverty ([19]). Moreover, the nature of employment, formal or informal, is equally important. Indeed, workers in the formal sector generally benefit from better pay and social protections ([8]). By contrast, informal workers have lower incomes, limited social protection, and greater vulnerability to economic shocks. In the same sense, the type of contract conditions the entry or exit of workers from poverty, as recalled by [17]: temporary and precarious contracts, seasonal or fixed-term contracts are often linked to unstable working conditions, reducing the prospects of escaping poverty.

Contextual factors dominated by socio-economic and institutional conditions in sub-Saharan African countries strongly influence working poverty. A distinction is made between access to social services such as health, education, social security and basic infrastructure (electricity, drinking water), which play a key role in reducing working poverty. Regions with better infrastructure and access to social services tend to have lower rates of working poverty ([25]). On the other hand, as pointed out ([24]), workers living in rural areas are more exposed to working poverty, due to low employment opportunities and dependence on subsistence farming, which generates low and irregular incomes.

2.3. Conceptual Model: General Specification and Selected Variables

In the context of this study, the objective is to identify the determinants of working poverty among employed workers in the WAEMU area. The analysis therefore seeks to explain the probability that an active employed individual is in a situation of poverty, according to a set of individual, professional and household factors.

Working poverty = f (Individual Characteristics, Professional Characteristics, Characteristics of household)

This problem calls for the use of a binary econometric model, the status of working poverty being a dichotomous variable: working poor or working non-poor:

Y i = 1 if the individual i , although active and employed, belongs to a household living below the national income poverty line Y i = 0 otherwise.

This formulation requires the use of a logistic regression model (logit) or probit, two classical approaches for estimating probabilities when the dependent variable is dichotomous. The logit model is preferred. This choice of the logit model is based on the binary nature of the status of working poverty which takes two modalities, excluding classical linear regression models (OLS), unsuitable for this type of data. This choice is also based on robustness to normality assumptions, in that the logit model is based on a logistic function, which is more flexible and has a better interpretability of coefficients in terms of odds ratios, insofar as these coefficients can be transformed into marginal effects to quantify the marginal impact of a factor on the probability of poverty. Finally, the logit model is suitable for large databases, such as HHLCSs, which have thousands of observations per country, allowing a robust estimation of the logit model.

Technically, the model is based on probability modeling P i that the individual i or a poor worker, conditional on the explanatory variables X i , according to the following formula:

P ( Y i = 1 | X i ) = e X i β 1 + e X i β

where:

Y i is the dependent variable (working poverty status: 1 = poor, 0 = non-poor); X i is the vector of the explanatory variables for the individual i ; β is the vector of the coefficients to be estimated.

This model can also be expressed in terms of the log-odds, i.e., the logarithm of the ratio of the probability of being poor to the probability of not being poor:

log ( P i 1 − P i ) = β 0 + β 1 X 1 i + β 2 X 2 i + ⋯ + β k X k i

where:

β 0 is the ordinate at the origin (constant); X 1 i , X 2 i , ⋯ , X k i are the different explanatory variables; β 1 , β 2 , ⋯ , β k are the associated coefficients.

The estimation is made by the maximum likelihood, making it possible to derive the probability that an individual is in a situation of working poverty according to their characteristics.

In this context, the explanatory variables selected in the model are as follows (Table 1):

Table 1. Explanatory variables of the model.

Table 1

Dimension	Variables	Modalities
Characteristics of household	Environment of residence	Urban/Rural
	Typology of household	Single Households, Single Parent Female, Single Parent Male, Single Household, Polygamous Household
	Household size
	Number of persons in employment
Individual characteristics	Age
	Sex	Male/Female
	Marital status	Single or common-law/Married monogamous/Married polygamous/Widowed/Divorced/Separated
	Level of education	No level/Primary/Secondary/Superior
Professional characteristics	Socio-professional category	Managers and related persons/Clerical and service employees/Skilled agricultural and manual workers/Unskilled workers, armed forces, other or unclassified
	Formality of the main job	Informal/Formal
	Status in the main job	Employee/Employer/Self-employed/Family or non-classified caregiver
	Institutional sector	Public/Private
	Branch of activity	Agriculture/Industry/Services
	Number of hours worked per week

Source: the authors.

2.4. Country Data and Context

The analysis is based on data from the Harmonized Household Living Conditions Surveys (HHLCS), conducted between 2018 and 2020 in eight WAEMU countries: Benin, Burkina Faso, Côte d’Ivoire, Mali, Niger, Nigeria, Senegal and Togo. These surveys, coordinated by WAEMU, the World Bank and national statistical offices, provide a comparable database across countries, with a rich range of modules covering sociodemographic characteristics, employment, living conditions, household expenditure and income.

The scope of the study is restricted to employed persons aged 15 years and over², to examine poverty among workers, whether employed, self-employed or caregivers.

Table 2.Main country indicators.

Table 2

Country	Poverty rate	Poverty line	Rate of urbanization 3	Share of employment in (%) 4			Share of employment in (%) 5
				Male	Female	Informal	Agriculture	Industry	Services
Burkina Faso	43.2	247861.50	32.0	51.7	39.4	96.0	31.0	23.0	46.0
Bénin	36.2	287187.30	58.0	76.6	73.6	96.3	28.0	22.0	50.0
Guinée-Bissau	55.2	298083.50	38.0	67.8	55.0	94.8	50.0	10.0	40.0
Mali	43.8	277172.50	34.0	76.2	40.5	95.5	68.0	10.0	22.0
Sénégal	37.5	369665.50	53.0	67.0	37.9	94.8	22.0	22.0	56.0
Côte d'Ivoire	37.5	369516.44	53.0	73.9	58.1	92.1	45.0	11.0	44.0
Togo	43.8	295182.06	43.4	71.4	67.6	95.6	30.0	20.0	50.0
Niger	41.2	195807.40	18.0	86.9	73.2	98.5	71.0	07.0	22.0

Sources: OCDE (2025); [2].

The review of the main socio-economic indicators of the WAEMU countries (Table 2) highlights significant disparities, while highlighting common structural trends that characterize the economies of the region.

Poverty rates, which remain high overall, reflect the persistence of household vulnerability. Guinea-Bissau, with 55.2 per cent, had the most critical level, followed by Mali and Togo (43.8 per cent) and Burkina Faso (43.2 per cent). In contrast, Benin (36.2%), Senegal (37.5%) and Côte d’Ivoire (37.5%) have slightly lower rates but are equally worrying. These differences must be interpreted in the light of poverty lines that vary greatly between countries, from 195,000 CFA francs in Niger to nearly 370,000 CFA francs in Senegal and Côte d’Ivoire, reflecting differences in the cost of living and differentiated levels of consumption.

Urbanization dynamics accentuate these contrasts: some countries, such as Niger (18%) and Burkina Faso (32%), remain predominantly rural, while Benin (58%), Senegal (53%) and Côte d’Ivoire (53%) are experiencing more advanced urban processes, generating both opportunities for economic modernization and strong pressures on social and productive infrastructure.

An examination of employment rates reveals a paradoxical situation: high overall employment rates are accompanied by persistent and often marked gender disparities. In Mali, 76.2 per cent of men are active compared to only 40.5 per cent of women, and in Burkina Faso the gap remains above 12 points, reflecting a clear underutilization of women’s potential in the economic sphere. Benin is a relative exception, with almost equal levels of employment for men and women.

Moreover, the labor market is still characterized by the predominance of informality, over 90% in all countries, even reaching 98.5% in Niger. This pervasiveness of the informal sector limits fiscal space, undermines social protection systems, and reflects the low absorption capacity of the modern sector.

The sectoral structure of employment reveals a low level of productive diversification: agriculture still accounts for 71% of employment in Niger and 68% in Mali, while services are the predominant sector in the most urbanized countries, with 56% in Senegal and 50% in Benin and Togo. Industry, on the other hand, remains marginal, rarely exceeding 15% of total employment, which testifies to an embryonic industrialization process.

In total, these findings suggest a triple vulnerability that combines the persistence of widespread poverty, the omnipresence of informality and excessive dependence on agriculture. These dynamics illustrate the need to jointly analyze the demographic, social and economic dimensions, while recalling that the structural transformation of the WAEMU area is an essential condition for meeting the Sustainable Development Goals and the ambitions of Agenda 2063, including decent work.

3. Results and Discussion

3.1. Descriptive Statistics

In the WAEMU area, working poverty affects almost one in three workers, the situation being more marked in Guinea-Bissau where one in two workers lives below the poverty line. A comparative analysis of the phenomenon in the area (Table 3) reveals significant disparities between countries. While overall rates remain high, exceeding 32 per cent in the majority of cases, there are particularly sharp peaks in Guinea-Bissau (50.0 per cent) and Mali (42.3 per cent). In contrast, countries such as Benin (32.7%), Côte d’Ivoire (32.8%) and Senegal (32.3%) have a relatively lower proportion of working poor. These contrasts reflect the strong heterogeneity of economic structures within the area, between less productive rural economies and more diversified economies.

The socio-professional categories also point to a number of discrepancies. Thus, the gender analysis reveals a slightly higher vulnerability of women, with the exception of Senegal where poverty affects men more. These findings reveal a general trend of increased vulnerability among women, confirming the persistence of gender inequalities in access to productive employment. However, the case of Senegal, where men are slightly more affected, illustrates the complexity of local dynamics and the need for in-depth contextual analyses. These gaps are consistent with the literature that highlights the concentration of women in subsistence agriculture and low-paid informal services.

Age also appears to be a discriminating factor, with young people aged 15 to 24 being the most affected segment, with a proportion close to 58.0% in Guinea-Bissau and remaining above 38% in almost all other countries. This situation is part of a dynamic already noted by [14], according to which employment growth in sub-Saharan Africa benefits young people little, often confined to precarious and informal activities. Workers aged 25 - 54 have lower but equally significant levels, while older workers, notably in Burkina Faso and Mali, remain highly exposed; this trend suggests persistent vulnerability over the working life cycle.

For its part, education is decisive in aversion to poverty among workers: lack of schooling is systematically associated with a high incidence of working poverty, greater than or equal to 45% in one country in two of the economic and monetary zone; on the other hand, the possession of a secondary degree significantly reduces this risk and the higher level leads to residual rates, close to zero in Niger and Burkina Faso, and less than 6% elsewhere, with the exception of Guinea-Bissau where 11.4% of workers with higher education are poor. These findings are a strong reminder of the central role of human capital investment in poverty reduction, converging with the work of [15], which emphasizes the importance of human capital in reducing poverty.

The place of residence is another divisive dimension. Rural areas are the main concentration of the working poor, with sometimes extreme gaps, such as in Côte d’Ivoire where working poverty reaches 47.8% in rural areas compared to only 17.7% in urban areas, or in Burkina Faso where it rises to 47.5% against 12.5%. To these spatial inequalities is added the impact of the formality of employment: informality, which is a constant throughout the region, is accompanied by rates above 33%, compared to less than 14% for formal workers, illustrating the structural link between informality and poverty already widely documented in the literature ([18]; [30]). Occupational status accentuates these differences, with family workers, who dominate in agriculture, representing the most exposed category, with rates above 55% in five of the eight countries of the zone, while for employees and employers, who are “better protected”, there are still pockets of vulnerability.

Finally, the sectoral distribution confirms the importance of agriculture in the reproduction of working poverty. Not surprisingly, the rates of working poverty in Guinea-Bissau exceed 60% and are close to or exceed 50% in several WAEMU countries. On the other hand, industry and, even more so, services, which are more developed in urban areas, appear to be relatively protective sectors, with rates generally below 25% for services.

In total, the results, highlighted with data from the HHLCSs, underline that working poverty in the WAEMU is the product of a structural combination between rural areas, low education, informality and dependence on subsistence agriculture. The findings are consistent with recent research findings that economic growth, in the absence of structural reforms and active policies, is not sufficient to reduce working poverty. The reduction of the phenomenon therefore calls for

Table 3.Distribution of the working poor by individual characteristics (sex, age group, level of education), household typology, place of residence and employment characteristics (formality, status, sector and branch)

Table 3

Indicators	Bénin	Burkina Faso	Côte d’Ivoire	Guinée - Bissau	Mali	Niger	Sénégal	Togo
Working poverty rate	32.7	37.9	32.8	50.0	42.3	37.6	32.3	37.7
Sex of the individual in employment
Male	32.2	35.6	30.7	47.9	41.6	36.3	34.3	36.1
Female	33.2	40.4	35.2	52.1	43.5	38.8	29.5	39.1
Age ranges
15 - 24 years	38.0	42.3	36.1	57.8	42.9	38.2	39.9	38.9
25 - 54 years	32.0	35.2	31.8	48.4	42.5	39.7	30.3	37.9
55 - 64 years	26.2	43.1	34.3	44.6	40.6	28.4	29.4	33.7
65 years and over	28.1	43.5	33.2	41.6	40.0	28.0	30.7	40.0
Level of education
None	39.9	45.0	42.0	60.9	49.7	40.9	41.8	52.5
Primairy	28.4	30.2	30.5	54.5	37.5	33.1	22.5	41.9
Secondary	20.3	24.3	18.7	39.2	23.1	24.3	18.9	27.2
Superior	3.5	0.3	2.9	11.4	5.2	0.0	4.2	5.0
Environment of residence
Urban	27.8	12.5	17.7	26.0	19.9	13.9	17.0	19.8
Rural	36.5	47.5	47.8	63.3	48.2	41.3	48.2	51.7
Formality of employment of the main job
Informal	33.7	39.8	35.1	52.3	44.0	38.1	33.8	39.1
Formal	6.5	2.1	5.9	13.6	7.6	4.6	6.4	11.9
Status in employment
Employee	24.3	19.8	19.0	30.0	25.6	24.1	21.8	23.1
Employers	28.1	9.9	14.2	22.3	16.2	25.8	12.9	12.9
Self-employed workers	30.7	36.6	35.0	51.6	41.9	36.7	33.8	39.3
Family or non-classified helpers	46.6	59.6	49.3	70.0	57.5	46.1	57.4	55.7
Institutional sector
Public	8.3	1.4	4.1	15.8	6.8	4.3	10.1	10.7
Private	33.3	39.1	33.8	52.4	43.4	38.2	33.3	38.8
Branch of activity
Agriculture	45.0	45.6	49.6	63.6	54.1	42.6	56.3	56.6
Industry	29.6	26.3	20.8	35.5	27.4	30.3	25.0	32.8
Services	20.9	13.5	17.2	24.8	20.1	20.8	19.5	22.3
N	1,761,429	2,250,566	3,782,201	351,192	2,794,024	3,426,050	1,714,083	1,258,603

Source: based on data from HHLCS in the WAEMU area.

ambitious educational policies, a sustained effort to formalize employment, the development of non-agricultural productive sectors and special attention to young people and women, whose vulnerability is systematically accentuated.

3.2. Econometric Analysis

Table 4 presents the results of the econometric analysis. Logit models have good explanatory capacity, with correct classification rates ranging from 70% to 74% and McFadden R² values ranging from 14% to 27%, levels considered satisfactory for this type of model, confirming the robustness of the estimates and the relevance of the explanatory variables mobilized.

The data reveal, first of all, a decisive role of the place of residence. All other things being equal, in almost all WAEMU countries, living in rural areas significantly increases the likelihood of being poor despite having a job. The effect is particularly marked in Niger (Odds Ratio = 3.47; p < 0.01) and Burkina Faso (Odds Ratio = 3.09; p < 0.01), while it reaches about 2.46 in Côte d’Ivoire and remains significant in Guinea-Bissau, Senegal and Togo. Benin is an exception: rurality appears to be slightly protective, although the effect remains statistically significant (Odds Ratio = 0.90; p < 0.05), reflecting structural specificities of the Beninese rural labor market. The Beninese exception could be explained by a relatively more structured organization of certain commercial agricultural sectors, notably cotton and cashew, favoring better integration into markets and a relative security of rural incomes, as [20] or the [29] believe. This singularity could also reflect a greater diversification of income sources through multiactivity strategies, which reduce exclusive dependence on agriculture ([11]; [4]). Finally, the role of internal and migratory transfers, as well as the impact of certain targeted agricultural and social policies, contribute to strengthening the resilience of rural households ([28]). These elements suggest that not all rural areas are equally vulnerable and that, in some contexts, agriculture that is better integrated into markets and supported by complementary activities can be a relative factor in protecting against working poverty.

Beyond the territorial effect, the family structure is a major factor in the vulnerability of workers. Compared to male-headed lone-parent households, female lone-parent households have a significant increase in risk in several countries, including Benin, Burkina Faso and Côte d’Ivoire, where the probability of working poverty is multiplied by factors between 1.7 and 2.4. On the other hand, this effect is not significant in Guinea-Bissau, Togo and Senegal, the latter being distinguished by a coefficient lower than unity (Odds Ratio = 0.835), suggesting a relative protection of Senegalese female lone-parent households. For workers in polygamous households, their profiles differ from country to country. They are associated with an increased risk in Guinea-Bissau (Odds Ratio = 1.508; p < 0.01), while they appear protective in Mali (Odds Ratio = 0.460; p < 0.01), Niger (Odds Ratio = 0.287; p < 0.01) and Togo (Odds Ratio = 0.458; p < 0.01), reflecting the importance of intra-family solidarity mechanisms in certain contexts. Moreover, household size has an aggravating effect everywhere: each additional person significantly increases the probability of being poor despite having a job. On the other hand, the number of employed persons in the household plays a systematically protective role in all the countries studied.

The results then confirm the highly protective effect of education. Primary schooling already reduces risk by 20% - 35%, secondary education halves probability, and higher education is emerging as a virtual bulwark against hard-working poverty. In some countries of the zone, the probability of working poverty is reduced by more than 85% with the increase in the level of education, such as in Burkina Faso (Odds Ratio = 0.01), Benin (Odds Ratio = 0.11) or Togo (Odds Ratio = 0.13). These results highlight the structuring role of human capital in access to decent and remunerative jobs.

Moreover, the characteristics of employment reinforce this reading. Membership in the formal sector reduces the risk of working poverty everywhere, with a particularly marked effect in Burkina Faso (Odds Ratio = 0.27; p < 0.01) and Benin (Odds Ratio = 0.38; p < 0.01). Employers appear to be better protected overall, while caregivers are the most vulnerable category, notably in Senegal, Niger, Guinea-Bissau and Mali (Odds Ratio = 1.77; p < 0.01). Self-employed workers have a more nuanced profile; they are sometimes significantly protected. As a follow-up to this analysis in relation to employment characteristics, industries also influence the probability of being poor despite having a job. Indeed, in most countries, working in industry or services reduces the risk of working poverty compared to agriculture.

With regard to the number of hours worked per week, the results indicate that this variable has an overall small and heterogeneous effect on the probability of being a poor worker in the WAEMU. When it is statistically significant, particularly in Benin, Burkina Faso and Togo, its impact remains marginal, with odds ratios very close to unity; this result suggests that an increase in hourly volume only minimizes the risk of poverty. In several countries, including Côte d’Ivoire, Guinea-Bissau, Niger and Senegal, the variable is not significant, while in Mali it appears even slightly aggravating. These results highlight a phenomenon of poverty despite a strong commitment to work, characteristic of contexts marked by informality and low productivity ([13]; [17]). This finding suggests that the volume of hours worked is an imperfect indicator of economic well-being, confirming the thesis that the reduction of working poverty depends more on the quality and productivity of employment than on the intensity of work.

Table 4.Determinants of working poverty in WAEMU.

Table 4

	Bénin		Burkina Faso		Côte d’Ivoire		Guinée - Bissau		Mali		Niger		Sénégal		Togo
Poverty Status	Odds	P- val	Odds	P- val	Odds	P- val	Odds	P- val	Odds	P- val	Odds	P- val	Odds	P- val	Odds	P- val
Constant	0.528	0.030	0.081	0.000	0.069	0.000	0.235	0.000	0.301	0.004	0.024	0.000	0.198	0.000	0.324	0.001
Residence (Reference: Urban)
Rural	0.904	0.015	3.091	0.000	2.456	0.000	2.169	0.000	1.661	0.000	3.472	0.000	2.083	0.000	2.240	0.000
Typologie du ménage (Reference: Monoparental Homme)
Individual households	0.205	0.000	0.173	0.011	0.166	0.000	0.017	0.000	1		0.193	0.034	1		0.116	0.000
Single Parent Woman	1.822	0.000	1.713	0.024	2.427	0.000	1.071	0.488	0.446	0.001	0.595	0.055	0.835	0.261	0.865	0.329
Monogamous household	1.234	0.081	0.706	0.111	1.729	0.000	1.164	0.087	0.588	0.010	0.624	0.077	1.134	0.391	0.704	0.021
Polygamous household	1.279	0.084	0.659	0.073	1.284	0.096	1.508	0.000	0.460	0.000	0.287	0.000	1.104	0.515	0.458	0.000
Household size	1.250	0.000	1.130	0.000	1.398	0.000	1.271	0.000	1.251	0.000	1.405	0.000	1.160	0.000	1.408	0.000
Number of persons in employment	0.865	0.000	0.901	0.000	0.717	0.000	0.872	0.000	0.754	0.000	0.813	0.000	0.890	0.000	0.699	0.000
Age	0.985	0.000	1.002	0.359	0.991	0.000	0.986	0.000	0.998	0.394	0.996	0.076	0.992	0.000	0.993	0.003
Sex (Reference: Male)
Female	0.852	0.001	0.838	0.015	0.937	0.169	0.790	0.000	1.121	0.110	0.826	0.006	0.844	0.002	0.921	0.215
Marital status (Reference: single or common-law)
Married monogamous	1.276	0.001	1.299	0.030	1.134	0.034	1.357	0.000	0.895	0.228	1.216	0.084	1.056	0.454	1.201	0.060
Married polygamous	1.297	0.015	1.449	0.023	1.595	0.000	1.533	0.000	1.319	0.024	1.524	0.008	1.023	0.818	1.152	0.387
Widower	1.334	0.020	1.156	0.488	1.457	0.003	1.907	0.000	1.097	0.684	1.103	0.625	1.379	0.024	1.234	0.151
Divorced	0.902	0.589	0.792	0.636	0.942	0.831	1.112	0.769	0.589	0.127	1.361	0.210	0.676	0.020	1.091	0.768
Separated	0.873	0.474	0.618	0.389	1.068	0.790	1.150	0.526	0.913	0.888	0.998	0.999	1.788	0.421	1.908	0.000
Level of education (Reference: No level)
Primary	0.684	0.000	0.647	0.000	0.731	0.000	0.801	0.000	0.778	0.001	0.665	0.000	0.615	0.000	0.777	0.000
Secondary	0.473	0.000	0.531	0.000	0.515	0.000	0.649	0.000	0.496	0.000	0.485	0.000	0.478	0.000	0.496	0.000
Superior	0.112	0.000	0.010	0.000	0.159	0.000	0.295	0.000	0.198	0.000	1		0.226	0.000	0.134	0.000
Socio-professional category (Reference: Managers and similar)
Office and service employees	1.370	0.037	2.052	0.069	1.239	0.398	0.790	0.119	0.824	0.397	1.670	0.110	1.134	0.445	1.493	0.027
Skilled agricultural and manual workers	1.353	0.048	1.605	0.246	1.210	0.459	0.819	0.193	1.007	0.979	2.161	0.022	1.551	0.011	1.825	0.001
Other workers	2.066	0.000	2.144	0.058	1.330	0.281	0.833	0.237	1.027	0.920	4.265	0.000	1.802	0.001	1.945	0.001
Formality of the main job (Reference: Informal)
Formal	0.379	0.000	0.268	0.003	0.465	0.000	0.540	0.000	0.426	0.000	0.455	0.174	0.408	0.000	0.580	0.003
Employment status (Reference: Employee)
Employers	0.985	0.934	0.269	0.043	0.527	0.004	0.517	0.063	0.502	0.074	0.619	0.010	0.389	0.001	0.265	0.000
Self-employed workers	0.823	0.004	0.748	0.004	0.926	0.243	1.125	0.099	1.351	0.000	1.116	0.402	1.025	0.699	0.711	0.000
Caregivers or not elsewhere classified	0.751	0.001	0.791	0.322	0.936	0.366	1.542	0.000	1.775	0.000	1.687	0.000	1.189	0.042	0.760	0.015
Institutional sector (Reference: Public)
Private	1.122	0.628	2.504	0.085	2.058	0.010	0.994	0.965	1.094	0.767	1.464	0.398	0.935	0.687	1.590	0.055
Branch of activity (Reference: Agriculture)
Industry	0.590	0.000	0.648	0.002	0.431	0.000	0.714	0.000	0.515	0.000	0.923	0.528	0.494	0.000	0.665	0.000
Services	0.440	0.000	0.378	0.000	0.404	0.000	0.433	0.000	0.418	0.000	0.584	0.000	0.468	0.000	0.536	0.000
Number of hours worked per week	0.991	0.000	0.991	0.000	0.998	0.070	1.000	0.743	1.003	0.040	1.000	0.982	1.000	0.988	0.995	0.001
N	5 386 634		5 938 168		11 531 102		702 384		6 605 258		9 111 836		5 306 758		3 338 469
R² Mc Fadden Ajusté	13.86202553		17.77195969		24.08992839		27.01769847		18.99554553		14.91914446		22.54588471		21.97133778
Sensitivity (Pr(+|D))	34.22795991		54.82209738		59.49652268		75.22998296		55.50075059		43.68817711		48.47574066		62.31941421
Specificity (Pr(-|~D))	88.15269486		82.00729927		80.34889333		72.57665678		83.34162108		85.43750000		84.50111857		77.31016924
Correctly classified	70.29808774		72.71053474		71.73438006		73.95962825		73.12504919		72.63469594		70.64199978		70.85251492

Source: based on data from HHLCS in the WAEMU area.

3.3. Discussion

The results highlight the decisive role of rurality in working poverty, largely confirming the findings of the literature. In sub-Saharan Africa, the [18] and the [2] recall that subsistence agriculture and informal rural employment concentrate the majority of the working poor. The fact that Niger, Burkina Faso and Côte d’Ivoire have risks multiplied by more than two or three for rural workers illustrates this correlation between low agricultural productivity, informality and poverty. The Beninese exception, where rurality seems slightly protective, is questionable: it could be explained by specific dynamics of social transfers, agricultural diversification or more profitable rural activities, which require additional research.

Family structure also appears to be a central factor in the vulnerability of workers in the WAEMU area, which corroborates the observations of [7] regarding the impact of social gender relations on African labor markets. The extreme vulnerability of polygamous households in Guinea-Bissau reflects the effect of negative economies of scale on the standard of living of workers: the proliferation of dependents weighs on disposable income. Similarly, workers from single-parent households headed by women have a much higher probability of poverty, except in Senegal where the opposite effect suggests the existence of compensatory mechanisms (social solidarity, urban opportunities, role of diasporas). These results confirm that working poverty is not only economic, but also structurally linked to family and gender dynamics.

The protective role of education is perfectly aligned with the work of [15]. In all WAEMU countries, primary education reduces the risk of economic vulnerability of workers, secondary education halves the probability, and higher education is a near absolute screen against working poverty. These results clearly show how human capital is a lever for mobility and for lifting people out of poverty, reminding us that economic growth alone is not enough to create decent jobs: it is educational investment and its articulation with access to formal employment that create the conditions for inclusive growth. Indeed, the residual existence of working poor even among higher education graduates, notably in Guinea-Bissau, underlines the limitations of education when the structure of labor demand remains constrained, confirming that educational investment must be accompanied by productive transformations.

The characteristics of employment reinforce this analysis. The fact that membership in the formal sector systematically protects confirms [16] and [30] findings on the role of formalization as an essential condition for decent work. The extreme vulnerability of caregivers, observed in Senegal, Niger, Guinea-Bissau and Mali, highlights the precariousness of these statuses, which often lack direct remuneration and social protection. Self-employed workers have more contrasting profiles across countries, confirming the heterogeneity of self-employment in West Africa, oscillating between subsistence entrepreneurship and more productive activities. Employer status reinforces this trend and provides greater resilience, reflecting the importance of access to productive assets and resources.

Finally, the sectoral analysis indicates that, in the majority of countries, industry and services reduce working poverty compared to agriculture, in line with the observations of [1]. However, these effects are not uniform and are highly dependent on the nature of the activities carried out. In some contexts, informal urban services can replicate forms of insecurity comparable to those observed in rural areas, as highlighted by [14]. In this regard, the HHLCS data show that the number of hours worked per week has a small overall effect on the probability of being a poor worker in the WAEMU, which is consistent with the literature on informality-dominated economies. Indeed, working long hours does not necessarily protect against poverty because of low labor productivity and income insecurity ([13]; [17]). This result is further supported by [15] and [18], who show that, once education level, sector of activity and household structure are controlled for, the effect of hours worked is generally marginal, suggesting that working poverty is driven primarily by deficits in job quality and productivity rather than by a lack of work.

In total, the results of the analysis of the data of the HHLCS in the WAEMU zone confirm the three main determinants highlighted by the literature: i) informality and rurality as the primary core of working poverty; ii) human capital as a key factor of protection; and iii) gender and family structure inequalities as cross-cutting dimensions often neglected. They also stress the need for multisectoral approaches, combining education, formalization of the economy, gender policies and transformation of productive structures, in order to achieve the decent work objective promoted by the ILO and enshrined in the SDGs and Agenda 2063.

4. Conclusion

The analysis of working poverty in the WAEMU area, based on data from the HHLCSs, highlights the complexity of the phenomenon of working poverty, which cannot be reduced to a simple income deficit. The results confirm that this form of poverty is rooted in productive structures dominated by informality and subsistence agriculture, combined with persistent inequalities in gender, education and territory. Rural areas appear to be a major determinant of vulnerability, while educational attainment and formal participation are key protective factors. Similarly, family structures, particularly polygamy and female-headed single-parent households, significantly increase workers’ exposure to working poverty.

These findings underline the need for a multidimensional approach that articulates education, formalization and social protection policies. The sustainable reduction of hard-working poverty requires the promotion of productive and decent jobs, but also structural reforms capable of transforming the economic and social foundations of the region. From this perspective, investment in human capital, modernization of social protection systems, gender equality and sectoral diversification appear to be key levers. These actions require strengthened governance, based on reliable data, regional coordination and rigorous monitoring and evaluation mechanisms to ensure the effectiveness and sustainability of public policies.

Ultimately, working poverty in WAEMU is a critical indicator for assessing the ability of economies to provide opportunities for mobility and dignity for workers. Its persistence challenges the region’s development trajectory and recalls that the goal of decent work, at the heart of the SDGs and Agenda 2063, cannot be achieved without an inclusive and equitable structural transformation.

Beyond these general policy orientations, the results show that female-headed single-parent households are particularly vulnerable. In several WAEMU countries, workers living in such households face a much higher risk of in-work poverty. This situation reflects limited access to the labor market, unstable incomes and heavy family responsibilities. Reducing this vulnerability requires stronger and better-targeted social protection, including cash transfers, family allowances and access to health and childcare services. In parallel, employment and training policies should give priority to women heads of households by improving access to formal, better-quality and more productive jobs. Such an integrated approach is necessary to reduce working poverty and prevent its transmission across generations in the WAEMU.

NOTES

*West African Economic and Monetary Union.

^#This article examines the determinants of working poverty in the WAEMU. It shows that, in this monetary and economic area, having a job does not guarantee escaping poverty, especially for rural, informal, low-educated workers, women and youth.

¹With regard to the monetary dimension, the definition of working poverty refers to the characteristics of the household. The working poor are individuals who live in households whose total income is below a given threshold of the country’s median ([22]). The underlying assumption is that income is pooled and equally shared among household members, in accordance with [5] unit model.

²For reasons of comparability, the study considers the definition of employment at international level and does not take into account the age of admission to employment determined by national legislation.

³OCDE et al. (2025). Dynamiques de l’urbanisation africaine 2025: Planifier l’expansion urbaine, Cahiers de l’Afrique de l’Ouest, Éditions OCDE, Paris, https://doi.org/10.1787/cb26f4e2-fr, Année: 2020.

⁴Banque Mondiale: Bénin (2022); Burkina Faso (2023); Côte d’Ivoire (2022); Guinée (2020); Mali (2022); Niger (2022); Sénégal (2022); Togo (2022).

⁵Banque Mondiale (2023), https://donnees.banquemondiale.org/indicator/SL.AGR.EMPL.ZS.

References

1.. AfDB (2021). African Economic Outlook 2021: From Debt Resolution to Growth. African Development Bank. 2021

2.. Banque Mondiale (2022). Jobs Diagnostic for Sub-Saharan Africa. World Bank. 2022

3.. Banque Mondiale (2024). Pauvreté, la prospérité et la planète: Les voies de sortie d’une polycrise. World Bank. 2024

4.. Barrett, C. B., Reardon, T., & Webb, P. (2001). Nonfarm Income Diversification and Household Livelihood Strategies in Rural Africa. World Development, 29, 315-331. Barrett, C. Reardon, T. Webb, P. 2001

5.. Becker, G. S. (1964). Human Capital: A Theoretical and Empirical Analysis, with Special Reference to Education. University of Chicago Press. Becker, G. 1964

6.. Caspersz, D. (2001). The Power to Choose: Bangladeshi Women and Labour Market Decisions in London and Dhaka by Naila Kabeer, London: Verso, 2000, 464 pp., ISBN 1-85984-804-4. Relations Industrielles, 56, 217. https://doi.org/10.7202/000153ar 10.7202/000153arhttps://doi.org/10.7202/000153ar Caspersz, D. Kabeer, L 2001 10.7202/000153ar

7.. Chant, S., & Sweetman, C. (2012). Gender, Poverty andLabourMarkets in Africa. Oxfam GB. Chant, S. Sweetman, C. Gender, P 2012

8.. Chen, M. A. (2006). Rethinking the Informal Economy: Linkages with the Formal Economy and the Formal Regulatory Environment. In B. Guha-Khasnobis, R. Kanbur, & E. Ostrom (Eds.), Linking the Formal and Informal Economy (pp. 75-92). Oxford University Press. https://doi.org/10.1093/0199204764.003.0005 10.1093/0199204764.003.0005https://doi.org/10.1093/0199204764.003.0005 Chen, M. Guha-Khasnobis, R. 2006 10.1093/0199204764.003.0005

9.. Deaton, A. (2003). Health, Inequality, and Economic Development. Journal of Economic Literature, 41, 113-158. https://doi.org/10.1257/jel.41.1.113 10.1257/jel.41.1.113https://doi.org/10.1257/jel.41.1.113 Deaton, A. Health, I 2003 10.1257/jel.41.1.113

10.. Doeringer, P. B., & Piore, M. J. (1971). Internal Labor Markets and Manpower Analysis. D. C. Heath. Doeringer, P. Piore, M. 1971

11.. Ellis, F. (2000). Rural Livelihoods and Diversity in Developing Countries. Oxford University Press. Ellis, F. 2000

12.. Elson, D. (2002). Gender Justice, Human Rights, and Neo-Liberal Economic Policies. In M. Molyneux, & S. Razavi (Eds.), Gender Justice, Development, and Rights (pp. 78-114). Oxford University Press. https://doi.org/10.1093/0199256454.003.0003 10.1093/0199256454.003.0003https://doi.org/10.1093/0199256454.003.0003 Elson, D. Justice, H Justice, D 2002 10.1093/0199256454.003.0003

13.. Fields, G. S. (1990). Labour Market Modeling and the Urban Informal Sector: Theory and Evidence. In OECD (Ed.), The Informal Sector Revisited (pp. 49-69). OECD. Fields, G. 1990

14.. Fox, L., & Kaul, U. (2017). The Evidence Is in: How Should Youth Employment Programs in Low-Income Countries Be Designed? U.S. Agency for International Development. Fox, L. Kaul, U. 2017

15.. Gindling, T. H., & Newhouse, D. (2014). Self-Employment in the Developing World. World Development, 56, 313-331. https://doi.org/10.1016/j.worlddev.2013.03.003 10.1016/j.worlddev.2013.03.003https://doi.org/10.1016/j.worlddev.2013.03.003 Gindling, T. Newhouse, D. 2014 10.1016/j.worlddev.2013.03.003

16.. ILO (2018a). Decent Work and the Informal Economy. International Labour Office.

17.. ILO (2018b). Women and Men in the Informal Economy: A Statistical Picture. International Labour Office.

18.. ILO (2023). World Employment and Social Outlook: Trends 2023. International Labour Office. https://www.ilo.org 2023

19.. Korem, A. (2021). Education et emploi en Afrique subsaharienne: Quelle éducation pour quel emploi? Revue d’Economie Théorique et Appliquée, 11, 167-186. https://doi.org/10.62519/reta.v11n2a5 10.62519/reta.v11n2a5https://doi.org/10.62519/reta.v11n2a5 Korem, A. 2021 10.62519/reta.v11n2a5

20.. Minot, N., & Daniels, L. (2005). Impact of Global Cotton Markets on Rural Poverty in Benin. Agricultural Economics, 33, 453-466. https://doi.org/10.1111/j.1574-0864.2005.00415.x 10.1111/j.1574-0864.2005.00415.xhttps://doi.org/10.1111/j.1574-0864.2005.00415.x Minot, N. Daniels, L. 2005 10.1111/j.1574-0864.2005.00415.x

21.. Munoz Boudet. A. M., Buitrago, P., De La Briere, B.L., Newhouse, D., Rubiano Matulevich, E., Scott, K., & Suarez-Becerra, P. (2018). Gender Differences in Poverty and Household Composition through the Life-Cycle: A Global Perspective (World Bank Policy Research Working Paper No. 8360). World Bank. http://hdl.handle.net/10986/29426 Buitrago, P. Briere, B.L. Newhouse, D. Matulevich, E. Scott, K. Suarez-Becerra, P. 2018

22.. Ponthieux, S. (2010). Assessing and Analysing In-Work Poverty Risk. In A. B. Atkinson, & E. Marlier (Eds.), Income and Living Conditions in Europe (pp. 307-328). Eurostat. Ponthieux, S. 2010

23.. Portes, A., & Hoffman, K. (2003). Latin American Class Structures: Their Composition and Change during the Neoliberal Era. Latin American Research Review, 38, 41-82. https://doi.org/10.1353/lar.2003.0011 10.1353/lar.2003.0011https://doi.org/10.1353/lar.2003.0011 Portes, A. Hoffman, K. 2003 10.1353/lar.2003.0011

24.. Ravallion, M. (2001). Growth, Inequality and Poverty: Looking beyond Averages. WorldDevelopment, 29, 1803-1815. https://doi.org/10.1016/s0305-750x(01)00072-9 10.1016/s0305-750x(01)00072-9https://doi.org/10.1016/s0305-750x(01)00072-9 Ravallion, M. Growth, I 2001 10.1016/s0305-750x(01)00072-9

25.. Ravallion, M. (2015). The Economics of Poverty: History, Measurement, and Policy. Oxford University Press. https://doi.org/10.1093/acprof:oso/9780190212766.001.0001 10.1093/acprof:oso/9780190212766.001.0001https://doi.org/10.1093/acprof:oso/9780190212766.001.0001 Ravallion, M. History, M 2015 10.1093/acprof:oso/9780190212766.001.0001

26.. Sen, A. (1997). On Economic Inequality (Expanded ed.). Clarendon Press. Sen, A. 1997

27.. Sen, A. (1999). Development as Freedom. Oxford University Press. Sen, A. 1999

28.. Taylor, J. E., & Martin, P. (2001). Human Capital, Migration, and Rural Population Change. In B. Gardner, & G. Rausser (Eds.), Handbook of Agricultural Economics(Vol. 1A, pp. 457-511). Elsevier. Taylor, J. Martin, P. Capital, M 2001

29.. World Bank (2019). Benin: Agricultural Sector Review. World Bank. 2019

30.. World Bank (2020). World Development Indicators 2020. World Bank. 2020