SOAS University of London

Chinese firms and employment dynamics in Africa: research findings

21 June 2019

This research project Industrial Development, Construction and Employment in Africa (IDCEA): A Comparative Analysis set out to explore comparative evidence on working conditions in emerging construction and manufacturing sectors in Africa. The initial aim was to respond to the lack of evidence on job creation and working conditions in Chinese firms in these sectors in Africa. The research concentrates on two sectors where the employment contributions of foreign, including Chinese, firms are particularly important: manufacturing and construction (for the latter, specifically in road infrastructure). An additional aim of the study was to analyse the characteristics of an emerging industrial workforce in countries with different economic growth patterns. Ethiopia and Angola provide two useful contrasting contexts for these questions. An international research team completed extensive qualitative scoping research, desk reviews and two large-scale surveys of workers in Ethiopia (837 workers) and Angola (682) in March-August 2017 and September 2016-March 2017 respectively. These workers were distributed in 37 firms in Angola and 40 companies in Ethiopia, roughly half of them being of Chinese origin. After four years of mixed-method research the project is coming to an end and we have some initial findings to share. Qualitative data collection continued until July 2018, firm management surveys were completed in 2018, and subsequently additional longitudinal data were collected through phone surveys in late 2018 and early 2019. The approach was to compare like with like as much as possible, selecting all the leading Chinese, other foreign, and domestic firms within specific and highly relevant sub-sectors. In other words, we compared workers in leading Chinese firms with the best of the rest, not with the average.

On the question of job creation and the proportion of national (Ethiopian and Angolan) workers in the labour force we find that workforce localization rates are substantially higher than usually assumed in media perceptions. In Ethiopia these rates were 90% of all workers (and 100% for low-skilled workers) and in Angola, where rates are usually much lower due to skill shortages, estimated rates were 74%. In Angola we found that localization had grown significantly in the previous 10 years as Chinese firms settled in that market context. Given that Chinese contractors have dominated the road construction market in Ethiopia and Angola, they have been the main contributors to job creation in absolute terms, especially in recent years.

We also compared take-home wages for workers by skill-group (low-skilled and semi-skilled) and by sector across different firms by origin. A combination of descriptive and statistical regression analysis reveals that there is significant variation in wages and that the main determinants are, for Ethiopia: skill level of workers; working in construction (higher wages); job tenure; education; work experience; socio-economic status and location, like being located in an industrial park, where wages are slightly lower. In Angola the main determinants of wages are: skill level of workers; job tenure; work experience; socio-economic status; specific location effects. Once we take all these factors into account the origin of a firm does not impact on wages on average. In other words, wages in sampled Chinese firms were broadly similar to other top firms in the same sectors, once other worker and company characteristics are taken into account. An interesting finding is that in Angola many Chinese firms adopted a migrant dormitory labour regime by employing relatively poorer workers from the Centre-South of the country, where employment opportunities are scarce. Therefore, these workers also obtained food and accommodation and managed to save more from their wages than workers employed in other firms, especially in Luanda, where living costs are high.

We find evidence that in Ethiopia sampled Chinese firms contribute to training and skill development at least as much as other firms in the same sector, but it is in the manufacturing sector where training is widespread and considered as much more necessary by firms. In Angola, all firms have to provide different forms of informal on-the-job training given the severe skill shortages in the country especially for workers lacking relevant experience and education, but national firms and some foreign firms tend to have some more formal types of initial induction training.

In relation to labour relations at the workplace we find evidence that generally leading national firms are more used to having trade union presence, whereas foreign firms, including Chinese companies, are more reluctant to engage with unions and adopt different management styles. However, we also found that there was no difference in the level of labour conflict across firms in both countries as all kinds of firms were similarly affected by strikes over wages, especially in the Ethiopian manufacturing sector. There were also no discernible differences between sampled Chinese and other firms in terms of safety, i.e. accidents or injuries at work, which were unsurprisingly more common in road construction in both countries.

Overall, therefore, the study shows that context, both national and sector contexts, as well as the circumstances of the economies and countries when surveys were conducted, are more important determinants of labour outcomes and labour relations than the origin of a firm.