Jump to ContentJump to Main Navigation
Manufacturing TransformationComparative Studies of Industrial Development in Africa and Emerging Asia$

Carol Newman, John Page, John Rand, Abebe Shimeles, Måns Söderbom, and Finn Tarp

Print publication date: 2016

Print ISBN-13: 9780198776987

Published to Oxford Scholarship Online: August 2016

DOI: 10.1093/acprof:oso/9780198776987.001.0001

Show Summary Details
Page of

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use. date: 03 April 2020

Industrial Policy and Development in Ethiopia

Industrial Policy and Development in Ethiopia

Chapter:
(p.27) 2 Industrial Policy and Development in Ethiopia
Source:
Manufacturing Transformation
Author(s):

Mulu Gebreeyesus

Publisher:
Oxford University Press
DOI:10.1093/acprof:oso/9780198776987.003.0002

Abstract and Keywords

Ethiopia’s economy has been on a continuous and high growth trajectory since the turn of the millennium, registering an average annual growth of over 10 per cent. This is about double the average growth rate recorded for sub-Saharan Africa over much the same period. Despite its recent impressive growth performance, Ethiopia remains among the world’s poorest countries. Agriculture continues to be the main source of livelihood employing 85 per cent of the labour force. The industry contribution to the economy remains at about 14 per cent. Ethiopia is also one of the few African countries to have formulated full-fledged industrial policy and has been aggressively pursuing it. This chapter looks to understand the policy choices and their implementation, as well as their efficacy, which is of paramount importance.

Keywords:   Ethiopia, growth, poor countries, agriculture, industrial policy

2.1 Introduction

The Ethiopian economy has been on a continuous and high growth trajectory since 2003/04, registering an average annual growth of 10.6 per cent between then and 2010/11. This is about double the average growth rate (5.2 per cent) recorded for sub-Saharan Africa (SSA) over the same period. The industry sector grew by more than 10 per cent annually averaged over the same period. Despite its recent impressive growth performance, Ethiopia remains among the world’s poorest countries. Agriculture continues to be the main source of livelihood employing 85 per cent of the country’s labour force. The industry contribution to the economy remains at about 14 per cent, far below the average for SSA countries.

Ethiopia is also one of the few African countries that has formulated a full-fledged industrial policy and aggressively pursued it over the last decade. Understanding the policy choices and the implementation as well as their efficacy is of paramount importance not only for providing feedback to the policy process in Ethiopia but also for drawing lessons from which other countries can learn. The current chapter provides a broad and historical account of industrial policies and performance in Ethiopia. Section 2.2 describes the evolution of industry and industrial policy in Ethiopia. Section 2.3 describes the current structure of the Ethiopian manufacturing sector. Section 2.4 discusses productivity across sectors, size categories, and time periods. Section 2.5 examines the current industrial policy of the country. Section 2.6 provides some industry cases on the implementation of the current industrial policy. Section 2.7 concludes with a discussion on some remaining and emerging challenges in realizing Ethiopia’s industrialization vision.

(p.28) 2.2 Evolution of Industry: Historical Perspective

A conscious move to stimulate industrial growth in Ethiopia began in the mid-1950s with the formulation of the First Five-year Plan (FFYP). Ethiopia has seen three different governments over the last eight decades. In keeping with the political ideologies governing the economic principles of the time, these successive governments adopted different policies for the development of industry in Ethiopia. In what follows, the salient features of the industrial policies and performances of these three periods (the Imperial regime, the Dergue, and the Ethiopian People’s Revolutionary Democratic Front (EPRDF)-led government) will be reviewed.

2.2.1 The Imperial Regime, pre-1974

The end of WWII marked in Ethiopian history the first attempt at guiding the economy through a comprehensive plan. The first implemented promotion of industrial development was the FFYP that covered the period 1958–62.1 The plan envisaged achieving industrial progress through substituting imports with the development of light industry producing consumer goods for the domestic market. In accordance with this strategy, the textile, cement, and food industries were given emphasis in the plan as they were believed to produce products in high demand in the local market which could be produced using the local resources abundantly available in the country.2

Two more five-year plans, the Second Five-year Plan and Third Five-year Plan, were launched between 1963 and 1973.3 During this period, the government extended the incentives to attract investors and continued to strengthen its presence in economic activities. The implementation of these initiatives envisaged in the three successive five-year plans (1958–73) attracted foreign investors and boosted the manufacturing sector. However, by the end of the plan period, the overall industrial base of the country had remained weak (World Bank, 1985).4

(p.29) 2.2.2 The Dergue Regime, 1974–91

In 1974 the Ethiopian Revolution erupted while the country was preparing the fourth five-year development plan. The military government nationalized most of the medium and large manufacturing enterprises and declared ‘a socialist economic policy’ (PMAC 1975). Industrial activities were reserved exclusively for the state. As a result, the manufacturing sector exhibited a sharp decline, particularly in the first few years following the revolution.

The government had no industrial policy per se until the mid-1980s. A Ten-year Perspective Plan comprising a macroeconomic framework—a Public Investment Programme containing an indicative portfolio of projects and production targets for the period 1984/85–1993/94—was formulated. This was modified in 1986 in the Ten-year Development Plan. According to World Bank (1985) the industrial sector was envisaged to grow at 15.4 per cent per annum over the plan period, which is three times the planned growth of the agriculture sector. Priorities were given to import substitution and labour intensive industries in an effort to address the country’s dependence on imported goods and inputs and to generate employment.

Table 2.1 Number of establishments, employment, and value added by ownership in the Ethiopian MLSM, 1979/80–2010

Number of establishments

Number of employments

Value added

Total

Public share (%)

Total

Public share (%)

Total (mill. Birr)

Public share (%)

1979/80

351

45.3

76,631

88.85

589.7

95.0

1985/86

369

48.8

90,845

93.33

715.3

95.2

1990/91

275

52.4

84,000

93.08

460.2

94.0

1995/96

627

25.7

90,039

86.25

1,593.8

87.9

1999/00

788

15.5

95,708

56.13

2,279.3

72.5

2004/05

1,207

10.4

110,160

48.70

3,030.6

N/A

2009/10

2,172

6.4

186,799

25.67

11,369.6

31.2

Source: CSA (various years(a), various years(b)); author’s calculations.

The nationalization and continued systematic restriction of the private sector from engaging in major economic activities had reduced the emerging vibrant sector into micro- and small-scale manufacturing activities. In contrast the state became the sole responsible organ owning and operating medium- and large-scale manufacturing activities. In 1985/86, one decade after the revolution, state-owned enterprises (SOEs) generated 95 per cent of the value added and 93 per cent of the employment of all MLSM (Medium and Large-Scale Manufacturing) enterprises (see Table 2.1).

Towards the end of the 1980s the SOEs’ financial position weakened, and government subsidies and overdraft facilities became instrumental for their survival (UNIDO 1991: 13). Because of the poor quality of products and lack of (p.30) experience in export markets, only 6 per cent of the total value of industrial output was exported by SOEs. During the last years of the Dergue regime there was a sharp decline in the Ethiopian economy, particularly in the manufacturing sector. The growth of the manufacturing value added started to fall in 1988 and reached −40 per cent in 1991.5 The industry value added at large had also exhibited a similar trend over this period. Policies hostile to the private sector coupled with substantial inefficiency in the public sector were the main causes of this decline.

2.2.3 The EPRDF Regime, post-1991

Soon after the EPRDF-led transitional government seized power it announced that the country would follow a market-led economic policy. The first decade of the EPRDF regime (1991–9) was marked by a series of reforms under the structural adjustment programme (SAP) with the aim of reversing the command economic system by way of fostering competition, opening up the economy, and promoting the private sector. In this period the government implemented three phases of IMF/WB sponsored reform programmes. The first phase of the structural and economic reform programmes took place during 1992/93–1994/95. A key measure undertaken during this period was liberalization in various markets.

The second phase of the economic reform programme (1994/95–1996/97) was aimed at limiting the role of the state in economic activity and the promotion of greater private capital participation. In 1996 the country entered a three-year Enhanced Structural Adjustment Facility (ESAF) arrangement with the IMF and began the third phase of the reform programme spanning the period 1996/97–1998/99. Under this arrangement, the government committed itself to achieving broad-based economic growth in a stable macroeconomic environment, while the liberalization measures were further strengthened.6

The favourable policy environment created by the economic reforms, coupled with macroeconomic stability, revitalized the manufacturing sector and the economy at large. The high growth period did not last long, however. The value added growth of the industry and the manufacturing sector in the period 1996–2003 averaged only 5 and 3 per cent per annum respectively, which is modest in comparison with preceding years.

(p.31) In 2002/03, the Ethiopian government formulated a full-fledged Industrial Development Strategy (IDS). The IDS was linked to various sub-sector strategies and to successive development plans such as the Sustainable Development and Poverty Reduction Program (SDPRP) 2002/03–2004/05 and the Plan of Action for Sustainable Development and Eradication of Poverty (PASDEP) 2005/06–2009/10. The first development plan put great emphasis on smallholder agriculture, while in the second plan policy scope was broadened to encompass the urban sector and the industrial sector. The industrial sector and the economy at large have shown impressive growth following the implementation of the PASDEP. Recent industrial policy and performance will be discussed in further detail in Sections 2.3–2.6. Table 1 in the working paper version of this chapter (Gebreeyesus 2013) provides a summary of the industrial development periods in Ethiopia.

2.3 The Current Structure of the Ethiopian Manufacturing Sector

Table 2.2 Ethiopia’s economic performance, 2000/01–2009/10

GDP

Agriculture value added

Service value added

Industry value added

Year

Growth (%)

Growth (%)

Share to GDP (%)

Growth (%)

Share to GDP (%)

Growth (%)

Share to GDP (%)

2000/01

8.3

9.6

50.9

5.3

38

5.1

12.1

2001/02

1.5

−1.9

49.1

4.3

38.6

8.3

12.9

2002/03

−2.2

−10.5

44.9

5.8

41.7

6.5

14

2003/04

13.6

16.9

47

6.0

39.7

11.6

14

2004/05

11.8

13.5

47.4

12.7

39.7

9.4

13.6

2005/06

10.8

10.9

47.1

12.8

40.4

10.2

13.4

2006/07

11.5

9.4

46.1

15.2

41.7

10.2

13.2

2007/08

10.8

7.5

44.6

15.3

43.5

10.4

13

2008/09

8.8

6.4

43.2

14.5

45.1

8.9

13

2009/10

10.1

5.8

42

15.2

46.1

8.8

13

Source: World Development Indicators, World Bank (2010).

The Ethiopian economy has recorded continuous and double-digit growth since 2003/04 (Table 2.2). The service sector outperformed other sectors in this period, and its share of GDP increased from 39.7 per cent in 2003/04 to 46 per cent in 2009/10. Despite 10 per cent annual average growth in the same period, the agriculture share in GDP declined from 47 per cent to 42 per cent. The industry sector also grew by about 10 per cent per annum between 2003/04 and 2009/10. However, its share in the economy remained relatively static (12–14 per cent) over the last decade.

(p.32) The construction and manufacturing sectors grew by 10 per cent and 11.8 per cent respectively between 2003/04 and 2009/10. In 2009/10, the construction and manufacturing sectors respectively accounted for 5.8 and 4.9 per cent of GDP, which is equivalent to 37.7 per cent and 44.5 per cent of the industrial value added.7

The Ethiopian Central Statistics Agency (CSA) has been conducting surveys on the country’s manufacturing activity since the mid-1970s. The coverage and frequency of the surveys differ by size and type of activity. First, the Survey on MLSM (CSA various years[a]) covers establishments that use power driven machines and employ ten or more workers. This survey has been conducted on a census and annual basis since the mid-1970s. Second, the Survey on Small-Scale Manufacturing (SSM) (CSA various[b]) covers establishments that use power driven machines but employ fewer than ten workers. The CSA has so far conducted four rounds of surveys (1994/95, 2001/02, 2005/06, 2007/08) all on a sample basis. Third, the Report on Cottage/Handicraft Manufacturing (CSA 2003) covers establishments performing their activity by hand or using non-power driven machines or tools. Only one survey of this industry, in 2001–2, has been conducted.

Because of a lack of recent data on the cottage/handicraft industry, the discussion below will focus on the SSM and the MLSM enterprises. However, we can shed some light on the relative importance of the cottage/handicraft industry to the Ethiopian manufacturing sector using the 2001–2 report. The cottage/handicraft sector was estimated to comprise a total number of 974,676 establishments generating employment for above 1.3 million people in 2001–2.8 In contrast there were only 31,863 SSM and 909 MLSM establishments employing respectively 97,781 and 98,986 people in that year. The cottage/handicraft industry thus accounted for about 97 per cent and 87 per cent of the total number of manufacturing enterprises and employment respectively. On the other hand, in terms of value added the cottage/handicraft sector contribution was only 21.5 per cent.

Subsections 2.3.1 to 2.3.6 describe in order the size distribution, sectoral composition, sunrise and sunset industries, ownership type, trade orientation, and geographical distribution based on the SSM and MLSM survey reports. When comparing SSM and MLSM we rely on the 2007/08 survey reports, the latest available for both size categories.

(p.33) 2.3.1 Size Distribution

The Ethiopian manufacturing sector is dominated by small and micro-firms.9 In 2007/08, for example, 43,338 (96 per cent) of the 45,268 manufacturing establishments that use power driven machineries were micro-firms employing fewer than ten people. When looking at the employment distribution we observe two strong modes, one at the micro-size firms and another at the large firms employing 50+ workers.

In terms of value added the size distribution is skewed to the largest size group (50+ employees). The large size group accounts for about 83 per cent of manufacturing value added, while the micro-firms contribute only 11 per cent of the manufacturing value added.

2.3.2 Sectoral Composition

Table 2.3 Sectoral distribution of the Ethiopian manufacturing sector (2007/08)

SSM sectoral share (%)

MLSM sectoral share (%)

ISIC code

Number of establishments

Employment

Value added

Number of establishments

Employment

Value added

15

food and beverage

56.76

53.8

47

25.1

31.2

42.3

of which grain mills

(53.2)

(50.4)

(42)

16

tobacco

0.1

0.9

3.8

17

textile

1.3

9.0

2.1

18

apparel

7.2

4.7

5.3

2.0

5.7

1.3

19

leather

4.3

6.5

4.0

20

wood products

3.6

2.4

0.6

21–22

paper and printing

7.4

6.7

5.1

24

chemicals

4.1

5.8

5.8

25

rubber and plastic

4.2

6.5

4.9

26

other non-metallic mineral

25.3

13.2

18.6

27

basic iron and steel

0.8

1.0

3.1

28

fabricated metal

10.1

11

15.6

5.2

3.9

4.8

29

machinery and equipment

0.2

0.2

0.1

34

motor vehicles, trailers, and semi-trailers

0.8

1.3

1.8

36

furniture

19.8

25

24.3

15.5

5.6

1.8

other

6

5

8

Total

43,338

138,951

1.14

(billion Birr)

1,930

133,673

9.17

(billion Birr)

Source: CSA (2010); author’s calculations.

Table 2.3 shows the separate sectoral composition for Ethiopian manufacturing establishments in 2007/08 for SSM and MLSM. The first three columns report the share of the different sectors of the SSM. The grain mills sector is the dominant industry of this size category and accounts for about 53 per cent of the total number of SSMs.

The last three columns of Table 2.3 report the sectoral composition of the MLSM. This formal sector is characterized by a high concentration of a limited range of light manufacturing activities such as food and beverage, textile, leather, non-metallic, and furniture. In 2007/08, the food and beverage sector accounted for about a quarter of the number of establishments, one-third of employment, and 42 per cent of the value added of the MLSM sector. The second important industry is the manufacture of other non-metallic mineral products. Some basic sectors, such as chemical, basic metal, and engineering, are as yet underdeveloped.

2.3.3 Sunrise and Sunset Industries

Table 2.4 Ranking of top ten manufacturing products in output and export contribution

Top 10 products produced (4-digit level) rank and share (%)

1995/96

2002/03

2009/10

Product type in order

Share (%)

Product type in order

Share (%)

Product type in order

Share (%)

1

malt liquors and malt

9.46

sugar and confectionery

12.40

sugar and confectionery

8.89

2

spinning, weaving, and finishing

9.42

cement, lime, and plaster

9.40

cement, lime, and plaster

8.37

3

sugar and confectionery

8.86

spinning, weaving, and finishing

8.42

malt liquors and malt

7.88

4

tobacco

7.65

malt liquors and malt

7.50

flour

6.95

5

tanning and dressing of leather

7.41

tanning and dressing of leather

7.26

plastics

6.83

6

flour

6.06

flour

5.15

soft drinks

5.85

7

basic iron and steel

5.03

soft drinks

5.08

structural metal products

5.80

8

cement, lime, and plaster

4.97

basic iron and steel

4.36

spinning, weaving, and finishing

4.98

9

soft drinks

4.70

publishing and printing

3.35

basic iron and steel

4.05

10

bodies for motor vehicles

4.50

plastics

3.33

soap detergents, perfumes

3.71

Top 10 share (%) sum

68.07

66.26

63.31

Total MLSM GVP (billion Birr)

5.53

8.95

42.0

Top 10 products exported rank and share (%)

1995/96

2002/03

2009/10

Product type in order

Share (%)

Product type in order

Share (%)

Product type in order

Share (%)

1

tanning and dressing of leather

91.27

tanning and dressing of leather

59.20

tanning and dressing of leather

34.86

2

spinning, weaving, and finishing

4.24

sugar and confectionery

20.17

spinning, weaving, and finishing

15.81

3

meat and fruit

1.65

spinning, weaving, and finishing

14.27

dairy

10.28

4

sugar and confectionery

1.27

food NEC

3.70

wearing apparel except fur

7.38

5

wearing apparel except fur

0.92

footwear

0.95

footwear

7.01

6

wood and wood

0.35

non-metallic NEC

0.66

flour

4.80

7

non-metallic NEC

0.09

meat, fruit

0.52

meat, fruit

4.67

8

footwear

0.06

malt liquors and malt

0.25

basic iron and steel

3.21

9

wine

0.04

edible oil

0.18

other fabricated metal prod.

3.01

10

furniture

0.02

wine

0.03

soap detergents, perfumes

2.16

Top 10 export share (%) sum

99.91

99.94

93.17

Total MLSM exports (billion Birr)

0.401

0.848

1.19

Source: CSA (various years), but own calculations.

Based on a disaggregated four-digit level of the International Standard Industrial Classification (ISIC) product classification we examined the growing and shrinking manufacturing industries in the MLSM over the last decade and a half. Table 2.4 reports the rank and share of the top ten industrial products in terms of output (gross value of production) and exports for three selected years. In 2009/10 the top ten products include in order of importance: (p.34) (p.35) (p.36) manufacturing of sugar, cement, liquors, flour, plastic, soft drinks, structural metal, textile, basic iron and steel, and detergent and soaps. In sum, they account for about 63 per cent of total MLSM production value. Emerging industries include the structural metal, detergent, and plastic sectors. There is some evidence that the textile industry and leather tanning are among the stagnant industries.

2.3.4 Ownership Types

Table 2.5 Paid-up capital and value added by ownership and industry group (2009/10)

Paid-up capital (mil. Birr)

Share in paid-up capital by ownership type (%)

Total

Public

Domestic private

Foreign

Food and beverage

5,192.3

15.9

75.3

8.8

Tobacco product

195.0

100.0

0.0

0.0

Textile

1,897.4

15.5

77.1

7.4

Apparel

361.9

0.4

64.3

35.3

Tanning and leather products

757.2

5.0

87.8

7.2

Wood products

1,757.7

93.0

6.7

0.3

Paper and printing

902.7

17.1

81.3

1.6

Chemical products

1,021.1

22.5

72.3

5.1

Rubber and plastic

1,673.0

9.1

73.3

17.6

Non-metallic mineral products

2,585.5

13.7

79.0

7.3

Basic iron and steel

710.6

13.8

83.7

2.5

Basic metal and engineering

1,823.1

22.2

72.6

5.2

Furniture

296.0

7.8

91.6

0.6

Total

19,173.5

23.0

69.5

7.6

Source: CSA survey 2009/10, but own calculations. Note that basic iron and engineering includes fabricated metal, machinery and equipment, and motor vehicle sectors.

The presence of foreign owned companies in Ethiopian manufacturing is generally limited. A recent (2011/12) sample based survey by the WB reveals that only 4 per cent of the sampled firms in Ethiopia are private and foreign owned (Geiger and Goh 2012). Table 2.5 reports the share of paid-up capital, by foreign and domestic private ownership. The foreign ownership share in terms of paid-up capital was only 7.6 per cent of the reported total in the MLSM sector. Foreign capital is relatively more visible in industries such as apparel (35.3 per cent), rubber and plastic (17.6 per cent), and food and beverage (8.8 per cent).

2.3.5 Trade Orientation

The volume of manufactured exports from Ethiopia is low. Only 4 per cent of manufacturing firms participate in export markets. The leather, textile, and (p.37) apparel sectors have the highest participation rates. The export to sales ratio is also correspondingly small (see Table 2.6). For example in 2010, the average share of export in total sales of the MLSM sector was only 3 per cent. This is much lower than in previous years. The export to sales ratio has also declined in the three most important export sectors—leather, apparel, and textile sectors—during this period.

Table 2.6 Export sales and imported raw materials by sector MLSM (2009/10)

Exports’ share in sales (%)

Imported raw materials share (%)

Exports coverage of imported raw materials (%)

2002/03

2007/08

2009/10

2002/03

2007/08

2009/10

2002/03

2007/08

2009/10

Food and beverage

6.00

3.14

1.78

22.03

27.91

24.8

79.88

28.62

16.29

Tobacco

0.00

0.31

0.45

79.84

13.8

5.5

0.00

8.61

12.41

Textile

17.24

14.06

9.19

29.04

29.64

37.0

93.52

80.24

47.60

Apparel

0.10

26.08

12.92

25.66

37.28

50.3

0.79

138.06

55.10

Leather

69.11

61.26

36.27

21.64

23.56

34.4

423.96

382.35

172.86

Wood products

0.00

0

0.05

54.63

23.87

21.1

0.00

0

0.39

Paper and paper products

0.00

0

0.00

75.36

81.54

59.5

0.00

0

0.00

Chemical

0.00

1.58

1.02

72.33

87.06

70.5

0.00

2.92

2.30

Rubber and plastic

0.00

0

0.13

85.00

77.22

92.3

0.00

0

0.27

Other non-metallic mineral products

0.57

0.13

0.07

29.79

19.84

58.1

9.35

4.13

0.26

Basic iron and steel

0.00

0

2.42

99.17

99.8

79.1

0.00

0

4.25

Fabricated metal

0.00

0

1.53

88.97

80.19

84.6

0.00

0

2.92

Machinery and equipment

0.00

0

2.04

92.32

97.68

85.1

0.00

0

4.25

Motor vehicles, etc.

0.00

0

0.00

93.39

97.13

98.5

0.00

0

0.00

Furniture

0.00

0.8

0.00

42.30

52.51

50.1

0.00

3.01

0.01

Total

9.95

6.05

3.03

43.52

53.61

51.0

52.93

24.42

11.65

Source: CSA (various years), but own calculations.

Table 2.6 also reports raw material import dependence and export coverage of imported raw materials. In 2009/10 imported raw materials accounted for about 51 per cent of the total use of raw materials in the MLSM sector. The dependence on imported raw materials is relatively lower in the agro-industries. The chemical, plastic and rubber, basic iron, and engineering industries depend heavily on imported inputs, ranging from 70 per cent to 99 per cent of total cost of raw materials.

The overall dependence of the manufacturing sector on imported raw materials has increased in comparison to 2002/03. The apparel, leather, and non-metallic mineral products which are customarily among the less dependent have shown increasing dependence on imported raw materials over time. In contrast, import dependence of wood and paper industries has declined. The declining share of export sales and a simultaneous increase in import dependence means the export coverage of imported raw materials of the manufacturing sector has fallen over time.

2.3.6 Geographic Distribution

Table 2.7 Geographical distribution of manufacturing enterprises

Top 10 towns share (%)

1995/96

Top 10 towns share (%) 2003/04

Top 10 towns share (%) 2009/10

Top 10 towns

# of est.

Employ

ment

Top 10 towns

# of est.

Employ

ment

Top 10 towns

# of est.

Employ

ment

Addis Ababa

67.2

60.4

Addis Ababa

55.8

53.6

Addis Ababa

40.3

45.9

Dire Dawa

4.0

7.0

Awassa

3.3

2.8

Awassa

3.6

2.5

Bahr Dar

2.6

3.3

Mekelle

3.0

1.7

Dire Dawa

2.8

1.9

Awassa

2.4

3.2

Bahr Dar

2.3

2.6

Mekelle

2.8

1.9

Nazreth*

2.3

1.5

Burayu*

2.2

1.4

Nazreth*

2.4

3.1

Jimma

1.9

0.4

Dire Dawa

2.2

3.9

Bahr Dar

2.3

1.4

Mekelle

1.6

0.3

Nazreth*

2.2

1.7

Debrezeit*

1.7

2.5

DebreZeit*

1.1

1.2

Debre Zeit*

1.9

2.8

Sebeta*

1.7

1.5

Harar

1.1

1.3

Sebeta*

1.7

1.7

Burayu*

1.6

1.3

Dessie

1.0

0.3

Dessie

1.5

0.4

Hosana

1.1

0.3

Top 10 towns sum (%)

85.2

79.1

76.3

72.6

60.5

62

Total

622

91,096

982

104,681

2,172

186,978

Note: (*) Indicates towns fewer than 100 km from Addis Ababa.

Source: CSA (various years), but own calculations.

Historically, MLSM establishments have been concentrated in a few large towns.10 Table 2.7 gives the geographical distribution of the Ethiopian MLSM establishments. In 1995/96 the top ten towns accounted for about 85 per cent of the total number of establishments and 79 per cent employment in the MLSM sector. In the same year Addis Ababa (the capital city) alone accounted for about 67 per cent and 60 per cent respectively. This concentration has shown a modest decline and in 2009/10 the top ten towns share reached 60 per cent and 62 per cent of the total number of MLSM establishments and employment respectively. The share of Addis Ababa also correspondingly declined to 40 per cent and 46 per cent respectively in the same year. New industrial towns such as Burayu, Sululta, Sebeta, Mojo, and Ambo are emerging around the capital city. Four of these were among the top ten (p.38) (p.39) (p.40) industrial towns in both 2003/04 and 2009/10. This brings the concentration in Addis Ababa and the surrounding towns (in a 100 km radius), for example in 2009/10, to above 48 per cent (number of establishments) and 52 per cent (in terms of employment).

Some of the current regional capitals such as Awassa in the south, Dire Dawa in the east, Mekelle in the north, and Bahr Dar in the northwest are also among the top ten towns hosting a substantial number of industrial establishments. The gap between the first and the next top ranking towns is, however, very large. For example in 2009/10, the first top town (Addis Ababa) hosted about eleven times more manufacturing establishments than the second largest industrial concentration town (Awassa).

2.4 Patterns of Industrial Productivity in the Ethiopian Manufacturing Sector

2.4.1 Productivity across Firm Size

Table 2.8 Productivity by size category for selected years (1995/96–2008/09)

Year

Value added per labour (1000s Birr) by employment size category

TFP by employment size category

<10

10–19

20–49

50+

10–19

20–49

50+

1995/96

9.44

16.96

28.25

−0.10

0.02

0.25

1999/2000

12.02

21.29

36.71

−0.15

0.00

0.28

2003/04

10.70

26.86

46.50

−0.20

−0.01

0.26

2007/08

8.2

10.84

21.63

42.59

−0.16

−0.03

0.17

2008/09

16.44

28.05

45.88

−0.15

−0.06

0.18

Avg. 1995/96–2008/09

12.85

34.02

42.63

−0.2

−0.016

0.22

Source: CSA (various years), but own calculations.

In this subsection we examine productivity difference by firm size. Table 2.8 gives the value added per labour across four broad size categories of the manufacturing enterprises for selected years between 1995/96 and 2007/08. The value added is adjusted for price change using a GDP deflator. The magnitude of the difference in labour productivity between the size groups is striking. For example in 2007/8, the value added per employee in large firms (employing 50+ workers) was about five times that of the micro-firms (employing fewer than ten workers), above four times that of the small size category (11–20 employees) and about double that of the medium sized enterprises category (21–50 employees).

The observed large difference in labour productivity is associated with differences in the proportion and quality of the input factors used in production. For example, large firms are more likely to use more capital than small (p.41) firms, and employ more skilled (and expensive) workers.11 To shed light on productivity differences while netting out the effects of input differences, we use total factor productivity (TFP) as an alternative productivity measure. The TFP is defined as the residual in a Cobb-Douglas production function for the MLSM sector.12 The estimation is based on firm level panel data covering 1996–2009. Similar to labour productivity, the TFP exhibits a big difference across the size categories in favour of the large size groups (see Table 2.8).

2.4.2 Productivity across Sectors

Table 2.9 Productivity and capital intensity in the MLSM, by industry

3-years average (2007/08–2009/10)

Industry

Value added per labour (000s Birr)

Capital per labour (000s Birr)

TFP

Basic iron and steel

126.97

102.19

0.64

Non-metallic minerals

85.35

82.15

0.44

Food and beverage

83.21

61.43

0.60

Fabricated metal

81.78

60.43

0.41

Rubber and plastic

76.75

42.94

0.53

Leather

73.13

26.74

0.61

Machinery and equipment

70.80

42.28

0.06

Motor vehicles

69.79

104.39

0.02

Chemicals

67.70

71.26

0.23

Wood

66.25

9.77

0.07

Apparel

56.75

12.87

0.57

Textile

50.71

21.31

0.30

Paper and printing

50.48

47.22

−0.01

Furniture

29.16

30.47

−0.33

All industries (mean)

72.3

52.5

0.30

All industries (median)

71.04

47.2

0.33

Source: CSA (various years); author’s calculations.

Table 2.9 reports average labour productivity and TFP in fourteen Ethiopian manufacturing industries based on the most recent three years average (2007/08–2009/10). There is a large difference in productivity across sectors. Industries such as basic iron and steel, other non-metallic minerals, food and beverage, and fabricated metals have relatively high levels of labour productivity and TFP. Furniture, paper and printing, textile, and apparel record low average productivity levels.

(p.42) 2.5 The Current Industrial Policy Framework

In 2002/03 the EPRDF-led government formulated a comprehensive IDS.13 IDS is based on the government’s broader development vision—Agricultural Development Led Industrialization (ADLI)—which was developed in the mid-1990s and subsequently elaborated. The IDS maintains that sustainable and fast industrial development can only be ensured if the sector is competitive in international markets. Hence, the export oriented sectors should lead industrial development and be given preferential treatment. The strategy explicitly further argues for a strong role from the state. Inspired by the East Asian experience the government has recently introduced the language of ‘developmental state’ as its policy principle regarding the public–private relationship. The strategy cites two important mechanisms in which the government could engage and promote the private sector, thus creating an environment conducive to and providing direct support for identified sectors. With regard to the first of these headings, the strategy paper identified the following intervention areas: (1) maintaining macroeconomic stability; (2) building a functioning and well-regulated financial sector; (3) creating dependable infrastructure services; (4) developing skilled and effective human resources; (5) creating an efficient civil service and legal framework; (6) developing industrial zones in major cities and towns with all required infrastructure facilities. The working paper version of this chapter examines these interventions in further detail (Gebreeyesus 2013). In what follows, we focus on initiatives aiming to support the identified sectors more directly.

The 2003 IDS declared priority sectors for government direct support which included textile and garment; meat, leather, and leather products; other agro-processing industries (e.g. sugar and sugar related industries); construction industry; and the micro- and small enterprises (MSEs). The list of priority sectors has been updated over time. For example the flower industry and some import substituting industries have been sequentially added. The targets and accompanying government supports were explicitly stated in the country’s five-year development plans, the so-called PASDEP (2005/06–2009/10) and the ongoing Growth and Transformation Plan (GTP) covering the period 2010/11–2014/15. The focus of this study is mainly on the former. The overall target for the industrial sector during the period of the PASDEP was to register an average annual growth rate of 11.5 per cent and thereby increase the sector’s share in overall GDP from 13.6 per cent in 2004/05 to 16.5 per cent by the end of 2009/10 (MoFED 2006). Targets were also set at sub-sector level with a particular focus on the selected export industries.

(p.43) In order to meet these targets the government provided extensive support, including economic incentives (credit, land, and duty and tax exemption), capacity building, and direct public investment. These support programmes have largely been directed at the exporting firms and industries. Exporters are eligible for various economic incentives. In addition to the efforts to boost the general human capital of the country the government has been involved in capacity building of the private sector particularly in the selected priority industries. In 2005, it launched an ambitious reform programme known as the Engineering Capacity Building Program (ECBP) with the aim of further enhancing the competitiveness of the private sector through improving workforce skills.

2.6 Industrial Policy in Practice—Some Industry Cases

Textile and leather are the two preferred industries receiving the most attention from Ethiopian policy makers for export promotion. The flower industry, which emerged spontaneously, is another celebrated industry. This section highlights the implementation and performance of industrial policy to date focusing on these sectors.

2.6.1 The Textile and Leather Sectors

According to the PASDEP five-year plan the textile and apparel sector was expected to generate foreign exchange of about US$500 million by the end of the plan period, i.e. 2009/10. It was envisaged that continuous investment support and expansion activities would be carried out by the government for more than 191 investors, with investment worth US$1.6 billion. Similarly, the leather industry was expected to generate US$500 million by the end of the plan period and at the same time to change the mix of exports towards processed and finished goods. The shoes sub-sector alone was anticipated to produce twenty million pairs of shoes and generate US$175 million and the other finished leather products to generate US$43 million by the end of the plan period.

In order to meet these targets the government made sector specific capacity building efforts in addition to the general support programmes given to all exporters. Two sector specific institutions, namely, the Textile Industry Development Institute (TIDI) and the Leather Industry Development Institute (LIDI), were set up to support, coordinate, and guide the private sector in these areas. These institutions have been implementing jointly capacity building programmes aimed at enhancing the competitiveness of the industries in international markets.

(p.44) Sector special training institutions were established under the guidance of these institutions. In 1998, the Ethiopian Leather and Leather Products Technology Institute (ELLPTI) was established to support the leather industry. ELLPTI provides training, laboratory testing, technical support, and consultancy to the sector. From 2003/04 to 2007/08 the institute provided training on footwear, leather, and leather products for more than 1,800 trainees. Another training centre, the Textile and Apparel Institute with facilities similar to those of ELLPTI, is also under establishment in support of the textile sector. At the time of writing, the scope of training is limited to basic tailoring skills but there are plans to expand the scope when complete implementation of the project is realized. To improve the international competitiveness of these industries the government introduced various additional support programmes, including benchmarking, institutional twinning, market research, and kaizen. In what follows we assess the implementation of these programmes with a focus on the textile and leather industries.

Benchmarking: In 2009, the government launched a programme called benchmarking with a view to upgrading technology and raising the capacity of the prioritized sectors, thereby increasing their competitiveness in the international arena. Under this programme selected textile and leather enterprises, both public and privately owned, will receive comprehensive direct support from internationally renowned firms sponsored by the government to alleviate their constraints in the areas of management, productivity, input supply networking, marketing of products, human resource development, and product quality.

Twinning programmes: The idea of the twinning programme is to establish long-term knowledge and experience sharing between the TIDI and the LIDI on the one hand, and other international and selected domestic and similarly renowned institutes on the other. In the leather sub-sector, a network has been established between the LIDI, Engineering Capacity Building, and selected foreign counterpart institutes (e.g. the Indian Leather and Leather Products Technology Institute). At the time of writing, this networking has enabled the LIDI to provide short-term training in the fields of leather technology and leather science.

Marketing and other support activities: Marketing products internationally is expensive and out of reach for many of the local industries. Moreover, whenever markets are available, the scale of operation of local firms is so limited that they cannot individually meet the order size from big companies. Realizing this constraint, the Ethiopian Embassies are endeavouring to promote the products and establish market links internationally; while at the same time the MoTI (Ministry of Trade and Industry) is coordinating and facilitating domestic firms to establish market networks for products and inputs among themselves through sub-contracting and other arrangements. (p.45) In this regard, domestic firms are linked with potential international buyers such as ARA of Germany, ADELCHI of Italy, ANNABELA of Italy, and the Brown Shoe Company of America.

Kaizen: The Kaizen scheme is a cross-sector initiative which was introduced in 2009 in Ethiopia as a two-year pilot project in thirty selected companies located in Addis Ababa and its environs. It is a Japanese management philosophy which aims at bringing incremental and continuous change. Encouraged by the results achieved in the selected companies in terms of increased productivity and improved quality, the government is envisaging a scale-up of its implementation to other companies. To this effect a Kaizen Training Institute was established by a regulation in October 2011.

Despite these efforts the textile and leather sectors’ export performance in the PASDEP period has been unsatisfactory. For example, the government envisaged US$500 million from exports from textile and garment industries by the end of the PASDEP period, but only about 8 per cent of this target (US$40.3 million) was actually achieved. Similarly, the plan for the leather industry was to generate US$500 million of which US$221 million was to come from shoe and other leather product sub-sectors. But only US$65 million worth of hides and skins and US$8.3 million worth of leather articles and footwear had been exported by the end of the plan period, respectively 23 per cent and 3.7 per cent of target.

2.6.2 The Flower Industry

In contrast to the textile and leather industries, the Ethiopian cut-flower industry represents an extraordinarily successful export diversification. In the mid-1990s two private domestic entrepreneurs started the experimentation by producing summer flowers and exporting to the EU market. A modern and large-scale floriculture industry began to emerge in the early 2000s. By 2008, the sector consisted of eighty-one farms that had created more than 50,000 jobs and generated more than US$100 million in foreign exchange. As a result the country became the fifth-largest non-EU exporter to the EU market of cut flowers and the second largest exporter in SSA, after Kenya. The sector continues to grow even further whereby the foreign exchange earnings from this sector have reached about US$189 million.14

The dynamics of the public–private interaction in the flower industry is different from other priority sectors. The flower industry was not even on the list of priority sectors at inception of the industrial policy document. Government was only lately aware about the export potential of this sector through (p.46) the experimentation of the private entrepreneurs and their effort to acquire government support. The early entrants faced a number of difficulties particularly related to logistics, land, and finance. In 2002, they formed an association and started to seek government support. The government responded quickly and positively, and finally at the end of 2002, the Prime Minister’s Office requested the MoTI to draw up a five-year plan of action for the sector, outlining the sector’s constraints and possible solutions. Based on the MoTI report, targets were set to put 1,000 ha under flower production by the end of five years.15

To meet these targets, the government came in with multi-faceted support starting from 2003, particularly focusing on: access to land, access to long-term credit, infrastructure, and air transport coordination. Prior to 2003, private investors in the flower sector were able to obtain land by leasing it from small farmers. With the government decision to engage in promotion of the sector, land held by the government was increasingly made available for flower farms within the vicinity of the airport in Addis Ababa, leading to the creation of flower enterprise clusters. The land was offered very cheaply with longer leases and payment periods. A survey carried out by this author in 2008 shows that the majority (83 per cent) of the source of farmland was by government lease. The government also made available cheaper loans with easier access for investors in the floriculture sector through the Development Bank of Ethiopia. Other forms of support include advocacy and capacity building.

2.7 Remaining and Emerging Challenges

The first decade of the EPRDF regime (1991–9) was marked by a series of reforms under the SAP to reverse the command economic system. In 2002/03, after a decade in power, the government formulated comprehensive IDS, which has been concretized into action by subsequent development plans and various sub-sector strategies. The government made notable efforts to implement the IDS. It particularly promoted selected export industries through setting targets and providing comprehensive support in meeting their targets.

Unlike many other reform policies that had to be agreed on with international financial institutions, the IDS was designed by the Ethiopian Government and based on its broad development vision, known as ADLI. The government has shown firm commitment and exerted maximum efforts in implementing it. (p.47) The strategy relies on the state playing a central role in guiding the private sector through the development process.

The industrial policy-making process in Ethiopia is characterized by more flexibility and scope for policy learning. In 1998, the government formulated an export promotion strategy which was narrowly defined as industrial policy. In 2002/03, a comprehensive industrial policy was designed, which has a list of several sectors for support including meat, leather and leather products, apparel and textile, other agro-processing sectors, construction, and small and medium enterprises (SMEs), and a package of incentives. Later, floriculture was added to the priority list after it emerged largely through entrepreneurial experiment from the private sector. The list of priority sectors has recently been adjusted to include import substituting sectors such as metal and engineering, chemical, and pharmaceutical industries. Over time, the government has also expanded its policy instruments to promote the selected industries.

The performance so far has been mixed and manufacturing export performance so far has not been satisfactory. Recent studies highlighted poor trade logistics and lack of quality inputs in the domestic market to be among the major constraints reducing global competitiveness of the textile and leather sectors in Ethiopia. The government has taken different measures to address these problems. For example, it implemented a wide range of reforms to modernize the civil service and in particular the customs administration. It invested in infrastructure in order to reduce the cost of doing business. Still, the poor trade logistics and lack of quality inputs continue to be a drag on the export sectors.

In contrast to the textile and leather, the Ethiopian flower export is booming and has become one of the major export sectors. Why has the flower industry, which was not even in the original government priority list, emerged to be so successful but not the leather and textile industries that long before received the most attention? This leads to the identification of potential products, an issue as prominent as the industrial policy itself. The Ethiopian case study suggests that although governments might have their own list of priority sectors a priori, they should allow for flexibility and revise their choices over time. To do so one needs to introduce a mechanism to elicit valuable information through continuous consultation with the private sector on potential industries. This is consistent with the Hausmann and Rodrik (2003) view that there is large element of uncertainty as to what a country will be good at producing, once we move beyond the broad aggregates (e.g. labour-intensive or natural resource-based) that are predicted by factor endowment models.

Several other issues arise regarding the effectiveness of the support programmes and the implementation of industrial policy. How much rent and how long should the private sector in the selected areas be given to bear fruit? (p.48) What form of relationship should be instituted between the government and the private sector? How do you create an environment that maximizes the social benefits while limiting rent-seeking? In this regard, Ethiopian industrial policy made a distinction between ‘developmental’ and ‘rent-seeking’ private sector; and vows to support the former and curtail the latter. It provides generous incentives and support programmes to build the private sector capacity and has also created public–private consultation forums (carrots). At times (particularly recently) it has introduced a number of measures (sticks) alleging to ‘discipline’ the ‘rogue’ private sector, which have resulted in increasing tension and policy uncertainties around the private sector. There are critics that maintain that these instruments (carrot and stick) are not transparent and policy makers tend to ‘patronize’ the private sector instead of encouraging competition and innovation. Failing to address these concerns might also damage the efficacy of industrial policy.

References

Bibliography references:

Central Statistical Agency (CSA) (various years[a]). ‘Report on Small-Scale Manufacturing Industries Survey’. Addis Ababa: CSA.

Central Statistical Agency (CSA) (various years[b]). ‘Report on Medium and Large Manufacturing and Electricity Industries Survey’. Addis Ababa: CSA.

Central Statistical Agency (CSA) (2003). ‘Report on Cottage/Handicraft Manufacturing Industries Survey’. Addis Ababa: CSA.

Federal Democratic Republic of Ethiopia (FDRE) (2002). ‘Ethiopian Industrial Development Strategy’. Ethiopia (Amharic version). Addis Ababa: FDRE.

Gebreeyesus, M. (2013). ‘Industrial Policy and Development in Ethiopia: Evolution and Present Experimentation’. WIDER Working Paper 2013/125. Helsinki: UNU-WIDER.

Gebreeyesus, M. and Iizuka, M. (2011). ‘Discovery of the Flower Industry in Ethiopia: Experimentation and Coordination’. Journal of Globalization and Development 2(2): 1–27.

Geiger, M. and Goh, C. (2012). Chinese FDI in Ethiopia: A World Bank Survey. Washington, DC: World Bank. Available at: <http://documents.worldbank.org/curated/en/2012/11/17073974/chinese-fdi-ethiopia-world-bank-survey>, accessed 10 July 2013.

Hausmann R. and Rodrik, D. (2003). ‘Economic Development as Self-discovery’. Journal of Development Economics 72: 603–33.

Imperial Ethiopian Government (IEG) (1957). First Five Year Development Plan (1958–1962). Addis Ababa: IEG.

Imperial Ethiopian Government (IEG) (1962). Second Five Year Development Plan (1963–1967). Addis Ababa: IEG.

Imperial Ethiopian Government (IEG) (1968). Third Five Year Development Plan (1969–1973). Addis Ababa: IEG.

Ministry of Finance and Economic Cooperation (MoFED) (2006). ‘Ethiopia: Building on Progress: A Plan for Accelerated and Sustained Development to End Poverty (p.49) (PASDEP) (2005/06–2009/10)’ Addis Ababa: MoFED. National Bank of Ethiopia (NBE) (2010), Annual Report 2009/10, available at: <http://www.nbe.gov.et/publications/annualreport.html>, accessed 10 July 2013.

Provisional Military Administrative Council (PMAC) (1975). Declaration of Economic Policy of Socialist Ethiopia. Addis Ababa: PMAC.

United Nations Industrial Development Organization (UNIDO) (1991). Ethiopia: New Directions of Industrial Policy. Industrial Development Review Series, Regional and Country Branch. Vienna: UNIDO.

World Bank (1985). Ethiopia: Industrial Sector Review, 1985. Washington, DC: World Bank.

World Bank (2010). World Development Indicators. Washington, DC: World Bank, available at: <http://databank.worldbank.org/ddp/home.do>, accessed 10 July 2013.

Notes:

(1) For the policy document see IEG (1957).

(2) The cottage and handicrafts sectors were also recognized in the plan as playing an important role for the development of manufacturing industry in the country.

(3) For the two policy documents see IEG (1962, 1968).

(4) World Bank (1985) provides one of the few detailed and available studies on the Ethiopian manufacturing sector carried out at that time. We would like to acknowledge that because of an absence of other sources, discussions in Subsections 2.2.1–2.2.2 rely heavily on this World Bank (WB) report.

(5) Source: World Development Indicators from the World Bank (WB 2010).

(6) For example, the maximum import tariff rate was further reduced from 60 per cent to 50 per cent; fertilizer price control was suspended; liberalization of the foreign exchange system was strengthened; and a foreign exchange retention scheme was introduced.

(7) Source: National Bank of Ethiopia (NBE 2010).

(8) The dominant activities in this sector were the food and beverage, textile, and non-metallic mineral products combined, accounting for 86 per cent of total cottage/handicraft establishments. About 63.2 per cent was located in rural areas with the remaining 36.8 in towns.

(9) The working paper version of this chapter (Gebreeyesus 2013: Table 3) shows the distribution of number of establishments, employment, and value added across the four size categories: micro (1–10 employees); small (11–20 employees); medium (21–50 employees); and large (over 50 employees); firms based on 2007/08 CSA surveys on SSM and MLSM.

(10) For example in 1980/81 the two provinces, Shoa and Eritrea, in which the largest two cities Addis Ababa and Asmara respectively are located, accounted for more than 80 per cent of the MLSM sector in terms of number of establishments, employment, and value added (WB 1985).

(11) Other reasons why large firms might be more productive than small firms include the fact that they are more likely to use better technology and management. They are also relatively more favoured in the capital market than are small firms.

(12) Micro-enterprises with ten or fewer employees are excluded from the estimation as there is an absence of time series data for the specified period.

(13) FDRE (2002). As far as we know the original industrial policy document is in the local language, Amharic; as yet there is no official English translation.

(14) Gebreeyesus and Iizuka (2011) have an excellent discussion of the emergence of the floriculture industry in Ethiopia. The discussion that follows relies heavily on that.

(15) These targets were updated in the PASDEP five-year plan.