Experimentation and Co-ordination as Industrial Policy

Examples from Ethiopia and Chile

Mulu Gebreeyesus and Michiko Iizuka

Industrial policy can be defined as the policies that stimulate specific economic activities and promote structural changes. It is not about industry per se but also includes non-traditional agriculture or service (Rodrik 2007). Pack and Saggi (2006) similarly defined industrial policy as any type of selective government intervention or policy that attempts to alter the structure of production in favour of sectors that are expected to offer better prospects for economic growth in a way that would not occur in the absence of such intervention. There is currently a revival of interest among scholars in identifying the right balance on the role of markets and the state and the debate on industrial policy is re-emerging; see for instance the articles by Adam Szirmai (WIDER Angle May 2009) and by Wim Naudé and Ludovico Alcorta (WIDER Angle January 2010).

One emerging view on a 'new industrial policy' is that while industrialization implies that a country moves into the production and export of new or non-traditional goods and services, the success of such new activities is unknown ex ante. So as, for instance argued by Hausmann, Rodrik and others (see further reading below), it is not known for certain what a country will be good at producing. The implication is that the discovery process requires entrepreneurial experimentation. Experimentation is rife with information externalities which lead to undersupply of entrepreneurs of this type. Producing new products needs complementary services and inputs (i.e. markets, physical and human capital, norms and institutions). The required inputs for new activities may be totally absent, or not well developed, and thus require co-ordination.

In this article we discuss the discovery of two new activities – namely floriculture in Ethiopia and salmon farming in Chile. These are non-manufacturing activities but involve highly sophisticated technical and organizational skills and knowledge which some manufacturing industries may not require.

Floriculture in Ethiopia

Ethiopia is endowed with the major ingredients for a successful flower industry. It has flatlands on higher altitudes, cool climate, low cost labour, proximity to major markets, and an international airport near the production areas. There was a limited and unsuccessful effort during the 1980s through state farms, to produce and export summer flowers to Europe. Following the change of government in 1991, Ethiopia implemented extensive reforms to transform the previous command economy to a market-oriented one.

In the early 1990s two domestic private entrepreneurs started the experimentation by investing in the flower business (mainly summer flowers) with export target. A modern and large-scale floriculture industry, however, began to emerge in the early 2000s after a foreign-company, heralded the production of roses in steel greenhouse and exports. However, until 2003 no more than five farms were involved in producing and exporting flowers. The land cover of the greenhouse was small (not exceeding 40 ha.) and the volume of export was less than US$ 2 million.

Photo: Greenhouse facilities for floriculture in Ethiopia, © Mulu Gebreyeesus
Photo: Greenhouse facilities for floriculture in Ethiopia, © Mulu Gebreyeesus

By the end of 2002, the government of Ethiopia became aware of the big opportunity the flower export sector could create. Then it decided to actively support the sector with multi-faceted incentive package and set a five-year target to scale-up the flower covered land to 1,000 ha. Following the government active involvement the sector started to take-off around 2004 and continued to grow fast. In 2008 the number of flower farms reached 81, a rise of about 16-fold in contrast to the number of farms in 2003. The land covered also reached about 1,200 ha, which is above the target set by the government. The industry generated permanent employment for about 50,000, of whom 70 per cent are women. Employment created is even larger when accounting for temporary employment such as construction and other related activities.

The industry became the fourth (following coffee, oilseeds and cereals) foreign currency generator for the country with about US$ 120 million, again a rise of about 240-fold comparing to the 2003 export value. In 2007, Ethiopia ranked the fifth largest non-EU exporter to the EU cut-flower market. Despite its late entry it surpassed many SSA countries (e.g. Uganda, Zambia, Zimbabwe, and South Africa) that started flower export early on, the country also became the second largest flower exporter in Africa (next to Kenya) in the same year.

The Chilean salmon industry

Chile is endowed with natural conditions favourable for salmon farming; i.e. clean water, longer luminosity, quiet environment, water temperature, relatively low-cost labour and good access to the important input, fishmeal.

With the development of fish farming technology in the 1960s at global level, the Chilean government became aware of its potential, given suitable environmental conditions in Chile. The government actively tried to evaluate the potential of creating this industry in the 1960s to early 1970s. In the mid 1970s, some entrepreneurs started to experiment and led to the emergence of salmon industry in Chile. It is also worthwhile to mention that emergence of these entrepreneurs coincides with the start of open policy towards trade and investment abroad in Chile.

The Chilean salmon industry has demonstrated strong export growth and the number of firms increased significantly since the mid-1980s. In 1985, 36 firms were already operating in Chile. This rapidly increased to 56 in 1987, with 117 farm sites. This further increased to 83 in 1990, 184 in 1994 and 219 in 1997.  During 1985-86, salmon exports reached over US$ 1 million, allowing Chile to be recognised as a world salmon producer for the first time. Although the number of firms in the 2000s declined due to mergers and acquisitions, export revenue increased substantially and reached US$ 2.392 million in 2008. The country is currently the second largest producer after Norway. The Chilean salmon industry had environmental problems in 2008 and as a result the volume of production declined. Here focus is placed on the process of emergence of new industry. The problem of environmental degradation in the Chilean salmon industry is treated in another study (see, Iizuka and Katz 2010). This fast growth is remarkable considering the fact that salmon is not native to Chile and the producers have had to learn the technique of production from the scratch.

In the process, related activities such as input suppliers and special service providers started to emerge. The number of supplier firms listed increased dramatically from 75 in 1993 to 461 in 2003. The increase in number of suppliers was strongest in the tenth region, where salmon farming concentrated, and rose from 14 to 228 firms. The number of foreign suppliers had also gone up from 20 to 96. The salmon industry in the southern region of Chile managed to create direct employment through core activity and indirect employment through creating associated industries.

The existing data on the number of firms and export pattern for Chilean salmon shows similarity to the case of flower industry in Ethiopia. The number of Chilean salmon farming firms increased significantly from the late 1980s and continued to rise until 2000s when it started concentration due to mergers and acquisitions in. Export volume has also shown a steady rise in this period following the surge of number of firms.

Comparison of the Ethiopian flower sector and Chilean salmon sector

We can compare and analyze these two successful cases of new industrial policy by using the 'functional innovation systems' framework developed by Bergek et al. (2008) and Jacobsson and Bergek (2006). According to this framework for innovation systems (IS) to evolve and perform well seven functional requirements must be fulfilled that includes (i) knowledge development and diffusion, (ii) entrepreneurial experimentation, (iii) influence of the direction of search, (iv) market formation, (v) legitimating, (vi) resource mobilization and  (vii) development of positive externalities.

Entrepreneurial experimentation: This was critical and a primary activity in the 'discovery' of the two industries. However, given the uncertainties associated with new activities the 'formative' stage was characterized by slow entry of entrepreneurs and low growth of exports in both cases. The length of the formative stage in Ethiopian flower industry was shorter (less than one decade) than the Chilean salmon industry (above two decades). With the support from respective governments (in terms of legitimacy and resource mobilization) then entry of firms increased and the sectors changed 'gears' and transformed to the 'growth' phase. In Ethiopia, the flow of FDI was instrumental in the transformation from formative to growth phase and as a result foreign-owned firms start to dominate in the later phase. In the Chilean case, the dominance of domestic firm continued throughout the growth phase until the mature phase in the 2000s when FDI flows intensified for merger and acquisition purposes. 

Knowledge transfer: In both cases the source of knowledge was from abroad. In Ethiopia, the main mechanism of the technology transfer was mainly through private acquisition; i.e. equipments have been bought and planting materials (new varieties) licensed from international companies. In fact in the 'formative' stage most firms also hired experts from abroad. FDI also became an important source of knowledge transfer particularly in the growth phase. In the early phase when only a few firms were operating and the association was not strong enough training was mostly given in-house. In the case of Chilean salmon industry, government played an important role in importing knowledge through bilateral technical cooperation, prior to any entrepreneurial activity. This initiative created domestic technicians who eventually became the entrepreneurs.

Guidance for the search: At the early stage the governments in both countries played important role in guiding the industry development mainly through resource mobilization. These supportive actions were crucial to reduce uncertainties and stimulate start-up in both industries at the early stage. They 'signaled' and promoted entrepreneurs to take action towards discovery process. The role of government then increasingly becomes regulator and facilitator. In the case of Ethiopia, at the initial stage the government provided various direct supports (e.g. land, finance, transport co-ordination, and other fiscal incentives) and set the target to stimulate the development of this sector. Later, its role started to change to conditioning the good business environment through actions such as increasing regulatory role (foreign exchange use, code of conduct, dispersion in location and environment), consolidating of higher education, strengthening institutional framework. The case of Chile is similar, i.e. government had taken developmental role to promote the industry in the formative phase and its role changed to regulator and coordinator in the later stage. Apart from government, there were key 'path finders' (such as, FundacionChile and Association of Salmon industry in Chile, and EHPEA, the industry association in Ethiopia) that were able to guide the industry via making consensus among participating stakeholders.

Resource mobilization: In both cases, governments mobilized resources. In the case of Ethiopia, land, finance and co-ordination of airfreight were critical in the early stage. In the growth phase, finance and land still continue to be important resources but the firms start to demand less regulation and a more liberalized airfreight market. Other resources such as human capital and infrastructure development become increasingly important for the sustainability of the sector. In the Chilean case, the resource mobilization clearly observed was finance. In the formative phase, financing was decided by the government and it was given when there was a clear alignment with the intention of government and will of entrepreneurs. In the growth phase, several financial schemes were established to meet the needs of entrepreneurs.

Legitimization: In both processes, industrial associations played an important role by involving government to further 'legitimize' the growth of industry. In the case of Ethiopia, at the early stage, the emphasis of the legitimacy process was convincing the government to get involved in the sector promotion. In the growth stage the association start to address concerns of the wider audience donors and the community and in fact buyers through the launching of a code of conduct of firms. In the growth phase new institutions including public, private and civil society emerged and relationships became increasingly formalized.  In the case of Chile, in the formative period, there was no clear consensus or consensus making mechanisms among stakeholders. The firms individually went through a trial and error process to figure out consensus as happened in the choice of technology; between net pen or ocean ranching. In the growth phase, the key path finders played a key role in determining the direction of the search through building consensus.

Market formation: In the formative stage, Ethiopian flowers were exported to two EU countries (Netherlands and Germany) and mostly through auction. Over time, as pressure for complying with EU standards mounted and diversification toward direct sales and other (non-EU) markets had also become necessary, association and government intensified market search. As a result, export through direct sales increased and market destination began to widen within the EU, and to non-EU countries such as Japan, the Middle East, Russia and others. In the case of Chile, an increase in production scale, and the need for the growing industry to sustain its position in the global market, to initiatives aimed at improvement of reputation via quality control as well as market diversification..

Development of positive externality: In the Chilean case, there was some knowledge exchange among the small circle of early pioneers and later the clustering of firms generated benefits of economies of scale. In Ethiopia, the expansion of the sector had positive externalities in terms of the propagation of planting materials, packaging factories, fertilizer and chemical suppliers. A few forwarding companies start to emerge. Most of the farms operating prior to 2004 were mainly engaged in rose production and export, but with the flow of foreign firms activities started to diversify and summer cut-flower producers emerged. Air cargo supply has also increased as a result of entry of foreign airlines. The skilled labour market has also increased.


Even though these two cases are different in many aspects (for example, geography, type of activity, developmental stage, and even the guiding philosophy of the governments), we found various similarities that tie these two successful cases together. The triggering factors for the emergence of the new activities were a combination of different factors including natural endowment and favourable climate. The entrepreneurial experimentation by private entrepreneurs was, however, critical for 'discovery' of the sectors in both countries.

One characteristics of the early stage in new activities is the existence of large uncertainty in technology, marketing, and infrastructure. The governments' selective support at the initial stage was equally critical in reducing these uncertainties. In both cases, the government role changed through phases of development of the sectors. At the early stage governments played a developmental role by providing some inputs and sharing costs (for example, finance and technical support in the case of Chile, and finance, land, and transport co-ordination in the case of Ethiopia). These helped for the success of the pioneers and entry of many other investors, thus created conditions for take-off. In the growth stage other forms of engagement such as increasing regulatory role, formalization of the interactions, and strengthening of institutions start to take place.

Another important lesson from both cases is harmonization between the governments and private sector in the sector building. This was made possible by the presence of pathfinder institutions that consistently pursue the development of the sector and co-ordinate activities accordingly. In Ethiopia, the industry association played the pathfinder role. In Chile, FundacionChile was the key institution from start, even though in the later stage the Association of Salmon Industry was also instrumental. In both cases the pathfinders play important roles in consensus-building between government and the sector, standard-setting (self-regulation), collective market search, developing capacity of members, promotion and legitimation of their respective sectors.

About the authors
Mulu Gebreeyesus
Mulu Gebreeyesus

Mulu Gebreeyesus is a researcher at UNU-MERIT, based in Maastricht, The Netherlands. He has participated in UNU-WIDER projects on Trade and Industrial policy and Promoting Entrepreneurial Capacity. He took part in the joint UNU-WIDER, UNU-MERIT and UNIDO workshop on New Pathways to Industrialization, held in Maastricht on 22-23 October 2009.



Michiko Iizuka
Michiko Iizuka

Michiko Iizuka is a researcher at UNU-MERIT, based in Maastricht, The Netherlands. She is a participant in UNU-WIDER's project on Trade and Industrial policy and participated in the joint UNU-WIDER, UNU-MERIT and UNIDO workshop on New Pathways to Industrialization, held in Maastricht on 22-23 October 2009.


Further reading

This paper is based on a paper presented by the authors at the joint UNU-WIDER, UNU-MERIT and UNIDO workshop on New Pathways to Industrialization, held in Maastricht on 22-23 October 2009. The full paper is available here.

Industrial Policy: Old and New Issues, WIDER Working Paper No. 2010/106, by Wim Naudé.
New Challenges for Industrial Policy, WIDER Working Paper No. 2010/107, by Wim Naudé.

Other sources

Bergek, A. S. Jacobsson, B. Carlsson, S. Lindmark, and A. Rickne (2008). 'Analyzing the functional dynamics of technological innovation systems: a scheme of analysis' Research Policy 37: 407-429.

D. Rodrik (2007). 'Normalizing industrial policy' Commission on Growth and Development Working Paper No. 3, Washington, DC, 2008.

Hausmann R., and D. Rodrik (2003). 'Economic development as self-discovery' Journal of Development Economics 72: 603-633.

Hausmann, R., and D. Rodrik. (2006). Doomed to choose: industrial policy as predicament. John F. Kennedy School of Government, Harvard University. Unpublished.

Hausmann, R., Hwang, J., and Rodrik, D. (2005). 'What you export matters.' NBER Working Paper No. 11905 and CID Working Paper No. 123. Kennedy School of Government, Harvard University 

Iizuka, M. and Katz, J. (2010). 'Natural resource industries, tragedy of the commons and the case of Chilean salmon farming', Globelics 2010 8th International Conference, 1-3 November, Kuala Lumpur, available here.

Jacobson, S., and A. Bergek (2006). 'A framework for guiding policy-makers interning in emerging innovation systems in 'catching-up countries', The European Journal of Development Research, Vol. 18, No.4 pp. 687-707.

Pack, H., and S. Saggi (2006). 'Is there a case for industrial policy? A critical survey,' The World bank Research Observer, 21(2) 267-297.

Rodrik, D. (2007). One economics, many recipes : globalization, institutions, and economic growth. Princeton, Princeton University Press.

WIDER Angle newsletter
September 2010
ISSN 1238-9544