STUDY LABOUR AND SOCIAL JUSTICE WORKING IN THE SHADOWS Informal Employment in Ethiopia Susanne Stollreiter and Rudolf Traub-Merz(eds.) May 2023 The informal economy is by far the largest provider of employment and livelihood opportunities in Ethiopia. The informal economy workers are exposed to severe decent work deficits, including the lack of access to public social protection. Ethiopia’s informal econo­my is well organized, with two thirds being member of an as­sociation, a cooperative, or an indigenous organization such as an Iddir (insurance) or an Eqqub(saving). Most informal workers join social groups for obtaining social services. Trade unions may provide organizing support. Many of the informal workers’ organizations are rather local. This limits their ability to take part in national policymaking, decision-making or social dia­ logue. Nevertheless, they are essential for civil society: they strengthen commu­nities, deliver economic and fi­nancial services and extend social protection.

LABOUR AND SOCIAL JUSTICE WORKING IN THE SHADOWS Informal Employment in Ethiopia

 Content EXECUTIVE 2 by Susanne Stollreiter FORMAL AND INFORMAL EMPLOYMENT – DEFINITION, STRUCTURE AND MAIN TRENDS IN ETHIOPIA 4 by Rudolf Traub-Merz  1 Defining Informal Employment 4  2 Labour Force and Employment in Ethiopia – Statistical Sources and Overview 5  3 Employment – Main Trends 5 INFORMAL EMPLOYMENT IN 9 by Rudolf Traub-Merz  1 Introduction 9  2 Informal Employment and Age Structure 9  Relations and Working Conditions 10  Distribution – Dominance of Agriculture and Services 16  Distribution 16 6 Gender Inequality 20  7 Living Outside Social Security Coverage 22 8 Summary 24 IS ETHIOPIA’S INFORMAL ECONOMY ORGANISED AND, IF SO, 26 by Jürgen Schwettmann  1 Purpose of this Chapter 26  2 Informal Economy Organisations: Types and Purpose 26  3 Survey Findings and Interpretation 33  4 Conclusions 43 INTEREST AMONG THOSE IN INFORMAL EMPLOYMENT IN TRADE UNIONS – THE CASE OF 46 by Rudolf Traub-Merz  1 Views of Informal Labour on Trade Unions 47  Points to Be Considered When Trade Unions Lay Out Strategies for Organising Informal Labour 56 TECHNICAL NOTES ON THE SURVEY 61 by Mulu Teka List of Figures 65 List of Tables 67

FRIEDRICH-EBERT-STIFTUNG – WORKING IN THE SHADOWS EXECUTIVE SUMMARY Work in Ethiopia is mainly informal. According to recent surveys, over 90 per cent of workers in Ethiopia are in the informal sector. That means that they are working and producing in precarious and vulnerable conditions and often are subject to poverty, insecurity and a severe lack of decent work. A high incidence of informality in an economy is a major challenge for sustainable development. States do not obtain tax revenues from people without work contracts, while workers do not benefit from public social protection schemes for health, unemployment, pensions and so on. The result is highly risky employment and an extremely vulnerable economy. The transition from an informal to a more formalised economy is therefore of strategic significance for both workers and governments. Before such a step is taken, however, knowledge must be acquired about the nature and characteristics of informal employment in a country in order to identify the right entry points and to discuss them at national level for adequate policy interventions. Informal employment exists in all countries, regardless of their level of socio-economic development. However, it is definitely more predominant in developing countries. In most African countries, more than 80 per cent and up to 95 per cent of the working population make a living by working informally. This impacts on societies beyond the economy. It shapes workers’ living conditions and most importantly their interaction and relationship with the state and society. As they do not benefit from public social protection informal workers usually have a more distanced relationship to the state and its institutions, as studies show. The Covid-19 pandemic dramatically highlighted what happens to the parts of society not covered by state intervention and protections. The number of people living in poverty increased dramatically during the pandemic and worsened informal workers’ living and working conditions. In Ethiopia, the pandemic, the war in Ukraine and internal factors, such as violent political conflicts, have all contributed to robbing the economy of its positive dynamic. Informal work was high even before the global health crisis, however. Economic growth in the past was not connected to an increase in formal employment, as, for example, an FES study 1 on employment proved. Minimum wage discussions in Ethiopia regained some momentum after the pandemic in 2023. One prominent reason given against a minimum wage is that only 8 per cent of the workforce would benefit from it. However, this does not take into account spillover effects from the formal to the informal sectors of the economy. Another interesting aspect of the minimum wage and other decent-work and labour-related discussions in Ethiopia is the fact that trade union social demands could gain momentum if the huge potential of(un)organised workers in the informal sector helped to increase their number and make their voices louder. Trade unions in Ethiopia, however, have not yet attempted to organise the informal sector in order to achieve a critical mass for social demands. This is a major opportunity that has not yet been sufficiently ­explored. In 2022 the Friedrich-Ebert-Stiftung(FES) published a study on the nexus between informality and social protection. The study was conducted by FES together with the German Institute of Development and Sustainability(IDOS) and the International Labour Organization(ILO). The team was joined by colleagues from Afrobarometer for methodological design and data collection. The study covered six African countries, including Kenya, Benin, Ethiopia, Cote d’Ivoire, Senegal and Zambia. The findings of the study are highly interesting and valuable for both researchers and policymakers. However, the six-country comparison also created a keen interest in obtaining a deeper understanding of the findings in some countries. That is why, together with the general interest in developing better insights and knowledge about the nature and character of informal employment in Ethiopia, FES-Ethiopia asked some of the researchers involved in the 2022 study“Working in the shadows” to take the data for Ethiopia from this study and to interpret them only for Africa’s most populated country. Using these findings FES-Ethiopia would like to increase knowledge about informal work in Ethiopia, its organisation and paths towards social protection. In addition, FES Ethiopia hopes that the study and its findings will stimulate a discussion for example in the labour movement in Ethiopia on some of the aspects mentioned, as well as with government institutions responsible for social protection. 1 Rudolf Traub-Merz, Manfred Öhm, Julia Leininger, Florence Bonnet and Henrik Maihack(eds.)(September 2022): A majority working in the shadows – a six country opinion survey on informal labour in sub-Saharan Africa, https://library.fes.de/pdf-files/iez/19558.pdf. The findings of the Ethiopia“Working in the shadows” study will be of immense value to both researchers and policymakers, as well as civil society and it is hoped that it will be widely consulted. 2

I sincerely congratulate the team of researchers on this great project and their commitment to bring it to a success and I look forward to the discussions and activities that this study might induce. Susanne Stollreiter Resident Representative FES-Ethiopia Executive Summary 3

FRIEDRICH-EBERT-STIFTUNG – WORKING IN THE SHADOWS FORMAL AND INFORMAL EMPLOYMENT – DEFINITION, STRUCTURE AND MAIN TRENDS IN ETHIOPIA Rudolf Traub-Merz 1 DEFINING INFORMAL EMPLOYMENT The concept of“informality” as a new approach to analysing economic and labour issues first appeared within the framework of the so-called“informal sector”, which was coined in 1972 by an ILO employment advisory mission to Kenya. 2 Later, the notion was expanded to“informal economy” as a concession to the fact that informality featured in all branches of the economy. In 2015, the Internal Labour Conference(ILC) adopted ILO Recommendation No. 204 and stipulated that the term“informal economy”“refers to all economic activities by workers and economic units that are – in law or in practice – not covered or insufficiently covered by formal arrangements”. 3 employment refers to a new socio-economic reality in which while a company is a formal(registered) enterprise, some or all of its employees may be working under informal conditions. Informal employment is a complex term in that it borrows its criteria in different ways for different groups in the workforce. Employers and own-account workers(self-employed) 5 are deemed to be informal if their economic units are not registered(in particular for tax purposes). In contrast, employees are informal if they enter into a wage relationship with an employer, but are not covered by social protection schemes. Workers contributing to a family business are a special category in that they are informal by definition. The informal economy covers a broad range of occupations. In rural areas, they are related mainly to agriculture, including livestock and fisheries, whereas in urban areas, we find occupations as diverse as parking-lot attendants, fuelwood carriers, money changers, gardeners, construction workers, pushcart transporters and waste recyclers, to name but a few. Whereas the informal economy has become a widely used concept, 4 its main focus is on enterprises and their relationship to the state and less to employment issues. As employment issues receded, 2003 witnessed a conceptual amendment when the term“informal employment” was introduced. The latest refurbishment responded to the growing trend for informality to cross over strongly into the formal economy. Large enterprises no longer followed the motto “everything is formal” and“all rules and norms” are followed. They took on informal elements into their undertakings while following legal norms in other regards. Informal “Formal” means that the current economic and social laws are applied and“informal” implies a disregard of them. But the binary use of formal and informal does not provide a full picture of the empirical reality of employment relations. Enterprises may follow some laws but not others, while their staff may be protected by some social regulations and not by others. Depending on what criteria are used, employment can be formal or informal, and switch to the opposite if the criteria are changed. The same can be said with regard to social regulations. An enterprise may provide employees with health insurance but not with an old-age retirement benefit scheme. It may pay the statutory minimum wage but ignore overtime payments and leave allowances. It may provide a written labour contract but not implement other legally mandatory provisions. Looking from the diverse application of legal provisions, employees may be formal in some regards but informal in others. One may thus talk of a formal–semi-formal–informal taxonomy and even expand it to include more zones in-between. 6 2 For an early report on institutional history, see Bangasser(2000). 3 The point is further made that illicit and illegal activities do not fall under the realm of the informal economy. 4 The 21 st session of the International Conference of Labour Statisticians(to be held in October 2023) is expected to adopt a new, much more detailed definition of the term“informal economy”. Given that informal-economy occupations account for the majority of the labour force in all developing countries and emerging economies the scarcity of references to informality in the UN’s 2030 Agenda for Sustainable Development(the SDGs) is striking. With the exception of SDG indicator 8.3.1(share of non-agricultural informal employment in total employment) the Agenda is silent about informality. This short discussion of operational problems involved in defining informal employment already points to the difficulty of agreeing on a common definition acceptable to all and applied in a uniform manner. Each study may decide which 5 The ILO(2018) uses the term“independent worker” as synonymous with own-account worker. 6 For a detailed discussion on the informal sector and informal employment, see ILO(2018) and Federation of Kenya Employers(2021). 4

Formal and informal employment – definition, structure and main trends in Ethiopia aspects of informality to select and which to ignore. 7 The operational definition will affect the results accordingly. But no matter how we look at the definition, informality features prominently in labour economics. listed and cannot conclude on the extent to which the relative decline in employment between 2013 and 2021 expresses a lasting trend or has been caused by temporary circumstances. 2  ABOUR FORCE AND EMPLOYMENT IN ETHIOPIA – STATISTICAL SOURCES AND OVERVIEW The Central Statistics Agency(CSA) of Ethiopia conducts national labour and household surveys at intervals of five to eight years. The most recent survey by the Central Statistics Agency(CSA) was implemented from 25 January to 24 February 2021. Some key findings are presented in a summary report(CSA 2021). For comparison, the report includes date from the surveys of March 1999, March 2005, June 2013 and February 2021. A superficial look at some labour trends shows dramatic changes over the years. The employment-to-population ratio declined from 76.2 per cent in June 2013 to 59.5 per cent in February 2021, pointing to deep-cutting disruptions in employment. The changes were restricted to rural areas, however. The employment-to-population rate for urban areas showed a slight increase(49.7 per cent vs 50.7 per cent), whereas the rural rate plummeted from 81.6 to 62.1 per cent. The decline affected both sexes but was stronger for women(76.3 per cent vs 53.0 per cent) than for men (86.9 per cent vs 71.1 per cent). The recent CSA survey must be assessed with four considerations in mind:(a) the survey was conducted after the post Covid-19 peak. The government of Ethiopia declared a five-month state of emergency in March 2020 to contain the virus. From November 2020 even though the state of emergency had been lifted and employment was on the path to“normalcy”, some employment effects from previous lockdown measures may have lingered on;(b) the survey was conducted in early 2021 after the military conflict in Tigray had erupted. The CSA report does not include data from Tigray but it remains unclear whether peace disturbances in other regions affected the data as well;(c) Ethiopia witnessed a swarm of locusts, described as the worst in 25 years. It lasted all through 2020 and began to decline only early in 2021. Locust swarms severely harm agriculture and may explain some of the employment disturbances in rural areas;(d) the CSA(2021: 23) defines persons in employment“as all those of working age 10 years and above who …. were engaged in any activity to produce goods or provide services for pay or profit”. The definition thus includes age groups otherwise considered school cohorts and excluded from the working population. When efforts to increase school attendance succeed for age groups 10+, the employment-to-population rate falls. We have no possibility of assessing the relative importance of the points The second source of labour data is the ILO. The ILO relies on CSA data but transposes them into“standard” categories for international comparison. It recalibrates the CSA working-age population 10+ to a working-age population 15+ and redefines the age group 10–14 as child labour, which for normative reasons is excluded. The ILO further looks at underlying trends between the CSA survey years, makes adjustments and provides extrapolations. The data are called“modelled ILO estimates” and are linked together as a historical time series. 8 The World Bank sources data from the ILO and publishes it in the World Development Indicators(WID). The most recent data are for 2019. Table 1 lists key employment and labour force data from the ILO model estimate for 2019. Population Figures are given in absolute numbers, while most other Figures are presented as share of population or employment. When transposed into absolute numbers, rounding errors arise which are left unadjusted. 3  – MAIN TRENDS The data in Table 1 refer to the national level and include formal and informal employment. Several points can be emphasised in reference to total national employment: – Agriculture is the largest sector in the economy (66.6 per cent). This share includes only work done for income and does not cover subsistence production for own consumption. The actual number of people working in agriculture is accordingly higher. – Services is the second most important provider of employment(24 per cent). One should add that services are more important in urban than in rural areas as they include the public sector, which is mostly urban-based. – Industry is made up of manufacturing, mining, utilities and construction and accounts for only 9.3 per cent of workplaces. In a historical-development perspective, this has been called the leapfrogging of manufacturing or hypertrophic development of services. – Total wage labour(formal and informal) is estimated to account for 8,454,541 people in 2019(ILOEST). The share of wage labour in the economy is low, representing only 15.9 per cent of total employment. National employment in Ethiopia is mainly(84.2 per cent) self-employment, either as employer or own-account worker. 7 The FES study uses membership of a pension insurance scheme as a proxy for social protection. 8 For details see ILO modelled estimates and projections(ILOEST), available at: https://ilostat.ilo.org/resources/concepts-and-definitions/ilo-modelled-estimates/ 5

FRIEDRICH-EBERT-STIFTUNG – WORKING IN THE SHADOWS Table 1 Ethiopia, labour force and employment 2019, based on ILO modelled estimates Ethiopia year 2019 Population Population total Population 15–64 Population 65+ Population 15+ Labour force& unemployment& employment Labour force(LF) as% of population 15+ Labour force(LF, total) Unemployment(% of LF) Unemployment(total) Employment(% of population 15+) Employment(total) Employment by sector Agriculture(% of employment) Industries(% of employment) Services(% of employment) Agriculture(total) Industries(total) Services(total) Employment by status group Wages and salariat workers(% of employment) Self-employed(% of employment) Employers(% of employment) Wages and salariat workers, total Self-employed, total Employers, total Own-account, total(self-employed less employers) Part-time employment Part-time employment(% of employment, 2013) Country 114,120,594 64,147,672 3,542,170 67,689,842 80.68 54,611,488 2.33 1,270,263 78.80 53,340,951 66.63 9.32 24.05 35,541,074 4,971,376 12,828,498 15.85 84.15 0.46 8,454,541 44,886,411 245,368 44,641,043 53.77 Source: World Bank, World Development Indicators, available at: https://datatopics.worldbank.org/world-development-indicators/ Male 57,375,433 32,082,743 1,624,542 33,707,285 86.63 29,201,296 1.74 506,934 85.12 28,694,674 73.44 9.51 17.05 21,073,369 2,728,864 4,892,442 17.97 82.03 0.72 5,156,433 23,538,241 206,602 23,331,639 42.91 Female 56,745,161 32,064,929 1,917,628 33,982,557 74.82 25,424,730 3.00 762,488 72.57 24,662,160 58.71 9.10 32.19 14,479,154 2,244,257 7,891,892 13.38 86.62 0.17 3,299,797 21,362,364 41,926 21,320,438 66.47 – The ILO estimate puts the share of employers in Ethiopia at 0.46 per cent of all employment, which appears to be too low. If we relate the shares of wage labour and employers(15.9 per cent/ 0.46 per cent) we get on average 34.6 employees per employer. This is on the high side and does not take into consideration that most employers are micro-entrepreneurs with fewer than five employees. The ILO estimate appears to take into account only businesses with ten or more employees. – Part-time employment rather than unemployment is a key feature of the labour market. More than half (53.8 per cent) of employment is provided as part-time employment, which the ILO usually defines as<36 weekly working hours. Even though the information is taken from 2013 and not available for 2019, there may not have been a substantial change since then. In contrast, unemployment affects just above 2 per cent of the labour force, a clear reaction to the social security 6

Formal and informal employment – definition, structure and main trends in Ethiopia situation. Where no unemployment insurance benefits are provided, people cannot be without income and accept part-time work whenever possible. – Substantial gender disparities can be observed for all labour and employment categories. Only 74.8 per cent of women look for gainful employment, whereas 86.6 per cent of men do so. This translates into female employment in the economy being about 4 million below male employment(24,662,160 vs 28,694,674). pushed groups from formal employment into informality, 48 million may be a reasonable Figure for 2020. Informality rather than formal employment is the“standard” in labour markets. While 85.2 per cent is clearly a weighty quantity in most regards, it is not possible to apply the informality rate in a uniform manner to all sub-categories. There are substantive structural disparities between formal and informal employment, of which three are immediately evident: – Female participation in employment is further limited in terms of part-time employment. The share of part-time employment for women is 66.5 per cent, whereas for men it is 42.9 per cent. If lower numbers in employment and lower working hours per person are taken together, the female employment participation rate is 25 per cent or more below the male participation rate. – Gender-based occupational segregation exists in regard to agriculture and services. There are substantially more men than women in agriculture(21.1 million vs 14.5 million), whereas women are more prominent in services(7.9 million vs 4.9 million). The reason for the higher female participation in services is trade, which in Ethiopia is primarily a female affair.  labour is the dominant employment form in the formal economy, usually accounting for 70 to 80 per cent. Within the informal economy, the prominent place is taken by self-employment, with wage labour a distant second.  has a higher employment share in the formal economy, mostly derived from utilities. Electricity and water are usually provided on the basis of formal employment conditions.  employment features more strongly in informal employment, whereas many workplaces in the formal economy follow the principle of“standard” work. – Gender segregation is visible likewise with regard to employers and employees. The ILO estimates that there are about five times more male than female employers (207,000 vs 42,000), but we should keep in mind the observation above that this may refer only to larger enterprises; in the same way wage labour is considered first and foremost a male occupation(5.2 million vs 3.3 million). In contrast, self-employment is shared equally between the sexes. 3.1  THE SHARE OF INFORMAL EMPLOYMENT 3.2  ESTIMATES AND FES SURVEY DATA – SOME CONSIDERATIONS It is not possible to directly compare the ILO modelled estimates with the findings of the FES survey, for reasons already indicated. Even though the period differs only by one year(2019 vs 2020), the effects on the labour market of the Covid-19 pandemic and the locust invasion are difficult to appraise. While they affect employment levels this is less the case for structural features. A change in employment of 10 or even 20 per cent will impact on the frequency of occurrence but not reverse the relative shares of the various groups. The data in Table 1 refer to the national level and include formal and informal employment. No data have been published that would allow us to determine the share of formal and informal employment in Ethiopia for all categories listed. Research in other countries indicates that the informal economy is by far the most dynamic provider of employment opportunities. The most comprehensive survey on a global scale has been done by the ILO(2018), which showed that for countries in Eastern Africa, the share of informal employment in total employment varied between 85 and 95 per cent(Table B.1). Ethiopia was not included. For 2019, however, the ILO announced the share of informal to total employment in Ethiopia to be 85.2 per cent. If we take 2019 as the most recent year of“normal employment” not affected by“external factors” such as war, pandemic or locust invasion and apply the 85.2 per cent-informality rate to total employment, the number of people in informal employment in 2019 would amount to 45.4 million. If we add a factor for population growth and assume that the pandemic With these caveats in mind, a look at selected labour features in the two labour data sets shows conformity and non-conformity, as indicated in Table 2. – The sectoral profile of employment is confirmed by both data sets. FES, however, indicates a higher share of employment in services and a lower share in agriculture and industries. This would somehow be in line with the CSA findings for February 2021, which indicated a serious fall in employment for rural areas. – The structural articulation into status groups is confirmed by both data sets. Self-employment is the dominant form, with wage labour a far-distant second. The FES survey, however, shows a lower share of employees and a higher share of self-employment. This may again reflect a labour market in crisis. When consumer demand falls, enterprises cut the number of employees, who turn to self-employment in search of a livelihood. 7

FRIEDRICH-EBERT-STIFTUNG – WORKING IN THE SHADOWS Table 2 Selected employment data(ILO 2019 total employment; FES 2020 informal employment) Employment by sectors Agriculture Industries Services Employment by status group Employee Employer Own-account* Part-time employment** Notes:*Includes family-support workers.  **ILO data for 2013; 2019 not available. All ILO FES 66.6% 9.3% 24.0% 61.9% 6.4% 31.7% Male ILO FES 73.4% 9.5% 17.0% 70.9% 7.8% 21.3% 15.9% 0.5% 83.7% 53.8% 10.3% 3.9% 85.7% 38.6% 18.0% 0.7% 81.3% 42.9% 11.2% 3.8% 85.1% 33.0% Female ILO FES 58.7% 9.1% 32.2% 50.2% 4.6% 45.3% 13.4% 0.2% 86.5% 66.5% 9.2% 4.2% 86.6% 45.9% – The gap in the employer category is tremendous and can hardly be linked to a labour market in crisis. As mentioned above, the divergence in their share is most likely caused by the use of different definitions. The FES defines as employers all those who employ at least one regular employee, while the ILO appears to set the number of employees higher. – The working time profile is confirmed by both records. Part-time employment is a substantive feature of the Ethiopian economy. The differences may be linked to the recording year(ILO 2013; FES 2020) or the definition of part-time work. Under ordinary circumstances, parttime work in informal employment should be relatively higher than in formal employment. It may well have a seasonal component or be linked to the restart of the economy after policy restrictions were removed. REFERENCES Bangasser P.(2000) The ILO and the informal sector: an institutional history. ILO: Employment Paper. 2000/9 CSA(2021) Central Statistics Agency[Ethiopia]. Labour Force and Migration Survey Key Findings(Addis Ababa, Ethiopia). Federation of Kenya Employers(2021). The Informal Economy in Kenya. March. ILO(2018) Women and men in the informal economy: A statistical picture. Geneva. – The gender dimension is re-confirmed with regard to sectoral employment. The FES data confirm that women have a lower participation rate in agriculture and a higher participation rate in services. The gender gap is more pronounced in the FES data than in the ILO data. – The gender gap in relation to employment status is less pronounced on the FES survey. Women have a similar employment profile to men in our three categories(employee, employer, own-account workers). We will show that gender differences appear when the size of business is considered and male employers own larger enterprises than female employers. – The gender gap in working hours is striking with both data sets. More female labour is employed in part-time work. 8

Informal employment in Ethiopia INFORMAL EMPLOYMENT IN ETHIOPIA Rudolf Traub-Merz 1  In many countries the informal economy is the most dynamic provider of employment opportunities. This is also the case for Ethiopia. 9 A detailed labour survey on informal employment in Ethiopia is not available, but for 2021 the ILO estimates the share of informal labour in total employment to be around 85 per cent. 10 If we take 2019 as the most recent“normal” year in terms of employment 11 and apply the informality rate of 85 per cent to total employment, estimated by the ILO to be 53,340,951, the number of people in informal employment would be 45.3 million. If we add a factor for population growth and assume that the pandemic pushed groups from formal employment into informality, 48 million may be a reasonable Figure for 2020. Informality rather than formal employment is the“standard” in the Ethiopian labour market. As part of a six-country survey project, which was implemented between 2018 and 2022, the Friedrich-EbertStiftung(FES, lead agency) in cooperation with the German Institute of Development and Sustainability(IDOS) and the International Labour Organization(ILO) conducted a country-wide representative study on informal employment in Ethiopia in October–November 2020. A technical protocol 12 was applied with regard to sample design, questionnaire and data collection to allow cross-country comparison, with findings from Kenya(2018), Benin(2018), Senegal(2019), Zambia(2019) and Cote d’Ivoire(2020). The survey was designed as an opinion poll, but extensive information had to be collected on employment and income in order to separate house­holds and persons in informal employment from other employment groups. The data are a source in their own right and can be used to analyse the social and economic diversity of informal employment in a country. This is what this chapter is about. Section 2 begins by looking at the age 9 Ethiopia was not included in the global ILO study on informal employment, but other countries in East Africa were. Their share of ­informal employment in total employment varied between 85 and 95 per cent(ILO, 2018: Table B.1.). 10 For a discussion of the concept of informal employment and labour survey data in Ethiopia, see the respective chapter in this report. 11 The years 2020 and 2021 are atypical in that labour markets were distorted to some extent by the Covid-19 pandemic and the war between Ethiopia and the Tigray People’s Liberation Front(TPLF). 12 For technical details, see the respective chapter in this report. structure. It provides an overview of the extent to which various age cohorts participate in informal employment. Section 3 identifies the main employment status groups and focusses on key structural features. Section 4 concentrates on the sectoral distribution of employment and outlines employment patterns between agriculture and services. Section 5 looks at income distribution and analyses aspects of poverty and inequality. While gender dimensions are reflected in all sections, Section 6 specifically measures the gender income gap in informal employment. Section 7 closes the overview with an evaluation of the extent to which informally employed people are covered by social security provisions. 2  EMPLOYMENT AND AGE STRUCTURE Informal employment involves all age cohorts and can be characterised as all-age inclusive. Ethiopia also has a high rate of child labour, which is not included in the analysis. The sample design of the FES study on informal employment followed the ILO definition for the working-age population, which uses 15 years as entry age and thus excludes child labour. The employment profile by age is arranged in age cohorts of 10 years and related to their share in the population(Figure 1). All age groups can be found in informal employment. However, some are overrepresented with regard to their population weight, while others are underrepresented. The most significant underrepresentation is found for the youngest cohort(15–24 years old). Compared with their relative share in the population(35.4 per cent) only about half(16.5 per cent) are engaged in informal employment. The other half are scattered throughout society, of which the most important sectors appear to be:(i) still in school;(ii) subsistence work or unemployment with no income at all;(iii) employment in the formal economy; and(iv) military services. Older age cohorts are overrepresented in informal employment. This implies their underrepresentation in other sectors of the economy. Our findings indicate a reversed population structure in the formal economy, with a higher share of younger employees and a low representation in particular for people older than 35 years of age. 9

FRIEDRICH-EBERT-STIFTUNG – WORKING IN THE SHADOWS Figure 1 Informal employment and population by age groups 40% 35.4 30% 30.4 25.6 20% 16.5 10% 25.4 14.8 16.2 10.8 0% 15–24 25–34 N= 2262. Source for population figures: UN population division. 35–44 Informal employment 45–54 Population 6.8 8.2 55–64 5.2 4.7 65+ 3  ABOUR RELATIONS AND WORKING CONDITIONS 3.1 P REDOMINANCE OF SELF-EMPLOYMENT A key feature of the informal work force is the predominance of self-employment. In Ethiopia, about 80 per cent of people who gain their income from informal employment do so in the form of self-employment; wage labour, which is the dominant mode of work in the formal economy, covers only 10 per cent or so(Figure 2). The predominance of self-employment is typical not only for Ethiopia but for other countries as well. Within our six-country study, only Kenya shows a significantly higher share of wage labour(34.3 per cent) and a correspondingly lower share of people in self-employment(55.9 per cent). But even there, self-employment stands out as the dominant employment pattern. Informal employment provides income opportunities mainly through self-employment, while employer–employee relations are a distant second. A look at gender composition does not disclose any striking disparity in Ethiopia(Figure 3). Female representation is close to male representation in all categories. Women have a slightly higher share as family-support workers and a slightly lower presence as wage earners. But the differences are not of sufficient magnitude that we can talk of substantive gender-distortions. 3.2  ICRO-ENTERPRISES – WITH LITTLE ADD-ON WAGE LABOUR Medium-sized enterprises tend to be classified as those with 50–249 employees, small enterprises as having 10–49 employees and micro enterprises 1–9. Following this categorisation, there is little informal employment in large and medium-sized enterprises or even small enterprises. Informal employment is made up of micro enterprises but even here, size tends to be at the low end. In Ethiopia, only 2.2 per cent of informal enterprises belong to the segment five to-nine employees. Most employers(82 per cent) hire only one or two regular employees(see Figure 4). The size of enterprises, more specifically their smallness strongly indicates that the separation of capital and labour has not progressed. Most enterprises sit on the borderline with self-employment; business proprietors remain owner-workers who add a(very) little wage labour to their own labour. Size of business is nevertheless gender-sensitive and ownership is marked by significant disparities(Figure 5). There are nearly twice as many male owners in enterprises with three or more regular employees(23 per cent vs 12.2 per cent). While a similar number of men and women become employers, they do so with diverging success. Female owners run smaller businesses and face more challenges expanding than men. 10

Informal employment in Ethiopia Figure 2 Informal employment, by status group 90% 80% 79.7 70% 60% 55.9 50% 40% 34.3 30% 20% 10% 0 10.3 7.4 3.9 Employers Senegal Employees Zambia Kenya Own-account worker Benin Cote d’Ivoire N(Ethiopia)= 2262. Figure 3 Informal employment, by status group and sex 90% 80% 80.0 79.3 70% 60% 50% 40% 30% 20% 10% 0 3.8 4.2 Employers N(male)= 1278; N(female)= 984. 11.2 9.2 Employees Male Own-account worker Female 11 6.0 2.3 Family worker Ethiopia 7.3 5.0 Family worker

FRIEDRICH-EBERT-STIFTUNG – WORKING IN THE SHADOWS Figure 4  of informal enterprises, by number of regular employees 90% 82.0 80% 70% 60% 50% 40% 30% 20% 15.7 10% 0 N(Ethiopia)= 89. 1–2 Senegal 3–5 Zambia Kenya 5–9 Cote d’Ivoire 2.2 Ethiopia 10 and more Figure 5  of informal enterprises, by sex of business owner 100% 90% 87.8 80% 77.1 70% 60% 50% 40% 30% 20% 10% 0 1–2 N(male)= 48; N(female)= 41. 18.8 12.2 3–5 Male Female 4.2 5–9 12

Informal employment in Ethiopia 3.3  MPLOYEES – WORKING CONDITIONS HIGHLY INSECURE Ethiopia has established legal provisions to regulate employer-employee relations and to provide employees with some level of security. In 2019, the Ethiopian government approved the new Labour Proclamation No. 1156/2019, 13 which replaced the labour law of 2013. Various clauses regulate working conditions, such as working hours(§61.2), overtime payment(§68), annual leave(§77;78), sick leave(§86), maternity leave(§88) and severance pay(§39; 40). Rather unusually, the Labour Proclamation states that“a contract of employment shall not be subject to any special form”(§5), thereby leaving it to the employer to decide the format. The proclamation nevertheless insists on written documentation:“Where a contract of employment is not made in writing, the employer shall, within 15 days from the conclusion of the contract, give the worker a written and signed letter containing the elements specified under Article 6 of this Proclamation”(§7.1). The letter shall“specify the type of employment and place of work, the rate of wages, method of calculation thereof, manner and interval of payment and duration of the contract”(§4.3). The proclamation does not consider an oral agreement as a substitute for a written document. Further13 See: Labour-Proclamation-No_-1156-2019.pdf more, no clause restricts coverage of the law to the formal economy and the Labour Proclamation applies to all employees whether in formal or informal employment. The FES survey did not inquire into the application of many labour regulations, but rather selected two as reference for employment security: the availability of a written contract and the duration of employment. The working conditions of informal employees are nothing if not insecure. In Ethiopia, only 16.2 per cent of wage earners could confirm they have a written contract. Most(61.1 per cent) identify their labour contract as an“oral agreement”, while 22.6 per cent state that they have no agreement at all (Figure 6). This is in contravention of the labour law, in which at least a letter is required to outline key provisions in writing. Without documentation, taking an employer to court for breach of agreement becomes difficult, if not impossible. The widespread absence of a written work contract is not unique to Ethiopia but typical of all countries in our survey. Flexibility for the employer and insecurity for the employee are the main features of the informal labour market. Another dimension of employment insecurity becomes visible by looking at length of employment contract. In Ethiopia, 28.7 per cent of informal employees identify their work contract as having no time restrictions and being permanent, Figure 6 Type of wage labour agreement for informal employees 70% 61.1 60% 50% 40% 30% 22.6 20% 16.2 10% 0 Written contract Oral agreement Senegal Zambia Kenya Benin Cote d’Ivoire N(Ethiopia)= 234. Question:“Do you have a written contract or oral agreement with your employer: YES, written contract/ Yes, oral agreement/ No, no agreement”. 13 No agreement Ethiopia

FRIEDRICH-EBERT-STIFTUNG – WORKING IN THE SHADOWS while 40.3 per cent are hired for a specified period and 30.9 per cent for the completion of a specific task(Figure 7). There are variations between countries and Senegal reports having half of informal employees hired for permanent jobs. But when most do not have a written contract, the permanence of the job may be more an illusion than reality. In most cases, employers offer jobs on condition of flexibility to fire. Employment insecurity does not have a particularly marked gender profile. The share of female employees with no written contract more or less matches the male share(16.5 per cent vs 16.1 per cent). Whereas in other countries, fewer women have access to written contracts, in Ethiopia their legal security is similar. It should be remembered that most employers engage only one or two regular employees. We did not inquire whether, previous to establishing a labour relationship, a personal acquaintance existed between an employer and an employee. But that may often be the case. Many employers do not recruit employees from an anonymous pool of wage labourers, but search within their neighbourhood, community or personal contacts. In this regard the employer’s power of control may be limited through the application of traditional norms and values, and even though employees work within a labour-law-free space, they may still enjoy some level of protection from community and kinship rules or personal network norms. 3.4  WORKING TIME Decent working time is an important indicator of the quality of employment. The first ILO Convention(1919) adopted a maximum of eight hours of work a day and 48 hours per week. Ethiopia subscribes to the ILO Convention by stating in the Labour Proclamation of 2019 that“normal hours of work shall not exceed 8 hours a day or 48 hours a week” (§61.2). The ILO defines“standard employment” as a working week between 35 and 48 hours. Below 35 hours is classified as “underemployment”, while above 48 hours is“overemployment”. We can further amend this taxonomy by calling “marginal employment” a weekly working time of 20 hours or less and“excessive employment” a working time of 60 hours or more. This classification provides a working time profile in informal employment as presented in Figure 8. The working time profile in informal employment can be summarised as follows: – Standard employment is the exception rather than the rule. Only 41.1 per cent of the informally employed have a working week of 35–48 hours. Some 58.9 per cent engage in work that deviates from“standard” as a social norm and are either underemployed(39.2 per cent) or overemployed(19.7 per cent). Figure 7 Duration of wage labour agreements for informal employees 70% 60% 50% 40.3 40% 30.9 30% 28.7 20% 10% 0 Permanent/ no time limit Senegal Zambia N(Ethiopia)= 181. With time limit Kenya Benin Cote d’Ivoire Completion of task Ethiopia 14

Figure 8 Working hours per week in main job Underemployment Informal employment in Ethiopia Standard employment Overemployment 90% 80% 70% 60% 50% 40% 30% 20% 10% 0 Marginal employment 16.0 < 21 hrs 39.2 < 35 hrs 41.1 35–48 hrs N= 2146. Question:“How many hours do you usually work per week in your main job? Number of hours” 19.7 > 49 hrs Excessive employment 6.8 > 60 hrs – Large groups of people can be found at the extremes of the working time profile: 16 per cent work 20 hours or less in activities to earn an income, while 6.8 per cent spend more than 60 hours a week to gain a livelihood. The distribution of working time should be related to the income it generates and how much a person needs to sustain a decent living. Working 60 hours or more a week may be problematic from a health point of view but not socially, if it is done freely and pays an income above basic requirements; marginal employment may equally reflect a“free choice” if a person is not willing to work longer than they need. Overand underemployment, however, express“anomalies”, if people are coerced into such work arrangements due to low income. Below we will show that underemployment or marginal employment are linked mainly to poverty, which explains why working shorter hours is not necessarily an expression of the high value a worker puts on leisure time, as economic-liberal labour market analysts suggest, but rather because of a lack of opportunities as a result of a shortage of monetary demand. Whereas underemployment is usually linked to low-level demand, overemployment is largely caused by low pay. People have to invest more working hours to earn a living income. A stall owner may sit with their merchandise from morning until late at night to wait for customers. A cleaner or restaurant worker may never follow a regular schedule but may work as long as there is something to do. A look at the gender profile of working time allotments indicates substantial disparities(Figure 9). The female share in underemployment is much higher than the male share (45.9 per cent vs 33 per cent). In the category of standard employment(35–48 hrs), the shares are reversed. Only 36.2 per cent of women find themselves in full-time work compared with 45.8 per cent for men. Up to this point of the analysis, working time allotment is the factor with the most distinct gender face. In terms of overall working time, women are less engaged in informal employment than men. This will emerge again as a structural factor when we look at gender income gap below. 15

FRIEDRICH-EBERT-STIFTUNG – WORKING IN THE SHADOWS Figure 9 Weekly working hours, by gender 50% 40% 30% 20% 10% 20.6 12.6 25.3 20.4 45.8 36.2 14.2 11.3 0 24 hrs and less N(male)= 1165; N(female)= 893. Question: see Figure 8 25–34 hrs Male 35–48 hrs Female 49–60 hrs 6.9 6.6 > 60 hrs 4  ECTORAL DISTRIBUTION – DOMINANCE OF AGRICULTURE AND SERVICES The Ethiopian economy is largely agriculture based, if looked at in terms of its employment pattern. Nearly two-thirds of employment opportunities are found in agriculture(62.6 per cent). Services – including private and public services – come second with a share of 30.9 per cent. Industry, which includes mining, manufacturing, water, electricity and construction, covers 6.5 per cent of jobs(Figure 10). If we break down the secondary and tertiary sectors into branches and also take into account the gender perspective, a variety of issues arise(Figure 11): – Agriculture is male dominated. Whereas 71.4 per cent of men engage in agriculture for their livelihood, just over half of women(51.1 per cent) are doing the same. If we consider further that more men than women can be found in informal employment, the difference between male and female informal employment in agriculture widens further. – In addition to agriculture, two other branches are male dominated: construction and transportation. They are less important in terms of number of jobs, but the gender disparity is striking. Construction provides four times more employment to men than women; transport can even be treated as a male monopoly in that there are twenty times as many male workers as female ones. Informal employment also has areas of female domination. This is particularly the case with trade, accommodation and food services. Our data do not allow us to separate trade into wholesale and retail trading, but trading in general is predominantly female. Catering is for women what transport is for men: a gender exclusive affair. It can be assumed that extending home-cooking to“external” clients provides an easy entry into informal employment for many women. 5  DISTRIBUTION Some introductory explanations are needed before we discuss the peculiarities of income, inequality and poverty. In our survey we asked respondents about their income in various ways. First, we asked the head of household or the most knowledgeable person about the total income of the household, asking them to consider the income of all members of the household and to add it up. Second, we asked the person selected for the second part of the interview, which could be the head of household or any other randomly selected member, to inform us about their individual income. We did not differentiate between sources of income (labour or other) but wanted to know the total income. Income was meant to be disposable income, namely after tax and any other contributions were deducted. Third, we did not ask respondents to provide the exact amount of their income, but to carry out a self-classification and select their income class from a prepared list. For 16

Informal employment in Ethiopia Figure 10 Informal employment, by sector and gender 80% 70% 71.4 62.6 60% 51.1 50% 40% 30% 20% 10% 0 Agriculture N(all)= 2256; N(male)=1277; N(female)= 979. 44.2 30.9 20.8 6.5 7.8 4.7 Industries All Male Female Services Figure 11 Informal employment, by branch and gender Agriculture 3.4 Industries 3.3 (without construction) 3.5 Construction 3.0 4.5 1.1 Trade 12.9 7.4 Transport 1.2 2.0 0.1 Accomodation and food services 5.4 1.9 10.0 Other services 11.4 9.5 13.9 20.2 0% 10% 20% All 30% 40% Male 62.6 71.4 51.1 50% Female 60% 70% 80% 17

FRIEDRICH-EBERT-STIFTUNG – WORKING IN THE SHADOWS surveys in other countries, we used the statutory minimum wage as reference and established income clusters as multiples of the minimum wage. This procedure was not feasible for Ethiopia as the country does not have a statutory minimum wage. 14 Instead, we selected a monthly income of 1,500 BIRR as a reference and established multiples of it with the following six income clusters: 0;<750; 750–1,500; >1,500–3,000;>3,000–12,000;>12,000. Looking at income in terms of income clusters comes with the disadvantage that it is not possible to directly calculate average income or inequality coefficients such as the Gini rate. To overcome this problem, we substituted the breadth of an income cluster with the middle income of the cluster. This method blurs the data to some extent as it implies that all persons in the same income cluster earn exactly the same (middle) income and therefore it ignores inequality within a specific cluster. The use of the middle income, however, is still a good approximation and provides meaningful insights with only low error. 14 There is an intense discussion in Ethiopia on the need for a statutory minimum wage and the Labour Proclamation of 2019 in§55, 2 states that“A Regulation of the Council of Ministers shall determine the powers and responsibilities of a Wage Board which shall comprise representatives of the Government, employees and trade unions together with other stakeholders that will periodically revise minimum wages based on studies which take into account the country’s economic development, labour market and other considerations.” 5.1  INEQUALITY WITHIN INFORMAL EMPLOYMENT Grouped by income clusters, Figure 12 provides an overview of the distribution of individual incomes within informal employment at the end of 2020. Nearly three-quarters(72.5 per cent) of people in informal employment earn a monthly income below 1,500 ETB, most of them(39.7 per cent) even less than 750 ETB. Just above a quarter(27.5 per cent) belong to the better-off income clusters of 1,500 ETB and more, and 7.9 per cent earn above 3,000 ETB. We have no similar income data for the formal economy and therefore cannot directly compare inequality between different segments of the economy. We can, however, state that inequality among people in informal employment is enormous. The Gini coefficient is 0.52. If we allow some adjustment to take into account the method of calculation, the Gini hovers between 0.50 and 0.55. Inequality is substantial; large groups barely manage to earn a decent living, but some enjoy decent, if not high incomes. Household income, which cumulates the income of all members, displays an even more unequal pattern. Two-thirds (67.1 per cent) of households have to be content with a monthly income of 1,500 ETB or less, most of them again (37.3 per cent) with less than half. At the other end, the picture is a bit rosier. The share of households above the inFigure 12 Monthly individual income in informal employment(in ETB) 50% 40% 39.7 32.8 30% 20% 19.5 10% 7.6 0% < 750 750–1,500> 1,500–3,000> 3,000–12,000 0.4 > 12,000 N=2236; GINI= 0.52. Note: Question directed to selected persons. Question:“Can you give me an estimate of your average monthly income in ETB(Note: after paying tax and after any contributions were deducted)[Read out options]: 0/> 0– <750/ 750–1500/>1500–3000/ > 3000–12000/> 12000. 18

Informal employment in Ethiopia Figure 13 Monthly household income in informal employment(in ETB) 40% 37.3 30% 29.8 21.4 20% 10% 9.8 0% < 750 750–1,500>1,500–3,000>3,000–12,000 1.4 >12,000–50,000 0.3 > 50,000 N=2313; GINI= 0.62. Note: Question directed to head of household or most knowledgeable person. Question:“Can you give me an estimate of your average household income in the last four weeks? Note: after paying tax and after any contributions were deducted)[Read out options]: 0/> 0– < 750/ 750–1500/ > 1500–3000/> 3000–12000/> 12000–50000/> 50000.” come threshold of 3,000 ETB increases to 11.5 per cent. A tiny number of households(0.3 per cent) manages to take home more than 50,000 ETB a month(Figure 13). Inequality between households is even worse than inequality between individual income earners(see Figure 13). At household level, the Gini coefficient increases to 0.62, substantially higher than inequality at individual level. The reason for this is obvious. When better-off households send a second or third member to engage in income-earning activities, they are much more“successful” than poor households. Members of poor households are trapped in a situation in which additional work scarcely provides them with more income. This is typical of economies in which not labour supply but aggregate monetary demand is the limiting factor for income generation. 5.2  ETWEEN EXTREME POVERTY AND MIDDLE-INCOME CLASS What can we make of these findings with regard to poverty and social class analysis? The World Bank provides a classification, distinguishing between“extreme poverty”,“moderate poverty”,“lower middle-income” and“higher middle-income”. The categories are listed in Table 3 and transposed to current monthly earnings in local currency with the help of the purchasing power parity calculator. Unfortunately, it is not possible to directly apply the World Bank poverty lines for assessing poverty groups in Ethiopia to the FES survey data on informal employment. The World Bank rates are based on household per capita income, but our approach of inquiring about household income by inTable 3 Poverty based on international poverty lines Poverty lines Extreme poverty Moderate poverty Lower middle-income class Upper middle-income class Income range $(per person, per day; 2011 PPP)* 0.0–1.90 1.91–3.20 3.21–5.50 > 5.50 Note:*PPP conversion rate for Ethiopia 2020= 34.9271654. Daily rates are transferred to monthly rates by the factor 30.5. 19 Income range ETB(monthly) 0–2,024 2,025–3,409 3,410–5,859 > 5,859

FRIEDRICH-EBERT-STIFTUNG – WORKING IN THE SHADOWS come clusters generates difficulties in calculating income per capita. In our survey, the average size of a household is 5.02 members. An educated guess would allocate most households with an aggregate monthly income of 3,000 ETB or less to the category“extreme poverty”, while many in the income range of 3,000–12,000 ETB would be assigned to the“moderate poverty”. Only about 2 per cent of households appear to qualify for the lower or upper middle-income class. Our income survey was conducted at the end of 2020 and reflects the earning pattern in the second half of the first pandemic year. It is difficult to know the extent to which income was still depressed from restrictive policy measures to contain the spread of the virus or whether employment opportunities had already returned to pre-Covid levels. Our own data point to a substantial increase in poverty during the pandemic, in particular in urban areas(Strupat 2022: 43– 44), but fall short of conclusions on inequality. Former nonpoor households were pulled into poverty and the pandemic policies produced a“new poor”. But the general trend in up- and downward movement of households is not documented and it remains unclear whether income depression affected all income groups in the same manner or whether some segments managed to shield themselves or even benefited during the crisis. With this uncertainty in mind, we abstain from applying a“correction” measure in response to the circumstances during the survey period and avoid conclusions on changes in inequality during the pandemic. A closing note on the informal labour force and belonging to the middle class is appropriate. While the total of 2 per cent of households qualifying as middle-income appears to be miniscule, we have to relate this number to the structure of the labour market. If formal employment accounts for 15 per cent and informal employment for 85 per cent of total employment, 2 per cent of informal employment earning an income at middle-class level or higher would be 1.7 (0.85*2) per cent of total employment. To get an idea of the extent to which people in formal employment belong to the middle-income class, we have to speculate about income within formal employment. If for the sake of argument we assume that 50 per cent of formal households, or 7.5 (0.15*50) per cent of total employment, have an aggregate monthly income of 12,000 ETB, which we previously defined as entry level, the middle-income class would be composed as follows: members in formal employment amount to 81.5(7.5/9.2) per cent vs those in informal employment who amount to 18.5(1.7/9.2) per cent.Admittedly, this is no more than a back-of-the-envelope calculation, but it highlights the point that the upper end of the income hierarchy in informal employment reaches deep into the middle-income class. Informal employment is not synonymous with poverty, but may also allow many to move up the income ladder in the country. 6 GENDER INEQUALITY A key determinant of income inequality is the gender income gap. The mean income of women is substantially less than what men earn. Taking all income categories within informal employment together, men on average earn 32 per cent more than women(Figure 14). Why do women earn less? The first reason becomes visible when looking at working time. On average, women spent 34.84 weekly working hours in their main informal job, Figure 14 Gender income gap in informal employment 1.40 1.32 1.20 1.00 1.00 0.80 0.60 0.40 0.20 0.00 mean monthly income N(male)= 1266; N(female)= 970. 1.11 1.00 working hours in main job Male Female 20 1.19 1.00 hourly income

Informal employment in Ethiopia while men work for 38.67 hours, or 11 per cent more. It should be considered normal that more work generates higher income. If we correct the gender pay gap by differences in working time, we get a net gender pay gap of 19 per cent(Figure 14) which is caused not by working time differences but by other reasons. against women in income matters. In our survey, we did not inquire into the volume of unpaid work and therefore cannot quantify it. We can, however, assume that the extent of unpaid working time is considerable and gender inequality is a factor that goes beyond what we measure with income indices. Differences in working time raise the question of why women should spend less time in informal employment and thereby“forgo” opportunities to earn more income? We have no reason to believe that the total working day for women is shorter than for men and if, for the sake of the argument, we just assume similar working time, we can conclude that women spend less working time earning income because they spend more working time in unpaid activities. What kind of working time could be devoted to activities which do not earn income? The first we should mention is subsistence work in production. This is particularly the case in agriculture, including cultivating plants in gardens, where food is produced for home consumption. This is further the case for domestic work, in particular cooking and cleaning. Maybe the largest part of unpaid work is devoted to care work. This includes child care, all levels of teaching and health care of all types, such as caring for the elderly. Unpaid care work can be seen as a factor that discriminates But where does the remaining 19 per cent income difference per working time unit come from. Two explanations suggest themselves: women are allocated to lower-paying economic sectors, or women do the same jobs as men but receive lower pay. “Same job but not the same pay” is at the heart of many gender inequality debates, but always hard to prove empirically, as many particularities have to be observed when “same work, different pay” studies are undertaken, such as working time, qualifications, job description and so on. On a more aggregate level, however, we can compare income levels by economic sector to see whether male dominated branches provide better income opportunities than branches with a large female share. The findings on income differences by branch are summarised in Figure 15. The data should be read in conjunction with the employment distribution by branch shown in ­Figure 11. Figure 15  average individual income, by branch and gender(ETB) 3,800 3,400 3,500 3,000 2,600 2,200 2,250 1,800 1,200 1,456 1,166 800 1.25 400 1,313 1.71 2,831 1,602 1,800 2.18 1.57 3,029 0 Agriculture N(male)= 1258; N(female)= 946. Industries Construction Trade Transport Male Female Male/Female 21 10.00 9.00 8.00 7.00 6.00 5.00 1,297 1,188 4.00 3.00 2.00 1.00 0.92 0.00 Accomodation and food services

FRIEDRICH-EBERT-STIFTUNG – WORKING IN THE SHADOWS – Gender income inequality is a general trend across the whole informal economy. Accommodation and food services is the only branch in which male and female income is at parity. In all other branches, male income surpasses female income considerably. – Norms appear to be in place that promote male employment and impede female employment in better-earning branches. Branches with the highest average income (construction and transport) are male-­dominated. The branch with the lowest income potential(accommodation and food services) is female-dominated. – Agriculture should be discussed separately from employment and income opportunities in other branches. Income distribution usually depends on access to land, which is often influenced by traditional inheritance. Our data suggest that occupational segregation exists which directs female employment into lower-earning segments of the economy, while keeping better-earning opportunities for men. The data are too highly aggregated to prove the case, however. But they do confirm that gender income inequality is a feature of most sectors of the economy. More research is needed to deepen our understanding of how income inequality and employment patterns interact. 7  IVING OUTSIDE SOCIAL SECURITY COVERAGE When individuals or households have low or no income, they may qualify for government support programmes or receive assistance from relatives and community-based sources. Individuals or households may benefit from government programmes such as cash transfers or food assistance; or rely on support from neighbours, relatives, kinship groups or other groups. In this section, we look at the various sources of solidarity and try to establish the extent to which they make a difference. Ethiopia has policies and legislation in place to anchor the provision of social protection. The Ethiopian Constitution guarantees all Ethiopians access to social security[Article 90(1)]. Programmes have been set up with the aim of extending social protection coverage. They differ according to social instruments, regional coverage and target groups and include urban and rural productive safety net programmes and the community-based health insurance scheme(CBHI). Of particular importance are the Rural Productive Safety Net Project(RPSNP), set up in 2005 and directed to drought-prone regions, and the Urban Productive Safety Net Project(UPSNP) established in 2016. They combine cash transfers to the poor with public works or direct help for special groups. The RPSNP is claimed currently to provide social assistance to 8 million beneficiaries, while the UPSNP provides social assistance to 0.6 million people in 11 cities(see: UNICEF; ILO Decent work; Bischeler et al. 2021). Impressive as these Figures may appear, public expenditure on social protection is low. In 2017, the latest year for which information on national coverage is available, aggregate social protection expenditure(excluding health) was a mere 0.65 per cent of GDP. Referring to the whole population, only 7.4 per cent 15 were covered by at least one social protection benefit(see ILO World Social Protection Data Dashboard: Ethiopia). When the first Covid-19 cases were detected in March 2020, the government declared a state of emergency which lasted five months from April until September. The government set up a social protection response to the effects of Covid-19, which included additional food distribution, the waiving of utility bills and a prohibition on companies laying off staff during the emergency. 16 Compared with the level of need, the social protection response to the pandemic was limited. Ethiopia failed to set up new programmes and instead of expanding horizontally to reach vulnerable households in hitherto untouched regions, it expanded the existing RPSNP and UPSNP vertically and improved support to some already covered households. The main constraint in providing a more extensive social protection response was the lack of financing. Assessing its impact, Bischeler et al.(2021: 36) conclude that“as a result, it is unlikely that Ethiopia’s social protection response offset any significant proportion of the estimated impact of COVID-19 on poverty”. WHO BENEFITS FROM GOVERNMENT AID POLICIES? To ascertain whether households in informal employment in Ethiopia actually benefitted from government aid we asked about inclusion in various social assistance programmes. The analysis of the answers, however, faces several challenges. Households may benefit from social assistance but not be clear about which“support line” they are receiving. They may further not be aware of the identity of the benefactor(government, church, NGO, other). And they may give rise to double-counting, if they state they are a beneficiary of the same items under several headings.“Free medical services” posed a further possibility of misunderstanding, when people do not count the payment of a premium to the community-based health insurance scheme as medical costs and call it“free” because they did not pay extra for it.“Free medical services” which was provided for in the questionnaire is ignored in the evaluation. 15 If we relate the 8.6 million beneficiaries of the RPSNP and UPSNP to the total population of some 112 million, we get coverage of some 7.7 per cent. The ILO calculation appears to take its guidance from these Figures. 16 The Covid-19 shock came on top of other shocks, such as a locust invasion and severe floods, all of which aggravated income pressure on many households. 22

Informal employment in Ethiopia Figure 16  aid coverage by selected assistance programmes 8% 7% 6.9 6% 5.7 5% 4% 3% 2% 1% 0% N= 2332. Cash transfer (PSNP) Food aid (community-based nutrition-CNB) 4.3 Cash for public works (PSNP) 6.3 2.4 Food for work (PSNP) School feeding Figure 17 Beneficiaries of social assistance programmes 100% 90% 80.6 80% 70% 60% 50% 40% 30% 20% 19.4 10% 0 No social assistance Social assistance support 13.9 Single social assistance programme 5.4 Several social assistance programmes Note: Figure excludes those who claim to have received free health services. With these caveats in mind, Figure 16 list the key social assistance programmes from which households in informal employment benefit. Following our classification, Ethiopia does not have a dominant support programme but offers a variety of support lines, even though most are linked to the PSNP. Most important are food aid(for 6.3 per cent of households), food-for-work(6.3 per cent) and cash transfers (5.7 per cent). Public works(4.3 per cent) and school meals (2.4 per cent) show significant coverage as well. The rates of the various programmes cannot be added together to get total coverage as households may benefit from various lines. To approximate total coverage, we distinguish between households that share in only one support programme and households that benefit from several. Figure 17 summarises the results. Overall, four out of five households(80.6 per cent) do not benefit from any social assistance programme. Of the remainder, 13.9 per cent benefit from a single support line, while 5.4 per cent were given access to more than one social relief measure. With the exception of school meals, other main support programmes were linked to the benefactor, in which the urban and rural PSNP(see above) featured prominently. To that extent, the bulk of the programme and the major part of the coverage may be classified as government ­assistance. 23

FRIEDRICH-EBERT-STIFTUNG – WORKING IN THE SHADOWS Social assistance coverage of 19.4 per cent for households in informal employment is higher than the coverage rate of 7.4 per cent estimated by the ILO in 2019 for the whole country. The FES data could thus be viewed as confirming the expansion of social security coverage during the pandemic. The Figures can not be compared directly, however, as the ILO refers to the total population while the FES reference is informal employment. If social security coverage is related to total informal employment of some 48 million, we obtain a Figure of 9.3 million beneficiaries. This is slightly above the RPSNP and UPSNP target of 8.6 million for the pre-Covid time. The marginal increase of only 0.7 million for 2020 would confirm what Bischeler et al. noted in their assessment of the pandemic-relevance of social security in Ethiopia. The government continued their pre-Covid programmes without major adjustments and only added a few vulnerable households. In terms of social security coverage, 2020 was treated more as a pandemic-free year than as a time of considerable income shocks. 8  SUMMARY The economy of Ethiopia is predominantly shaped by informal employment, which accounts for some 85 per cent of total employment. Income, working conditions and structural features of the informal work force were investigated in a countrywide, representative survey of 2,400 households, which was conducted in October and November 2020 by the Friedrich-Ebert-Stiftung(FES) in collaboration with the German Institute of Development and Sustainability(IDOS) and the International Labour Organization(ILO) as part of a six-country survey project. Framed in terms of a uniform technical protocol to allow cross-country comparison, the findings on Ethiopia can be related to the outcome in the reference group, which is made up of Kenya, Zambia, Senegal, Benin and Cote d’Ivoire. In general, informal employment in Ethiopia features the same characteristics which are predominant in the other countries. Labour and capital are not(yet) disjoined. Self-employment is by far the prevalent form of earning an income and employer–employee relations come a distant second. Where they exist, they feature microenterprises in which the owner is worker-employer with very little add-on wage labour. 0.60, depending on whether individual or household income is referred to, highlights the point that the upper end of the income hierarchy in informal employment reaches deep into the middle-income class. For some, informal employment is the ladder by which they may move up the income hierarchy in the country. Informal employment has a gender face which is most manifest in the gender income gap. The gap is partly caused be factors which keep women in unpaid work activities and partly by lower earnings for same working time. There are indications that societal factors guide women into lower remunerated economic sectors of the informal economy, such as petty trading and food processing, but available data do not allow a detailed disaggregation into“same job, same payment” analysis. Social protection may be highly in demand but is out of reach for most. At the end of 2020, more than 80 per cent of households did not benefit from any social support programme. This hardly changed during the pandemic period. The survey confirms other studies which showed that the government added only a few vulnerable households and in terms of social security coverage, 2020 was treated more as a pandemic-free year than as a time of considerable income shocks. A note of caution is appropriate in closing. The survey in Ethiopia was conducted in October–November 2020, after the government had lifted the state of emergency for the implementation of pandemic policies. Specific negative circumstances, such as restricted market opportunities, may have still impacted on employment and income. Even though it is difficult to exactly quantify their significance, the survey team nevertheless believes that the main structural features disclosed in the survey were unaffected by the pandemic and the survey findings are a good reflection of the standard conditions of informal employment in Ethiopia. Work is characterised by high employment and income insecurity. For most, the day is not structured by standard working time(35–40 hrs per week) but by conditions of underemployment, if not marginal employment and for others by overemployment to the point of excess. Employment is the result of low monetary demand for products and services, which translates into high labour competition and low income. Most households in informal employment have to be grouped as extremely income-poor but this does not mean that all informal employment can be equated with poverty. Income distribution is rather unequal, and a Gini of 0.50– 24

REFERENCES Bangasser, P.(2000): The ILO and the informal sector: an institutional history. ILO: Employment Paper, 9; available at: https://www. ilo.org/wcmsp5/groups/public/---ed_emp/documents/publication/ wcms_142295.pdf. Bischler, J., Asheber, T., Hobson, M.(2021): Towards shock-responsive social protection: lessons from the COVID-19 response in Ethiopia, Shock-­ responsive social protection responses during COVID-19. Research Report: Oxford Policy Management, Oxford, March; available at: https://www.unisdr.org/preventionweb/files/77049_towardsshockresponsivesocialprotect.pdf. ILO  Women and men in the informal economy: A statistical picture. Geneva. – (2021) : Decent work country programme for Ethiopia 2021–2025. May; available at: file:///C:/Users/User/Desktop/Traub%20(stand%2006-052019)/1%20-%20FES%20-%20ab%202016/2018-FES/2018-Umfrageprojekt%20Informal%20Economy/2023-WS%20Ethiopia%20&%20 Reports/Population-Labour%20data/2022-ILO-Ethiopia-decent%20 work%20programm.pdf.  Social Protection Data Dashboard: Ethiopia, available at: https://www.social-protection.org/gimi/WSPDB.action?id=13. Strupat, C.(2022): Economic Effects of the COVID-19 Pandemic on the Informal Economy: Who has Fallen into Poverty, in: Traub-Merz, R./Öhm, M./Leininger, J./Bonnet, F./Maihack, H.(eds): A Majority Working in The Shadows. A six-country opinion survey on informal labour in sub-Saharan Africa. September. Traub-Merz, R./Öhm, M./Leininger, J./Bonnet, F./Maihack, H. (2022): A Majority Working In The Shadows. A six-country opinion survey on informal labour in sub-Saharan Africa. September. UNICEF(n.d.): Ethiopia. Social Protection; available at: https://www. unicef.org/ethiopia/sites/unicef.org.ethiopia/files/2020-01/National_­ Social_Protection_Budget_Brief_2017_18.pdf. UN(2020): United Nations Ethiopia, SOCIO – ECONOMIC IMPACT of COVID-19 in Ethiopia. Addis Ababa, May; available at: file:///C:/Users/ User/Desktop/Traub%20(stand%2006-05-2019)/1%20-%20FES%20 -%20ab%202016/2018-FES/2018-Umfrageprojekt%20Informal%20 Economy/2023-WS%20Ethiopia%20&%20Reports/Reports-Statistics/2022-Ethiopia-UN-Socio-Economic-Impact-Assessment-FINAL.pdf. Informal employment in Ethiopia 25

FRIEDRICH-EBERT-STIFTUNG – WORKING IN THE SHADOWS IS ETHIOPIA’S INFORMAL ECONOMY ORGANISED AND, IF SO, HOW? Jürgen Schwettmann The habit of share and share alike is easily understandable in a community where everyone is likely to find himself in difficulties from time to time, for it is scarcity not sufficiency that make people generous, since everybody is thereby ensured against hunger. He who is in need today receives help from him who may be in need tomorrow.  Evans-Pritchard(1940, p. 210), regarding his studies of the Nuer people in Sudan. Life in Ethiopia is dependent on your neighbours and friends; you must live a social life or Mahiberawi nuro(literally meaning“collective life”), and do so through membership in various social associations(e. g. solidarity groups, redistributive groups, work groups, religious groups, funeral associations). When you yourself or a member of your family becomes sick or die your association will help you manage socially, financially and morally … they are your social security. (Kebede, 2018, p. 214) Collective action has intrinsic value. Being part of a group and participating in meeting common objectives provide direct benefits to individuals. In addition, collective action has an instrumental value; it can help individuals, groups, and communities achieve common goals. In Ethiopia, a study found that one form of collective action, Iddir, or burial societies, helped households in the study areas attenuate the impact of illness. (IFPRI/CAPRI, 2010, p. 171) 1  URPOSE OF THIS CHAPTER The survey-cum-opinion poll on the characteristics and perceptions of informal economy actors in six African countries. 17 included a series of questions that sought to explore whether and how informal economy actors are organised into groups, associations, cooperatives, or other forms of organisation. The present chapter presents the findings of this particular element of the survey, and contextualises them in the context of Ethiopia’s informal economy. The chapter includes extant academic and“grey” literature on the subject, and discusses them in the context of the FES survey findings. 17 Benin, Cote d’Ivoire, Ethiopia, Kenya, Senegal, Zambia. 2  ECONOMY ORGANISATIONS: TYPES AND PURPOSE 2.1  HAT KIND OF ORGANISATIONS? In sub-Saharan Africa during the 1980s and 1990s, many informal market associations, credit associations, trade organisations, cooperatives, farming and marketing groups, ethnically based development associations, and other societies emerged to give it a political voice, especially at the local level(Tripp, 2001). This development continues until today, also in Ethiopia(Kelkil, 2015). This section provides an overview of organisational diversity in Ethiopia’s informal economy. 26

Is Ethiopia’s informal economy organised and, if so, how?  social and solidarity economy as an ­umbrella concept economic system. Figure 18 shows the rapid growth of academic literature on the subject. The many different types of organisation that exist in the informal economy – be it in Ethiopia or elsewhere – share certain characteristics, values and principles, which are those of the“social and solidarity economy”(SSE). The 110 th International Labour Conference(ILC, 2022) adopted the following definition of the term“social and solidarity economy”:  The SSE encompasses enterprises, organisations and other entities that are engaged in economic, social, and environmental activities to serve the collective and/or general interest, which are based on the principles of voluntary cooperation and mutual aid, democratic and/ or participatory governance, autonomy and independence, and the primacy of people and social purpose over capital in the distribution and use of surpluses and/ or profits as well as assets.[…]. According to national circumstances, the SSE includes cooperatives, associations, mutual societies, foundations, social enterprises, self-help groups and other entities operating in accordance with the values and principles of the SSE. (ILO, 2022) While the social and solidarity economy, as an umbrella concept that brings together various forms of member-based, democratically controlled organisations, is relatively recent, its values and principles have been applied for centuries. The term“social and solidarity economy” has gained in popularity since the global financial crisis in 2007/08, which was caused by the failures of a neoliberal Since 2015, some 30 countries around the world have adopted framework laws related to the social and solidarity economy(Socioeco, 2022), and the African Union recently adopted a ten-year(2023–2032) SSE Strategy for Africa (ILO, 2022). The OECD recently issued a recommendation on the social and solidarity economy, while the UN General Assembly adopted a resolution on the subject in April 2023. With the exception of social movements, the various organisational types introduced in this section observe the value and principles contained in the above definition and could therefore be considered part of the Ethiopian social and solidarity economy(even though the term is not yet common in the country). 2.1.2  Indigenous forms of cooperation in Ethiopia In Ethiopia, traditional civil society organisations continue to play an important role in communal life and grassroots collective activities. They are also known for promoting a culture of self-help and social solidarity during emergencies, food crises, conflict and political instability, most notably in and around peripheral areas and marginalised rural and urban communities(Kelkil, 2015). This section briefly introduces the main types of such indigenous organisations in Ethiopia, which are known under different denominations. 18 18 Table adapted from (Kelkil, 2015). Figure 18  articles on the social and solidarity economy 12,000 10,000 8,000 6,000 4,000 2,000 0 1990 1995 2000 2005 2010 2015 Social and solidarity economy Economía social y solidaria Economie Sociale et Solidaire Total 2020 27

FRIEDRICH-EBERT-STIFTUNG – WORKING IN THE SHADOWS Table 4 Indigenous organisations in Ethiopian languages Type Labour cooperation Mutual support and funeral insurance Social welfare and assistance Oromo Dabbo Dabare Eeba, Ogdi Kambatta Gezma Hera Gogotta Amhara Wonfel Iddir Mahiber Gurage Gaze Iddir Kicha Sidama Dabo Iddir The Iddir(or Edir) are burial societies that are almost ubiquitous throughout Ethiopia(Aredo, 1993). They organise burial ceremonies and offer financial, material and moral assistance to the surviving relatives. Iddir obtain their legal personality from the Ministry of Justice by paying a registration fee(Veerakumaran, 2007), meaning that Iddirs must be considered to belong to the formal sector. It was also reported that substantial formalities were observed in the operations of Iddirs. Regular meetings, the keeping of minutes of meetings, and the establishment of rules to govern how funds will be managed(collections from members and disbursements to those who suffer shocks), and how fines will be imposed are all examples of formalisation(Howlet, et al., 2021). In addition to their primary function as mutual insurance, Iddirs are places in which women share experiences and discuss issues of pressing concern. Iddirs provide opportunities for social interaction, risk sharing, dispute resolution, information dissemination, reciprocity and mutual assistance(Teshome, et al., 2014). Of the many types of Iddir, which may be aligned with ethnicity, sex, age, workplace or residence, the most dominant is the community Iddir, which is formed by households living in the same neighbourhood(Yitbarek, 2008). These Iddirs, therefore, could be considered neighbourhood groups. Iddirs are not necessarily organisations of the poor, as has been found by relevant studies. 19 The functions of Iddirs are not limited to the provision of insurance and psychological support for members. They are often involved in community development programmes, such as the construction of roads and schools, as well as the installation of public utilities. They have also facilitated the introduction and expansion of community-based health insurance(CBHI) schemes(Teshome, et al., 2014). Indeed, it was reported that the promoters of such schemes used the Iddirs(and Eqqubs) as examples to explain their own func19 For example, a survey of 1,033 farm households in the major coffee-growing areas of Ethiopia showed that, of the very poor households, 71 per cent reported membership of Iddir as compared with 85 per cent for the rich and 95 per cent for the very rich(Aredo, 2010). According to the World Values Survey of Ethiopia carried out in 2007, 60.6 per cent of the sample of 1,500 households said they were members of an Iddir(Pankhurst, 2008). Other sources report a participation rate in Iddirs ranging from 50 to 90 per cent throughout Ethiopia, while the latest financial inclusion report(based on a sample of 6,770 rural and urban households) indicates a 50 per cent participation rate(World Bank, 2021). tioning, and to encourage rural households to join a scheme. In more general terms, social relations and social networks, such as Iddirs and Mahibers[see below], are essential for the creation and sharing of knowledge among farmers, and offer a more pluralistic approach to agricultural extension as well as facilitating farmers’ access to extension agencies (Gebremariam, et al., 2021). The Eqqub(or Ekub) is a rotating savings and credit association(ROSCA) whose members pay a fixed amount of money each week or month into a common fund and then cast lots to decide who should take the collected money. 20 Someone who has won the lottery once cannot do so again until all members have had their turn. Eqqubs provide an ideal network for collecting cash for entrepreneurs requiring capital for business start-up or expansion. The Eqqub system indeed operates on the basis of mutual trust and confidence, and members mostly share a common bond determined by family ties, ethnic origin, social class or professional occupation(Kebede, 2018). ROSCAs similar to the Eqqub system exist in virtually all African countries, as well as in Asia, the Caribbean and many other parts of the world; they are widespread in Ethiopia as well. 21 Debo / Wenfal / Lefenty are forms of indigenous cooperation and mutual assistance that exist mostly in rural areas. Debo is a system of farmer’s cooperation during the period of farming, weeding, harvesting, threshing and house construction and so on, as they exist in many African countries (and elsewhere). These groups do not have a formal system of administration; they provide a mechanism whereby all farmers help each other on a rotating basis to overcome seasonal labour peaks, thereby saving time and increasing labour productivity. 20 Eqqubs generate a“Pareto improvement” in that the pooling of resources reduces“waiting” time before purchase for all participants except the one who is last to collect the kitty(who nevertheless does not have to wait longer than if he had saved alone)(Aredo, 1993, p. 11). 21 Based on a sample of 1,500 rural households in three regions(Oromia, Amhara, SNNPR) the penetration rate of Eqqubs in rural areas was estimated at 24.1 per cent(Kedir, 2005). A study carried out among 102 manufacturing SMEs in Addis Ababa found that 34 per cent of them had relied on Eqqubs to obtain loans(Fanta, 2015). The World Bank financial inclusion report(2021) reports that 63.1 per cent of urban households and 29.4 per cent of rural households do hold savings. Of those, 52.6 per cent put their savings in informal. 28

Is Ethiopia’s informal economy organised and, if so, how? Indigenous forms of cooperation continue to flourish in Ethiopia today, where they co-exist with more formal cooperative or mutual organisations. They are also of great importance to the Ethiopian informal economy because informal entrepreneurs greatly depend on the presence of densely knit family and kinship ties(Kebede, 2018). Traditional organisations are explicitly excluded from the provisions of Ethiopia’s Civil Society Proclamation N° 1113/2019, and therefore are not to be confused with civil society organisations. 2.1.3  Faith-based and neighbourhood groups These types of groups, which are prevalent in Ethiopia, share many features, and are often related to Iddirs and Eqqubs and, in some cases, to the defunct peasant associations. 22 There is thus a considerable amount of overlap between the indigenous forms of cooperation and the faithbased and neighbourhood groups in Ethiopia, in terms of both membership and the services they render. A religious group is, as its name implies, a group formed by members of the same faith who also often share additional common bonds such as gender, location or age group. The main function of these groups is the organisation of collective worship, often followed by a social gathering. In addition, many religious groups operate social assistance funds, or organise care and support should a member fall sick. In many African countries, religious groups play a key role in providing health care and in extending social protection to households in the informal economy. In Ethiopia, religious associations are known as Mahiber(or Senbete), which have been described as being akin to Iddirs(Aredo, 2010). They are particularly widespread among the followers of the Orthodox religion(Habtu, 2012). The four main faith-based organisations in Ethiopia are: the Ethiopian Evangelical Church Mekane Yesus(EECMY), the Ethiopian Orthodox Church(EOC), the Ethiopian Catholic Secretariat (ECS) and the Ethiopian Islamic Affairs Supreme Council (ELASC). Neighbourhood associations are probably the most difficult to define because they exist under many different names(farmers’ association, village association, urban market association, water users’ association), and serve many different purposes, including solid waste management, community development, mutual social assistance, protection against criminals, involvement in local politics, self-regulation of markets, collective farm work such as land clearing, information-sharing). The collective identity among members is central to these associations because the neighbourhood(which could be a street, a village, a quarter of a town) serves as the common bond uniting its members. 22 Peasant associations were introduced in 1975 by the Derg regime. By 1990 there were more than 20,000 of them throughout the country, where they represented the lowest level of government administration which operated under the authority of the state and the(now abolished) All-Ethiopia Peasants’ Association(AEPA)(Berry, et al., 1991). Figure 19 is an attempt to illustrate the overlap and relationships that exist between the various forms of indigenous organisation in Ethiopia. Iddirs and Eqqubs can derive their common bond both from their location and from the shared faith that unites their members(plus other factors). Figure 19  overlaps and relationships Neighbourhood associations Eqqubs Iddirs Faith-based associations 29

FRIEDRICH-EBERT-STIFTUNG – WORKING IN THE SHADOWS Many religious groups are de facto also neighbourhood groups. And in some cases, a certain group may encapsulate all four identities. 2.1.4 Cooperatives Cooperatives correspond to a universal definition agreed upon by the International Cooperative Alliance(ICA) in 1995 (and confirmed by ILO Recommendation 193 of 2002): “The term ›cooperative‹ means an autonomous association of persons united voluntarily to meet their common economic, social and cultural needs and aspirations through a jointly owned and democratically controlled enterprise.” Whereas in most African countries the national cooperative movements are still strongly influenced by the cooperative traditions of the former colonial powers, principally the British and the French administrations(Schwettmann, 2022), Ethiopia, as a sovereign nation, was able to develop its own cooperative variant, which, while being inspired by European models, was not imposed from the outside, as in most other countries of sub-Saharan Africa. The history of formal or modern types of cooperatives in Ethiopia started during the era of Emperor Haile Selassie I between 1950 and 1974. The first cooperative law was promulgated in 1960, initially meant for poor farmers only, followed by a Cooperative Proclamation issued in 1966, which embraced the universal principles of cooperation and catered for any type of cooperative. The growth of the movement was slow; in 1974, when the Derg overthrew the imperial regime, Ethiopia counted just 149 cooperatives(multipurpose, savings and credit, consumer, handicraft)(Bezabih, 2009). The Derg government(1975–1991) in Ethiopia established an extensive network of socialist agricultural cooperatives throughout the country to organise the peasants, control agricultural prices, levy taxes and extend government control to the local level. These cooperatives were characterised by mandatory membership, quotas for grain to be delivered to the government, and boards of directors and managers appointed by the ruling party. Farmers came to view these as synonymous with government oppression(Veerakumaran, 2007). Cooperatives were controlled by five different sectoral ministries and by the National Bank of Ethiopia. By the end of the Derg regime the country counted some 10,524 primary cooperatives, with 4,529,259“members”, most of which were dismantled immediately after the overthrow of the Derg regime in 1991. The subsequent government embarked upon a massive cooperative development programme which was grounded in the universal cooperative principles, and saw the numbers of primary societies grow from 7,366 in 1991 to 24,167 in 2007 and 92,755 in 2021. According to official statistics, total membership in cooperatives amounted to 21,043,370 in 2021; there are also 131 cooperative unions, but no national apex cooperative organisation(Bezabih, 2009; ICA Africa, 2021). These cooperatives are engaged in 36 different types of activities, including grain marketing, coffee marketing, saving and credit services, vegetable and fruit production and marketing, dairy production and marketing and livestock marketing, among others. More than three-quarters of all cooperatives are concentrated in three regions, namely Addis Ababa(43 per cent), Oromia(19 per cent) and Amhara(15 per cent)(Lemma, 2008). The cooperatives are regulated by Cooperative Societies Proclamation No. 985/2016 and regulated by the Federal Cooperative Agency, as well as by a Cooperative Advisory Board. Cooperative promotion agencies exist also at the regional and zonal( Woreda) levels. Several Ethiopian universities, such as Mekelle and Haramaya University, have cooperative development departments. Cooperatives play a significant role in Ethiopia’s economic development, in particular with regard to the marketing of agricultural produce. A case in point is the Oromia Coffee Farmers Cooperative Union(OCFCU), which consists of 407 primary cooperatives representing 557,186 small coffee farmers. The OCFCU exports organic coffee through fair trade arrangements; the fair-trade premium then finances rural development projects in Oromia, such as schools, water supply systems and dispensaries. The OCFCU has established its own bank, the Cooperative Bank of Oromia, which now has more than 10 million account holders, as well as a cooperative insurance company. 2.1.5  Credit unions Although credit unions, also known as savings and credit cooperatives(SACCOs) are cooperative societies as well, they fall, in many countries, under a separate law, and are monitored by a different authority than“ordinary” cooperatives. In Ethiopia, credit unions were, until 1998, registered, supervised and audited by the National Bank of Ethiopia. The 1998 Cooperatives Proclamation transferred these functions to the(general) Federal Cooperative Agency(Bayabil, 2021). The development of credit unions in Ethiopia started in the mid-1960s. The first was pioneered by the employees of Ethiopian Airlines in 1964. In 1973 the then 28 SACCOs formed their own national apex body, known as Ethiopia Thrift and Cooperative Societies Ltd(ENTACCS). The Derg regime abolished ENTACCS but did not interfere with the development of SACCOs in urban areas, 23 so that by the end of Derg rule in 1991, the country had 495 SACCOs with a total membership of 119,799(Tesfamariam, 2015). Since then, the Ethiopian SACCO movement has experienced steady and substantial growth: at present, there are 21,328 primary SACCOs with 5.4 million members, and cumulated assets of 635 million USD(WOCCU, 2022). This represents around 2 per cent of the total assets of the Ethiopian banking system. Moreover, there are two and a half times more SACCOs than the current 8,944 bank branches(National Bank of Ethiopia, 2022), which means that SACCOs contribute very significantly to financial intermediation, especially in rural areas. The primary SACCOs have formed 83 regional unions, but not yet a national apex organisation. 23 Under the Derg SACCOs were not allowed in rural areas. 30

Is Ethiopia’s informal economy organised and, if so, how? The success of the SACCOs might be due to two factors:(a) their structure, membership and modus operandi is similar to that of the aforementioned Eqqubs, and(b) formal financial institutions are seldom accessible to the rural poor and the informal economy operators in cities and towns(Aredo, 1993). 2.1.6  Community-based health insurance Better health services constitute, alongside improved water supply, a top priority expressed by the respondents in the FES survey. Community involvement stands at the centre of Ethiopia’s universal health strategy, both with regard to the provision of health services and the organisation of health insurance. The latter relies on Ethiopia’s version of community-based health insurance(CBHI). Such insurance, which targets primarily the informal economy, is to be complemented by a Social Health Insurance(SHI) regime for civil servants and formal sector workers. In 2008, the Ethiopian Council of Ministers approved a Health Insurance Strategy with the aim of making CBHI and SHI schemes operational within a year. While this goal was not achieved entirely, the CBHI began scaling up in 2013, expanding from 13 pilot woredas(districts) to currently 827 woredas. However, some Ethiopian regions, such as the Somali and Gambella regions, are just about to be covered by the scheme(EHIA, 2020). In 2015, the MoH launched the Health Sector Transformation Plan(HSTP), which targeted 80 per cent health insurance coverage by 2020. While this target has been missed, Figure 20 illustrates the impressive growth of the CBHI scheme since its launch(on a pilot basis) in July 2011. Given an average household size of 4.6 in Ethiopia, the scheme is believed to cover 32 million people, or 28 per cent of the population of 115 million in 2020(UNFPA, 2021; EHIA, 2020). A similar coverage rate(28.1 per cent) was found in 2019 through a survey of 8,366 households, however with huge differences between administrative regions, ranging from 3.4 per cent in Dire Dawa to 55 per cent in Amhara  24 (Merga, et al., 2022). In comparison, health insurance coverage was just 1.25 per cent in 2011–2012(CSA, 2021). By the early 2000s, the only insurance provider was the Ethiopian Insurance Corporation(EIC), which covered a mere 0.02 per cent of the population(Lavers, 2019, p. 63). The Ethiopian version of CBHI is unusual in that it consists of a yearly contract concluded between the member and the insurance scheme, upon which the member makes an annual advance payment of the insurance premium. Existing CBHI members are expected to renew their contract before its expiry. The renewal rate rose from 54 per cent in 2015 to 82 per cent in 2020, indicating a high degree of satisfaction with the health insurance scheme(EHIA, 2020). 24 This difference is partly explained by the delay in implementation of the CBHI scheme in regions of lower coverage. Figure 20  Ethiopia: Member Households 8,000,000 7,000,000 6,000,000 5,000,000 4,000,000 3,000,000 2,000,000 1,000,000 0 2012 2013 2014 Paying members 2015 2016 2017 Indigent members 2018 Total 2019 2020 31

FRIEDRICH-EBERT-STIFTUNG – WORKING IN THE SHADOWS CBHI members pay a 240 birr(US$ 6.86 at the average 2020 exchange rate) annual premium per household, with additional payments for adult children. 25 This premium includes a 25 per cent federal government subsidy. Regional and woreda(district) governments meanwhile cover premiums for a small proportion of“indigent” households deemed unable to pay(Lavers, 2019). In 2020, the entire scheme mobilised 1.64 billion Ethiopian Birr, equivalent to around 47 million US$, of which 80 per cent represented member contributions. The undeniable success of Ethiopia’s community-based health insurance(CBHI) is due, in addition to the factors mentioned above, to its anchoring in traditional Iddirs(see above)(Bekerie, 2003). The functioning of CBHI at the local level resembles the modus operandi of the Iddir, while offering additional advantages, such as enlarging the risk pool from the local community to the entire nation. Government-mobilised faith-based organisations, revolving credit associations( Eqqub) and burial associations( Iddir) to promote CBHI(Lavers, 2021). In fact, it was reported that previous positive experience by community members with solidarity-based community-based organisations and associations outside of health(such as Iddirs and Eqqubs) contributed to enhanced community participation and enrolment in CBHI schemes(USAID, 2015). Such community participation is naturally limited by the need to ensure uniformity of the rules and procedures governing CBHI throughout the country. This dilemma – the desire to ensure strong community involvement, coupled with the need for centralised oversight and decision-making – are inherent in hybrid CBHI schemes, which are, however, the most effective means to achieve the goal of universal health coverage(Schwettmann, 2022). Member-based organisations in the informal economy create economic opportunity through collective action and the pooling of resources, including finance, by providing goods and services to their members, and/or to the community at large. This may include pre- and post-production services such as agricultural marketing and supply, financial intermediation through Iddirs, Eqqubs or credit unions, job creation through workers’, labour contracting and platform cooperatives, community-based initiatives to improve infrastructure or organise water and electricity supply, income generation through fair-trade arrangements, and many others. As an example, it was reported that 95 per cent of the total agricultural inputs supplied to the rural community in Ethiopia was distributed through cooperatives(Kelemu, et al., 2014). As mentioned above, Eqqubs and credit unions have provided the essential start-up capital for thousands of informal businesses, which, in the absence of creditworthiness, would not have been able to secure a loan from a commercial bank. Informal economy organisations extend social protection through social support arrangements based on the principles of mutuality and solidarity, such as community-based health insurance(see above), also known as“ mutuelles” (mutuals), as well as care-giver associations, burial societies ( Iddirs), mutual insurers, social cooperatives that foster the social inclusion of marginalised population groups, health service cooperatives, community food banks, foundations, and similar groups. The social function is of great importance to the various groups and organisations operating in Ethiopia’s informal economy, even if their primary function might be of an economic or financial nature. 2.1.7  Social movements A social movement is loosely defined as“collective action taken by a group of people to tackle social problems”(Makini, n.d.). In recent years, social movements have arisen around the world to demand democratic reforms and the abolition of military or single-party regimes(for example, the“Arab Spring”). Other social movements have called for better housing, health care, water supply and so on, either demanding action by the state, or solving the issues themselves through organised self-help. 2.2  WHAT FUNCTIONS? The collective organisations of the informal economy empower individuals and communities by providing them with voice and representation. This may happen at the local level through professional associations of artisans, motor-cycle taxi drivers, subsistence farmers or street vendors; at the national level through national federations of these primary organisations, such as cooperative apex organisations or informal sector umbrella bodies; and at the global level through international NGOs such as the International Co-operative Alliance(ICA), the International Association of Mutual Benefit Societies(AIM), or the International Domestic Workers Federation(IDWF). This societal function of informal-economy organisations could be seen as the equivalent of the role of trade unions in the formal economy. The organisations introduced above exercise four types of functions in support of their members: they create economic opportunity, they extend social protection, they empower individuals and communities, and they protect the environment. This section briefly discusses these four functions in the Ethiopian context. 25 It was reported that in 2011 the premiums amounted to about 2–3 per cent of household monthly income, see Mebratie, A. D. et al.(2019):“The impact of Ethiopia’s pilot community based health insurance scheme on healthcare utilization and cost of care”, in: Social Science& Medicine 220, pp. 112–119. Finally, informal-economy organisations contribute to environmental protection through corrective and preventive measures, including waste removal and recycling by waste pickers’ associations(ILO, 2019), community-based reforestation, cooperatively-organised water management, renewable energy generation(ILO, 2013), climate-change adaptation, and similar initiatives related to environmental protection. The Former Women Fuel Wood Carriers Association of Addis Ababa is an example of such environmental action. More than 15,000 women in Addis Ababa make their living from illegally collecting fuelwood from the protected eucalyptus grove atop Entoto Mountain. The association was es32

Is Ethiopia’s informal economy organised and, if so, how? tablished to provide its members with alternative sources of income by teaching them skills such as weaving, embroidery, knitting and various handiwork, thereby helping to protect the Eucalyptus forest on the mountain(Arefaynie, 2013). In many instances, informal-economy organisations exercise more than one function simultaneously. An Eqqub, for example, will have been established originally to accumulate financial resources, but very often operates in parallel with a social fund to assist members in emergency situations. Iddirs play a societal role, in addition to their social and financial functions. A street vendors association may carry out economic, social and societal functions at the same time. The four functions described above, which are interdependent and mutually reinforcing, empower social-and-solidarity-economy(SSE) organisations to reduce sustainable development and decent work deficits, including in the informal economy, in a manner that complements action by the state, the commercial private sector and the donor community. 3  FINDINGS AND INTERPRETATION 3.1 DENSITY AND TYPOLOGY The proportion of survey respondents who are members of a group of any type is shown in Figure 21. Two-thirds of all respondents declared that they are a member of at least one informal economy group; this is the second highest proportion among the six countries covered by the survey(Senegal came first), and considerable higher than the six-country average of 54.3 per cent. Figure 21 Organisational density 80% 10.00 70.4 70% 60% 50% 40% 30% 20% 1.81 10% 55.5 1.41 55.1 1.14 41.9 1.23 36.0 1.63 9.00 66.0 8.00 7.00 6.00 5.00 4.00 3.00 1.84 2.00 1.00 0 Senegal Zambia Kenya Organized Benin Cote d’Ivoire Groups per person 0.00 Ethiopia 33

FRIEDRICH-EBERT-STIFTUNG – WORKING IN THE SHADOWS Figure 22  membership per respondent 40% 36.0 33.2 30% 20% 19.0 10% 0% Not organized 1 group 2 groups 7.1 3 groups 2.1 4 groups 1.1 5 groups 1.5 6+ groups Many individuals are members of several groups at the same time. Multiple membership is particularly widespread in Ethiopia(on average, 1.85 groups per person 26 ); this is the highest rate of multiple membership among the six countries. 27 Figure 23 28 indicates the popularity of the different types of groups. Surprisingly, savings clubs(or ROSCAs), which are the most popular member-based organisations in the other countries covered by the survey, emerge as the least attractive in Ethiopia. This finding is not consistent with the extant literature, which affirms that between 20 and 50 per cent of all Ethiopians are members of an Eqqub(with a higher rate in urban than in rural areas). An obvious explanation for this discrepancy could be that the respondents were not familiar with the term“savings club”, and therefore preferred to tick other types of groups, in particular“neighbourhood associations”, because, as mentioned above, most Eqqubs are de facto neighbourhood associations. A similar observation can be made with regard to Iddirs. Membership of a burial society was not included as an option in the questionnaire, and so respondents might have chosen other options, such as“religious group”,“neighbourhood association” or“cultural group”. In fact, as mentioned earlier, the distinction between these types of groups and the Eqqubs and Iddirs is rather blurred. Taken together, these five types 29 involve around 70 per cent of respondents who declared being a member of any type of group. This would mean that around 26 1.88 for male participants, 1.78 for female. 27 This observation is consistent with the findings of the related literature, which reports that people may join several Iddirs or Eqqubs. Aredo(1993), for example, found that, on average, participants in such groups had joined 1.4 of them, while Pankhurst(2008) observed that between 20 and 40 per cent of households belong to more than one Iddir. 28 The total in this Figure exceeds 100 per cent because it includes multiple memberships. 29 Eqqubs; Iddirs; neighbourhood associations; religious associations; cultural groups. 46 per cent of all respondents belong to one of these types, a ratio which is consistent with the findings of the relevant extant literature. The rate of credit union membership(5.3 per cent of“organised” respondents, or 3.4 per cent of all respondents) is lower than the“penetration rate” of 8.0 per cent reported by the World Council of Credit Unions(WOCCU, 2022). However, WOCCU calculates the penetration rate by dividing the total number of credit union members by the economically active population age 15–64 years old. This age group represents 57.2 per cent of the Ethiopian population. Therefore, the penetration rate of the total population would be 4.6 per cent, a ratio close to that reflected in the survey. Regarding cooperatives, the survey noted a“cooperative density” 30 of 11.8 per cent(or 7.5 per cent when all respondents are considered), whereas the membership data cited in Section 2.1.3 suggest a density of 13.0 per cent. This difference could be due to the fact that official statistics may include credit unions and cooperatives that are dissolved, dormant or abandoned, or for which membership data are outdated. In fact, Cooperative Societies Proclamation No. 985/2016 did not require the re-registration of the then existing cooperatives, an exercise which would have automatically removed non-active cooperatives from the registry. Moreover, the official statistics may include multiple memberships(people who are members of more than one cooperative), which would artificially increase cooperative density. The rate of membership of mutual benefit groups determined by the survey is, at 2.8 per cent, surprisingly low, especially because the FES-DIE-ILO survey reports a 34.9 per cent health insurance coverage among the informally em30 Membership of cooperatives(not including credit unions) divided by the total population. 34

Is Ethiopia’s informal economy organised and, if so, how? Figure 23 Group typology Neighbourhood association Religious group Cooperative 11.8 Social movement 7.4 Credit union 5.3 Cultural group 4.6 Political party 3.0 Mutual health insurance 2.8 Professional association 1.9 Trade union 1.3 Saving clubs 1.2 Any other 4.1 0% 10% 20% 47.9 31.0 30% 40% 50% 60% ployed in Ethiopia, who are mostly, if not exclusively insured through community-based health insurance(CBHI), which is de facto mutual. We assume that most of the respondents were not familiar with the term“mutual”, which is not very common in the English-speaking world. Because CBHI works closely with Iddirs and Eqqubs, we further assume that CBHI members lie“hidden” among the members of religious groups and neighbourhood groups. Moreover, it was reported that the survey respondents considered CBHI as a government-induced scheme, not as a genuine selfhelp organisation, and therefore refused to classify it as a “mutual”. in recent years. Other prominent social movements include the Solidarity Movement for a New Ethiopia, which seeks to unite the many Ethiopian ethnic groups in their quest to hold government to account for allegations of human rights violations, or women’s rights associations such as Setaweet, Yikono or the Yellow Movement(Solés i Coll, 2020). As the survey unfortunately does not indicate to which social movement the concerned respondents belong one can only speculate that some of those who were motivated to join a group by societal/political reasons(see 3.3) are members of a social movement. According to the survey, a relatively important minority of 6 per cent of respondents with group membership confirmed that they were affiliated with a social movement in Ethiopia. This is almost double the percentage found, on average, in the six survey countries. Despite the limitations imposed by government 31 social movements are flourishing in Ethiopia. For example, among the Oromo people the issue of land rights and the lack of fair compensation for forced evictions has been an essential social mobilisation element 31 For example, the highly restrictive Charities and Societies No. 621/2009 Proclamation, which has been replaced by the much more liberal Organization of Civil Societies Proclamation No. 1113/2019, https://en.unesco.org/creativity/policy-monitoring-­ platform/revision-civil-societies-law. 3.2  HIGHER-LEVEL STRUCTURES The formation of higher-level structures(unions, federations and so on), and/or the creation of networks enhance the bargaining power and political clout of the affiliated primary groups; they can also generate economies of scale and scope when they carry out business operations. This process is known as vertical and horizontal integration. Horizontal integration refers to the cooperation between different types of cooperatives and other SSE organisations, for example, agricultural marketing cooperatives and consumer cooperatives, whereas vertical integration means the formation of higher-level structures, such as unions and federations, by primary groups of the same type, as illustrated in Figure 24. 35

Figure 24 Horizontal and vertical integration Agricultural Cooperative Union Coffee marketing coop Dairy coop FRIEDRICH-EBERT-STIFTUNG – WORKING IN THE SHADOWS VERTICAL INTEGRATION National Cooperative Federation Consumer Cooperative Union Consumer coop Addis Consumer coop Mekelle HORIZONTAL INTEGRATION Financial Cooperative Union Teachers’ SACCO Ethiopian Airlines SACCO Financial Cooperative Marketing Cooperative Consumer Cooperative Transport Cooperative 36

Is Ethiopia’s informal economy organised and, if so, how? The majority of countries in which credit unions and cooperatives exist have established national umbrella organisations, 32 whereas such an apex body has yet to emerge in Ethiopia, even though Article 17 of the Cooperatives Proclamation of 2016 provides for the establishment of a“Cooperative Societies League”, which would represent the entire movement. Moreover, there are no national associations or federations of Iddirs and Eqqubs, as they exist in South Africa, for example. 33 Regional federations of Iddirs do exist, however. A case in point is the Tesfa Social Development Association, a network of 105 Iddirs with 8,526 member-households which have come together to drive community development activities(Kelkil, 2015). the term“higher-level structure”, or because they did not have the necessary information. In any case, the survey shows that the Ethiopian member-based organisations are not as well integrated vertically and horizontally than those in other East African countries, meaning that indeed relatively few primary organisations belong to a high-level structure. The absence of national federations of the different types of self-help groups is another reason why their horizontal integration is weak, because these federations could initiate and facilitate cooperation and exchanges between different groups and organisations. In the FES-ILO-DIE survey, the question of whether their group belonged to a higher-level structure was affirmed by just 58 respondents; more than 500 of those who belonged to a group did not answer this question at all, presumably because they might not have been familiar with 3.3 MOTIVATION TO JOIN Why do people join a group, an association, a cooperative? We sought to answer this question by asking which type of services they expect from their group. 32 Such as KUSSCO Kenya, CamCCUL Cameroon and ESASSCO Eswatini, as well as the continent-wide network ACCOSCA. Most anglophone African countries have formed national cooperative federations(for example: Ushirika Tanzania, CFN Nigeria, GCC Ghana, and the continent-wide network ICA-Africa; the latter has three member organisations in Ethiopia, but none of them has national outreach. 33 The National Stokvel Association of South Africa(Nasasa) and the South African Federation of Burial Societies(SAFOBS). Figure 26 provides strong evidence that the great majority of respondents(72.4 per cent) join a group in order to obtain social services. While the social-services motivation, with an average of 38.2 per cent, predominates across all six countries, it is much stronger in Ethiopia. We must therefore conclude that Ethiopians join neighbourhood and religious groups(which, as we have argued above, are likely to inFigure 25  structures 40% 35.3 30% 20% 10% 0% Zambia 14.1 13.9 9.7 8.6 Kenya Cote d’Ivoire Senegal 37 Benin 6.2 Ethiopia

Figure 26 Reasons for joining a group Social services 72.4 FRIEDRICH-EBERT-STIFTUNG – WORKING IN THE SHADOWS Financial services 17.7 4.6 5.0 Economic services Societal/Political functions Others clude Eqqubs, Iddirs, and CBHI groups) primarily for their social purpose, even if these groups provide financial services as well. This then would mean that, for most people, the Iddirs’ insurance function is less important than their social assistance function. The financial services function constitutes, at 17.7 per cent, the second most important motivation to join the group. This would include the members of credit unions, ROSCAs and those Eqqub/ Iddir members who see those groups primarily as a financial instrument. The economic function comes third, at 4.6 per cent. One can assume that the members of cooperatives would have chosen this option. Finally, the societal function ranks fourth with 3.4 per cent. One would assume that affiliates of social movements, of political parties and of trade unions would prioritise this function. In fact, 12.5 per cent of social movement affiliates have retained this option. 3.4 GROUP CHARACTERISTICS The groups’ membership size and geographical outreach have a considerable influence on their governance structure, which in most cases consists of an elected board and a chairperson. A membership size of 100 to 150 may still be small enough to ensure a sufficient level of social cohesion as a basis for direct democratic governance. Larger groups may have to rely on indirect governance structures(such as delegates elected by certain categories of members) which ensure efficiency but weaken collective identity within the group. Smaller groups can assess the needs of members by examining individual cases, while larger groups are more likely to enforce rules-based regimes and may be less inclined or able to consider individual cases. Ethiopia’s member-based organisations have a slightly larger membership than those in the other survey countries; in Ethiopia, 73.5 per cent of all groups count 100 members or Table 5 Members per group Option Less than 10 10 to 50 51 to 100 101 to 1,000 Above 1,000 Total Ethiopia Groups 63 573 456 344 49 1,485 Percentage 4.2% 38.6% 30.7% 23.2% 3.3% 100.0% Other five countries Groups 274 2,435 1,075 829 254 4,867 Percentage 5.6% 50.0% 22.1% 17.0% 5.2% 100.0% 38

Figure 27  territorial coverage A village or a quarter of a town Is Ethiopia’s informal economy organised and, if so, how? A certain market or part of a market The entire nation A federal state A district(Woreda) less, around 4 percentage points less than in the other five survey countries. The literature reports that membership of Iddirs ranges from 10 to 600 households, with most of them having between 100 and 200 member households (Pankhurst, 2008). Eqqubs seem to have far fewer members, with a median membership of 24(Kebir, 2011). There are, however, much larger Eqqubs, especially those established by“big traders”. Membership of cooperatives averages at 219 households and those in credit unions at 253, 34 but some of them have a membership of several thousand. Figure 27 above confirms that the vast majority of Ethiopian member-based informal economy organisations are locally rooted; 77 per cent of groups confine their membership to a village, a town quarter or a market. This is consistent with groups’ relatively small membership size. A further 16 per cent operate at the level of a district( woreda, of which there are eight hundred in Ethiopia). We can assume that the few groups(84 in total) that have a nationwide outreach are either a trade union, a political party or a social movement. countries. The survey also showed that 186 of the Ethiopian groups were 50 years of age or older. As an example, some of the 105 member- Iddirs of the Tesfa Social Development Association(mentioned above) were established as far back as 1930(Kelkil, 2015). Informal economy groups, associations and cooperatives can build a bridge between the informal and the informal economy of a country, provided they are officially registered and/or recognised. In that case, the group itself enters the realm of the formal economy, while its members remain in the informal. The proportion of groups that are registered depends on their type: ROSCAs and religious groups may not need(or desire) to register, while cooperatives and credit unions must register to be able to function. We have already mentioned that in Ethiopia, Iddirs must be registered; this is not the case with Eqqubs. We can also assume that all community-based health insurance(CBHI) groups are registered because they fall under the umbrella of the Ethiopian Health Insurance Authority. One can assume that the longevity of an informal economy group can serve as a proxy indicator for its performance and stability; one would further expect that groups operating in the informal economy have a rather low survival rate because they rarely get any government support, and they are often managed or led by individuals with no managerial experience or expertise. The survey falsified the latter assumption, revealing that Ethiopia’s religious and neighbourhood groups have a median age of 10 years; cooperatives have a median age of five years whereas savings clubs tend to live two years. Similar results were found in the other five survey 34 Calculated on the basis of the data provided in paragraphs 3.1.3 and 3.1.5 Furthermore, informal-economy groups that keep accounts can be considered more“formal” than those that do not. Certain categories of groups, such as cooperatives and credit unions, are obliged by law to keep accounts, while one would assume that others, such as religious associations, which do not carry out any financial transactions, would not see the need to keep books. A third aspect indicating a certain degree of formality is the stability of informal-economy groups. The survey revealed that the great majority of groups operate permanently; the number of groups that function seasonally only(for example, during the harvesting season), or those that come together on an ad-hoc basis(for example, to assist a member in case of an emergency) is quite low. 39

FRIEDRICH-EBERT-STIFTUNG – WORKING IN THE SHADOWS Table 6 Formality indicator Registered or formally recognised Account keeping Permanent operations Average Degree of formality of groups Ethiopia 56.0 75.3 74.7 68.7 Six-country average 58.2 79.2 81.5 73.0 Taking this into account the survey delivers three indicators that can measure groups’ degree of formality and stability (Table 6). Ethiopian informal-economy groups have a lower“formality indicator” than the six-country average, but still higher than the respective values for Benin and Cote d’Ivoire. We can deduct that indeed most Ethiopian member-based groups effectively build a bridge between the formal and the informal economy. 3.5 MEMBER CHARACTERISTICS The survey shows that in Ethiopia female informal economy actors are slightly more“organised” than their male counterparts. In terms of absolute numbers, however, male participation predominates because considerably more respondents were men than women. The female participation rate(as a percentage of respondents) is higher in all types of groups, except for the more “formal” types, such as cooperatives, trade unions and political parties. We find a similar pattern in Senegal and in Cote d’Ivoire, while in Zambia, males are more“organised” than women. Interestingly, it was found in Ethiopia that in both rural and urban areas, women often have separate Iddirs which cater to their specific needs(Aredo, 2010). Moreover, it was reported that females are not only more likely to join an Eqqub, but that they also tend to make higher monthly Eqqub contributions(Kedir, 2005). In a study covering 1,850 self-help groups operating in rural and urban areas of Oromia and the Southern Nations, Nationalities, and Peoples’ Region(SNNPR) it was found that out of a total membership of 29,744, 22,670 were female. Moreover, among the 1,850 groups, 960 were women-only(Yntiso, 2015). Men are more likely than women to exercise leadership roles in their groups; 5.9 per cent of responding men and 3.3 per cent of responding women declared that they had served as a board member; the corresponding rates for roles as chair, leader or president of the group were 2.7 per cent(male) and 1.5 per cent(female). Table 7 Membership by gender Organised Not organised Total Group membership by gender (in per cent and number of respondents) Male Number Per cent 827 64.70 452 35.30 1,279 100.00 Female Number Per cent 666 67.70 318 32.30 984 100.00 Both Number Per cent 1,493 66.00 770 34.00 2,263 100.00 40

Is Ethiopia’s informal economy organised and, if so, how? Table 8 Membership by location Member of any group Not a member of any group Group membership by location Urban 51.3 48.7 Rural 67.7 32.3 Both 64.00 36.00 Table 8 shows a significant difference in the propensity to join a group, according to location; citizens in urban areas are considerably less likely to join a group than those residing in rural areas. In both rural and urban areas, the faithbased and neighbourhood groups are the most important organisational types. Cooperatives have a greater presence in rural than in urban areas, as was to be expected. 35 With an average age of 37.6 years the“organised” respondents were slightly older than those who do not belong to any group(35.2 years). The likelihood of joining a group increases with age; in the 15–24 years of age cohort, 46.9 per cent of respondents are members of a group; this indicator reaches 72.7 per cent in the 55–64 cohort. The education level of the respondents does not seem to influence group membership: whether“no school” or“secondary education”, the membership ratio ranges between 65 and 70 per cent. However, it is far lower among university graduates (42.2 per cent). The survey found that income levels had little bearing on respondents’ degree of organisation. Three-quarters of the“organised” respondents earn less than 1,500 birr per month(around 43 US$ at 2020 exchange rates). This proportion reaches two-thirds among the non-organised. This difference is plausible because most Ethiopian member-based groups can be classified as self-help organisations, which are typically formed by poorer households. Respondents’ employment status has little influence on their propensity to join a group. Whether working as an own-account worker, employer or family member, more than twothirds of respondents are members of a group. Employees, however, are, at 47.7 per cent, less organised. This may be because some types of organisations, especially cooperatives, are of little interest to persons who do not themselves operate a business or a farm. It was also found that migration status 36 had no significant impact on group membership. 35 Because the majority are engaged in agriculture. 36 Whether a respondent was born in the location where the interview took place. Figure 28 Income level of respondents 50% 40% 30% 39.2 31.8 20% 10% 0 3.3 2.2 0>0– <750 32.6 33.3 750–1,500 organized 23.4 17.5 9.1 6.8 >1,500–3,000 non-organized >1,500–3,000 0.6 0.1 >3,000–12,000 41

FRIEDRICH-EBERT-STIFTUNG – WORKING IN THE SHADOWS Considering the above it appears that, in general terms, the overall effect of socio-demographic and socio-economic variables on group membership is negligible. Member-based groups are popular throughout Ethiopia, in rural and urban areas, among women and men, and among the poor and the rich. 3.6  ATISFACTION WITH GROUP PERFORMANCE Are members satisfied with the performance of their groups and the leaders of these groups? This aspect was measured through four indicators, relating to the usefulness of group services, their timely delivery, and the honesty and competence of group leaders. Figure 29 shows that a large majority of members expressed satisfaction with groups’ performance. This may not be surprising because(a) membership of groups is voluntary, meaning that unhappy members would simply leave; and(b) group leaders are elected democratically(in most cases at least), so that underperforming leaders would not be re-elected. Across all four indicators, over 80 per cent of all respondents were satisfied, if not very satisfied with the performance of their groups and their leaders. However, the degree of satisfaction is, in Ethiopia, significantly lower than in the other five survey countries. Whether this is due to a more critical attitude of the Ethiopian public, or to a relative underperformance of the groups compared with those in the other countries, has yet to be determined. 3.7  ROLE OF GROUPS IN CRISIS RESPONSE A sizeable body of literature from disciplines such as anthropology, sociology and economics has shown that informal risk-sharing arrangements are very common in developing nations. Traditional social protection systems based on the solidarity principle have always existed in Africa, and informal insurance arrangements are still prevalent in modern society(Howlet, et al., 2021; Parnell, 2001). As pointed out earlier, such risk-sharing arrangements relying on reciprocity, mutuality and solidarity are widespread in Ethiopia as well. A total of 73 per cent of respondents who belong to a group confirmed that their group had provided help and assistance in the event of a health shock or other crisis. The majority received financial aid, either as a cash transfer or as a loan; others were supported through labour support or in-kind contributions. The percentage of groups providing crisis-related support is higher in Ethiopia than in other countries covered by the survey(an average of 61.2 per cent). This is consistent with the findings of the relevant literature, which reports that Iddirs, for example, assist their members in case of adverse shocks such as asset losses(due to theft or fire) or the death of crucial livestock, such as oxen, or in the case of health problems, when Iddirs cover medical expenses during illness(Habtu, 2012). In doing so Iddirs serve the local community as an insurance mechanism. Other types of groups, including faith-based and neighbourhood associations, Eqqubs, cooperatives and credit unions, often include Figure 29  with group performance 100% 90% 80% 90.5 84.0 70% 60% 50% 40% 30% 20% 10% 0 Servic1e–s2are useful 86.9 80.4 90.0 82.2 Services are timely Ethiopia Leadership is competent 6-country average 89.4 82.0 Leadership is honest 42

Is Ethiopia’s informal economy organised and, if so, how? emergency funds that provide financial support to their members in times of crisis, or offer in-kind support, including labour. It is obvious that in a country whose social protection system is rather patchy group-based self-help mechanisms are of the utmost importance. 4  CONCLUSIONS The FES-ILO-DIE survey and the review of official statistics and extant literature confirm that the majority of Ethiopian informal-economy operators are indeed organised in diverse groups and associations; the organisational density in Ethiopia’s informal economy is much higher than that observed in the other five survey countries. Of particular importance are the indigenous forms of cooperation and mutuality, namely the Eqqubs, the Iddirs and the Mahibers. Not only are these traditional institutions pervasive among the Ethiopian population, but they also often serve as a nucleus or springboard for the formation of“modern” organisations, such as cooperatives, credit unions and community-based health insurance groups. The majority of groups are very“local”, covering just a village, a market or a town quarter; they are relatively small, with three-quarters of the groups having less than one hundred members. Households in rural areas are more“organised” than those in cities, which may indicate that the process of urbanisation weakens traditional organisational structures. The motivation for joining a group of any type is to a lesser extent influenced by gender, age, location, education, employment status or income levels. The overwhelming majority of respondents declared that they had joined their group in quest of social services, even if the primary function of the group might be of an economic or financial nature, thereby indicating that most groups pursue multiple purposes simultaneously. Moreover, it is not uncommon for households to join more than one group, be they of the same or different types. We can conclude that the quest for social security through risk-pooling, mutual assistance and solidarity, coupled with the desire for social interaction and togetherness, constitutes a cross-cutting concern that drives the self-organisation of Ethiopia’s informal economy. While both the survey and the literature come to the same conclusion regarding organisational density in Ethiopia’s informal economy, they differ with regard to organisational types. Whereas the survey found that neighbourhood groups, religious groups and cooperatives are the three most important types, the relevant literature reports that the traditional forms of organisation – Iddirs, Eqqubs and Mahibers – are the most prevalent organisational types. This difference, however, is superficial because the three traditional organisations can be subsumed under category “neighbourhood and religious groups”. The members of the various organisations operate in the informal economy, whereas the organisations themselves exhibit a high degree of formality; this enables the latter to connect their member-households with the formal sector, be they business entities or local and municipal authorities. This means that member-based organisations can build a bridge between the informal and the formal economy. The survey further revealed that the vast majority of respondents expressed a high degree of satisfaction with the services their group provides, and with the competence and trustworthiness of the groups’ leaders. Informal-economy groups seldom experience governance issues because they are based on a collective member identity(derived from a common faith, kinship, political orientation, professional occupation, ethnic origin, age group or income level), voluntary membership, active member participation, and democratic management structures. Ethiopian informal-economy organisations exhibit a low degree of horizontal and vertical integration; none of the major types of groups has formed a national association or federation, which limits their ability to take part in national policymaking, decision-making or social-dialogue processes. The great majority of informal-economy actors work on their own account; those who work for others, or employ others (including family members), are a minority. Informal-economy operators are generally excluded from formal social security schemes, which means that life risks must be minimised through the establishment of self-help groups and mutual benefit associations, such as Iddirs, Eqqubs and Mahibers. Ethiopia’s community-based health insurance scheme has been successful because it is embedded in these traditional, local structures, and mobilises them for the collection of fees, as well as for health communication and prevention. Although built on such a decentralised foundation the scheme is nonetheless grounded in nationwide, standardised rules and regulations, and applies centralised risk-pooling. Government partly subsidises the scheme and bears the entire premium for the poorest households. The Ethiopian CBHI therefore constitutes a hybrid model of health insurance in which three layers cooperate: the local member-based organisations which collect fees and play a role in health communication and prevention; the local authorities which manage the local supply and demand for health services, and, if necessary, refer patients to higher levels of the health service hierarchy; and central government, which is responsible for risk pooling and the application of country-wide rules. In conclusion, we can establish that Ethiopian informal-economy organisations are essential to their members because they provide services(such as universal health coverage) that the state should provide, but does not because of financial constraints or other factors. The organisations in the informal economy strengthen communities, prevent risks, respond to disasters, extend social protection, build trust, deliver economic and financial services, and provide voice and representation, especially at the local level. In a neoliberal economic environment which, de facto exists in Ethiopia, such organisations fill a void left by the state and the private sector. However, their political and societal influence is limited and mostly confined to the local level. What is missing in Ethiopia is a national, representative organisation of the informal economy which would represent and defend the interests of informal-economy actors at the highest level of state and government. 43

FRIEDRICH-EBERT-STIFTUNG – WORKING IN THE SHADOWS REFERENCES Aredo, D. – ( 1993) . The informal and semi-formal inancial sectors in Ethiopia: a study of the iqqub, Iddir, and savings and credit co-operatives, Nairobi: Africa Economic Research Consortium. – ( 2010) . The Iddir: an informal insurance arrangement in Ethiopia. ­ Savings and Development, 34(1), pp. 53–72. Arefaynie, F.,(2013). The fuelwood carriers of Addis Ababa.[Online], Available at: https://voicesofafrica.co.za/fuelwood-carriers-addis-ababa/ [Accessed 16 February 2023]. Bayabil, M.,(2021). Regulation of Savings and Credit Cooperative Societies in Ethiopia, Addis Ababa: Addis Ababa University, School of Law. Bekerie, A.,(2003). Iquib and Idir: Socio-Economic Traditions of the Ethiopians.[Online], Available at: http://www.tadias.com/v1n6/OP_2_ (2003)-1.html,[Accessed 14 April 2021]. Berry, T., Ofcansky& LaVerle,(1991). Peasant Associations. In: Ethiopia. Washington: Library of Congress. Bezabih, E.,(2009). Cooperatives: a path to economic and social empowerment in Ethiopia, Dar es Salaam: ILO. CSA,(2021). Sustainable Development Goals.[Online] Available at: http://www.dataforalldemo.org/dashboard/v1/ethioinfo/ ethioinfo#/,[Accessed 05 February 2021]. EHIA – ( 2020) . CBHI Members’ Registration and Contribution 2011–2020 GC; CBHI Bulletin September 2020, Addis Ababa: Ethiopian Healt Insurance Agency. – ( 2020) . Over 32 million citizens Covered by Community Based Health Insurance.[Online], Available at: https://ehiagov.com/node/381 [Accessed 01 May 2023]. Evans-Prichard, E.,(1940). The Nuer. Oxford: Clarendon Press. Fanta, A. B.,(2015). Informal finance as alternative route to SME access to finance: Evidence from Ethiopia. Journal of Governance and Regulation, 4(1), pp. 94–102. Gebremariam, Y. et al.,(2021). Determinants of Farmers’ Level of Interaction with Agricultural Extension Agencies in Northwest Ethiopia. Sustainability, 13(6), pp. 1–24. Habtu, K.,(2012). Classifying Informal Institutions in Ethiopia, Wageningen: Wageningen University. Howlet, A. B., Syed, M. A.& Habeebullah, Z.,(2021). Informal Insurance Arrangements in Ethiopia: A Literature Review. Asian Journal of Business and Management, 3(3), pp. 210–219. ICA Africa,(2021). Mapping of key Figure s – Ethiopia, Addis Ababa: ICA Africa. IFPRI/CAPRI,(2010). Collective action and vulnerability: Burial societies in rural Ethiopia. In Resources, rights, and cooperation. In: A sourcebook on property rights and collective action for sustainable development. Washington D.C.: IFPRI, pp. 171–174. ILC,(2022). Resolution concerning decent work and the social and solidarity economy. Geneva, ILO. ILO – ( 2013) . P roviding clean energy and energy access through cooperatives, Geneva: ILO. – ( 2019) . Waste pickers’ cooperatives and social and solidarity economy, Geneva: ILO. – ( 2022) . Decent work and the social and solidarity economy – International Labour Conference, 110 th Session, 2022, Geneva: ILO. – ( 2022) . The Ten Year(2023–2032) Social and Solidarity Economy(SSE) Strategy for Africa.[Online], Available at: https://www.ilo.org/global/ topics/cooperatives/sse/WCMS_861451/lang--en/index.htm [Accessed 16 February 2023]. Kebede, G. F.,(2018). Social Capital and Entrepreneurial Outcomes: Evidence from Informal Sector Entrepreneurs in Ethiopia. The Journal of Entrepreneurship, 27(2), pp. 151–319. Kebir, A. – ( 2011) . ROSCAs in Urban Ethiopia: Are the Characteristics of the Institutions More Important than those of Members?. Journal of Development Studies, 47(7), pp. 998–1016. – ( 2005) . The Economics of Rotating Savings and Credit Association: Evidence from Ethiopia, Michigan: Center for African Development Policy Research. Kelemu, K. et al.,(2014). Revitalizing the role of cooperatives in Ethiopian economy. International Journal of Development Research, 4(10), pp. 2019–2023. Kelkil, F. T.,(2015). Civil Society Organisations and Societal Transformation in Africa: The Case of Ethiopia, Pretoria: University of South Africa. Lavers, T. – ( 2019) . Towards Universal Health Coverage in Ethiopia’s‘developmental state’? The political drivers of health insurance. Social Science& Medicine, Volume 228, pp. 60–67. – ( 2021) . Aiming for Universal Health Coverage through insurance in Ethiopia: State infrastructural power and the challenge of enrolment. Social Science& Medicine, Volume 282, pp. 1–10. Lemma, T.,(2008). Growth without Structures – the Cooperative Movement in Ethiopia. In: p. Develtere, I. Pollet& F. Wanyama, eds. Co-operating Out of Poverty: The Renaissance of the African Co-operative Movement. Dar es Salaam: ILO, pp. 128–152. Makini, K.,(n.d.) Social Movements.[Online] Available at: https://www. ide.go.jp/English/Research/Topics/Pol/Social_movements/overview.html, [Accessed 24 February 2023]. Merga, B. T., Balis, B., Bekele, H.& Fekadu, G.,(2022). Health insurance coverage in Ethiopia: financial protection in the Era of sustainable development goals(SDGs). Health Economic Review, 12(43), pp. 1–7. National Bank of Ethiopia,(2022). NBE News.[Online] Available at: https://nbe.gov.et/banks-in-ethiopia-continue-to-grow-despite-challenges/,[Accessed 09 February 2023]. Pankhurst, A.,(2008). The Emergence, Evolution and Transformations of Iddir Funeral Associations in Urban Ethiopia. Journal of Ethiopian Studies, 41(1/2), pp. 143–185. Parnell, E.,(2001). The Role of Cooperatives and other Self-Help Organizations in Crisis Resolution, Geneva: ILO. Schwettmann, J., – ( 2022) . Cooperatives in the social and solidarity economy: Sustainable development and decent work in Africa’s informal economy(PhD thesis). Manchester: Manchester Metropolitan University. – ( 2022) . Towards universal health coverage: the Cases of Benin, Cote d’Ivoire, Ethiopia, Kenya, Senegal and Zambia, Berlin: Friedrich-Ebert Stiftung. Socioeco,(2022). SSE Legislations.[Online], Available at: https://www.socioeco.org/app_legislation_en.html#geo_1,[Accessed 25 February 2022]. Solés i Coll, G.,(2020). Social Movements fighting for Human Rights in Ethiopia. Bilbao: Centre d’Estudis Africans i Interculturals(CEAI). Tesfamariam, K.,(2015). Savings and Credit Cooperatives in Ethiopia: Development, Challenges and Proposed Intervention. International Journal of Cooperative Studies, 4(1), pp. 1–8. Teshome, E., Zenebe, M.& Metaferia, H.,(2014). Participation and significance of self-help groups for social development: exploring the community capacity in Ethiopia. Springer Plus, 3(189). Tripp, A. M.,(2001). Non-formal institutions, informal economies, and the Politics of Inclusion. UN-Wider discussion paper, Issue 108. 44

Is Ethiopia’s informal economy organised and, if so, how? UNFPA,(2021). World Population Dashboard Ethiopia.[Online] Available at: https://www.unfpa.org/data/world-population/ET, ­[Accessed 12 April 2021]. USAID,(2015). Ethiopia’s Community-based Health Insurance: A Step on the Road to Universal Health Coverage, Addis Ababa: USAID. Veerakumaran, G.,(2007). The Ethiopian Cooperatve Movement – An Explorative Study, Mekelle: Mekelle University. WOCCU – ( 2022) . Credit uniion statistics 2021, Madison: WOCCU. – ( 2022) . Statistical report 2021, Madison: WOCCU. World Bank,(2021). Financial Inclusion in Ethiopia – Key findings from the Ethiopia Socio-economic survey 2018–19, Washington: World Bank. Yitbarek, E.,(2008). The Role of Iddir in Neighbourhood Upgrading in Addis Ababa, Ethiopia. Journal of Ethiopian Studies, 42(1/2), pp. 187–197. Yntiso, G.,(2015). The Self-Help Groups Approach in Ethiopia: Promising Achievements and Formidable Challenges. Journal of Ethiopian Studies, Volume 48, pp. 33–60. 45

FRIEDRICH-EBERT-STIFTUNG – WORKING IN THE SHADOWS INTEREST AMONG THOSE IN INFORMAL EMPLOYMENT IN TRADE UNIONS – THE CASE OF ETHIOPIA 36 Rudolf Traub-Merz In many countries in sub-Saharan Africa, the argument is gaining momentum that trade unions are losing organisational relevance in the formal economy and can regain transformational power in socio-economic affairs only if they increase their membership base and open up to informal labour. Trade unions should follow the shifting frontiers between formal and informal employment and when the formal economy remains small or even shrinks, informal labour should be the future recruitment terrain for trade unions. 37 Informal labour differs from the formal workforce in that it is not covered by the social protection shield provided by the state to the formal economy. In response, the informal workforce builds its own protection network. Informal labour actors establish associations, cooperatives and a variety of self-help groups, all of which share as overriding principle an interest in obtaining social security for their members. In Ethiopia, about 66 per cent of the informally employed are organised, 38 a level that probably exceeds organisational density in the formal economy. Where neighbourhood groups, religious associations, cooperatives and credit unions feature prominently, trade unions have a shadowy existence. According to the FES survey of 2020, a mere 1.3 per cent reported membership of a trade union, a clear indication that unions have not(yet) become an organisational power to be reckoned with within the informal workforce. Opening up to a segment of the economy that is built on different rules and in many regards stands outside the scope of social and labour laws poses challenges for trade unions. In the formal economy, they engage in collective bargaining, represent wage labour in dispute resolution and participate in tripartite social dialogue. Are such skills and capacities in demand among informal labour? Or do trade unions have to come up with new services and find 37 The terms“informal employment”,“informal labour” and“informal workforce” are used synonymously in this chapter. Informal employment does not include“unemployment”. In economies where no unemployment benefits are provided, the differences between the different concepts are marginal. Unemployment is“non-affordable” and“underemployment” becomes the main social challenge. 38 For details, see the respective chapter in this report. new ways and strategies for interest representation if they want to become a force in the informal economy? The history of trade unions in Ethiopia is rather recent, starting from the first central trade union confederation, CELU, formed in 1963. Under the Derg regime, trade unions were transferred into mass organisations within the Leninist framework of Ethiopian socialism and tasked“to assure social control of the workforce and to whip up productivity”(Admasie 2022: 176). Following Admasie’s historical phasing(2022), central trade union organisations remained state-captured during the EPRDF regime from 1993 to around 2010 and continued to function as yellow unions. Over much of its history, Ethiopian unions have been guided by the state to maintain industrial peace and demobilise workers to prevent labour unrest from challenging political rule. Only from 2010 onward did a“progressive weakening of ruling party/government control over the confederation”(Admasie 2022: 177) allow a new batch of trade union leaders to embark on a path of union revitalisation and growth. Since then, a“gradual emergence of greater relative autonomy and growing assertiveness of the trade union movement” can be observed (2022: 177). This is not the place to discuss the appropriateness of the historical classification and the degree to which trade unions in Ethiopia have achieved autonomy and can be seen as fully independent. We furthermore do not look at the question of whether trade unions in Ethiopia have actually embarked on an expansive strategy and see it as a complementary task to enter the informal economy and become a voice to be reckoned with for informal labour. Whether trade unions in Ethiopia are“crossing the divide” 39 is a matter of their internal organisational debates and the dynamics of the socio-economic environment within which they operate and beyond the subject of this chapter. This presentation rather crosses the lines. Instead of looking from trade unions to informal labour, it seeks to establish the views of informal labour on trade unions. Part I presents the opinions of a representative sample of informal workers and tries to answer the following questions: What does in39 This is the title of a book on this issue. See Webster et al.(2017). 46

Interest among those in informal employment in trade unions – the case of Ethiopia formal labour expect from trade unions? Is informal labour interested in trade union membership and accepting trade unions as cooperating partners? In Part II, the presentation changes perspectives again and returns to that of the trade unions. Re-reading some of the findings, structural features of informal labour are reconceived as organisational challenges for trade unions. By putting informal labour’s expectations in line with trade union capacities some parameters are identified that trade unions should consider when approaching the informal economy with a view to organisation. mal economy, they are seen mainly as“newcomers” and do not enjoy the benefits of already being well-known(Figure 30). Rather surprisingly, the awareness profile does not change significantly if we look at it in terms of the location of respondents. Trade unions are usually urban-based and their activities focus on an urban clientele. We should therefore expect an urban bias in knowledge of trade unions in urban settlements and a much lower level of awareness in rural areas. This is indeed the case in our five-country comparison group, but not in Ethiopia. Informal labour in towns and cities is as uninformed about trade unions as are rural dwellers. The survey of informal labour in Ethiopia is part of a survey project implemented in six countries of sub-Saharan Africa between 2018 and 2020. The project covered Kenya(2018), Benin(2018), Senegal(2019), Zambia(2019), Cote d’Ivoire (2020) and Ethiopia(2020). To allow a comparative perspective, the findings of the surveyed countries, excluding Ethiopia, are included as the five-country average group. 1  IEWS OF INFORMAL LABOUR ON TRADE UNIONS  UNIONS ARE FAIRLY UNKNOWN In general, members of the informal labour force have a very low level of knowledge about trade unions. An overwhelming majority of 86.7 per cent stated that they had never heard of trade unions, leaving a mere 13.3 per cent with some degree of awareness. Knowledge of trade unions in Ethiopia is far lower than in the other countries. In our five-country comparison group knowledge varied between 61.7 per cent(Cote d’Ivoire) and 23 per cent(Zambia), with an overall average of 44.9 per cent. When trade unions in Ethiopia enter the inforHow can we explain this low level of knowledge? Trade unions are(or should be) special interest groups that act first and foremost in the interest of their members. There is a further structural distance in that unions are concerned with wage labour, while informal labour is mostly composed of the self-employed. To that extent, there is no immediate reason why informal labour should take note of what trade unions do when acting on behalf of their members. The situation changes when unions step out of their immediate employment sphere and enter a wider public space to pursue their aims. This is the case when they engage in strikes beyond single companies, when they look for allies in support of their protest activities and when they engage the media to communicate the reasons for their doings. When unions take a national stance and engage in political campaigns, such as in defence of democratic rights and values, their profile becomes popular and their national outlook may make them attractive to some but even hated by others. Whatever the case, they become known to the general public and their activities are, at least for a time, the talk of the day. From this standpoint, the extremely low level of awareness is indicative of the fact that trade unions have not been sufFigure 30  of trade unions 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0 86.7 88.0 82.2 Never heard about trade unions 8.7 10.7 8.2 Heard about trade unions but no idea what they do 3.1 5.7 2.3 Know little about trade unions 1.5 1.4 1.6 Know about trade unions and what they do All Urban Rural N= 2237. Question:“Have you ever heard about trade unions and what they do?(1) I have not heard about trade unions;(2) I have heard about trade unions but do not know what they do;(3) I know a little about trade unions;(4) I have heard about trade unions and I know what they do.” 47

FRIEDRICH-EBERT-STIFTUNG – WORKING IN THE SHADOWS ficiently active in the recent past in influencing public opinion on socio-economic and political reform. They appear rather to have acted in a solitary arena separated from informal labour and not to have communicated their objectives and points of view. Judging from the current status of knowledge about their activities, if trade unions want to collaborate with informal labour or even recruit for membership, they should concentrate on the small circle(4.6 per cent) of those who are familiar with them. Or they could first come up with an information campaign to widen the circle and attract a larger spectrum for cooperation. Sample adjustment In the following the sample of respondents has been adjusted. Those who stated that they had“never heard about trade unions” did not continue the interview and played not further part in the opinion poll. The data from here on reflect only the views of the 13.3 per cent of the full sample who stated that they knew about unions to some extent, knew a little about them, or knew them by name but not what they do. A share of only one in seven or eight respondents appears small, but it has to be related to the overall size of the informal workforce. If informal employment represents 48 million people, as discussed earlier, 13.3 per cent of that is still 6.4 million people who have at least some knowledge of trade unions. This is a sufficiently broad base to guide unions’ thinking about their future plans for informal labour. 1.2  WHO KNOW ABOUT TRADE UNIONS SHARE A POSITIVE VIEW, WITH SOME RESERVATIONS Respondents were confronted with six categorical statements which covered sentiments of support or rejection for trade unions. The findings are shown in Figure 31. The sentiments are arranged by level of support. For a better understanding of divergent views in other countries, the answers for Ethiopia are shown next to those of the five-country average. The five-country reference group covers Senegal, Benin, Cote d’Ivoire, Kenya and Zambia. cent called them inefficient, and a similar share“corrupt”. Interestingly, Ethiopia’s corruption rating is falling, which implies that here more respondents believe in the honesty of trade unions. – A spike occurs on the statement that“unions should be banned”. It was thought to be the polar opposite to “unions should become stronger” and in the comparison group, this played out. Only 10 per cent called for such a ban on unions, which reconfirms the broad support for stronger unions. In Ethiopia, however, 39.5 per cent expressed support for a ban, even though 72.8 per cent want to see“stronger unions”. Obviously, numerous respondents did not consider the two statements as mutually exclusive, implying that they wanted to have a ban on unions to make them stronger. To make sense of such a combination, a ban appears not to have been conceived as closing down trade unions and ending their existence, but rather banning certain things unions were doing. We do not know what kind of things respondents wanted to see ending but their views must be placed in the category of“reservations” against (some) trade union actions. Overall, similarities can be observed between Ethiopia and the comparison group with regard to fundamental attitudes of informal labour towards trade unions. A majority rates them as important organisations which help to improve the social situation in the country. In all countries, at the same time there are substantial reservations about or dissatisfaction with their performance. Informal labour in Ethiopia differs from the corresponding view in the other countries, in that the positive rating of trade unions is slightly lower and it is more reasonable to talk of a split. We will come back to these differences when we inquire more into the views of informal labour on trade unions. 1.3  INDEPENDENCE: ARE TRADE UNIONS PERCEIVED AS AUTONOMOUS OR STATEDOMINATED? To obtain an understanding of how informal labour conceives trade unions’ operational basis we asked whether people thought that trade unions operate independently of government or whether they are dependent on government. There are three main statements that profile the views of informal labour on trade unions: – A majority rates trade unions as“important”(69.1 per cent); they“should become stronger”(72.8 per cent) and they“improve the social situation”. Approval of trade unions in the comparison groups is similar, but slightly higher. – Approval is not unqualified and substantial reservations exist in all countries when unions are assessed with regard to efficiency and corruption. More than 40 per In the comparison group, opinion was split right down the middle. About 40 per cent saw trade unions as acting independently of government, while a similarly large group regarded them as independent actors that enjoy organisational autonomy. Informal labour in Ethiopia, by contrast, tended to regard trade unions more as organisations under government control. A large majority(65.6 per cent) considered unions as strongly influenced by government and only 20 per cent identified them as independent actors(Figure 32). We did not unearth more opinions on organisational autonomy and cannot say whether the vote for dependence or in48

Interest among those in informal employment in trade unions – the case of Ethiopia Figure 31  of informal labour: support for and rejection of trade unions 90% 80% 70% 77.1 72.8 82.1 69.1 60% 50% 40% 59.0 56.1 43.8 40.7 39.5 30% 47.9 35.8 20% 10.1 10% 0 TU should become stronger TU are important TU improve social situation TU are inefficient TU should be banned TU are corrupt Ethiopia 5-country average N(Ethiopia):(1)=275;(2)=258;(3)=271;(4)=246;(5)=276;(6)=266. Note: The five-country average covers Kenya, Zambia, Senegal, Benin and Cote d’Ivoire. Question: We have listed below several statements about trade unions and want to know if you agree or disagree. Please tell us for each statement if you: fully agree; mostly agree; partly agree/partly disagree; mostly disagree; fully disagree:(1) in general, trade unions are important organisations;(2) trade unions in Ethiopia are inefficient;(3) trade unions in Ethiopia help to improve the social situation for many people;(4) trade unions in Ethiopia are corrupt;(5) it would be good if trade unions in Ethiopia became stronger;(6) trade unions in Ethiopia should be banned. Figure 32 Perceived trade union independence from government 70% 65.6 60% 50% 40% 43.9 42.2 30% 20% 19.7 10% 0 N(Ethiopia)= 259. 5-countries Trade unions independent of government Ethiopia Trade unions not independent of government 49

FRIEDRICH-EBERT-STIFTUNG – WORKING IN THE SHADOWS dependence was influenced by legal matters, financial procedures, recruitment and political posture of staff and leadership, or other things. Our data shows views in the aggregate and indicate a perception that governments at certain times and in certain ways have influenced trade union behaviour. Ethiopia is unique in our sample in that a large majority believes that the unions are not independent. We do not know on what this opinion is based and cannot say whether it is a historical remnant of the years when trade unions were under the tutelage of the state or whether it is informed by more recent affairs. It is, however, important to record that trade unions are not(yet) seen as representative bodies enjoying organisational autonomy, but rather as existing in the proximity of the state. Informal labour in the five-country comparison group assigns trade unions to a partisan profile. Unions are considered as helping some groups and not others. Key messages from the survey in the comparison group are as follows: – a strong majority believes that trade unions help government employees, private sector employers and the government; – a good majority believes that trade unions usually do not help informal workers, farmers, the poor and own-account workers; – an overwhelming majority believe that trade unions are far from helping the unemployed or everybody. 1.4  EMBEDDEDNESS: WHOM DO TRADE UNIONS SUPPORT? Trade unions are member organisations which act or should act in the interest of their members. But when they engage on issues that reach beyond their members’ immediate interest, the spectrum of potential beneficiaries of their action widens. One example is their call for social policy reforms, such as in health care and education, from which non-members also gain. When unions engage in political issues without an immediate social agenda, they may even become allied to groups which otherwise are not close to the trade unions. How does informal labour classify trade unions as benefactors? Which social groups win from what trade unions do? How is the class profile of unions perceived? We framed nine statements and asked whether trade unions help to improve the situation of a specific group. The list includes groups from within the“traditional” trade union clientele, as well as groups deemed to be rather distant from them. The list also includes groups usually seen as institutional adversaries to trade unions and reaches beyond particular interests by invoking the abstract category“everybody”. The following groups were selected: – government employees – workers in the informal economy – own-account workers – farmers – the poor – the unemployed – government – employers in the private sector – everybody. The findings are shown in Figure 33. To simplify presentation, the responses are shown as net answers, in that rejections are subtracted from approvals(agree minus disagree). A column is positive if more agree with a statement than disagree; a column is negative if more reject than approve it. The level of approval – the gap between agreement and disagreement – is indicated by the height of the column. The answer profile of informal labour in Ethiopia differs completely. Basically, trade unions are seen as discriminating against nobody and working for the benefit of all. A majority of informal workers subscribe to the view that unions support all groups no matter who they are and where in the economy or in society they are placed. No partisan profile can be identified. Being perceived as above partisan politics and in support of all groups implies that unions are not viewed as conflict agents which struggle with other classes for the benefit of their members. In contrast, they are benefactors, with no discriminatory agenda. They are not embedded inside a society with different interests but stand above it and work for the benefit of all. They stand where the state is supposed to be: above particular interests and caring for all. This underlines the previous view that trade unions are not seen as enjoying organisational autonomy but are bodies which are influenced if not directed by the government. What role on behalf of government are unions perceived to play? The highest approval rating is given to the statement“unions improve the situation of the poor”. This points to an understanding that unions are part of the social policy agenda of the state. 1.5  PROXIMITY TO TRADE UNIONS Personal experience is one way of developing knowledge of trade unions and becoming familiar with their aims and activities. If personal contacts with trade unionists and participation in trade union activities are taken as indicators of personal experience, it can be concluded that knowledge of trade unions is not obtained through direct contacts. Only some 13.2 per cent report of having met a trade unionist and only 10 per cent had participated in trade union activities (Figure 34). Those who had heard of trade unions(13.3 per cent) obtained their information from other sources, such as media or hearsay. Knowledge is not drawn from personal experience. If we recall that these rates of personal experience refer to only 13.3 per cent of total informal labour, we can further conclude that trade unions are indeed absent from the in50

Interest among those in informal employment in trade unions – the case of Ethiopia Figure 33 Groups perceived to benefit from trade union activities Values express differences(%): agree(fully + mostly) – disagree(fully + mostly) Trade unions help … … government employees … private employers … the government … informal workers –11.9 12.8 9.8 28.3 28.0 27.2 22.4 21.3 … farmers –18.5 28.5 … the poor –15.1 … own-account workers –21.5 … everybody –40.6 –42.3 … the unemployed –50% –40% –30% –20% –10%    0% 34.3 24.2 16.0 27.5 10% 20% 30% 40% 50% Ethiopia 5-country average N(Ethiopia):(1)274;(2)=268;(3)=267;(4)=269;(5)=257;(6)=254;(7)=261;(8)=244,(9)=262 Question:“Do you agree that trade unions improve the situation of the following groups:(1) the poor;(2) workers in the informal economy;(3) farmers;(4) the unemployed;(5) government employees;(6) the government;(7) employers in the private sector;(8) own-account workers;(9) everybody? Answer options: fully disagree/mostly disagree/ partly agree-partly disagree/mostly agree/ fully agree/ don’t know-no answer.” Figure 34 Personal experience with trade unions 30% 27.2 27.2 20% 10% 13.2 10.0 11.0 10.2 0 Ever been contacted by a trade unionist Ever participated in a trade union activity If a trade union calls for a demonstration, could you imagine to participate 5-country average Ethiopia N(Ethiopia): Ever contacted=303; Ever participated=305; Imagine to participate=302. Question: Talking specifically about yourself:(1) have you ever been contacted by a trade unionist?(2) Did you ever participate in any trade union activity?(3) Do you see a benefit for yourself in what trade unions are doing?(4) If a trade union called for a demonstration, would you consider participating?(5) Did you ever consider becoming a member of a trade union? 51

FRIEDRICH-EBERT-STIFTUNG – WORKING IN THE SHADOWS Figure 35 Personal benefits as reasons for attending trade union activities 80% 70% 69.3 Respondents who are willing to participate in a future trade union activity state that they see personal benefit/ no personal benefit in what trade unions do. 60% 50% 40% 30% 55.3 44.7 30.7 20% 10% 0 N(Ethiopia)=301. Question: see Figure 34, option(3). 5-country average Personal benefit Ethiopia No personal benefit formal economy. Out of all informal workers, only 1.8 per cent had ever been contacted by a trade unionist and only 1.4 per cent had ever participated in a trade union activity. Informal economy and trade unions exist in separation. Ethiopia is not unusual in relation to the five-country comparison group. A lack of personal experience does not mean that there is no interest in unionism. When people are asked about their openness to attending a union activity in future the picture becomes more favourable.“If a trade union called for a demonstration, could you imagine participating?” receives a “yes” from 27.2 per cent in Ethiopia. This is nearly three times higher than the number of those who have attended an activity in the past. A similar gap is manifest in the comparison group. Informal labour appears to be willing to move closer to trade unions and unions would gain more attention within informal labour if they increased their activities. 1.6  IN TRADE UNION MEMBERSHIP When reflecting on strategies for expanding their organisational base to the informal workforce, trade unions may either want to recruit individuals or look for affiliates and take on board established groups. For this reason, respondents were asked to disclose their interests twice: first, whether they were interested in joining trade unions as individual members; secondly, whether they would be in favour of their group affiliating with a trade union. The first question was directed to all those who had heard of trade unions, while the second was addressed only to those who are members of a group. 1.6.1  membership Interest in individual membership is shown in Figure 36. Ethiopia registers a surprisingly high 31 per cent of respondents who would consider membership. In relation to the reference group, Ethiopia comes second, behind Kenya, but ahead of the other countries. There are various ways of reading these Figures. Interest in membership is around two to three times as high as the level of personal experience with trade union affairs. Non-familiarity with union affairs will not prevent many from moving closer to trade unions or considering joining them. Interest in membership among the informal labour force as a whole, however, is very low in Ethiopia, at 4.2 per cent, and the country lies at the bottom of the reference group. Why are people open to attending a trade union demonstration or even becoming a member if they are not motivated by personal exposure? What kind of expectation drives them closer to unions? In inquiring about the reasons, we focus on individual gains and expect the utility calculus of personal benefits to be positively corelated with willingness to join as a member. In asking“do you see a personal benefit in what trade unions do?” and correlating the answers with interest in membership, we get different result. In the reference group, a large majority(63.3 per cent) of those interested in membership believe in personal gains from unionism. In Ethiopia, however, the picture is otherwise. A majority(63 per cent) are willing to become union members even though they currently do not see a personal advantage in what trade unions do(Figure 37). Joining a union may be done for one of three reasons:(a) you want to show solidarity;(b) you hope for personal gains when being a union member;(c) you feel pressure to do so. 52

Interest among those in informal employment in trade unions – the case of Ethiopia Figure 36  in individual trade union membership Cote d’Ivoire 16.1 9.9 49.5 Kenya 7.8 Senegal 7.5 23.2 Zambia 4.7 22.8 27.7 Benin 4.2 31.0 Ethiopia 4.2 0% 10% 20% 30% 40% 50% Interest in membership of respondents that had heard of trade unions Interest in membership as share of all informal labour Ethiopia): N(interest in membership)=303; N(all informal labour)=2237. Question: See Figure 5: option 5. Note: Kenya’s figure of 7.8 per cent is too low because the base reference is different. The actual number is estimated at 10–12 per cent. Figure 37  in trade union membership and perception of personal benefits from trade union activities 70% 63.3 63.0 60% 50% 40% 30% 36.7 37.0 20% 10% 0 N(Ethiopia)=92. 5-country average See personal benefits from trade unions Ethiopia See no personal benefits from trade unions 53 60%

FRIEDRICH-EBERT-STIFTUNG – WORKING IN THE SHADOWS A final answer is difficult but there are good reasons to argue that the motivation to seek membership of a union is based on an amalgam of expected personal gains and perceived political pressure. As previously shown, a majority consider that unions help all groups, including the poor. This could be translated as“whoever achieves membership gets benefits”. Memberships privileges people. Unions, furthermore, are not appreciated as autonomous organisations, but as being directed by the government. When unions call for membership, this could be understood as an invitation by the government, which has to be honoured. It is important to recall that a majority do not think this way and decline membership of a trade union outright, but a surprisingly large minority are ready to join. Hope of personal gains and perceived pressure to do so may by the reasons for this attitude. 1.6.2  ffiliation of groups er they would support the idea of joining a union as a group and requesting affiliation. 40 It must be emphasised from the beginning that interviewing individual members of groups does not establish the“corporate or collective interest” of a group. Members’ opinions are one factor but there are group structures, leaders and statutes which contribute to frame a collective identity. By looking at members’ views only, we are still far from ascertaining what a group may decide. But members’ perceptions are an important factor that group leaders have to take into consideration when deciding about group issues. The opinion poll thus cannot draw conclusions about what groups may decide if they are offered affiliation by a trade union. However, it can show what members think if such offers are made. Figure 38 shows split opinions in Ethiopia. Just under half want their group to affiliate to trade unions, while the other half speak against it. Ethiopia differs again from the referTrade unions may be less interested in winning over individuals as new members but more interested in affiliating already established groups as a collective. Accordingly, we approached those in the informal labour force who had heard of trade unions and are members of a group wheth40 As shown in the respective chapter of this brochure, 66 per cent of informal labour are organised in groups. Putting the question only to organised labour downsizes the sample. Respondents qualify if (a) they have at least heard of trade unions and(b) they already have membership of a group. Figure 38  of group members in affiliating with trade unions 70% 61.4 60% 50% 40% 30% 43.3 44.3 32.0 20% 10% 0 5-country average Ethiopia Affiliation not wanted Affiliation wanted Question directed to those who(a) know about trade unions, and(b) are members of a group. N(Ethiopia)=194. Question:“We want to know your views about possible cooperation between your group and trade unions: Do you think that your group should become an affiliate to an umbrella trade union?” 54

Interest among those in informal employment in trade unions – the case of Ethiopia ence group, where more than 60 per cent register a no to trade unions while a third want to affiliate. These votes can be taken as indicating a stronger demand for autonomy in the reference group and a broader openness to connecting with unions in Ethiopia. A rate of those wanting to join of 44.3 per cent in Ethiopia does not imply that 44.3 per cent of all groups are interested in union affiliation. The opinion poll counts the number of members, not of groups. If the answer profile reflected sentiments within each and every group, 43.3 per cent of members would speak out against affiliation, which may be treated as a veto. In any case, split opinions increase the role of the leadership in deciding the fate of a group. The already high level of interest in affiliation appears to be a positive signal for unions. If they approach established groups, they may have a good chance of winning them as affiliates. 1.7  TO PAY MEMBERSHIP FEES TO TRADE UNIONS Self-financing is central to membership organisations. A union that offers affiliation and provides services to members without demanding monetary compensation in the form of dues risks draining its resources. Providing services for a nominal or no fee at all may be acceptable for certain reasons, but rendering services in general without requesting monetary support from members will either lead to financial collapse or the need for external funding. Respondents were asked to take a position. Should their group(when affiliated) pay full membership fees or should it negotiate for a discount or even free services? The answers are summarised in Figure 10. Ethiopia deviates significantly from the reference group. In other countries, an overwhelming majority(78.4 per cent) would try to bargain with trade unions for a discount or even request free services. Only a minority(21.6 per cent) would be ready to pay the normal group fee. In the case of Ethiopia, a slight majority (52.9 per cent) register the view that their own group should pay in full and not demand any reduction(Figure 39). There are again various ways of reading these answers. In Ethiopia, there is a well-established understanding that organisations need funding and if members do not pay for services, the quality of services will suffer or not be forthcoming. If we read the answers in the light of other responses, a different interpretation becomes more likely: when unions are perceived as performing under government directives and if affiliating to unions is done at the invitation of the government, members should accept what has been set out by the government. 1.8  SUMMARY The survey on the views of informal labour on trade unions disclosed some remarkable features and particular points, in which opinions in Ethiopia differ from opinions in the reference group, which consists of informal labour Figure 39 Willingness to pay membership fees if affiliated 90% 80% 78.4 70% 60% 50% 40% 52.9 47.1 30% 21.6 20% 10% 0 5-country average Ethiopia YES – my group should pay full membership fee NO – my group should demand reduction in membership fee Ethiopia N=206; 5-country group N=1085. Question:“If your organisation would become an affiliate of a trade union, what should be the financial relation between your group and the trade union? Answer options: My group should pay member fees to the trade union and not demand a reduction: YES/ NO”. 55

FRIEDRICH-EBERT-STIFTUNG – WORKING IN THE SHADOWS in Senegal, Benin, Cote d’Ivoire, Kenya and Zambia. The informal labour force in Ethiopia has(a) the lowest level of awareness of trade union affairs;(b) among those who know about trade unions there is the highest level of reservations against union activities;(c) there is an understanding that trade unions are non-partisan and help all groups in society;(d) there is a larger than average interest in joining unions as members or affiliates; and(e) there is a high preparedness to pay unions full membership fees if affiliated. These sentiments not only set informal labour in Ethiopia apart from sentiments articulated in the reference group. They equally point to internal contradictions and demand in particular an explanation of how a low level of knowledge of and experience with trade unions can be brought into line with a fairly high interest in membership. The views of the informal work force on trade unions are polarised in two regards. The overwhelming majority are absolutely ignorant of trade unions and consequently did not participate in the part of the survey dealing with more nuanced views on trade unions; among those who are aware of the existence of trade unions and were selected for interviews, some two-thirds rejected individual membership of trade unions and about half rejected affiliation to a union as a group. This, however, leaves a surprisingly large number of people who are willing to join a union, even though they lack knowledge and experience of trade unions. This group may be willing to accept membership because they do not regard trade unions as independent organisations but as directed by the government, if not part of the governance structure. Unions are not viewed as partisan organisations anchored in a class-divided society and fighting for specific interests, but rather as being delegated by government to help everyone, in particular the poor. Joining a union in that case means accepting an invitation from the government. In one regard it is an instruction to obey; in another regards it is a privilege that comes with benefits. The views of the group interested in trade union membership may have their foundation in the past, when unions were embedded in government structures, and under the Derg, even existed as socialist mass organisations. It is not the objective of this report to discuss whether such views adequately reflect the current trade union situation in the country. It is important to note, however, that views on trade union non-partisanship do exist in some segments of the informal workforce and may become an issue when in future, trade unions and groups of informal workers meet to discuss cooperation projects. 2  ASIC POINTS TO BE CONSIDERED WHEN TRADE UNIONS LAY OUT STRATEGIES FOR ORGANISING INFORMAL LABOUR cooperation. These reflections emphasise the pros and cons of organising in a rather aggregate or abstract way and cannot be seen as a blueprint for an organising strategy. Such a strategy should be directed to the grassroots and include an analysis of the specific circumstances on the ground. The issues raised are guided and informed by the FES survey on informal labour, and do not include views, statements or expectations of trade unionists in Ethiopia on the informal workforce. The points raised are not connected to any current deliberations within Ethiopian trade unions and are inspired by general views on affinities and divergences between trade unions and the informal workforce. 2.1  TRADE UNIONISM HAS ITS ORIGINS IN INFORMAL LABOUR In a historical perspective, trade unions and informal labour have never been disconnected spheres. Trade unions originate from informal labour and were established by segments of the informal workforce in the struggle to improve working conditions and wages. The success of these struggles brought groups in the informal workforce into employment contracts that contain social security insurance, for example, for health, unemployment or pensions. In a historical sense, trade unions are“formalisers” of employment relations. In struggling for betterment, they have led labour from informality to formality. The condition of informality was the key reason for establishing trade unions. Any discussion of trade unions that fails to consider the informal workforce as their clientele is a “non-starter”. 2.2  ETTING THE FIGURES RIGHT ON THE INTEREST OF INFORMAL WORKERS IN TRADE UNION MEMBERSHIP The FES survey revealed that some 4.2 per cent of the informal workforce in Ethiopia are open to trade union membership. As small as this Figure may appear, it translates into some 2 million people, if related to the estimated total informal labour force of 48 million. If unions initiate information campaigns on social aims and what they stand for, they will be in a position to raise the interest of informal labour in unionism much more. The potential demand for trade union membership from informal labour is huge and far above current trade union membership. According to estimates, current membership in trade unions in Ethiopia is below one million. 41 If unions approach the informal workforce, they will find a recruitment potential that surmounts current membership several times over. Part II tries to point to some basic parameters that trade unions should take into consideration when deciding whether and how to approach the informal workforce for 41 In its membership list for 2019, ITUC notes CETU with 203,560 members. Admasie(2022: 180) points to 751,800 members in 2020. 56

Interest among those in informal employment in trade unions – the case of Ethiopia The key questions are: should trade unions that otherwise operate in the formal economy mobilise for membership from the informal workforce, and if so, how should they mobilise? 2.3  TRADE UNION MEMBERSHIP TO ALL OR GOING FOR SELECTIVE RECRUITMENT? The informal workforce is a very heterogeneous group, with a wide diversity of conditions of work, levels of earnings, types of occupation and employment statuses. Unions have to decide whether they should be selective in recruiting or opening up to all types of informal labour. The internal composition of informal labour leans towards self-employment, while the“union-friendly” employer–employee relationship comes a distant second. It is important to note that the respective roles are not firmly cemented and that informal operators may easily switch status from employer to own-account worker or employee, or inversely, depending on the business cycle. The FES survey found that around 10.3 per cent of the informal workforce were employees, which translates into 4.9 million informal wage workers. The share of employers was 3.9 per cent or about 1.87 million enterprises. The overwhelming majority of employers are microentrepreneurs that hire only one or two workers. Only a tiny group(2.2 per cent) of some 40,000 enterprises employ more than five workers. If trade unions opt to organise large-scale wage labour, remaining on familiar ground, they would be focusing only on a few hundred enterprises. Summarising, then, the choice boils down to: stay on home ground, concentrate on larger agglomerations of wage labour but limit yourself to a tiny share of the Ethiopian informal workforce; or expand to encompass a larger part of the informal work force and approach groups of own-account workers for cooperation. 2.4  RGANISING BACKGROUND: BETWEEN SCATTERED WORKPLACES AND LABOUR AGGLOMERATIONS The spatial dimension of work is of paramount importance for organising. Agglomerations of work and channels for personal communication provide fertile ground for trade unions to organise. This is the case in economies with a high concentration of capital and built on factories and large infrastructural and services outlets. The informal economy is based mainly on low concentrations of capital and misses out on the agglomeration of work in some areas but not in others. Rural economies feature a high share of isolated workplaces but in larger urban settlements, the economy produces labour agglomerations in particular at market places and transport hubs. When labour density matters, urban organising may be more promising than rural recruitment. 2.5  GROUND – LABOUR LAW OR PUBLIC SERVICES Self-employment differs from wage labour in respect of “common ground”. For informal employees, labour law may be a core reference for improving working conditions, but this is hardly the case for own-account workers. Myriad petty traders, motorbike riders, artisans and food providers engage in exuberant competition but they share common ground in regard to safety conditions, supply of water and electricity, access to customers and other things. Self-employment takes place under conditions set by the state and is structured through road networks, provision for market places, access to utilities and other state services. Improvement of working conditions for the self-employment often depends on positive response from public authorities. 2.6  BARGAINING WITH EMPLOYERS OR NEGOTIATING WITH PUBLIC AUTHORITIES In many cases public authorities and not private employers are the negotiation counterpart. Lobbying policymakers on access to infrastructure, property rights, rental agreements, taxation, security and hygienic conditions and pressuring public administrators on the implementation of concessions already granted are central. 42 Collective bargaining with employers without the right to strike is collective begging. Is a“labour conflict approach” feasible with public authorities? Closing down workplaces and withholding services appear to be equivalent to strikes and means of applying pressure. The difference, however, is that such actions do not harm public administrators or policymakers but rather customers or service recipients. If a public authority has to be pressed to supply services with protest action, it needs to be well communicated to the wider public to obtain its support. This may open a window of opportunity to win allies from the church, civil society groups, political parties, or others. Popular support for labour and business concerns may become a bargaining card against public authorities. Threats to withhold votes in a forthcoming election may be a lever to gain attention and overcome political and administrative indifference. 2.7  PUBLIC SERVICE DEFICITS AND NEEDS The demands of the informal workforce for public services differ strongly. The FES listed eight important public services by priority and identified the ranking of top priorities in six countries as shown in Figure 40. The need for infrastructural improvement with regard to water supply, roads and electricity in Ethiopia is higher than in the reference group, where the demand for social services, such as better health 42 See Schmidt(2023). 57

FRIEDRICH-EBERT-STIFTUNG – WORKING IN THE SHADOWS Figure 40 Ranking of the demand for public services(by first priority) Better health services 20.0 33.9 Better water supply 20.7 12.8 Better roads and bridges 18.5 8.6 Better schools and education 15.1 25.1 Better electricity 13.3 supply 5.2 Better food programs in times of crisis 6.0 9.0 Better police services 4.0 2.3 Better pensions for the elderly 2.4 3.2 0% 10% 20% 30% Ethiopia 5-country average Question: If the government wants to improve services to the people, what are in your opinion the sectors that the government should focus on?/ Please rank the sectors you have highlighted above? 40% services and better schools, comes top. This is not necessarily the case for each and every location. When trade unions want to organise own-account workers, determining the public service deficits for special groups may be paramount. 2.8  UNIONS ARE NOT FIRSTCOMERS AND INFORMAL LABOUR IS CHARACTERISED BY A HIGH ORGANISATIONAL DENSITY WITH A PLURALITY OF GROUPS When unions approach the informal workforce, they do not find unorganised terrain but an organisational density that may outnumber the organisational level in the formal economy. According to the FES survey, 66 per cent of the informal workforce are self-organised in a variety of groups, of which neighbourhood groups and religious associations are the most prominent, with 23 million and 15 million members, respectively. Cooperatives/credit unions come third with 8.2 million members and feature mainly in agriculture and trade, whether as producer or marketing cooperatives. Trade unions have to decide from which angle they want to approach the informal workforce:(a) offer something new (do what others have not yet done);(b) try to substitute for existing groups(do the same but do it better); or(c) find ways of cooperation with existing groups(make existing groups stronger). 2.9  INDIVIDUALS OR AFFILIATING GROUPS Opening up trade unions for individual membership may easily be done but may not always be advisable. Individuals may want to join unions for material benefits but bargaining for better conditions or negotiating for improved services may not be possible if members are thinly spread throughout the economy. To avoid dissipation or exhaustion of trade union energies, targeting larger groups may be advantageous. If trade unions opt for group membership, they may want a group to dissolve after affiliation and acquaint members with their own practices. Many initiatives for the self-organ58

Interest among those in informal employment in trade unions – the case of Ethiopia isation of the informal labour force, however, produce fairly stable structures with high satisfaction rates and little preparedness to see their own group dissolved. In many cases, trade unions will not be able to substitute their own activities for the existing functionality of new affiliates. Instead, they may have to provide services as add-ons and accept that affiliates will maintain their own organisational life. There is a major difference at the entry point: when individuals join unions, they cannot bargain for what services they want; when groups join as members, however, they may want to bargain the conditions of their membership. Trade unions may have to consider special bargaining packages when approaching groups as potential affiliates. REFERENCES Admasie, S.A.(2022): Trade Union Resurgence in Ethiopia, in: Global Labour Journal, 13(2): 171ff. Webster, E./ Britwurm, A.O./ Bhowmik, S.(2017): Crossing the ­Divide. Precarious Work and the Future of Labour. University of KwaZulu-Natal. Schmidt, V./ Webster, E./ Mhlana, S./ Forres, K.(2023) Negotiations by workers in the informal economy. ILO Working Paper 86, January. 2.10  INFORMAL LABOUR IN TRIPARTITE DIALOGUE Most groups in the Ethiopian informal workforce are small, counting 100 members or less and only 3.3 per cent have 1,000 members or more. Most are locally rooted and confine membership to a village, a town quarter or a market. Only a few have national outreach. Trade unions may have superior knowledge in establishing higher-aggregate organisational structures. Offering support in this regard may be a key lever in assisting informal workforce organisation in establishing wider-reaching structures. Ethiopia has signed ILO Convention No. 144 on tripartite consultation, which includes trade unions but excludes representation from the informal workforce. Trade unions could either consult with groups of informal labour and include their demands when presenting their own views; or they could mobilise employers and the government to expand membership and include one or several seats for self-representation of the informal work force. 2.11  LABOUR MEMBERSHIP MAY BECOME A STRAIN ON FINANCES Most informal labour is income-poor and increases in membership may not improve a trade union’s financial situation. Even though the FES survey found a high level of readiness among informal workers to pay membership fees when joining, we doubt the robustness of such declarations, as they may be based on wrong assumptions or expectations. If trade unions are confronted with a dire financial situation, and want to expand membership to improve finances, they may soon be confronted with a dilemma: provide resources for the extension of services to newly recruited members or lose new members again fairly soon. 59

Technical notes on the survey TECHNICAL NOTES ON THE SURVEY Mulu Teka PROJECT TEAM The original survey project in 2020 has been realised as a joint project between the Friedrich-Ebert Stiftung(FES) as lead institution, and the International Labour Office(ILO) and the German Institute of Development and Sustainability (IDOS) as cooperating partners. The implementing partners in the survey countries were national survey institutes(NSI) that are part of the Afrobarometer network. Additional technical support, including data management, was provided by the Institute for Development Studies(IDS), University of Nairobi. Members of these institutions met on various occasions to jointly develop the questionnaire and to agree on details of the survey protocol. For Ethiopia, the data collection in 2020 was undertaken by ABCON Research and Consulting, Afrobarometer’s national partner in Ethiopia. Mulu Teka, Director and Principal Researcher of ABCON, was in charge of implementing the survey protocols for Ethiopia. He also took care of the data presentation for the Ethiopia-only study in this current report. OBJECTIVES OF THE SURVEY The main objectives of the survey were to obtain a better understanding of the nature and situation of the informally employed with regard to social security, in particular health issues, views on trust in state and government, self-organisation and interest in trade unions. OPERATIONAL DEFINITION OF INFORMAL EMPLOYMENT The ILO(2019:15) 1 provides a definition of informal employment for three categories of workers:  are considered to have informal jobs if their employment relation is, in law or in practice, not subject to national labour legislation, income taxation, social protection or entitlement to certain employment benefits(advance notice of dismissal, severance pay, paid annual or sick leave, etc.”  and own-account workers are considered to be in informal employment when their economic units belong to the informal sector. The informal sector is a subset of household enterprises(not constituted as separate legal entities, independent of their owners) that produce goods or services for sale in the market, and that do not have a complete set of accounts and/or are not registered under national legislations.”  family workers are, by definition, informally employed, regardless of whether they work in formal or informal sector enterprises.” The definition of informal employment rests on the definition of employment. The definition of employment was changed in 2013 with the adoption of the 19 th ICLS resolution I. Employment became more closely linked to remuneration, and own-use production of goods was excluded from it. It was also recognised as one of five forms of work. Our survey is aligned to these changes, and own-use production of goods – that is, farming, raising animals or fishing intended to be consumed only or mainly by the household – is listed as subsistence production and excluded from the survey and the concept of informal employment. To identify informal employment and its various categories, the survey used the following operational definitions. Informal employees: Reference is to the main job. If their employers did not pay contributions to a public or private pension scheme, employees were grouped as informal. Informal employers and own-account workers(including agricultural activities intended mainly for the market): informality is defined by non-registration in the national registry used for company taxation purposes. Contributing family workers: defined by default as having an informal job because of the informal nature of the tasks such workers tend to perform. In cases of multiple jobs: the main job is defined as the one in which the respondent usually works the highest number of hours for pay or profit. Only the main job was considered for identifying informal jobs and secondary jobs were not considered. THE QUESTIONNAIRE The questionnaire originally consisted of 143 main questions, broken down into several sections. The key groups are: 1 ILO(2019) Interactions between workers’ organizations and workers in the informal economy: a compendium of practice. – Personal and sociographic data, such as age, sex, status within the household, education, respondents’ employment situation and income, household assets. 61

FRIEDRICH-EBERT-STIFTUNG – WORKING IN THE SHADOWS – Health issues: respondents’ experiences with health services; respondents’ resources in paying for medical treatment; health insurance, including reasons for joining/not joining. – Trust in state/government: respondents’ expectations with regard to services provided by the state; as well as respondents’ views on the state’s capacity and willingness to provide services; paying taxes and fees in exchange for services; and social inequality, social justice and the role of social policy. – Self-organisation and interest representation: where, why and how do respondents organise themselves in groups? Do respondents feel that their interests are represented by their groups? What are the respondents’ views on trade unions? With the outbreak of the Covid-19 pandemic, some questions were added about how people responded and government lockdown policies. A standard questionnaire, which contains identical or functionally equivalent questions, was used to allow comparison across the six African countries originally involved. However, in order to fit the local context, the questionnaire was indigenised, both in Ethiopia and the other five countries. This involved inserting response categories, terms, nomenclatures/designations that are specific to Ethiopia. The questionnaire was then translated into four major languages: Afan Oromo, Amharic, Af-Somali and Tigrigna. 2 SAMPLE SIZE A national sample of 2,400 households was used for the survey in Ethiopia. A sample of this size is large enough to make inferences about all informally employed persons who are 15 years of age or above with an average margin of sampling error of no more than plus or minus 2 per cent at a 95 per cent confidence level. SAMPLE DESIGN The sample was designed to provide a representative cross-section of all informally employed citizens aged 15 or above. Every Ethiopian citizen who corresponds to the criteria of age and informal employment had an equal and known chance of being selected for interview. To achieve this, the sample was selected by random selection method at every stage of sampling and the application of probability sampling based on population size. provide a list of individual citizens, the sample was primarily clustered into geographic units for which there is reliable population data. Primary sampling units(PSUs) – sometimes referred to as enumeration areas(EAs) – are the smallest geographic units with such data. A total of eight households(HHs) were clustered in each enumeration area for logistical efficiency and to lower the cost of contacting the sample. The clustering resulted in a sample of 300 PSUs for the survey. 3 In order to increase the precision of the survey estimates, the sample was further stratified using first-order administrative division, that is, regions and city administrations. The latter were further classified into area of residence, urban or rural. With nine regions and two city administrations, the sample for Ethiopia was classified into 21 urban and rural strata. 4 The PSUs were then allocated to each stratum based on share in national population. SAMPLE SELECTION The sample was selected in four stages. At the first stage, 300 urban and rural PSUs/EAs were randomly and independently selected from each of stratum, based on probability proportional to population size and using the latest sampling frame 5 available at the Central Statistics Agency (CSA). A corresponding number of digital EA maps were then procured from the CSA. At the second stage, sampling starting points(SSPs) were randomly selected for each of the PSUs/EAs using their digital maps. This was the next best method because no complete list of households was available from which the sample could be randomly drawn. The randomly selected SSP was marked on the map and field teams located it before moving to household(HH) settlements. A second SSP was also selected as a reserve in case the first could not be located on the ground due to demolition or other reasons. At the third stage, once the SSP is located, a fixed number of HHs – eight – were randomly selected within each of the PSUs using walk pattern at an interval of 5/10 HHs. The interviewers start walking away from the physical start point, with interviewer 1 walking towards the sun, interviewer 2 walks in the opposite direction, interviewers 3 and 4 at a 90-degree angle to the right and left. With this walking pattern, all four directions are covered. By counting HHs on both sides of the walking path, HH no. 5 is selected as the first HH for the interview and HH no. 15 for the second. If the interview could not take place because nobody was at home or the interview started but could not be finished, the walk continues to the next HH on the same side of the road or opposite(HH no. 6), while the second interview was done in HH no. 16. The sampling process was guided by a clustered, stratified, multi-stage design. Because census data do not 2 The survey protocol of the project Informal Employment, Social Protection& Political Trust(IESPPT) requires that the questionnaire be translated into a local language spoken by at least 5 per cent of the national population. These languages meet this criterion. 3 With a cluster size of eight HHs per EA, a sample size of 2,400 translates into 300 PSUs(2400 divided by eight households). 4 Because Addis Ababa does not have a rural designation. 5 A sampling frame is a complete list of the target population from which a sample is selected. The frame contains the latest list of PSUs/EAs prepared for the upcoming population and housing census in Ethiopia. 62

Technical notes on the survey At the fourth and final stage, once at the selected HH, interviewers register men and women household members who are informally employed and aged 15 or above. A computer program installed as part of CAPI on tablets then randomly selected the respondent using simple random sampling(SRS). ics the questionnaire in tablets. A total of 75 fieldworkers, composed of 60 interviewers and 15 field supervisors, were deployed for the fieldwork. SAMPLE SIZE ACHIEVED AND RESPONSE RATE TRAINING, PRE-TEST, AND DATA COLLECTION Training of fieldworkers was conducted from 13 to 17 October 2020 and was followed by a pre-test in rural and urban areas around Addis Ababa, and a final pre-departure orientation. The fieldwork for interviews took place from 22 October to 16 November 2020. The data was collected electronically in CAPI 6 mode using tablets and the face-to-face interview method. The data collection platform was developed using the Survey-To-Go(STG) application which mim6 CAPI= Computer-assisted personal interviews. CAPI involves the use of tablets for interviews in the field and a server at the centre for transfer and storage of data. A total of 2,332 interviews were completed out of the targeted 2,400. The response rate for the survey was thus 97.3 per cent. The shortfall of 2.7 per cent or 64 interviews come from the Tigray and Amhara regions. Military conflict erupted in Tigray during the fieldwork. As a result, 46 per cent or 62 interviews out of the targeted 136 could not be conducted. In the Amhara region, two interviews out of the targeted 568(0.4 per cent) were missed due to the refusal of respondents. RESPONDENT PROFILES The survey drew respondents from diverse demographic and social backgrounds. Table 9 Sample size achieved and response rate Region Tigray Afar Amhara Oromia Ethiopia Somali Benishangul-Gumuz SNNP SIDAMA Gambela Harari Addis Ababa Dire Dawa Total Planned Sample Size(N) Primary Sampling Units(PSUs) Interviews 17 136 6 48 71 568 108 864 18 144 3 24 48 384 11 88 2 16 1 8 13 104 2 16 300 2,400 Sample Achieved(n) Completed Interviews % Completion 74 54.4% 48 100% 566 99.6% 864 100% 144 100% 24 100% 384 100% 88 100% 16 100% 8 100% 104 100% 16 100% 2,336 97.3% 63

Table 10 Respondent profiles Gender Men Women Area of residence Rural Urban Age 15–20 21–30 31–40 41–50 53–60 >=61 Education No formal education Informal schooling only Primary+ some secondary Secondary+ some secondary Post secondary FRIEDRICH-EBERT-STIFTUNG – WORKING IN THE SHADOWS N 1,593 743 N 1,810 526 N 77 593 669 496 286 215 N 881 163 851 297 144 % 68.2 31.8 % 77.5 22.5 % 3.3 25.4 28.6 21.2 12.2 9.2 % 37.7 7.0 36.5 12.7 6.1 64

LIST of FIGURES LIST OF FIGURES 10 Figure 1 Informal employment and population by age groups 11 Figure 2 Informal employment, by status group 11 Figure 3 Informal employment, by status group and sex 12 Figure 4  of informal enterprises, by number of regular employees 27 Figure 18  articles on the social and solidarity economy 29 Figure 19  overlaps and relationships 31 Figure 20  Ethiopia: Member Households 33 Figure 21  density 12 Figure 5  of informal enterprises, by sex of business owner 13 Figure 6  of wage labour agreement for informal employees 14 Figure 7  of wage labour agreements for informal employees 34 Figure 22  membership per respondent 35 Figure 23  typology 36 Figure 24  and vertical integration 37 Figure 25  structures 15 Figure 8 Working hours per week in main job 38 Figure 26  for joining a group 16 Figure 9 Weekly working hours, by gender 39 Figure 27  territorial coverage 17 Figure 10 Informal employment, by sector and gender 41 Figure 28  level of respondents 17 Figure 11 Informal employment, by branch and gender 42 Figure 29  with group performance 18 Figure 12  individual income in informal employment 47 Figure 30  of trade unions 19 Figure 13  household income in informal employment 49 Figure 31  of informal labour: support for and rejection of trade unions 20 Figure 14  income gap in informal employment 49 Figure 32  trade union independence from government 21 Figure 15  average individual income, by branch and gender(ETB) 23 Figure 16  aid coverage by selected assistance programmes 23 Figure 17  of social assistance programmes 51 Figure 33  perceived to benefit from trade union activities 51 Figure 34  experience with trade unions 52 Figure 35  benefits as reasons for attending trade union activities 65

53 Figure 36  in individual trade union membership 53 Figure 37  in trade union membership and perception of personal benefits from trade union activities 54 Figure 38  of group members in affiliating with trade unions 55 Figure 39  to pay membership fees if affiliated 58 Figure 40 Ranking of the demand for public services FRIEDRICH-EBERT-STIFTUNG – WORKING IN THE SHADOWS 66

LIST OF TABLES 6 Table 1  labour force and employment 2019, based on ILO modelled estimates 8 Table 2  employment data(ILO 2019 total employment; FES 2020 informal employment) 19 Table 3 Poverty based on international poverty lines 28 Table 4 Indigenous organisations in Ethiopian languages 38 Table 5 Members per group 40 Table 6 Formality indicator 40 Table 7 Membership by gender 41 Table 8 Membership by location 63 Table 9 Sample size achieved and response rate 64 Table 10 Respondent profiles LIST OF TABLES 67

IMPRINT ABOUT THE AUTHORS IMPRINT Dr. Jürgen Schwettmann is a former ILO-official and COOP Chief, who works now as independent consultant with 44 years of work experience in international development cooperation. His work focuses on: cooperatives and the social and solidarity economy, decent work, development cooperation, informal economy and rural development. Mulu Teka serves as the Director and Principal Researcher for ABCON – Research& Consulting. He has over 16 years of work experience as analyst and in national survey designing. ABCON is Afrobarometer’s national partner in Ethiopia. Dr. Rudolf Traub-Merz is a consultant and currently works as coordinator of the FES-IDOS-ILO survey project on informal employment, Social Security and Political Trust in countries in sub-Saharan Africa. Previously, he served as Resident Representative of FES offices in Lagos, Harare, Manila, Shanghai and Moscow. Susanne Stollreiter is Resident Representative of Friedrich-Ebert-Stiftung in Ethiopia since 2022. Before she worked for FES in Germany, Latina America and Central Africa. Publisher: Friedrich-Ebert-Stiftung Ethiopia Office P.O. Box 8786 Addis Ababa Ethiopia Responsible: Susanne Stollreiter, Resident Representative FES-Ethiopia Tel.+251 11 123 3245 info.ethiopia@fes.de Editing: James Petterson Design/Layout: pertext, Berlin| www.pertext.de ISBN 978-99990-996-0-8 The views expressed in this publication are not necessarily those of the Friedrich-Ebert-Stiftung(FES). Commercial use of media published by the FES is not permitted without the written consent of the FES. Publications by the FES may not be used for electioneering purposes. © 2023 www.fes.de/bibliothek/fes-publikationen 69

WORKING IN THE SHADOWS Informal Employment in Ethiopia In 2018 the Friedrich-Ebert-Stiftung(FES) in cooperation with the International Labour Organization(ILO) and the German Institute of Development and Sustainability(IDOS) launched a research project on»Informal Employment, Social Security and Political Trust in sub-Saharan Africa« which includes an opinion survey of views on access to health services, on political trust and reasons for joining groups, including interest in trade union membership. The opinion surveys were conducted as country-wide representative surveys with a uniform research protocol to allow cross-border comparisons. The report presents findings from surveys carried out in Kenya(2018), Benin (2018), Senegal(2019), Zambia(2019), Côte d’Ivoire(2020) and Ethiopia(2020). The study can be found in internet under https://library.fes.de/pdf-files/iez/19558. pdf. In 2023 FES-Ethiopia asked some of the researchers involved in the 2018–2022 study“A majority working in the shadows” to interpret the data from this study only for Ethiopia. The result of this is the study“Working in the shadows – informal employment in Ethiopia”. With the findings of the 2023 study FES-Ethiopia would like to increase knowledge about informal work in Ethiopia, its organization and paths towards social protection. In addition, FES Ethiopia hopes that the study and its findings will stimulate discussions in the labour movement in Ethiopia on some of the aspects mentioned, as well as with government institutions responsible for social protection. The report focusses on important aspects of informal work and income including work security, income inequality and gender pay gap and examines the extent to which informal households are covered by social security provisions. It then looks at organisational issues from two angels. It analyses the types of organisations the informally employed are organised and inquires on motives for organizing and satisfaction with group membership. It further investigates the views of the informally employed on trade unions and their interest and readiness to enter into a cooperation with them. Further information on the topic can be found here: https://ethiopia.fes.de/