Information - ARRIDE - IDE-JETRO

Transformation of a Woodworking and 

TitleDistrict in Kampala, Uganda: Dichot 

SME Cluster and Large Firm Sector 

Author(s) Yoshida, Eiichi 

Citation IDE Discussion Paper. No. 171. 2008 

Issue Date 2008-10 

URL 

http://hdl.handle.net/2344/790 

Rights 

http://ir.ide.go.jp/

INSTITUTE OF DEVELOPING ECONOMIES 

IDE Discussion Papers are preliminary materials circulated 

to stimulate discussions and critical comments 

 

IDE DISCUSSION PAPER No. 171 

Transformation of a Woodworking 

and Furniture Industrial District in 

Kampala, Uganda: 

Dichotomous Development of SME 

Cluster and Large Firm Sector 

Eiichi YOSHIDA* 

Abstract 

Clustering small manufacturers are believed to attain various types of collective 

efficiency. A woodworking and furniture SME district in Uganda has created a 

learning environment for artisans to start up their own workshops. In the district 

workers can access various managerial information including business skills and 

input materials easily than outside. Hence it attracted new entrants to follow and 

district growth continued. On contrary large firms are locating separately and 

dispersedly from SME district and have a negative image to SME. This dichotomy 

has been created partly through spatial division of two sectors and partly through 

policy favouritism toward large firms. 

Keywords: SME, cluster, agglomeration, incubation, woodworking, furniture, Uganda, 

Kampala 

* Research Fellow, African Studies Group, Area Studies Center, IDE 

(Eiichi_Yoshida@ide.go.jp)

The Institute of Developing Economies (IDE) is a semigovernmental, 

nonpartisan, nonprofit research institute, founded in 1958. The Institute 

merged with the Japan External Trade Organization (JETRO) on July 1, 1998. 

The Institute conducts basic and comprehensive studies on economic and 

related affairs in all developing countries and regions, including Asia, the 

Middle East, Africa, Latin America, Oceania, and Eastern Europe. 

The views expressed in this publication are those of the author(s). Publication does 

not imply endorsement by the Institute of Developing Economies of any of the views 

expressed within. 

INSTITUTE OF DEVELOPING ECONOMIES (IDE), JETRO 

3-2-2, WAKABA, MIHAMA-KU, CHIBA-SHI 

CHIBA 261-8545, JAPAN 

©2008 by Institute of Developing Economies, JETRO 

No part of this publication may be reproduced without the prior permission of the 

IDE-JETRO.

1. Introduction 

Implementation of industrial cluster policy requires that the government prioritize specific 

industrial sectors and industrial space. In the early period after independence, many Sub-Sahara 

African (SSA) countries pursued industrialisation through large-scale parastatal industries; however, 

this policy resulted in many liquidations in the public sector, which proved that the policy direction 

was not suitable for industrialisation in these countries. Since then, many SSA countries have not 

had any particular policy focusing on specialisation in an industry sub-sector, primarily because a 

policy consensus was reached with donor countries that focused on poverty reduction and AIDS 

care in recent years and on post-civil war restructuring in many countries, including Uganda. Due 

to this, laissez-faire has remained the basis of governments’ industrialisation concepts, and the 

domestic industrial environment has remained exactly as it was in the 1960s and 1970s. 

Uganda, one of the Least Developed Countries (LDC) and Heavily Indebted Poor Countries 

(HIPCs) in SSA, had not paid much attention to industrial priorities or to priorities in economic 

planning, other than structural adjustment. The industrial areas were zoned, and some estates, 

which were developed in the early years after independence, were occupied by the parastatal 

industries and public investment. Some parts of these plots became vacant following a series of 

divestures and privatisation failures. On the other hand, as in many other SSA countries, small 

industries in Uganda have grown substantially and have come to dominate the urban industrial 

landscape. 

In the industrial landscape of Uganda, small firms in a similar production range created industrial 

districts which are called clusters or agglomerations; whereas large firms in a similar production 

located themselves in a dispersed manner in the industrial zones developed around the city. 

However, the current concept of spatial planning and zoning has not taken into account the 

slum-like growth of SMEs’ spatial concentration. Hence, SME sub-sectors including furniture, 

metalworking, and other industries were excluded from industrial policy formulation. Instead, 

information from external sources made the local governments aware of the significance of the 

geographical concentration of small firms as a form of industrial cluster. Even in Uganda, despite 

the fact that a visible agglomeration of small firms in the woodworking and metalworking sectors 

exists, the government recognised neither the roles these firms played nor the significance of 

prioritisation of industry or industrial space. 

This paper deals with the significance of clusters made up mainly of SMEs in an African LDC 

context. As these are clusters formed without relevant cluster policy or industry prioritisation policy, 

they may present a slightly different context than other cases in this publication. This study selected 

1

Uganda as a typical LDC in the SSA region and chose the woodworking and furniture industry as a 

typical SME cluster sector in SSA. There are many small industry clusters or districts in SSA 

without any relevant policy guidance and which maintain no mutual relations with large firms. In 

this sense, it is significant to examine how the dominant SME clusters transformed separately from 

the large firms without policy guidance and to study what kind of intervention should be proposed 

in the event that this trend strengthens. 

This paper begins by introducing the global trend in the woodworking sector and attempts to 

map out where the Ugandan woodworking sector might be located in this global trend before 

looking at woodworking cluster formation. Then, we see how the trajectory of sectoral dynamism is 

separate from the mushrooming of woodworking SMEs. As mentioned above, SMEs form clusters 

and large firms are dispersed, so this study discusses the two sectors separately. In contextualising 

the SMEs’ location preference within and outside of the districts, this study examines how the SME 

clusters maintained distance and stood apart from large firms in their institutional aspects, 

particularly from an industrial spatial planning point of view. Finally, the chapter concludes by 

discussing the problems caused by this dichotomous development of SMEs and large firms. 

2. Global Trends in Furniture Production and Trade 

A major change has been emerging during the past decade in the traditional location patterns of 

furniture and woodworking industries throughout the world, due to changes in global demand. 

These changes have been caused by the globalization of modern urban lifestyle, the growth of a 

mass urban middle classes in developing countries, and modernization of product distribution 

systems. It is also becoming difficult in recent years to distinguish between furniture designs of 

different countries when they are exhibited side by side, for example at the ‘Milano Salone’ 

International Furniture Trade Show or the International Furniture Fair Tokyo. Along with the 

expanding distribution of modular furniture, production and exports have been dramatically 

increasing in China and East European countries where supplier companies are concentrated. South 

Africa, being rich in forest resources, has been partly incorporated into the furniture value chain 

developed by the global woodworking industry, as components processed in South Africa are 

moved on to further processing outside the region. In terms of the trade volume of African 

countries, distribution has been increasing in East African countries such as Malawi, Uganda, and 

Kenya (Tables 1, 2, 3) (Kaplinsky et al. 2002, 2003; Kaplinsky and Readman 2004; Loebis and 

Schmitz 2005). 

However, these international trade statistics do not show the overall trend of the entire 

woodworking industry in these African countries. The reality in these countries is that small, 

2

medium, and micro enterprises in the sector generate the majority of the industry’s production; 

however, they have been excluded from this healthy trend of global production growth and have in 

fact been facing a severe struggle for survival in a less dynamic domestic consumer market in 

recent years. Small, medium, and micro enterprises (SMEs) involved in woodworking, as in many 

other developing countries, form spatial concentrations or agglomerations, even in poor countries 

like Uganda. However, between SMEs and large firms in Uganda’s woodworking sector, there 

exists not only a spatial division but market segregation as well, with the upper market reserved for 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

large firms and the rest for SMEs. Needless to say, export markets are hardly accessible to SMEs, 

which were accustomed to shipping their handcrafted products to neighbouring Sudan, Rwanda, 

and Congo before regional conflicts disrupted regional trade routes. 

The above-mentioned shift in the international trade of furniture and woodworking products has 

two primary causes. The first of these is a change on the consumer side (i.e., global urbanization), 

and the other is a change on the supplier side (logistical system development). Due to the 

worldwide trend toward urbanization, along with the information diffusion through globalized 

media like satellites, television, and the Internet that typifies modern urban lifestyle, furniture 

consumers have become more familiar with modern design styles that are simple and minimalist as 

well as lightweight. These urbanized consumers have less residential space than their predecessors, 

and their household units are of a wider variety, including not only those who are unmarried and 

married-with-children, but also those who are single, living with younger and older generations, 

single mothers, pensioners both single and couples, divorced, and widowed. This wide variety of 

households leads to different lifestyles such as multi-habitation, co-habitation, connected housing, 

collective housing, same-sex habitation, and so on. The demand for household space and furniture 

has this become more complex, demanding flexibility on the producer side. 

3

Modular furniture has also enabled the globalization of distribution. Modular furniture 

(assembled by consumers) distributed through mass merchandise chain stores such as IKEA, as 

well as sales through online catalogues such as Belle Maison in Japan, have succeeded in entrusting 

the task of assembly to consumers. This has reduced the bulkiness of furniture transportation, 

leading to significant cost reduction in furniture production and distribution. This situation has 

promoted the globalization of supply management, which in turn has promoted the globalization of 

the supply chain network. 

Consumers, mainly those in the urban middle class in developing countries, have also begun to 

show strong preferences for global, modern designs over traditional, heavy, extravagant styles. The 

traditional industries of furniture, woodworking, and woodcraft even in the developing world have 

been influenced by these changes. Also, due to the above-mentioned systematization of production 

and globalization of distribution, distribution costs for modular furniture were reduced, thereby 

enabling transactions with distant lumber sources and woodworking production centres. Some 

non-western home ware products played a role in this; witness how East Asian furniture, including 

"futons" (Japanese sleeping mattresses), was modified and adopted into the modern western 

lifestyle in the 1980s and 1990s. With this development, influenced by the so-called “ethnic boom,” 

traditional design models were applied to meet the demands of modern western lifestyle. In Japan, 

demand increased for Balinese furniture, Vietnamese furniture, and wood products to fit into an 

ordinary Japanese room size. These various furniture products were increasingly exported through a 

number of different channels. 

On the production side, evidence of this global trend can be observed in the growth of exports 

from transitional economies, in terms of production volume and added value in the overall 

4

manufacturing industry, starting from 1993. This growth was strongly displayed in 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

the furniture and woodworking sector. Growth of manufacturing in developing areas, excluding 

transitional economies, was powered by production of radios, televisions, communications 

equipment, and automobiles; however, the growth in furniture and woodworking was somewhat 

slower. 

Nevertheless, the furniture and woodworking sector in East Africa also showed solid 

development, as shown in Table 4. Annual economic growth in the 1990s exceeded six percent in 

Uganda, and industrial production in Uganda exceeded the average growth rate for all 

manufacturing industries in Sub-Sahara Africa, except communications equipment. The growth of 

furniture production maintained a similar level. Similarly, the increase in production in the furniture 

and woodworking sector in Kenya was 2.5 times greater than the African average. Although 

manufacturing businesses stagnated in Kenya during this period, the furniture and woodworking 

industry showed the largest increase in production volume among all industries, with a 6.5 percent 

growth rate (Table 5). 

In terms of trade volume, African countries in general are not in the top tier in worldwide 

exports, and South Africa ranks the highest, at 28th. However, these statistics include only a very 

limited number of business types. In the Ugandan Industrial Production Index, for example, data on 

the nation’s furniture sector was sourced from only seven companies, and so the data does not 

cover numerous SMEs that belong to Uganda’s aforementioned industrial agglomerations. 

According to the Uganda Business Register, which includes SMEs, there are more than 3,000 

business establishments in the furniture and woodworking sector in Uganda, including 

family-operated backyard businesses as well as small roadside workshops. The discrepancy 

between these two data sets indicates that most business establishments in the furniture and 

5

woodworking industry are SMEs that do not appear in the rosy international trade statistics of 

recent years. 

3. Historical Overview of the Furniture and Woodworking Industry in Uganda 

In the central part of Africa, the vast tropical rain forest spreads from the Congo Basin into 

Uganda, which is endowed with abundant forestry resources including mahogany, mvule, mgavu, 

and other hardwood species. Since there are also extensive forestry resources along the coast of 

Lake Victoria, Uganda has acted as a supply centre of railway sleepers and other components since 

the colonial period, ever since the colonial railway was laid between Kampala and Mombasa. Prior 

to Uganda’s independence in 1962, the domestic industry was mainly composed of colonial 

enterprises that farmed and processed traditional products including sugar, tea, and coffee. There 

was limited interest in the localization of manufacturing industries among colonial settlers, other 

than the above-mentioned industries for small local consumption markets, which were supplied by 

Indian immigrants. Craftspeople from South Asia played an important role in this early stage. 

Compared with other former British African colonies, where there were restrictions on the 

residency of indigenous African people in the cities, Uganda‘s land system or tenure was unique in 

that the Buganda Kingdom was allotted a parcel of land by the colonial authority to administer 

autonomously. The city of Kampala, which was the capital city of Uganda Protectorate, and the city 

of Mengo, which was the capital city of the Buganda kingdom, overlapped to form a twin city 

much like Budapest. So, it was not impossible for Africans to secure the freedom to live in the city 

6

if their residence was on the Buganda kingdom side. However, for Ugandan people, starting 

businesses was, in general, extremely difficult, as access to technical education as well as finance 

was extremely limited. 

The scale of the woodworking industry sector in Uganda prior to independence was extremely 

limited compared to the scale of the plantation sector. The woodworking sector mainly produced 

the railway sleepers necessary for laying the rail of the Uganda Railways to connect the Indian 

Ocean and the inland plantations, The woodworking sector also covered the needs of the increased 

settler population; to meet the settlers’ needs, the woodcraft jati/caste from Gujarat, India, and a 

group of South Asian Muslim woodcrafters who settled in Uganda played an important role in 

importing woodworking skills and technology (Ministry of Information, Broadcasting and Tourism 

1964). 

After independence from the UK in 1962, restrictions on migration to, and employment in, 

cities affecting Ugandan people were eased, and the environment for business start-ups was 

dramatically changed; however, most of the woodworking technology and its management were 

still in the hands of the Indian immigrants. The problems associated with industry ownership by 

Indian immigrants after independence led to the idea that the Indians should hand over the means of 

production to the native people of Uganda, which consequently resulted in the forcible realization 

of the so-called Africanization of the economy in 1972. In August 1972, the Idi Amin regime 

ordered foreign expatriates, including Indians as well as anyone who carried a British passport, to 

leave the country within three months, resulting in the departure of Indian craftspeople and foreign 

owners from the woodworking industry and the abrupt devastation of the sector. 

Even in the furniture and woodworking sector, large firms and small workshops were handed 

over to indigenous Ugandans. Some owners legitimately purchased their holdings from deported 

Indian immigrants, while others were given these as gifts by the dictatorial president, Idi Amin. A 

small number of remaining settlers and Ugandan craftspeople who formerly worked in the industry 

were forced to become independent craftspeople or spin off their own companies. However, the 

domestic economic environment, including the woodworking industry as well as the overall 

manufacturing sector, was plunged into long-term stagnation from 1972 onward, which persisted 

through the period of intermittent civil war between 1968 and 1986. A new industrial landscape did 

not arise, and agglomeration did not develop. 

Since Yoweri Museveni, the president to date, assumed power and stabilized internal affairs in 

1986, the economic environment for the manufacturing and woodworking industry has changed 

considerably. Development assistance funds began to flow in, and thereafter private investment 

increased gradually. Deported Indians began to return, spurred by the government's decision in 

1992 to return the property and assets of deported former residents. This accelerated the inflow of 

private funds, which led to an increase in domestic investment as a whole. 

7

The increased inflow of development assistance funds, supported by the stable government, 

helped expand the amount of investment funds flowing into the cities, initiating a boom in the 

housing and construction sector and in the promotion of business in general. 

The business environment for SMEs in Uganda thus went through four basic stages: 

1. The period of colonization, when the inflow of Ugandan people into cities was restricted 

and free business promotion was impossible; 

2. The early period of civil war, when arguments over the ownership of production 

heightened and few spin-offs from Indian production facilities were possible; 

3. The period of turmoil when the economy deteriorated, and business management as well as 

registration of business establishments were extremely regulated; and 

4. The current period, in which the free movement of people and the conducting of private 

business have become largely possible. 

4. Methodology 

This chapter examines how the locational concentration of SME clusters has been transformed, 

using the Ugandan woodworking and furniture industry as an example. In order to make the 

historical transformation visible, this study employed the flowchart approach to industrial cluster 

transformation, which sets up a time series framework for analysing industrial cluster development 

(Kuchiki 2005). 

In order to characterise how the concentration had been chosen by SMEs and how the 

stand-alone position had been selected by large firms, we conducted our own survey by collecting 

data from major furniture and woodworking firms as well as from SMEs in the agglomerated areas. 

For comparison purposes, data was collected from dispersed firms in small towns outside the 

market thresholds of the above-mentioned agglomerations. 

This survey was conducted in the Bwaise area in the city of Kampala, the largest woodworking 

SME agglomeration, from January through August of 2002. At the same time, research was 

conducted on 30 workshops in the Natete Area, a smaller agglomeration in one of the suburban 

commercial centres of Kampala. For the purpose of comparison between agglomerated and 

dispersed firms, data was collected in 2004 on approximately the same number of companies in 

Kabale, a local town of 30,000 inhabitants and the district administration centre of one of Uganda’s 

58 local districts. 

This study collected data from firms in the woodworking and furniture industry districts where 

SMEs are concentrated. Moreover, data was separately collected from large firms which are located 

in a dispersed manner and are distant from these SME agglomerations. The above-mentioned two 

areas are easily recognisable along the major traffic route in Kampala, capital city of Uganda. More 

8

contextual explanation of the examined areas will be provided below. This range of data allowed us 

to compare the characteristics of agglomerated firms and non-agglomerated ones. 

Firms studied were engaged in actual woodworking processes, including machine work and 

manual work, and/or designing work. Firms excluded from the survey were those who supply 

timber, fabrics, and products other than furniture products for wholesale and retail. In-depth 

interviews were conducted in 2002 with the help of interpreters fluent in 

Luganda/Nyankore/Bakiga and English. Data was collected on the general characteristics of the 

firms, general locational preferences, investment motivation, and managerial problems faced by the 

firms. 

Apart from the above-mentioned three SME cases, interview surveys were arranged with five 

large woodworking firms out of 13 firms in the Uganda Business Directory, all of which were 

located individually in different industrial zones. So, this study assumes that the data from the large 

firms does not reflect the character of the SME industry clusters. 

5. Locational Preference among Woodworking and Furniture Firms 

In this section, the decision-making process of large firms will be examined. Large-scale 

woodworking companies (hereinafter referred to as "major woodworking firms") in this case are 

member companies of the Uganda Chamber of Commerce (UCCI) or the Uganda Manufacturers' 

Association (UMA), or else they were selected based on their being listed in the Uganda Business 

Directory issued by Monitor News and being equipped with corporate landline phones (Monitor 

2002). 

Major woodworking companies are individually located in different industrial zones, including 

the Kampala Industrial Area, Kireka, Kawenpe, Bwaise, Nalukorongo, and Port Bell. These areas 

are all in the statutory zoned areas created under the Kampala Planning Act. In the Kampala 

metropolitan area, the land zoning concept is strictly applied and trade licenses are rarely granted 

outside of the designated zones; therefore, location other than in areas designated for an industry 

would be difficult (USAID Presto Project 1999). 

In an attempt to compare the locational preferences of major companies, this study utilized the 

responses from five out of the 13 companies which cooperated with our interviews in Kampala. 

Four out of five of these companies were recipients of foreign direct investment (FDI). The history 

of each company’s investment is relatively short; the first company arrived in 1993, one year after 

the return of assets to Indians was publicized. 

The results of our interviews indicated that the most important factors affecting location 

preference were the rent, price of plots, and the availability of skilled labour, followed by the costs 

of infrastructure, such as factory construction, utilities, etc. Other significant factors included road 

9

infrastructure, access to financial institutions, and the existence of local consumer markets, in 

addition to access to suppliers of raw materials and good living environments. 

Factors of minor importance included levels of corruption, personal influence, availability of 

unskilled labour, proximity to companies of similar origin, and availability of knowledge, 

information, and technology, etc. The access to export markets as well as subcontractors around 

Kampala was also of limited importance. 

The availability of subcontractors is usually of high importance to non-vertically-integrated 

industries, but this matter was not given much attention in research conducted by the Uganda 

Investment Agency concerning domestic FDI companies' awareness of the Ugandan investment 

environment. In this Uganda Investment Authority study, the stabilization of political affairs was 

overwhelmingly mentioned as a motivating force for investment. Uganda’s period of civil war in 

the 1970s and 1980s resulted in withdrawal of investments and temporary abandonment of assets. 

Many returning Indian managers who actually experienced this period mentioned the critical 

condition of internal affairs during that time. Next in importance as an agglomeration factor was the 

size of the domestic market, followed by physical infrastructure, including telephone lines, roads, 

factory plots, and electricity (Uganda Investment Authority 2003). 

In other research by UIA, power supply issues such as power outages, supply shortages, and 

unstable voltages were raised as issues for the management and administration of companies of all 

sizes. Lack of landline phones, maintenance delays, and poor roads were raised next as obstacles to 

development (Table 7). Based on the above results, stabilization of domestic politics was found to 

be the top priority for promotion of industrial development. However, once the political situation is 

stabilized, the availability of industrial sites becomes important for SMEs, and stable and 

inexpensive infrastructure becomes necessary for large firms. Thus, access to suitable industrial 

plots seems to be critical for firms of all sizes. 

The following section deals with this real estate issue from the standpoint of SME managers 

10

and large-firm investors in Uganda. A central question is: “Why is access to industrial plots raised 

as such an important item?” 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

11

6. Woodworking and Furniture SMEs in Urban Industrial Districts and Local Towns 

6.1 Characteristics of the Surveyed Areas 

Uganda’s manufacturing industry is primarily concentrated in four sectors: metal processing, 

woodworking, leather processing, and food processing. 1 These sectors, as seen in Tables 6 and 7, 

are dominated by SMEs. This implies that people can start these businesses with little initial capital, 

limited manual tools, and small workshop space. Among these sectors, the metal processing and 

woodworking businesses are known for their obvious agglomeration along major roads in Kampala 

(Tables 8 and 9). 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Through in-depth interviews and with paper questionnaires, the interviews were conducted in 

the largest agglomeration of furniture and woodworking firms in the country, in the 

Makerere-Kavule/Bwaise area, and in a smaller agglomeration in the Natete area chosen for 

comparison purposes. Also, an area of dispersed firms in the town of Kabare was covered for 

comparison purposes. 

The characteristics of three areas are as follow. 

Area B (Bwaise/Makerere-Kavule, Kawempe Division, Kampala City, Uganda) contains the 

1 The formation of a stable administration in 1986 ushered in a period of free personal and residential movement and promotion 

of private business, which carried vast consequences for Ugandan SMEs. Despite historically restrictive polices governing 

Ugandan SME start-up, structural adjustment policies from the late 1980s through the early 1990s resulted in a massive number of 

people leaving the public sector to start businesses, along with Ugandan returnees from Japan and western countries. These 

returning migrants’ remittances also encouraged those who received the funds to start SMEs. These tendencies are especially 

observable in the manufacturing sector. 

12

largest number of woodworking firms in the entire Great Lakes region. Located between the city 

centre and the suburbs, it has become an un-zoned de facto industrial area for small and medium 

businesses, and the area also accommodates a university, primary schools, and churches and 

mosques. Firms and workshops are spread along the two major paved roads, which meet at a 

roundabout, with the heaviest concentration of firms observed in the space surrounding the 

roundabout. Seventy-five workshops are located in this area. The 70 that responded to our survey 

are clustered in a space of about two square kilometers. The study thus regards Area B as 

representing a large and concentrated industrial district. 

Area N (Natete, Kampala City, Uganda) is located along the major national highway leading to 

the western region of the country and continuing into neighbouring Rwanda. The area 

accommodates about 40 workshops. The 31 of these that responded to us are clustered in a space of 

about four square kilometers. The study thus assumes that the firms in Area N represent a 

smaller-scale agglomeration. 

Area K (Kabare Town, Kabare District, Uganda) accommodates about 40 workshops and firms 

in an area of 30 square kilometers. Thirty of these were successfully surveyed. These firms and 

workshops are spread out across the town, with some located in a group of several workshops and 

some standing alone. The study assumes that the firms in Area K represent dispersed 

(non-agglomerated) firms (Table 10). 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6.3 Site Selection Process during Agglomeration Formation 

The survey indicated several main reasons why the woodworking SMEs chose the 

agglomeration sites as their locations. 

The highly agglomerated Area B was selected by SMEs primarily because it lies along a major 

traffic road, allowing access to transportation and contact with customers (creating publicity for the 

firms). The next most common reason, cited by approximately one-third of the respondents, is that 

13

there is already a permanent pool of customers in the area. A similar response was that that there is 

a market close to the commercial centre, implying that locating in the area creates high publicity for 

the firm. Taken together, these two responses represent the single most important factor in the 

firms’ decision to locate in the region. Other responses included ease of obtaining plots of land for 

industrial use (Table 11). 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

In contrast, in Area N, there was no reason for agglomeration given by more than half of all 

firms surveyed. The ease of securing a location was cited by about one-third. As in Area B above, 

another reason was the presence of customers/market/publicity in the agglomeration site. Other 

reasons included the region’s location along a major intercity road as well as the ease of obtaining 

input goods. 

In Area K which contains the dispersed firms, two-thirds listed the region’s proximity to a 

central commercial district as a reason for locating there, and low rent was cited by nearly half. 

Agglomeration-related reasons, such as the presence of a market, were not mentioned very often. 

Judging from the responses given in these three areas, the availability of relatively inexpensive 

workshop plots seems to be a key issue. Since major traffic highways are cited as being more 

important than local road infrastructure, it seems that the publicity effect resulting from location 

along a major paved highway is critical for a small furniture workshop. Existence of a market or 

proximity to the centre of a commercial district also means that the area’s centrality attracts a 

general mass of shoppers to the area, who pass by the woodworking shops on the main roads. 

The agglomerations discussed here were transformed substantially over the 1980s and 1990s. 

In the beginning, some series of trigger events induced gradual agglomeration of companies in the 

spaces, and various forces made this agglomeration self-reinforcing (discussed further below), 

leading to the expansion of the agglomerations. Companies which located in the region at the early 

stage of the agglomeration process (the “early group”) might perceive the area differently, and also 

may have utilized different decision-making factors regarding location preference, than those who 

located there after the agglomeration grew sizeable enough to be well-known by customers and 

producers outside the area (the “middle stage group” and “late group”). 

Until the deportation of foreign settlers in 1972, there were at least two furniture factories 

14

managed by Asian/Indian crafters in Area B, and Asian owned sawmill and Asian timber 

wholesalers were located in Area N. It is likely that there were timber wholesalers and sawmills, or 

furniture plants owned by Indians in various parts of the city. This indicates that woodworking 

businesses were formerly dispersed throughout the city and that agglomeration was initiated by 

factors in the 1980s. In fact, one company in each area indicated in an open-ended response that 

they initially chose the area because there were fewer woodworking businesses there in earlier days 

and that they then preferred a dispersed location. 

Companies in Area B’s early group, which located there at the earliest stage of agglomeration, 

state as their reasons both the area’s location along a highway and the existence of a market; 

however, the group of firms present in the area before 1986 (i.e., the group that located there prior 

to economic reconstruction) overwhelmingly chose only the area’s location along a major traffic 

zone (Table 12) (Ministry of Commerce, Industry and Tourism 1971). 

In Area N, each group cited the availability of workshop space, and this was the case for the 

early group as well as the newcomers. On the other hand, the mid-stage and late groups also cited 

the publicity associated with locating in Area B, which was not a reason given by the early group. 

The middle stage group also mentioned the existence of customers in the vicinity, which was not 

mentioned by the early group either. In Area N, the middle stage group mentioned easy access to 

input goods and the existence of a market; however, there was no definite time-dependent trend in 

reasons given for location preference. 

These results appear to indicate that firms in Area B seemingly valued the area’s proximity to 

CBD and the low cost of plots, along with the heavy traffic through the area. Moreover in recent 

years, the areas’ reputation as a furniture production centre has been established, and this has 

attracted SMEs looking for advantageous locations. In the case of the smaller agglomeration in 

Area N, it appears that the agglomeration has not grown enough to attract firms en masse to the 

area. 

Groups arriving in the early stages of agglomeration were drawn by the areas’ proximity to the 

commercial centre and location along a major road. However, these conditions are met not only in 

these two places, but in a vast number of alternative locations. What, then, was the deciding factor 

that promoted agglomeration in these specific areas during the early stages? 

Statistics on business establishments in 1969 indicate there were two furniture and wood 

workshops managed by Indians in Area B at that time. Also in Area B there was a wood workshop 

at Makerere University, which taught techniques for building and repairing, as well as a few timber 

wholesalers. The interviews proved that woodworking shops managed by Indian craftspeople 

played a significant role in technology transfer, as supported by the fact that all three of the oldest 

managers in the area obtained their skills from Indian craftspeople and then started independent 

businesses (in 1969, 1972, and 1974 respectively). Business promotion in the early stage of 

15

agglomeration in the Bwaise area seems to have been led by Indian craftspeople until the time of 

their deportation. Spin-offs from a small number of workshops owned by Indian craftspeople 

resulted in the sprouting of local agglomeration (Ministry of Commerce, Industry and Tourism 

1971). 

At the time of the spin-offs, however, managers who did not have plots would have had the 

option to remain or to shift to another area. Managers of some early-group companies – for 

example, 6 out of 13 companies that located in the Bwaise area prior to 1986 – actually owned land 

in the relevant area. In Area N as well, two members of the early group owned land in the area. In 

fact, according to the Kampala survey, there are several areas where around 10 workshops stand 

16

side by side in addition to the above-mentioned two areas, but woodworking shops independently 

operating in an area other than agglomerated areas are extremely rare. 

In terms of ownership of land and sheds by managers, all surveyed companies constructed 

sheds for their workshops at their own cost; however, only eight percent of the companies owned 

land, and members of the late group are overwhelmingly renting plots. There are only three 

landlords with signed rent contracts in the overall Bwaise area because many managers rent land 

from the same landlord. Woodcrafters who had woodworking skills and techniques or whose 

relatives owned land were thus given chances for business promotion and choice of location before 

the newcomers. This shows that infrastructure, i.e., existence of roads and access to land, was also 

essential in determining SME location preference (Figure 1) (Yoshida 2005). 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

source: Prepared by author. 

17

7. Institutional Endowment that Guided SME Cluster Formation 

7.1. Land Administration Policies and Land Markets 

When a small-scale manufacturer decides on a location for a workshop or factory in Uganda, it 

is necessary to obtain a trading license or a permit from the local authority, regardless of the 

enterprise’s scale. This means that land use acts and zoning regulations virtually institutionalize 

spatial patterns of industry location in the city, dictating what industry should be located where. 

Land zoning regulation is complex in Uganda, operating according to two basic concepts that 

operate in parallel: (1) the modern zoning concept, which is strictly applied to an area under the 

Kampala Planning Act, and (2) relatively loose traditional common law regarding land utilization. 

The former strict concept originated in the legacy of British-style zoning during colonial 

administration, which is based on the concept of dividing and categorizing areas depending on use 

and in which spaces for production and residence were supposed to be separated. Parallel with this 

concept, a very different land title system originating in the Buganda Kingdom, where land used to 

be leased or sold to private entities on the basis of permission, is still effective in the country 

(Mugambwa 2002a, 2002b). 

Due to the coexistence of these systems, Kampala, a city with 1,200,000 inhabitants and about 

380 square kilometers of built-up land, is divided by two different planning methods. The different 

land use regulations influence the location of woodworking businesses and agglomerations of 

SMEs. In reality, SME agglomeration has been formed according to the traditional land tenure 

system, or Mailo Land zones. 

The urban land market in Uganda has been generally energized due to the economic 

reconstruction that has taken place since 1986. The property market, including the rental market, 

has been consistently expanding in transaction volume since the current government’s declaration 

in 1992 to return the foreign-owned assets from the pre-Amin period. Overall, however, 

information on land value and rent has been difficult to obtain in a market where bidding actions 

remain opportunistic and speculative transactions are prevalent. 

Plots where major woodworking industries are located are usually transacted at a per-acre unit 

through an agent, with the involvement of the Uganda Investment Authority (UIA) and the former 

Uganda Development Corporation (a divested quasi-state investment authority). In these 

transactions, bidding price information is publicized and opportunistic trading or lease agreements 

are difficult. In contrast, rental markets for small plots of 20 to 30 square meters, where SMEs 

prefer to locate, are not open. Formal markets for smaller plots are difficult to establish, due in part 

to the Ugandan land act, created to conform to the British-era land policies. 

In reality, multiple small sheds made simply by nailing boards together share large plots of land. 

18

Subdivision of a plot is only a matter of a rental agreement between the landlord and lessee. If there 

is no problem in terms of land utilization and there is no change in land ownership, then as long as 

these plots are subdivided and used under a verbal voluntary lease agreement, the use complies 

with the relevant land act, although the issue of building regulation still remains. Also, in the case 

of Mailo Land, compliance regarding the building code was not traditionally strictly enforced, so 

slum-type buildings have developed in this part of the city. Forcible eviction by the planning 

administration has not been vigorously pursued on Mailo Land, and so the planning administration 

is considered to have been very permissive in regard to subdivision and usage for the purpose of 

workshops on Mailo Land. This made it possible to legally obtain business permits for the 

formation of small workshops in an SME agglomerated area. Furthermore, zoning for land 

utilization was not strict in Mailo Land law; therefore, the lease market for small plots was tacitly 

accepted. 

7.2. Land Market Information 

Property markets in clustered areas, however, seem to create unfavourable conditions for 

workshop owners, as they preclude the common business practice such as the exchange of 

agreements or a basis for shed ownership. New entrants might anticipate risks at the time of the 

"contract," such as future rent hikes or requests for relocation. In Uganda, where the city land 

market is imperfect, information on small sized plots for workshop use is very limited, and it is 

difficult for managers of SMEs to collect rent plot information. As shown in Table 14, of the 

surveyed workshop owners, 16 obtained land through introduction by relatives or friends in Area N, 

and 5 companies utilize their own or relatives' land. In Area B, 23 workshops stated that they had 

previously known their landlords, and 32 workshops replied that they were introduced by relatives 

or friends. The absolute deficit of small plots in the city land market, as well as limited options in 

site selection in areas other than industrial agglomerated areas, are also factors in the selection of 

agglomerated areas. 

In the land market in Uganda, business promotion and participation by SMEs is restricted due 

to zoning regulation. However, as groups of SMEs in the same industry began to spin off and the 

increased demand for workshops became obvious, intensive land utilization was enabled by 

subdividing and leasing sections in the same neighbourhood. Furthermore, as landlords recognized 

that local land authorities were receptive to land subdivision and utilization, workshop managers 

became preferred as tenants, and a citywide rent market began to form in which small plots sites 

circulated. 

To gain such tenants, landlords in SME agglomerated areas pursued intensive land utilization 

that ignored the building code by subdividing and leasing land sections to groups of small 

19

workshops. Agreements without written contracts or official registration are favourable to landlords, 

because eviction is easy when landlords want or need to shift land use. On the other hand, landlords 

have difficulty accessing information when seeking to verify the reliability of tenants, and there is a 

risk involved in exchanging verbal agreements with walk-in tenants. However, managers who have 

established their reputations as craftspeople in the community can build a reputation as reliable 

tenants. In the vicinity of agglomerated areas, where there exists a constant labour pool preparing to 

start businesses, craftspeople’s decision concerning whether or not to rent has become easier. 

Twenty-three firms in Area B replied that they had known their landlords before the contract, as 

they were mostly spin-offs from other firms in Area B and had met or contacted local landlords 

during their apprenticeships. Landlords' risks are thus reduced by selecting tenants based on their 

reputation as reputable craftspeople or managers, or on referral from well-known craftspeople or 

reliable relatives and acquaintances in the vicinity (Table 14). 

 

 

 

 

 

 

 

 

 

 

8. Conclusion-What Policy is Needed for Upgrading Clustered SMEs? 

The agglomerations of woodworking production in Kampala originated with the location of 

workshops of Asian craftsman along a major highway. These workshops played the role of small 

anchor firms. They gave others the information necessary to exploit the opportunities of those 

specific locations, resulting in SME clusters as shown in the flow chart in Figure2. Early entrants 

found these locations to be advantageous not only because of the proximity to the CBD, but also 

because publicity was available due to the heavy traffic through the areas. Early factors that 

encouraged agglomeration after this initial stage included the asymmetry of land market 

information for spun-off medium and small workshops. Access to cheap small plots for small-scale 

businesses was critical. Spatial expansion of the agglomerations resulted from the location 

preference expressed by medium and small companies as they attempted to reduce the risks of 

entry and take advantage of information asymmetry as well as of the large existing markets in the 

areas. 

The locational behaviour of firms transformed the SME areas into de facto SME industrial 

zones without proper institutional recognition or infrastructure provision. These de facto zones 

localized necessary utilities and institutions such as micro finance lenders. 

20

Cooperative relationships among firms in the agglomerated areas have been found to be active, 

but cooperation between these firms and the newly established major woodworking industries has 

not developed at all. In fact, with regard to vertical labour specialization with SMEs, some larger 

firms, including the one that graduated from the aforementioned SME Cluster B, responded that 

this was "impossible" because of differences in skill levels. The outsourcing ratio among the five 

larger companies surveyed was extremely small. Production is basically self-sufficient and 

vertically integrated, except where the timber processing is outsourced to mills. On the other hand, 

in the SME agglomerations, specialization of production processes among SMEs is active in Areas 

B and N, and cooperative frameworks other than labour specialization have also been established. 

In particular, tool sharing, labour pooling, and even money lending has developed among 

latecomers. SMEs also took advantage of the pool of idle skilled and unskilled labour craftspeople 

in agglomerated areas. 

Among major woodworking industries, the issue of location preference in Kampala is strongly 

related to the issues of securing an experienced workforce and reasonably priced property. Large 

firm investors are unaware of the massive pool of skilled labour in the vicinity of local 

woodworking cluster. In this regard, there is a good possibility for the two sectors to develop a 

mutually cooperative framework through labour market integration. This interaction between two 

sectors, and movement of skilled labour between sectors, can develop if SMEs succeed in making 

technical upgrades (Figure 2). 

However, the limited or non-existent cooperation between SMEs and large firms originates in 

the ways in which these two sectors emerged in the country. The government has not recognized 

the potential of this sub-sector and has always favoured large firms in its policy by offering 

investment incentives and tax holiday packages without any specific priority policy for the 

sub-sector or industrial space. SMEs have been ignored for long periods of time, and until recently 

lacked any incentives or proper financial schemes. Indivisible plots in industrial zones did not 

favour small-scale operation. 

The market for these two sectors is also divided, as large firms aim mainly at upmarket 

clientele and SMEs aim for the middle and working class market. This dichotomy has created 

prejudices among the large firms, who distrust SMEs’ skills and competence, despite the proven 

fact that these smaller firms possess a variety of skill levels. SMEs often view large firms with 

disappointment and prejudice, because SMEs suspect that government officials always favour large 

firms due to collaboration and corruption. 

Now, both sectors should recognize the doors of new opportunity opening for cooperation 

because regional market integration resumed in 2004 and the development of a long-awaited export 

processing zone is nearly complete. To maximize the emerging opportunity mentioned in the 

beginning of the paper in the global shift of production in the woodworking and furniture sector, 

21

the state policy needs to institutionalise the existing SME clusters and incorporate the large firms 

into the network through, for example, co-bidding a government tender or exchange of technical 

expertise. As shown in Figure 3, we should draw, at the moment, two different flow charts to show 

the trajectory of SME cluster formation and the trajectory of the large firms’ sector formation in 

Uganda. It is vital to assist the integration of the two different flows and to assist the merger of the 

existing SME clusters with large firms in order to form a comprehensive woodworking industry 

cluster. However, without the provision of some kind of support to remove the bias and distrust 

toward SMEs’ technical capabilities in the minds of major woodworking companies who mostly 

consider vertical disintegration to be a sort of taboo, it will be difficult for the industry sector as a 

whole to develop as a comprehensive woodworking industry cluster. Uganda’s woodworking 

industry has the opportunity to rise as a leader from among the many other apparently stagnant 

Marshallian-type woodworking agglomerations in Sub-Saharan Africa by establishing a 

cooperative framework for industry clustering and meeting the technical needs of supplier-seeking 

global firms. 

22

23

Bibliography 

Bank of Uganda, 2004, Private Sector Investment and Investor Perception in Uganda 2003 Report. 

Kampala: Bank of Uganda. 

Kaplinsky, Raphael, Mike Morris and Jeff Readman, 2002, The Globalization of Product Markets 

and Immiserising Growth: Lessons from the South African Furniture Industry. Institute of 

Development Studies, University of Sussex: Brighton. 

(http://www.ids.ac.uk/idS/global/pdfs/RKJRupgradingMarch%2004.pdf) 

Kaplinsky, Raphael, Olga Memodovic, Mike Morris and Jeff Readman, 2003, The Global Wood 

Furniture Value Chain: What Prospects for Upgrading by Developing Countries, The Case of 

South Africa. Sectoral Studies Series, United Nations Industrial Development Organisation: 

Vienna. 

(http://www.unido.org/file-storage/download/?file_id=11904) 

Kaplinsky, Rafael and Jeff Readman, 2004, Globalization and Upgrading: What Can (And Cannot) 

Be Learnt From International Trade Statistics in the Wood Furniture Sector? Institute of 

Development Studies, University of Sussex: Brighton. 

(http://www.ids.ac.uk/ids/global/pdfs/RKJRupgradingMarch%2004.pdf) 

Kuchiki, Akifumi, 2005, Theory of a Flow Chart Approach to Industrial Cluster Policy, Institute of 

Developing Economies Discussion Paper Series: no36. 

(http://www.ide.go.jp/Japanese/Publish/Dp/pdf/036_kuchiki.pdf) 

Loebis, Lienda and Hubert Schmitz, 2005, Java Furniture Makers: Winners or Losers from 

Globalization? Institute of Development Studies, University of Sussex: Brighton. 

(http://www.ids.ac.uk/ids/global/pdfs/RKJRupgradingMarch%2004.pdf) 

Ministry of Commerce, Industry and Tourism, 1971, Manufacturing Industry of Uganda in Figures 

1969 and 1970, Kampala, Uganda: Ministry of Commerce, Industry and Tourism. 

Ministry of Economic Planning and Development, 1964, The Economic Development of the 

Kingdom of Buganda. Mengo, Uganda: Uganda Bookshop Printing Department. 

Ministry of Finance and Economic Planning, 1993, Census of Business Establishments 1989 

Manufacturing Sector Report. Entebbe, Uganda: Bureau of Statistics. 

Ministry of Information, Broadcasting and Tourism, 1964, UGANDA 1964, Entebbe, Uganda: 

Uganda Government Printer. 

Ministry of Planning and Economic Development, 1970, Index of Industrial Production. Entebbe, 

Uganda: Bureau of Statistics. 

Monitor, 2002, The Monitor Business Directory, Kampala, Uganda: Monitor. 

Mugambwa, John T., 2002, Source Book of Uganda’s Land Law, Kampala, Uganda: Fountain 

Publishers. 

24

Mugambwa, John T.,2002, Principles of Land Law in Uganda. Kampala., Uganda: Fountain 

Publishers. 

Uganda Bureau of Statistics, Statistical Abstract, Entebbe, Uganda. 

Uganda Bureau of Statistics, 2003, A Report on the Uganda Business Register 2001/2002, Entebbe, 

Uganda: Bureau of Statistics. 

Uganda Investment Authority, 2003, Private Sector Investment Survey 2003 Report in Uganda, 

Kampala, Uganda: UIA. 

UMACIS, 1995, Uganda Private Enterprise Survey Report: World Bank Private Sector 

Development Department. Kampala, Uganda: Uganda Manufactures Association. 

UNIDO, UNIDO Data Base: United Nations Industrial Development Organization. 

(http://www.unido.org/data) 

United Nations Industrial Development Organisation, 2003, Africa Foreign Investor Survey 2003: 

Motivations, Operations, perceptions and Future Plans-Implications for Investment Promotion. 

Vienna: United Nations Industrial Development Organisation. 

USAID Presto Project, 1999, The Business Real Estate Survey Uganda. Kampala: Presto. 

Uganda Small Scale Industry Association (USSIA), 1999, USSIA Membership Staff and 

Investment.: Kampala: Uganda Small Scale Industry Association. 

Yoshida, Eiichi. Industrial District as a Localised Incubation System in Developing Economies, 

Woodworking Industrial District in Uganda. Bonn: 11 th General Conference, European 

Association of Development Research and Training Institute. 

(http://eadi.org/gc2005/paperswgopen/yoshida_rev.pdf) 

25

Information - ARRIDE - IDE-JETRO

You also want an ePaper? Increase the reach of your titles

Delete template?

Save as template?