The potential for manufacturing to create jobs and stimulate economic growth through structural change is well documented (McMillan and Rodrik 2011). It is also well known that countries in East Asia, most recently China and Vietnam, achieved rapid growth in part due to the expansion of labor intensive manufacturing for export. The share of manufacturing in output and employment in most African countries is very low (Diao et al 2017). Growing wages in China and the governments ‘go out’ policy have lead some observers to speculate that China could export up to 60 million manufacturing jobs to Africa (Lin 2011). Such a boom in manufacturing in Africa has the potential to be transformational. We have seen however that China still accounts for a relatively small share of global FDI in African manufacturing. Nevertheless, China’s share of investment in African manufacturing has been growing and there are reasons to believe that the trend upward will continue.
In this section, we present results from recently completed (or ongoing) – mostly unpublished - research that takes an in-depth look at Chinese investment in manufacturing in four of Africa’s fastest growing economies: Ethiopia, Ghana, Nigeria and Tanzania. All of this evidence is based on firm level interviews with the exception of Ethiopia, where we have some evidence based on Ethiopia’s manufacturing census. We examine the types of manufacturing investments made by Chinese investors in these countries and the extent to which these investments are linked with local economies. Since Special Economic Zones (SEZ) have been closely associated with China’s strategy for investing in African manufacturing, we begin with a review of what we know about SEZs set up by China in Africa.
China’s Special Economic Zones in Africa
The establishment of Special Economic Zones has been a major component of China’s domestic growth that has also featured in their outward economic strategy. Of 19 overseas zone proposals that were approved in 2007, five are in Sub-Saharan Africa. These include a mining-focused zone in Zambia, a textile-focused zone in Mauritius, and zones hosting various manufacturing sectors in Ethiopia and two Nigerian states. Most of these zones, with the exception of the Mauritian one, have attracted a few tenant companies but are not yet at capacity nor have they realized their committed levels of investment. Although employment data are tentative, it appears that these companies have generated substantial employment, much of which has been local; in the three zones for which data are available (Zambia, Ethiopia and Ogun, Nigeria), there were reportedly 1,849 Chinese employees compared to 11,192 African employees. However, substantial local linkage and technology transfer benefits do not appear to be occurring outside of Nigeria, where there has been some degree of local contracting, as companies have not been willing or able to tap into local input supply for various reasons and few local firms have joined the zones (Brautigam and Tang 2014). Similar issues are reflected in the evidence from cases studies below. Nevertheless, the potential for technology transfer and local linkages exist if local firms can upgrade their production and/or enter the zones (Brautigam and Tang 2014).
Firm level evidence from case studies
Researchers conducted scoping studies on actual Chinese manufacturing investment in selected entry-level sectors (leather processing, textiles, shoes, plastic goods, agro-processing, etc.) in four African countries (Brautigam, Tang and Xia 2017). To select these firms, they obtained and coded official data (in Chinese) from the Chinese Ministry of Commerce on investment approvals. They then selected four countries with significant Chinese interest in the sectors mentioned above: Ethiopia, Ghana, Nigeria and Tanzania. In each of these countries they attempted to match the data from MOFCOM with official data on registered firms from the relevant investment approval authorities.
In each country, they conducted a subsector census by tracking down and interviewing as many of the existing Chinese firms in the selected sectors as they could reach. Consistent with research by Shen (2015) the work revealed many more projects on the books than were operational. This exercise also revealed that nearly twice as many firms were registered with local authorities in these four countries than with MOFCOM, although these firms were typically much smaller than those registered with MOFCOM. Moreover, manufacturing firms made up a much higher share (82%) of locally registered projects than they did in MOFCOM-registered projects (39%). The researchers also interviewed African firms and institutions in these subsectors: suppliers, subcontractors, government promotion offices, industry associations, etc. Using semi-structured interviews, they focused on linkages and learning opportunities.
To the best of our knowledge, this research is the first to focus solely on Chinese manufacturing investment in Africa. The researchers interviewed over 90 Chinese firms, some with multiple factory investments in entry-level manufacturing and agro-processing in the four countries. These firms had hired over 20,000 African workers and also employed over 1,300 Chinese staff. The average value of these investments was slightly above $10 million. A significant percentage (at least 28%) of Chinese firms had originally come to Africa as traders and later decided to invest in production. The motivations for these investments varied by country, though local market access tended to play a major role while access to resources was minor. However, Chinese manufacturing firms in Ethiopia’s leather and textile sectors were much more export oriented, given the nature of these sectors.
A large majority of firms sold their output primarily in local markets. Yet Chinese factories reported that their main competitors were other foreign firms in Africa or imports, not local African firms. In many cases, firms were pursuing import substitution and taking advantage of protectionist policies imposed by host governments, while exporters often sought to take advantage of the African Growth and Opportunity Act (AGOA) and the Everything But Arms (EBA) deal to gain preferential access to American and European markets.
With regard to local training, skill diffusion, and technology transfer, they found this occurring mainly through local worker training. There were almost no significant joint ventures between African and Chinese investors. A small number of Chinese firms had sent local staff to China for training, but at least 70% reported onsite training of local workers. Several had employed local technical school students as interns. There were also a number of African as well as non-Chinese foreign firms that had “technical partnerships” with Chinese companies that supplied expertise and temporary technical assistance on a contract basis.
However, they found few examples of technology transfer through backward linkages so far, except in Nigeria, where a number of local firms have contracted with Chinese suppliers. Although Chinese firms obtain significant inputs locally, only a handful work with these suppliers to improve quality. Still, by establishing quality standards and insisting that local firms figure out how to meet them, Chinese factories have pushed some local companies to invest in new machinery and methods. In other cases (plastic recycling in Ghana and Tanzania), they saw a segmentation of the market, where local firms now dominate in collection and initial processing, and Chinese firms dominate in the manufacture of plastic products from recycled plastics.
The African government statistics and the MOFCOM OFDI project data both indicate considerable interest on the part of Chinese firms in manufacturing investment. Yet there remain significant challenges to investors: as we saw in the case of Ethiopia, only 5% of planned projects materialized (EIC, 2016). In addition, the small size of the projects not captured by the MOFCOM OFDI project data suggests that the investment figures reported in the MOFCOM OFDI flow data are likely to be somewhat accurate. However, the MOFCOM OFDI project data does not capture the full number of Chinese firms and their distribution across locations and sectors. Nevertheless, the takeaways from this research and the official data remain similar: although the Chinese may be investing in manufacturing in Africa, their share of manufacturing investment and employment in Africa is still rather modest.
However, the researchers report that they expect Chinese investment in manufacturing to grow. In general, Chinese firms were optimistic about the opportunities for production in Africa. For example, one large firm invested heavily in Ghana’s two paper mills. Back in China, the company was operating steel plants, cement factories, wood processing plant and other business. With rising production costs, overcapacity, and a saturated market in China, the owner reported that he planned to close these factories down in the next few years and relocate them to Africa. This sentiment is common among Chinese investors in Africa’s manufacturing sector. The discrepancy between planned projects and operational projects also reveals an unmet demand for OFDI in African manufacturing by Chinese investors. Finally, the Chinese government is very supportive of OFDI to Africa. For example, at the December 2015 Forum on China Africa Cooperation (FOCAC) in Johannesburg, South Africa, the Chinese government officially committed to assisting in African industrialization (Eom, Hwang, Xia, and Brautigam, 2016).
Evidence based on Ethiopia’s manufacturing census
Abebe et al (2017) use Ethiopian manufacturing establishment data from 1997 to 2013 combined with a technology transfer survey module designed by them in cooperation with Ethiopia’s Central Statistical Agency to assess the nature of linkages between foreign and domestic firms in Ethiopia’s manufacturing sector. The identification problems associated with this type of work are well known (Harrison and Rodriguez-Clare, 2009). Abebe et al (2017) establish causality by using an event study design that exploits the exogenous placement and timing of the opening of large greenfield FDI plants. This event study design allows them to examine trends in the outcomes of domestic plants both pre and post the opening of the large FDI plant. The relatively small number of Chinese firms in the census – 19 in total - combined with the nature of the research design does not allow for an explicit test of the impact of Chinese FDI on domestic plants.
Nevertheless, the survey data does allow us to say something about which sectors Chinese firms invest in, how many domestic and foreign workers they employ and the wage rates paid by Chinese firms relative to domestic firms. In total, Chinese firms in Ethiopia’s manufacturing sector employed a little over 5,000 workers, the vast majority of whom were Ethiopian. Entry by Chinese firms into Ethiopia’s manufacturing sector came 4-5 years later than other foreign investors and only started to pick up around 2009. Thus, the Chinese firms are still smaller on average than other foreign firms in terms of employment size. A regression of the log of real wages on an indicator variable for Foreign_other and Foreign_Chinese indicates that like other foreign firms, Chinese firms pay a wage premium relative to domestic firms. Also, like other foreign firms, Chinese firms source roughly 70 percent of their inputs locally. Chinese firms export slightly more of their total output than other foreign firms. Finally, the Chinese firms in the sample are primarily invested in textiles, leather products and plastics.
As noted, there are not enough Chinese firms in the sample to make conclusions about linkages between Chinese firms and domestic firms. However, since the Chinese firms look similar to the rest of the foreign firms along many dimensions, it may be reasonable to think that the conclusions reached for foreign firms in general would also hold for Chinese firms. Abebe et al (2017) first show evidence from the technology transfer module clearly indicating the existence of knowledge transfers via labor flows from FDI to domestic plants, communication externalities, and (to a lesser extent) backward and forward linkages in the supply chain. They then evaluate the changes in total factor productivity (TFP) and the rate of entry of domestic plants when a large FDI plant is opened in a locality. Their empirical strategy exploits the government designation of locations for large greenfield FDI plants, in combination with an event study research design. Their estimates suggest that in the three years starting with the year of the event, domestic plants have on average around 16 percent higher total factor productivity. They also find that the number of local domestic plants is around 25 percent higher after the opening of the large FDI plant. These results demonstrate that attracting quality FDI can generate spillovers to the domestic economy and may therefore be viewed as an important component of a developmental strategy of industrialization. Their work may also be viewed as a validation of a cornerstone in the Ethiopian government’s industrial policy, which has prioritized attracting foreign direct investment to its manufacturing sector.
