MERIT-Infonomics Research Memorandum series

Foreign direct investment through acquisitions and

implications for technological upgrading:
Case evidence from Tanzania

Brian Portelli & Rajneesh Narula

2004-008

MERIT – Maastricht Economic Research International Institute of Infonomics

Institute on Innovation and Technology

PO Box 616 c/o Maastricht University

6200 MD Maastricht PO Box 616
The Netherlands 6200 MD Maastricht
T: +31 43 3883875 The Netherlands
F: +31 43 3884905 T: +31 43 388 3875
F: +31 45 388 4905

http://www.merit.unimaas.nl http://www.infonomics.nl
e-mail:[email protected] e-mail: [email protected]
 Foreign direct investment through acquisitions and implications for technological
upgrading: Case evidence from Tanzania

Brian Portelli and Rajneesh Narula

Abstract

Using a case-study approach, this study examines the role of privatisation on the industrial
landscape of Tanzania. We examine the impact of FDI through acquisitions on technology
transfer within the acquired firm as well as the development of linkages to other firms based
in the host country. Our results suggest that technological upgrading has occurred following
FDI, the intensity of which reflects the type of firm-specific assets of parent MNE, as well as
the pre-acquisition state of these industrial activities. Improved backward linkages are also
evident with local economic agents, but their type and extent – reflecting Tanzania’s
comparative advantage in the primary sector - confirm that capabilities both within the
acquired firms and also in the industrial base of the host country greatly influence the
magnitude and intensity of technological upgrading. ‘Narrower’ technology gaps between the
MNE affiliate and the domestic sector are more likely to result in backward linkages and
determine the type of technological content of inputs sourced locally rather than within the
MNE.

1. Introduction
This article seeks to contribute to our understanding of the implications of FDI on the
upgrading prospects of host developing countries, based on primary research through case
studies of FDI through privatisation in Tanzania. It is generally accepted from a
developmental perspective that FDI flows are crucial because they represent an important
source for technological spillovers in the host country.
In developing countries, structural adjustment programmes prominently integrate
macroeconomic stabilization policies alongside policies to promote increased inward FDI.
The most important element of these programmes is economic liberalization and the
divestiture of state-owned industrial concerns. Particularly in the least developed economies,
such as those in Sub Saharan Africa (SSA), the scope for FDI has increased in the context of
government liberalisation and privatisation programmes (Lall 2002, Pigato 2001). Although
not the only means available, FDI spillovers are regarded as one of the most practical and
efficient means by which industrial development and upgrading can be promoted (Narula and
Dunning 2000). It is argued that FDI potentially increases the rate of technical progress in the
host country through a ‘contagion’ effect from the application of more advanced technology

1
and management practices used by foreign affiliates on to other host based firms (Findlay
1978). Most notable of these externalities is the potential contribution of FDI to the industrial
and technological upgrading in the host country or efficiency spillovers starting from the
industrial set-ups they acquire.
The focus in this paper is on two inter-related aspects of the potential FDI-led
upgrading process in a developing country. The first focus is on the impact of FDI on
technology transfer or absorption in new acquired enterprises, i.e. within the acquired firm, as
a result of the technology and knowledge transferred from the acquirer. The second focus is
on the potential of technology spillover in the host country through vertical backward
linkages, i.e. with other firms based in the host country1. An analysis of local value chain of
MNE affiliates is not enough to inform us on the potential for technology spillovers in the
host country Per se. Backward linkages also serve as a conduit for MNE technology
spillovers. However it is not the scope of the present paper to analyse the type of knowledge
transfer through linkages, but only to explore the potential for these spillovers to occur.
The empirical evidence presented in this paper is based on primary data on FDI
activity in Tanzania. Tanzania represents a typical case of a developing country that has been
undergoing an extensive economic liberalization process. Privatisation of state-owned-
enterprises represents a major pull factor for FDI inflows2. The only empirical evidence from
Tanzania pertains to the macro impact of acquisitions (Mbelle and Shitundu 1999,
Mwandenga 2000). Little has been written on the process of FDI-led industrial upgrading and
technological development both within the affiliate and through linkages with host-based
firms.
The rest of the paper is organised as follows. First, we review the literature and
empirical evidence on the impact of FDI through M&As in host developing countries and on
MNE linkages and spillovers with reference to their respective determinants. Next, we use
this theoretical framework to inform our research propositions. After a brief description of the
research methodology, we present two cases of FDI through acquisition in Tanzania and after
that the main findings pertaining to our study. The final section presents some conclusions
and policy implications.

2. Background and review of relevant literature

The increase in privatisation related FDI is reflected in the rise of foreign exchange
revenues from privatisation deals which increased from 1.1 billion US dollars in 1991-1994
period to 3.2 billion US dollars in 1995-1998 period3. During the 1990s, FDI represented

2
about 60 per cent of foreign exchange raised in all developing regions and a higher proportion
of 80 per cent in Sub Saharan Africa (SSA). In an important study, Liberatori and Pigato
(2000) show that between 1990 and 1998, approximately 14 per cent of FDI was privatization
related. In particular, the increase in privatisation-related FDI has nurtured cross-border
acquisitions in SSA. Between 1997 and 1999, almost 40 per cent of FDI inflows into Africa
came in the form of cross-border M&As. On a global scale, however, Africa’s share in cross
border M&As is negligible at less than 0.5 per cent in 1999 (UNCTAD 2000) reflecting the
fact that very few domestic enterprises possess firm-specific assets – for example in the form
of new technologies, well-known brand names or strong presence in attractive markets that
could make them interesting targets for a take-over by foreign firms.
FDI represents a “combination of capital, stock, know-how and technology” (De
Mello 1997). Host countries increasingly view FDI as a potential source of bridging the gaps
in their capability development, exploitation of resources and participation in the international
market. Although the literature of the impact of FDI is rather inconclusive, what seems to
emerge is that there is a potential direct contribution to economic growth, whether through
additions to the capital stock and/or technological base of countries. However, such potential
contribution is dependent on the type of FDI involved and the absorptive capacity of the host
country. For developing countries, FDI is an most important means of acquiring new
technology, but it is evident that FDI is not a sine qua non for industrial and technological
upgrading.
Cross-border M&As can act as a tool for competitive upgrading (UNCTAD 2000).
FDI through acquisitions can influence the rehabilitation of industrial activity, the injection of
new technology and spillovers of knowledge into otherwise stagnant industries. It is argued
that M&As represent an important means how to restructure and upgrade industrial
capabilities (Maucher 1998, UNCTAD 2000). This may happen in two ways. First by
knowledge transfer to the acquired firm, and second, by the creation or upgrading of linkages
with the domestic economy. The next section deals with each of these separately.

Intra-firm knowledge transfer

An acquired enterprise may benefit from technological upgrading, comprising both
‘hard’ (physical investment, plant and equipment) and ‘soft’ (knowledge, management
systems, processes) technologies and may lead to rapid changes in the host country,
particularly with respect to market orientation, local skills and capabilities, as well as
corporate strategies to make the transition up the value chain. Upgrading in this way is

3
confirmed by varied empirical evidence (see for example, Zhan and Ozawa 2000,
Chudnovsky et al 1995, Mortimore 1998, Moden 1998). This upgrading includes the
increasing capabilities of acquired firms because of the greater technological strengths that
foreign investors potentially bring in as a result of the firm-specific assets of parent MNEs.
This is also reflected in the knowledge flows from the acquirer to the acquired unit (Bresman
et al 1999). Several studies show that FDI tends to raise output and/or productivity, through
introduction of new technologies (e.g, Barrel and Pain 1997; Djankov and Hoekman 1999).
Evidence also suggests that MNEs introduce new or improved management techniques to the
acquired firms (Chudnovsky et al 1995, Allard and Lundborg 1998). Foreign affiliates are
often at the forefront of new management and organizational techniques, quality management,
standards, training and marketing methods, reflecting home country practices introduce new
concepts to move the acquired firm towards the technological frontier.
However, it is important to note that acquired enterprises may not always experience a
net technology inflow in a post-acquisition scenario. This depends inter alia upon the motive
of the investment, and the role of the newly acquired subsidiary within the structure of the
MNE. Once the decision to enter a given market through FDI is taken, the kinds of activity
and the level of competence of the subsidiary are co-determined by the nature of the location
advantages of the host location. That is to say, while MNE internal factors such as their
internationalization strategy, the role of the new location in their global portfolio of
subsidiaries, and the motivation of their investment are pivotal in the structure of their
investment, all these issues are dependent on the available location-specific resources which
can be used for that purpose. Indeed, the host country’s location advantages play an important
role in determining the level of embeddedness of the subsidiary (Benito et al 2003), and this is
the primary determinant of the quality of the FDI. In some instances, MNEs may divest part
of their operations because other subsidiaries in other countries within the MNE network are
better positioned to perform these tasks. MNEs sometimes seek to rationalise their operations
on a regional or global basis, and therefore reduce (or increase) the intensity of operations of
an acquired subsidiary by lowering the level of competence and/or scope of their subsidiary
(Narula 2003). Thus, whether an acquired firm is upgraded or downgraded upon becoming a
part of an MNE is not always clear, since it is a function of not just the competences and
resources that are available to the MNE, but also the MNE’s global or regional strategy and
how the newly-acquired subsidiary fits into its existing structure (Benito et al 2003).
If upgrading is analysed within the context of a categorization provided by Humphrey
and Schmitz (2000), we look at process upgrading (firm processes- transforming inputs into

4
outputs more efficiently by re-organising the production system or introducing superior
technology), product upgrading (firms upgrading by moving into more sophisticated product
lines – defined in terms of increased unit values), and functional upgrading (firms acquiring
new functions in the value chain, broadening the scope of their industrial activity). While
process upgrading refers to doing the same industrial activity more efficiently, product and
functional upgrading refer to a repositioning by the industrial activity on to new, higher value
added levels. In this paper, we take the approach that whereas each category of upgrading is
equally important for the industrial and technological upgrading processes of the firm, it is
best to analyse this with an interdependent dynamic perspective. Hence, an industrial
restructuring process presupposes that there is process upgrading in the improvement of
production processes towards more efficient and economically viable production. The basis
for product upgrading, either brand development or repositioning, is determined by of process
upgrading comprising investment in tangible and intangible technology. Improved production
process also implies a strong element of functional upgrading in the enterprise.
The role of human capital is very important across forms of industrial upgrading. All
forms of upgrading involve extensive reliance on the human capital. Upgrading results from
both transfer of ‘hard’ and ‘soft’ technology and knowledge from the parent MNE to the
affiliate through the application of firm-specific assets in the form of embodied and
disembodied forms. Hard technology is said to include aspects of embodied knowledge,
embodied in the machinery and equipment. Soft technology is said to include aspects of
disembodied knowledge as a result of the transfer of operative skills and design.

Technology Diffusion through backward linkages

The perennial question posed in the literature on FDI and development is whether
MNEs create linkages and become integrated in the host economy, or whether they operate in
so-called ‘enclave sectors’. When MNEs establish subsidiaries in a particular location they
need to build linkages with domestic agents in order to carry out their operations, and these
linkages constitute one of the ways in which skills and technological transfer is thought to
disseminate to the rest of the economy.
The attention in various studies has been extensively laid upon the relationships
between MNE affiliates and local suppliers, i.e. backward linkages of MNE affiliates in the
host economy.4 Various determinants of these backward linkages with local suppliers have
been identified in the literature. There are factors related to the MNE and those associated
with the host country. For example, the ability of the domestic economy to benefit from MNE

5
investment crucially depends on the relative technological capabilities of the recipient and the
transmitter. The presence of externalities per se does not imply that the domestic economy
can or will internalize them, nor does it imply that these externalities have to be significant in
quantity or quality. Ceteris paribus, MNEs prefer to use technologies that are suited to their
own needs, and the purposes for which they have made the investment. MNEs generally do
not make available their proprietary assets available at the whims of governments; rather they
tailor their investment decisions to the existing market needs, and the relative quality of
location advantages, especially skills and capabilities in which the domestic economy has a
comparative advantage (Lall 2000)5.
Again, the MNE investment motive and its overall strategy are important factors to
consider. For example, domestic market oriented affiliates generally purchase more locally
than do export oriented firms because of lower quality requirements and technical
specifications (Reuber et al 1973, Altenburg 2000). As a result, MNE affiliates are more
likely to be integrated backward in the host country when they source relatively simple inputs
(Ganiatsos 2000, Carillo 2001). For example, in the case of FDI in agro-based industries,
there is a greater likelihood for affiliates to be integrated backward, especially given the early
stage of development of the host country. Rodriguez-Clare (1996) argues that more linkages
are created when MNE production uses intermediate goods intensively, when communication
costs between parent and affiliate are large and when the home and host markets are not too
different in terms of intermediate goods produced. Affiliates established through M&As are
likely to have stronger links with domestic suppliers than those established through greenfield
investment (UNCTAD 2000, Scott-Kennel and Enderwick 2001) since such FDI can find
established linkages upon acquisition that are likely to be retained if they are efficient6. Most
importantly, linkages vary by industry. In the primary sector, the scope for location-specific
vertical linkages is often limited, due to the continuous production processes and capital
intensity of operations. In manufacturing, the potential for vertical linkages are broader,
depending on the extent of intermediate inputs to total production and the type of production
processes (Lall 1980). Blomstrom and Kokko (1997) suggest that ‘some of the host country
characteristics that may influence the extent of linkages – and thereby in the longer term the
extent of spill-overs – are market size, local content regulations and the size and technological
capability of local firms’. They argue that there is a propensity for linkages to increase over
time, as the skill level of local entrepreneurs grows, new suppliers emerge and local content
increases. The time factor is highlighted also by Rasiah (1994) and is related to the experience
and integration of a MNE affiliate in the host country through greater ‘indigenization’ of

6
operations in terms of management, knowledge about their location and operations7. The
embeddedness of firms is often (but not always) a function of how long the MNEs have been
present in the host country, since firms tend to build incrementally8.
Host countries need to have the basic endogenous absorptive capability structures
(Abramovitz 1995) to be able to assimilate and take advantage of foreign technology made
available in the host country. Knowledge accumulation is much more rapid once the initial
threshold level of absorptive capacity exists (Criscuolo and Narula 2002). Technology
diffusion through backward linkages presupposes that, first, domestic firms in the industry
exist, and second, these possess the capacity to usefully internalise the knowledge being made
available by the MNE. Diffusion to the rest of the economy may not occur because of
deficiencies in the institutional capability systems of the host country or other deficiencies in
the absorptive capacity of domestic economic agents in the host country. Learning and the
acquisition of knowledge themselves require skills and abilities that are non-obvious.

3. Research questions, propositions and methodology

We reiterate our research hypotheses. Our first research focus relates to the analysis of
the impact of FDI through technology transfer in acquired enterprises.

H1; FDI through acquisition leads to technology transfer within the firm, the intensity
of which is determined by the absorptive capabilities and know-how present in the acquired
firm.

The second research focus refers to the need to analyse the potential for MNE
technology spillovers in the host country through evidence of vertical backward linkages of
the acquired firms with other firms in the host country.

H2; FDI through acquisition leads to the establishment of more backward linkages
and the nature of locally sourced inputs is determined by the technological capability of the
linked firms.

The empirical evidence presented here is based on primary data collected from a study
of FDI and MNE activity undertaken in Tanzania. This investigation was undertaken during
2002 and 2003 and comprised the use of exploratory questionnaire, as well as a
comprehensive collection of data and related information on the FDI sector in Tanzania.

7
Company visits and semi-structured interviews with management in a sample of foreign
companies have been undertaken for this purpose9. This article focuses on two of the most
important case of MNE affiliates in Tanzania whose acquisition took place in the early 1990s.
On average, this time line permits us to have a seven-year period over which to examine their
operations after acquisition.
Case study analysis was deemed the best method that could be used to provide an
insight on the upgrading process within these MNE affiliates10. Studies of industrialization in
Africa have not paid much attention to firm-level activities and industrial upgrading process
which influence the path the various firms have followed and how they have been coping in
changing world technological and market conditions (Wangwe 1997). We are fully cognizant
that the case study approach has its limitations when it comes to making generalizations. Case
studies are generalizable to theoretical propositions rather than to populations or universes.
Unlike survey research, which relies on statistical generalization, our emphasis is more on the
analytical generalization. We stop short of making generalizations from the case studies since
these are made on the basis of cases which have been selectively sampled. The scope of this
paper is more to gain an insight into the industrial upgrading process than in producing
statistically significant outcomes11.

5. Case studies and empirical evidence

This section starts with a brief description of the two acquisitions studied and proceeds with
an analysis of the empirical evidence in the context of the proposition posed earlier.

5.1 Case studies

Tanzania Breweries Limited
Tanzania Breweries Limited (TB) was privatised in 1993, after 16 years in state
ownership, when it was bought by South African Breweries (SAB plc) in 199312. TB is
engaged in the manufacture, sales, distribution of beer, alcoholic fruit beverages (AFB’s) and
non-alcoholic beverages within Tanzania. The TB group is listed on the Dar es Salaam Stock
Exchange, and has a market capitalisation of over US$ 470 million, with over 22,000
Tanzanian shareholders. It employs more than 1,250 people and is represented throughout the
country with three breweries in Dar es Salaam, Arusha and Mwanza, ten distribution depots, a
traditional brewery and a distillery13. TB is also engaged in barley farming for use at its
malting plant in Moshi. TB’s mission is to be a successful brewer and profitable marketer of
quality beers while being recognized as a good corporate citizen to stakeholders. The

8
Company’s vision is to be recognized as one of the most successful and respected commercial
enterprises in East Africa by the year 2005 and aims to optimise the creation of wealth, and
give a fair reward for the contribution of stakeholders. Their strategic objective is to grow
shareholder value through enhancing volume and productivity growth, seeking opportunities
to enhance its position on the local and export market, growing brands and achieve and
maintain operational efficiency work hard to strengthen regional brands and market positions.
Tanzania’s beer market has potential for growth given its population size. However beer
consumption levels are comparatively low when compared to neighbouring countries14. TB
has a sales turnover of around US$174 million (2003) of which around 5 per cent is exported
to neighbouring countries. Key data for TB are shown in Table 1 below.

Table 1

Tanzania Cigarette Company Limited

Tanzania Cigarette Company Limited (TCC) was formerly established in 1961 by the
British American Tobacco. Its principal business is the manufacture, marketing and
distribution of cigarettes. Between 1967 and 1975, the Government of Tanzania (GOT)
completely nationalised the company. In 1995, as a result of Government’s privatisation
policy, 51 per cent of the company was sold to RJ Reynolds Tobacco Company, a subsidiary
of RJR Nabisco, for US$ 55 million. In May 1999, Japan Tobacco (JT) acquired all non-US
tobacco operations of RJR Nabisco including its 51 per cent stake in TCC and in September
2000, JT increased its shareholding in TCC to 75 per cent. TCC became a public limited
company in August 2000, with shares listed on the Dar Stock Exchange in November 2000.
As a result, the shareholding structures stands at JTI 75 per cent, general public 19.5 per cent,
GOT 5.5 per cent. TCC is currently the only cigarette manufacturer in Tanzania mainland and
its sales turnover is around US$ 100 million (2002), 10 to 15 per cent is exported to
neighbouring countries within the parent network. TCC enjoys a local market share of over
95 per cent The company currently employs 700 people. TCC has a network of 680
wholesalers and 40,000 retailers. It has direct links with around 98,000 tobacco growers. Data
pertaining to TCC are illustrated in Table 2.

Table 2

5.2 Empirical evidence

9
 Next, we present case evidence pertaining to the analysis of our propositions. We do
this in the following manner. First we highlight the main challenges for industrial upgrading
emerging in the aftermath of the parastatal period. Afterwards, the process of technology
transfer for restructuring taking into account the various forms of upgrading. Next, we
examine the outcome of this restructuring process and evidence of vertical backward
integration of these firms in order to highlight the potential for inter-firm technology
spillovers in the host country. The underlying feature in this entire exercise is the emphasis on
the nature and extent of absorptive capabilities within firms (both within the acquired firms
and vertically linked firms) as a determinant factor for technology transfer in the former and
technology spillovers in the latter.
Any examination of industrial upgrading and intra-firm technology transfer must be
analyzed within the context of operative conditions in the parastatal period and hence the
requirements for industrial restructuring emerging from such contexts. Both TB and TCC,
were loss-making parastatals, depending extensively on public funds for operations,
uncompetitive industrial activities with poor product quality and deteriorating market shares.
This situation was the result of a number of structural deficiencies, mainly massive over
employment, the utilization of obsolete technology, productive process with outdated work
practices. For example, prior to 1993, TB’s production facilities were characterized by old
plant and equipment, archaic production methods which resulted in low capacity utilization,
productive inefficiencies and deteriorating beer quality. The company was incurring
continuous heavy losses in local market share (just over 40 per cent in the early 1990s) which
served to exacerbate the precarious financial position15. A similar story can be told for TCC.
While the company had a virtual monopoly of the protected domestic cigarette market, it was
making substantial losses. TCC suffered from over-employment and weak technological and
management capabilities and was losing market share to products from neighbouring
countries, particularly Kenya. In general, the post privatisation experience is characterised by
industrial and technological (including know-how) upgrading in different forms. This
upgrading has occurred in all activities at TB (preparation, brewing, fermentation, packaging
and sales) and TCC (processing, manufacturing, packaging and sales).

5.3 Intra-firm technology transfer

The immediate strategy post-privatisation at TB has been to undertake extensive
‘physical’ investment in plant and equipment with the objective to regain lost market shares
through the modernization of production facilities, increase production capacity and improve

10
efficiency and product quality. This process involved the rehabilitation of the main breweries
in Dar es Salaam and Arusha, as well as setting up of a new brewery in Mwanza. This
investment programme enabled TB to start utilizing state-of-the-art technology in its brewing
process, i.e. more automated equipment particularly in the brewing and fermentation stages
of the production process and bringing the company in line with technological standards
envisaged by the parent MNE for best manufacturing methods in the industry. The
rehabilitation of the Dar and Arusha plants has been done in a modular manner. Overall
around US$100 million had been spent in this modernization process and this capacity
expansion programme has been completed in 2001, timed with the expiration of the
company’s tax holiday under Tanzania’s fiscal incentives for FDI. Other modernisation has
been undertaken in the packaging related machinery as for example, bottling and canning
lines, labelling as well as in the upgrading of TB distribution facilities16.
As a result of this investment, TB has considerably improved its product efficiency,
became more cost effective, improved the reliability of plant and machinery and reducing
down time17. More automation has improved the production process, resulting in faster return
on investment due to inceased production, reduced energy consumption and lower staffing
18
levels . Flexible production has resulted in the decrease of the turn down ratio to under 20
per cent of nominal capacity. Modern productive facilities have resulted in improved batch
control that has improved product traceability19. The modernisation of the production process
has also led to important improvements in the environmental impact of the breweries. For
example, less energy consumption and liquid waste and increased recycling of waste from
production are the most notable results in this respect20. Technology transfer has also taken
place in subsidiary companies in the TB group. For example, at Tanzania Distilleries and Dar
Brew, plant and machinery rehabilitation was undertaken within a restructuring programme
aimed to increase capacity, productive efficiency, revamp traditional brands and undertake
innovation in the product mix21.
In addition to physical investment, ‘soft’ technology transfer has involved the
application of parent MNE ownership advantages, particularly in the embodied in the
production systems and methods. For example, quality checks from the raw material, input
stage to the finished product stage have become more systemized, so as to determine the
highest quality of beer produced as well as the determine the beer age in the trade (from more
than 16 weeks to less than 12 weeks). Since acquisition, there has been a general emphasis on
the improvement of the quality of beer produced. TB’s sales and marketing departments have
been overhauled, drawing on the extensive expertise of foreign personnel from the parent

11
MNE. Greater focus has been put on brand research and development which has been
dramatically improved in an effort to consolidate traditional beer brands whilst at the same
time develop new ones. Foreign expertise has been brought in the company to effect changes
in the production process and so that the functional departments meet the new strategic
objectives of the company. The numbers of brands in the TB stable have increased from 2 in
1993 to 10 in 200322. New product launches, such as ‘49er’, a non-alcoholic beverage have
been affected. ‘Soft’ technology transfer has occurred with respect to greater utilization of
information technology when compared to the parastatal period. Prior to 1993, information
and methods of its generation was not time-based, not seen as a priority for bringing about
efficiency in business operations and limited to particular administrative tasks. Privatization
necessitated the generation of more quality and timely information to be used by various firm
stakeholders. Through the use of information technology, TB has in place a standardised
control system for process units that can be configured to specific needs in the achievement of
performance criteria. All this has enabled TB to achieve results in a relatively short time with
return on investment increased rapidly after acquisition in 1993.
In the aftermath of privatisation, efforts were undertaken to arrest the under-utilisation
of human capital. Training and human resource development was given utmost priority as a
way to make the staff understand the new rules of a competitive business. Training was tailor-
made to suit the operations and included short-term courses in plant training seminars as well
as overseas training23. In particular, a major in-house training programme was launched in
1994, with two main objectives to retrain production floor employees as well as to mould the
more experienced, local capabilities into the MNE core competencies. Employment levels
were decreased from over 4,000 in 1993 to 1,250 in 2003. This dramatic reduction is
attributed to the technology changes (production is now computerised), outsourcing of
services such as gardening, clearing, laundry, cargo loading, offloading and security, closure
of certain distribution outlets and the upgrading of packaging lines which use two shifts not
three. Forced retrenchment has been exercised and in other cases, jobs not requiring
permanent employment have been identified and transferred into casual employment24. The
number of expatriates amounts to just 5 per cent of total employment and reflects key
managerial and technical posts. It is noteworthy that locals are well placed within the
company managerial hierarchy. Expatriate personnel are required to pass on their knowledge
to locals and total local staff complement is expected to be realised by 2007. Since
acquisition, expatriates in management positions complemented local management to ensure
that the new applied management systems and methods maximized the effectiveness of

12
industrial restructuring being undertaken. Local capabilities are considered vital for the
running of TB’s operations since they involve extensive know-how of the industry (as a result
of being employed in the company for many years) and intimate knowledge of market place.
The new owners have recognized the importance of retaining these capabilities, as well as
recruiting local talent in a bid to understand more the local culture, tastes and preferences.
After acquisition, there has also been an overhaul in human resource development, through a
re-organization and streamlining of operating structures to ensure cohesion and better
coordination among various departments in line with respective roles vis-à-vis the company's
core business25. A strong element of disembodied technology has been to focus the human
capital to best complement and optimise the new technology being utilized26.
SAB’s competitive advantages in innovation processes and operative synergies have
introduced new ways of working based on the undertaken of cross functional teams for shared
decisions rather than through the isolated decision making process as was the norm prior to
privatisation. As a result more delegation of power in decision-making at all operative levels
has taken place:

“…an executive board under the executive managing director (comprising executive
directors of the various operative departments meet regularly to make decisions and review
company performance. They [the board] have been mandated by the board of directors to
make decisions on its behalf since the directors board meets only four times a year to take
major strategic policy decisions. Like this the delegation of powers has been pushed down the
company hierarchy emphasising the importance of accountability” (TB company interview).

Therefore, outmoded management hierarchies were replaced resulting in better

coordination and communication across the business section. This was facilitated by the
concurrent specific training to selected employees in areas such as strategic quality
management, cost control and operational efficiency. Important changes have been
undertaken in the way the company measures employee performance to dispel the ‘parastatal
habit’ of automatic salary increases with no relation to productivity27. TB now measures
performance at the enterprise level rather than on individual basis and it sets annual
productivity targets and bonuses are paid according to achievements. These changes have
been aimed at instilling job responsibility and commitment28 and ensuring that production
targets are met and new management practices effective.

13
 As a result of technology transfer, TB has therefore registered an overhaul in its
productive activities, achieving a turn around in performance to the extent that production
costs have been sufficiently reduced and permit a profit to be made on a reduced sale price of
beer29. TB performance since 1993 shows that there do exist opportunities for a profitable
industrialized country brewing operation, even in developing countries (like Tanzania) where
consumer purchasing power is relatively low, given that allowance is given to local conditions
and through a gradual but intensive restructuring process. TB was transformed into a profit
entity in just under two years: profit after tax amounted to US$ 34.3 million in 2003.
Accordingly, since 1993, the company has registered constant growth in operating margins,
from 18.4 per cent in 1995 to 27.4 per cent in 2003. TB’s operating margin stands above the
best performing margin benchmark for the parent network of 26 per cent. At the basis of this
performance lies the increase in sales turnover from an average of around US$25.0 million
prior to 1993 to over US$130 million to date. The substantial plant and equipment investment
led to an increase in production capacity from 0.4 million hl in the immediate aftermath of
privatization to the current. 2.3 million hl. This increase was corresponded by an
improvement in productivity indicators. For example, packaging efficiency and product
quality standards by an average of 11 per cent, considerably cutting on waste and down-
time30. The number of productive man hours increased from 30 per cent (1993) to around 80
per cent (2002). Capacity indicators for TB’s brewery landmark in Dar es Salaam (which
produces 55 per cent of the national volumes) reflect an approximate capacity utilisation rate
of 70 per cent up from below 50 per cent levels in the early 1990s. Productivity per employee
increased by 6 per cent exceeding targets set by management of 3 per cent per annum.
Concurrently, employee salaries registered a tenfold increase from the early 1990s. The
financial soundness of the firm made it possible for the management to offer an attractive
financial package to its workers an incentive for further for further enhancements of
productivity. TB has also positively contributed to the government budget through taxes and
dividends. Estimates indicate that TB’s tax contribution increased from US$ 12 million in
1993, to US$ 63 million in 2002. Table 3 illustrates TB’s key performance indicators before
and after privatisation in 1993.

Table 3

In the case of TCC, all nine operational departments31 have registered improvements
in process, product and functional upgrading. Of primary importance to cigarette production

14
is the activity undertaken in the leaf and manufacturing departments32 where substantial
upgrading in operations, mainly in leaf procurement planning (cigarette component blending,
processing, packing, qualification and purchasing), leaf logistics (including warehousing), and
ongoing cigarette blend maintenance and related primary production issues have been
registered. Furthermore, synergies have been established across functional departments, such
as sales and marketing, product development and quality assurance departments to streamline
the production process. Between 1995 and August 1998, TCC invested around US$26 million
in extensive factory renovation, commissioning of new plant and machinery and restructuring
its primary processing, cigarette manufacturing and packaging activities. As a result, TCC’s
plant in Dar es Salaam has become more automated incorporating equipment that utilizes
state-of-the-art technology such as ‘programmable logic control’ and ‘mean weighted control’
systems. For example, the adoption of the ‘mean weight control’ system which is an
automated standard weight of tobacco in each cigarette stick has led to approximately US$ 23
million in savings on raw tobacco.
In addition to the investment in ‘hard’ technology, the new owners implemented a
number of process innovations in the various stages of the cigarette manufacturing process.
For a start, the system to check the quality of sourced raw tobacco has been upgraded through
more efficient plant layout and utilization of new machinery. As a result of the utilization of
new automated equipment, substantial yield efficiencies in tobacco registered. The plant
layout and the linkages between different operational departments has been streamlined to
result in a leaner production process with, less slack, wastage and more productive efficiency.
Since acquisition, the leaf and manufacturing department have been ISO 9002 certified and
this has invariably led to improved product quality assurance. Cigarette production has to
undergo rigorous quality control checks according to MNE parent quality standards before
being sent to the packing machine stage. In addition, the company has been moving towards
ISO 2000 certification, regarding health and safety and work environment standards.
Investment in fixed assets particularly aims to maintain the equipment modern and up to
international standards, maintain business continuity, loss prevention and safety.
The manufacturing process utilising modern and efficient machinery has led to
substantial uniformity in product quality, improved cash flow efficiency and general product
quality and consistency and less machinery down time. Invariably, the utilization of this
technology has resulted in more uniform quality products, less wastage of raw materials,
leading to substantial cost savings. Indeed, as a result of the investment in modern plant and
equipment, productive efficiency has improved from less than 60 per cent in the early 1990s

15
to over 85 per cent. Productivity measured by cigarettes produced per man-hour has increased
from around 3,300 in 1997 to around 8,400 in 2000. Machine downtime has been reduced by
more than 80 per cent since 1997. TCC has increased its production capacity from 1.5 to 5
billion cigarette sticks. Currently, the department manufactures 3.3 billion cigarettes annually
amounting to less than 70 per cent capacity utilisation. Operational efficiency has increased
from 70 per cent to 85 per cent. For example, 1 shift used to have 100 persons on board now
has only 54 people, 2 shifts are operated in the production floor. Cigarettes are produced more
efficiently as a result of new industrial machines capable of producing 8000 cigarettes per
man hour. Leaner production methods have led to cuts in manufacturing costs. Modern
equipment has increased yields from both leaf and NTMs (non tobacco materials).
Investments in manufacturing technology and processes, quality assurance, new products,
brand equity, distribution network and computerization are expected to continue to strengthen
competitiveness and increase efficiencies.
In the emerging context of technology transfer from the parent MNE, TCC has thus
been able to improved product quality and strengthen and reposition its brand portfolio. There
has also been a refocusing of export business in markets with significant long-term growth
opportunities. TCC can now produce to an international quality standard and has a cigarette
portfolio of 11 brands directed at various consumer segments33. In November 1996, Aspen
became the first international JTI brand to be produced by TCC and has since been
successfully exported to neighbouring countries. TCC launched Winston and Winston Lights,
in 1998 and 1999, respectively, the Winston brand being one of JT’s flagship brands34.
Upgraded technological capabilities has thus permitted TCC to undertake the production of
both Virginia and American blend cigarettes to international standards. Sportsman and Sweet
Menthol brands remain the biggest selling brands accounting for over 70 per cent of total
sales volume. In September 2002, TCC introduced Embassy Kings. The company invested
over US$ 2 million in 2002 in sales and promotion to support and to strengthen brand equity.
In the context of the 10 to 20 per expanding its export orientation, TCC has acquired the total
share capital of RJR Kenya Limited and renamed it as TCC (Kenya) limited.
Yield efficiency stands at a 98 per cent. Through ‘mean weigh control’ technological
systems the exact weight of tobacco in the cigarette stick is implemented and the variation
between one stick is now at most 20 mg, ensuring savings to the tune of $20 to 23 million.
This has raised production from less than 3,000 sticks per minute to 8,000 sticks per minute.
By the fiscal year ending December 2001, TCC’s turnover had increased by 85 per cent to
reach US$ 88.5 million from US$ 47.9 million in 1985. Total tax contribution (VAT and

16
excise) to the Treasury reached a record high of US$ 40 million In 2001. TCC’s exports both
regionally and internationally have been steadily increasing and represented about 10 per cent
of sales volume in 2001. TCC is the largest agro-based industry in Tanzania adding
substantial value to one of the country’s main commodities. Key performance indicators of
TCC are equivalent to those in JTI subsidiaries in Europe and the Middle East. The progress
achieved in terms of cost and productive efficiencies has enabled TCC to prepare the
groundwork for the implementation of JTI recently launched global enterprises resource
planning initiative (ERP) initiative. TCC has been undertaking this exercise since 2003 and
runs until 2004. In the same vein, TCC is implementing the ‘Rhino’ project running from the
second half of 2002 until 2006, aiming at continuously improving the company’s product
quality and achieve costs efficiencies through reorganization of the factory layout,
optimisation of the headcount and employee development. Key performance indicators for
TCC in the aftermath of privatisation are illustrated in Table 4.

Table 4

As in the case of TB, the human capital component was vital in the achieving various
forms of upgrading and benefiting from the forms of technology transfer from the parent
company. In the immediate aftermath of privatisation, staff complement downsizing was
undertaken35. The workforce was reduced from 1300 (750 on the production floor) to 730
(300) over four years. The main downsizing was undertaken in the production floor, as a
result of extensive automation of plant and equipment which led to a drastic decrease in
manual jobs. There are only 3 expatriates in the company who are employed in key executive
and technical roles. Locals are employed in key management positions (such as technical,
administrative and sales and marketing positions) as a result of the extensive capabilities and
host country experience they possess.
TCC gives considerable importance to human resource training and development. It
has put in place extensive internal and external training programmes. TCC is one of the main
employers in the host country, and seeks and retains the best young graduates in Tanzania,
providing them with career advancement opportunities as well as external training and
secondments to other JTI plants around the world. TCC set up an on-site training centre late
1997 to spearhead this strategy. The initial post-privatisation training mainly focused on
generic training to enhance employee awareness of organisational change, professionalism
and life skills. Substantial changes to the work ethic inherited from parastatal period was

17
required. The continuing training initiatives addressed employees’ individual development
needs and increase effectiveness, particularly of those employees at the production floor
without basic skills but who were deemed to be trainable. Other training programmes have
been aimed at broadening managers’ international exposure within the parent network and
the MNE training centre in St. Petersburg, Russia. For example, a system of secondment of
TCC personnel to sister affiliates has picked momentum in recent years and a number of local
personnel from middle management upwards have already benefited. These training
programmes are emphasised for the development of senior management, i.e. supervisory and
technical staff. A threshold level of capability for production floor workers was important as
the company has been modernising its plant and equipment. For example, suppliers provide
training on specific machinery prior to commissioning so that when the actual machinery is
installed in the Dar es Salaam plant, it can be utilised immediately without undue work
stoppages. Normally, employees short term ad-hoc training courses with direct relevance to
on-the-job specific tasks.
As a result of the considerable staff training and the introduction of new management
methods, extensive manpower efficiency has been achieved. The upgrading process is
expected to continue in the future, as future human resource objectives will focus on
organisational development through management development and job-specific training.
Within line management, strategies currently in the pipeline include the development of areas
of management in accordance to the ‘frontier’ parent standards to ensure a talented pool of
candidates ready for progression within the company and further ‘indigenize’ the affiliate. A
noteworthy change in the human resources has been the re-focusing of individual contribution
to corporate goals. Parastatal working practices have been changed and every stage in the
production process has become systemized according to ISO 9002 certified specifications. In
contrast to the parastatal period, every task in the company is now defined, with detailed job
description and performance is evaluated accordingly36. A global suggestion system for
empowering shop floor workers to contribute to and take ownership for operational
improvements was implemented in 2002. A significant number of high quality suggestions
were received from personnel and resulted in improved processes and cost savings. It is
noteworthy that a number of production floor suggestions have been also diffused throughout
the parent network and adopted as best practice.
The utilisation of information technology has also been an important change. For
example, TCC receives all the information from its branches scattered all over the Tanzania
mainland within two days at most and this information is processed for submission to the

18
management who in turn transmit it to the parent company head office in Geneva.37 Systems
that generate timely data and information have also been implemented with considerable
success at TB.
Hence, prior experience in the business or industry is regarded as an important in
providing present owners with knowledge of current operations. Without the support of
previous managerial experience for innovation, it was difficult to develop a successful
innovation programme. The experience and exposure of the managers at TB and TCC are
shown to have been a dynamic source of innovation, high productivity and competitiveness.

MNE vertical backward linkages

Next we analyse the vertical backward integration of the cases being studied. We
recall our earlier proposition that seeks to highlight the potential for inter-firm technology
spillovers in the host country as a result of backward linkages established by the newly
acquired firms. As highlighted earlier on, our premise here is that vertical backward linkages
of the acquired firms with host-based firms serves as a medium for MNE technology
spillovers in the host country. We draw attention to three main aspects. First of all there is a
need to ascertain whether MNE affiliates have established vertical backward linkages in
Tanzania. Second, there is need to analyse whether, as the literature suggests, local sourcing
has increased as a result of the MNE acquisitions. Finally it is important to analyse the
technological characteristics of local sourced inputs, as this should shed light on the
technological capabilities of the linked firm. In this context, we distinguish between resource
based inputs, low and medium technological intensive manufactured inputs.
Following acquisition by SAB, TB initiated a comprehensive programme of local
sourcing. Local sourcing makes also strong economic and logistical sense for the company.
In the immediate aftermath of privatisation almost all inputs to the production process were
imported from within the parent network. However, this started to change as a result of a
number of factors, notably the new strategic objectives of safeguarding profits by reducing
costs (greater cost control can be exerted on locally procured inputs) and improving the
corporate citizen image of the firm in the host country by generating and visibly benefiting the
host economy through linkages. In addition, there are important logistical considerations for
local sourcing. There is a six week lead time from SAB’s plc main parent sourcing hub in
Durban, South Africa, to Dar es Salaam, for imports of key production inputs and any
stoppages occurring at source can therefore cause stoppages in TB’s production plants in
Tanzania. Hence proximity to suppliers is deemed also important in terms of logistical and

19
business continuation. In the context of this strategic development therefore, TB identified a
number of inputs to the production of beer that could be sourced locally from the primary
inputs (barley) to other intermediate inputs in the packaging process. For example, currently,
TB sources approximately 30 per cent of its barley requirement from domestic sources. For
this purpose, TB has set up a maltings plant in Moshi to systemize the purchase of barley
from over 500 farmers in the region. In 2000, approximately US$ 2.4 million were paid to
farmers for the crop, representing the single largest source of direct income for farmers in the
region. In the near future, TB plans to meet the entire barley requirement from local sources
(increasing from 6,000 to 26,000 tonnes) through further investment in research and
development of the crop to improve yields and deepening of the present linkages with the
farming community. Another important backward linkages established by TB is the sourcing
of locally manufactured glass beer bottles. The supplier, KIOO Glass Ltd, an MNE affiliate,
has established close technical collaboration with TB so as to meet production quality
requirements38. Another supplier in Tanzania for TB is Carnaud Metal Box, manufacturer of
metal crown corks. This supplier is also a long established foreign affiliate and was the first
supplier to enter into a technical collaboration with TB after privatisation in 1993. In addition,
TB sources plastic crates and shrink-to-fit packaging from Simba Plastics and is currently
sourcing some of its bottle label requirements from Tanzania Printers, a local printing
company (labels). TB has also established strong backward links with Showerlux Ltd,
manufacturer of industrial chemicals. Around 36 per cent of TB’s inputs are sourced locally
and plans are underway to increase this percentage to 50 per cent in the next 3 to 5 years.
TCC industrial restructuring since 1995 has catalysed investment in upstream
industrial activities particularly the farmers and growers who supply the raw tobacco. During
the 1996-2000 period, the revitalization of TCC has also served to ‘crowd in’ other foreign
investment in Tanzania in tobacco processing activities. Indeed, around US$ 65 million was
invested in the sector by leading world tobacco merchant MNEs39. Two of the largest tobacco
leaf processors and sellers in the world, Universal and Dimon, have invested over US$60
million in new factories in Tanzania, following the restructuring of TCC. Total tobacco
processing plants have a combined processing capacity of 78 million kg. per annum and these
operations are vertically linked with TCC. These tobacco processors source raw tobacco from
farmers, process the raw material which then they sell to TCC for the manufacturing of
cigarettes40. On the one hand, there is TCC’s ongoing investment and requirements for quality
tobacco, as well as the ‘crowding in’ of this investment in tobacco processing. Annually, TCC
buys approximately around 1,800 tonnes of raw tobacco for the purpose of manufacturing

20
cigarettes and 2,000 tonnes for the purpose of exports to other plants within the parent
network. This backward linkage makes TCC one of the leading agro-based local value added
manufacturers in Tanzania. In addition to the raw tobacco, TCC has also established
noteworthy vertical ties with printing packaging suppliers.
Both TB and TCC are the main active partners in the so called, Tanzanian Private
Sector Initiative (PSI). PSI has been established to develop and expand the role of the
corporate private sector in Tanzania in a bid to increase the opportunities for small local
business to contribute to, and participate in the pro-poor growth of the Tanzanian economy,
i.e. to promote more linkages as part of successful business operations. In other words, the
PSI has been intended as a vehicle for large companies (mainly foreign affiliates in Tanzania)
to work collectively in partnership with each other, donor agencies and government. It thus
serves as a catalyst to cause greater involvement of the private sector engaged in local
enterprise development and social investment towards effecting sustainable development and
poverty reduction41. Since privatisation, both companies have been at the forefront of vertical
integration in the host country, by increasing the nature and extent of backward linkages with
host-based firms.
On the basis of the above evidence, three main points regarding linkages emerge., i.e.
the evidence of vertical backward linkages, the hypothesised positive impact of the
acquisition on establishing new linkages and technological characteristics of sourced inputs. It
is clear that in both cases, backward linkages in the host country have been established and
there is evidence that there is a potential for technology spillovers to emerge between the
affiliates and their suppliers42. FDI through acquisitions has resulted in more backward
linkages being established when compared to the situation in the pre-acquisition period. The
deepening of local value chains has primarily been the result of the affiliates’ growth and
turnaround in productive performance. Whereas in the formerly stagnant, under capitalized
and almost declining industries, the scale and scope for vertical linkages was almost non-
existent, this is not the case following privatisation. Nevertheless, it is the prerogative of new
foreign owners to decide whether to source locally or from abroad, for example through the
parent network. Production volumes have increased, new products and brands developed and
this has led to an increased input requirements to respective production processes of beer and
cigarettes. TB has increased or in almost all cases, started backward linkages with locally
based firms. This has been the result of a comprehensive effort by the affiliate to ‘look around
for and develop technical collaborative agreements with locally based suppliers’ (Company
interview, 2002). In the parastatal period, local sourcing was not the norm and enterprises

21
preferred to import intermediate inputs from neighbouring countries. Cost, competitive and
logistical considerations were not considered important at that time. Following the
liberalisation of the economy and the restructuring of TB, cost and logistical factors assumed
greater importance in the pursuance of the company’s strategic objectives. Hence, as a
procurement strategy, local sourcing was deemed as ‘making a lot of sense’ (Company
interview 2002). This is also the case with TCC. In the pre-liberalisation period, production
volumes were abysmally low and this had a negative impact on the tobacco cultivation and
processing industry. After liberalisation and TCC’s restructuring programme, the tobacco
industry picked up, reinforcing the local value chain.
From the evidence of vertical backward linkages, it emerges that MNE affiliates
source different inputs (with different technological characteristics and/or intensity) from
different firms simultaneously. This information is important since it sheds light on the
potential technology spillover effects from the MNE to its local suppliers. It is widely
accepted that sourced intermediate inputs reflect different technological characteristics in
terms of the nature and extent of production process involved in their production, the physical
and human capital involved in their production as well as the material which they are
produced. Through the overview of the MNE affiliates locally sourced inputs, we can
distinguish between resource-based, low and medium technology inputs. For TB and TCC,
the ‘lion’s share’ of inputs reflect primary, resource-based inputs. Since acquisition, both TB
and TCC have developed vertical backward linkages with local farmers in the barley and
tobacco sectors, respectively43. Indeed, these linkages reflect the ‘match’ existing between
local economic activities; i.e. those found in agriculture and the agro-food manufacturing
industry in which the two MNE affiliates operate44.

7. Conclusions and implications

Using a case-study approach, this study examines the role of privatisation on the
industrial landscape of Tanzania in the agro-foods sector. The main findings suggest that on
the whole technological upgrading has occurred following the acquisition by MNEs. We
analysed the potential of technology diffusion through backward linkages and concluded that
there is evidence of improved backward linkages with local economic actors. Capabilities in
the host country context, both within the acquired firms and also in the supplier base of the
host country, matter for the magnitude and intensity of technological upgrading. It is worth
emphasising that both of the cases were in an industrial sector which match the comparative
advantage of the host country. Domestic capabilities in other, more technology-intensive

22
sectors would be considerably lower, and it would unreasonable to expect such positive
results. It is therefore important to emphasise that our results should only be generalised with
great caution. ‘Wider’ technology gaps between the domestic and foreign-owned sector are
more likely to result in fewer backward linkages as well as the type of technological content
of inputs sourced locally.
It is important to highlight some caveats that emerge from the analysis which may be
relevant to the discussion of upgrading. The sustainability of the acquired firm’s operations
given the multitude of constraints in the local market is another factor worth considering
when assessing the implications of vertical linkages. It is true that greater integration of
foreign affiliates is a positive implication for the host country through the potential spread of
technological spillovers and demonstration effects of acquired firms to their suppliers and
customers. The relevance of this issue is exacerbated by the dearth of an indigenous industrial
base generating economic activity.
There are also a number of caveats regarding MNE backward linkages in the host
country. However, while vertical linkages with other foreign affiliates based in the host
country may be important, these linkages may not offer the same type of potential for
externalities in the host countries. Here, the issue revolves around the presence of a
competitive and capable domestic industrial base able to link up with MNE affiliates. In the
cases examined, it is evident that in the case of backward linkages reflect primary activities in
the case of local suppliers and more manufactured inputs with various types of technological
capabilities in the case of foreign suppliers. This represents a notable variation to consider
when promoting policies for the ‘indigenization’ of FDI and MNE activity. Although the
formation of backward linkages is a noteworthy, positive development, the size and market
power of foreign affiliates in the host country can affect negatively local suppliers as
highlighted in Altenburg (2000) and Brimble (2001)45. In the absence of a realistic
counterfactual, one can argue that the MNE vertical linkages established with the present
suppliers are leading to externalities that would have otherwise not existed. Probably in the
short and medium term it is ‘easier’ for government to promote further an enabling
investment climate so that these FDI activities are consolidated further, because in the present
circumstances these vertical linkages (foreign to foreign) represent the sole available mode of
industrial upgrading and capability development in Tanzania being diffused to other host
based economic activity. As long as industrial and technological upgrading happens and spills
over to more than one firm, it does not matter who the beneficiary is if it serves to embed
further the MNE affiliate in the host country. However, this outcome should not be taken as

23
an end in itself but a means to an end. Further research is needed to gather more insights on
the technology transfer aspects of FDI through acquisitions, as well as regards the
technological and knowledge spillovers of MNE affiliates to their local suppliers.

24
References

Abramovitz, M. (1995), ‘The Elements of Social Capability’, in Perkins, D.H. and Koo, B
(eds.), Social Capability and Long-term Growth, Basingstoke: Macmillan Press.

Allard, P. and Lundborg, L. (1998), ‘Foreign acquisitions of Swedish companies: a study of

internal and external effects’, ISA Studies on Foreign Direct Investment, No.5 (Stockholm:
Invest in Sweden Agency).

Altenburg, T. (2000), ‘Linkages and Spillovers between Transnational Corporations and

Small and Medium-sized Enterprises in Developing Countries: Opportunities and Best
Policies’, in UNCTAD, TNC-SME Linkages for Development: Issues-Experiences-Best
Practices. New York and Geneva: United Nations.

Barrel, R. and Pain, N. (1997), ‘Foreign Direct Investment, Technological Change and
Economic Growth Within Europe’, Economic Journal, v.107 (445).

Belderbos, R., Campanelli, G. and Fukao K., (2001), ‘Backward vertical linkages of foreign
manufacturing affiliates: Evidence from Japanese multinationals’, World Development, 29, 1,
(189-208).

Benito, G., Grogaard B., and Narula, R. (2003), ‘Environmental influences on MNE
subsidiary roles: Economic integration and the Nordic countries’, Journal of International
Business Studies, 34, (443-456).

Blomstrom, M. and Kokko, A. (1997), ‘How Foreign Investment Affects Host Countries’,
Policy Research Working Paper, The World Bank.

Bresman, H., Birkinshaw J., and Nobel R. (1999), ‘Knowledge Transfer in international
acquisitions’, Journal of International Business Studies, 30, 3 (439-462).

Brimble, P. (2001), ‘The Thai Hard Disk Drive Industry’, UNCTAD (Geneva: UNCTAD),
mimeo.

Carillo, J. (2001), ‘Foreign direct investment and local linkages: experiences and the role of
policies. The case of the Mexican television industry in Tijuna’, UNCTAD (Geneva:
UNCTAD), mimeo.

Castellani, D and Zanfei, A. (1998), ‘Multinational growth and the creation of linkages with
local firms: evidence from the electronics industry”, paper presented at the IRD&P workshop
on the Economics of Science and Technology: Micro-Foundations and Policy (Italy,
University of Urbino), 5-6 June.

Chudnovsky D., Lopez, A. and Porta, F. (1995), ‘New foreign direct investment in Argentina:
privatization, the domestic market and regional integration’, in Manual R. Agosin, ed.,
Foreign Direct Investment in Latin America (Washington, D.C.:Inter-American Development
Bank), (39-104).

25
Criscuolo, P and Narula, R. (2002), ‘A Novel Approach to National Technological
Accumulation and Absorptive Capacity: Aggregating Cohen and Levinthal’, MERIT
Research Memorandum, 2002 – 16.

De Mello, L.R., Jr. (1997), ‘Foreign direct investment in developing countries and growth: A
Selective Survey’, Journal of Development Studies, Vol. 34, (1-34).

Djankov, S and Hoekman, B. (1999), ‘Foreign Investment and Productivity Growth in Czech
Enterprises’, World Bank Economic Review, Vol. 14, (49-64).

Driffield, N and Mohd Noor, A. (1999), ‘Foreign direct investment and local input linkages in
Malaysia’, Transnational Corporations, 8, 3 (1-24).

Findlay, R. (1978), ‘Relative backwardness, direct foreign investment and the transfer of
technology, a simple dynamic model’, Quarterly Journal of Economics, Vol, 92, (1-16).

Ganiatsos, T. (2000), ‘Global component outsourcing in developing countries’ electronics and

automotive industries’, paper submitted to the UNCTAD Expert Meeting on the Relationship
between SMEs and TNCs to ensure the competitiveness of SMEs (Geneva), 27-29 November,
mimeo.

Gorg, H and Ruane F. (1998), ‘Linkages between multinationals and indigenous firms:
evidence fro the electronics sector in Ireland’, Trinity Economic Papers, Technical Paper 13
(Dublin: Trinity College).

Government of Tanzania (1999), The Tanzania Development Vision, 2025, Planning

Commission, The United Republic of Tanzania, Dar es Salaam.

Humphrey, J and Schmitz, H. (2000), ‘Governance and upgrading: linking industrial cluster
and global value chain research, IDS Working Paper 120, Brighton, Institute of Development
Studies.

Jo, S. H. (1980), ‘Direct foreign private investment’, in Chong Kee Park, ed., Macroeconomic
and Industrial Development in Korea (Seoul: Korea Development Institute), (129-182).

Kumar, N. (1990), ‘Multinational Enterprises in India, (London: Routledge).

Lall, S. (2002), ‘Implications of Cross-Border Mergers and Acquisitions by TNCs in

Developing Countries. A Beginner’s Guide’, QEH Working Paper Series, No. 88, June.

Lall, S. (2000), ‘Transnational Corporations and Technology Flows’, in D Nayyar (ed), New
Roles and Functions for the United Nations and Bretton Woods Institutions, Helsinki; UN
University WIDER.

Lall, S. (1980), ‘Vertical Inter-firm Linkages in LDCs: An Empirical Study’, Oxford Bulletin
of Economics and Statistics (August 1980) (209-222).

Landi, J, (1986), ‘The sourcing policies of MNEs: a case study of Nigeria, (University of
Reading), PhD thesis.

26
Liberatori, M and Pigato M. (2000), ‘FDI Through Privatisation in sub-Saharan Africa’,
World Bank, Washington, D.C.

Lim, L and Pang, E.F. (1982), ‘Vertical Linkages and Multinational Enterprises in
Developing Countries’, World Development, Vol. 10 No.7 (585-595).

Maucher, H.O. (1998), ‘View: Mergers and Acquisitions as a means of restructuring and
repositioning in the global market – business, macroeconomic and political aspects’,
Transnational Corporations, 783), (153-83).

Mbelle, A and Shitundu J, (1999), ‘The Management of Industrial Reforms in Tanzania.

Perspectives and New Challenges’, mimeo, Economics Department. ERB, Dar es Salaam.

McAleese, D and McDonald, D. (1978), ‘Employment growth and the development of

linkages in foreign-owned and domestic manufacturing enterprises’, Oxford Bulletin of
Economics and Statistics, November, (321-339).

Moden, K. M. (1998). ’Foreign acquisitions of Swedish companies: effects on R&D and

productivity’ (Stockholm: Research Institute of Industrial Economics), mimeo.

Mortimore, M. (1998), ‘Getting a lift: modernizing industry by way of Latin American

integration schemes. The examples of automobiles’, Transnational Corporations, 7, 2
(August), (97-136).

Mwandenga, A. (2000), ‘Socio-economic impact of privatization: The Tanzania Experience’,

African Journal of Finance and Management, Vol, 19, No. 1, July 2000.

Narula, R. (2003), ‘Switching from import substitution to the ‘New Economic Model’ in
Latin America. A case of not learning from Asia’, MERIT Research Memorandum Series,
2002-032.

Narula, R. and Dunning, R. (2000), ‘Industrial Development, Globalisation and Multinational

Enterprises: New Realities for Developing Countries’, Oxford Development Studies 28(2)
(141-167).

Noorbaksh, F, Paloni, A and Youssef, A. (2001), ‘Human Capital and FDI Inflows to
Developing Countries: New Empirical Evidence’, World Development, Vol. 29, No. 9, (153-
1610).

Pigato M. (2001), ‘The foreign direct investment environment in Africa’. Africa region
working paper series No.15. World Bank, Washington DC.

Rasiah, R. (1994), ‘Flexible production systems and local machine tool subcontracting:
electronics component multinationals in Malaysia’, Cambridge Journal of Economics Vol. 18:
(279-98).

Reuber, G.L., Crookell, H., Emerson, M. and Gallais-Hamonno, G. (1973), ‘Private Foreign
Investment in Development’, (Oxford: Clarendon Press).

27
Rodriguez-Clare, A. (1996), ‘Multinationals, linkages, and economic development’,
American Economic Review 85, (852-73).

Scott-Kennel, J and Enderwick, P. (2001), ‘The degree of linkage of foreign direct investment
in New Zealand industry’. (Wellington: Victoria University of Wellington), mimeo.

UNCTAD, (2001), World Investment Report 2001: Promoting Linkages. New York and
Geneva: United Nations.

UNCTAD, (2000). World Investment Report 2000: Cross-border mergers and acquisitions
and development. New York and Geneva: United Nations.

Wangwe, S. (1997). ‘Economic Reforms and Poverty Alleviation in Tanzania’, Discussion

Paper No. 16, Economic and Social Research Foundation, Dar es Salaam.

Yin, R.K (1984). Case study research: design and methods. Beverly Hills, CA: Sage
Publications.

Zhan, X. J and Ozawa, T. (2000), ‘Business Restructuring in Asia: Cross-border M&As in

Crisis affected Countries’ Copenhagen Business School Press.

28
 TABLE ONE

Tanzania Breweries Limited - Basic indicators

1994 1997 2002 2003

Turnover (in US million) 29.0 119.2 135.0 174.0

Profit before tax (in US million) -1.0 19.0 34.3 47.6
Cash flow from operations (in US million) 0.6 17.2 30.1 43.2
Net cash invested to expand operations (in US million) 6.0 10.5 3.3 1.1

Total number of employees 3,000 1,250

Number of persons receiving training in Tanzania - 746 1,201 1,089
Number of persons receiving training abroad - 23 31 31

(in 000 US dollars) 1997 2000 2002 2003

Cash derived from sales 119.200 169.900 166.300 208.300

Cash paid to suppliers 47.400 63.100 57.000 64.900
Cash value added 72.200 107.100 109.300 143.500
Cash utilised to
Remunerate employees for their services 9.100 13.200 9.900 10.800
Pay direct taxes to Government 600 600 7.200 11.200
Pay excise duty and Value Added Tax 44.000 62.500 62.200 77.800
Cash disbursed among stakeholders 65.500 100.800 103.300 142.400
a
Cash retained for re-investment 6.700 6.300 6.000 1.1000
a
Cash retained to fund replacement of assets and to facilitate further
growth

Source: Field Interviews and Company Annual Reports

29
 TABLE TWO

Tanzania Cigarette Company Limited - Basic Indicators

1996 1999 2001

Sales turnover (in US million) 61.7 77.6 88.5

Profits before tax (in US million) 7.7 15.3 21.6
Profits after tax (in US million) 5 15.3 22

Cigarettes manufactured per man hour 3,000 6,000 8,400

Machine efficiency (percentage) 51 76 78

Total number of employees 1,205 730 735

Sales 269 279 286
Administration 232 192 194
Manufacturing 704 259 255

Source: Field Interviews and Company Annual Reports

TABLE THREE

Performance indicators for Tanzania Breweries Limited

Pre - privatization Post - privatization

Sales turnover (in US$ million) 29.0 135.0

Production level (in hecto litres) 0.4 million 2.3 million
Capacity utilisation (in per cent) 25 80
Number of product brands 2 10
Local market share (in per cent) 31 85
Exports (in per cent) 1.5 5.0
Number of employees 3,050 1,250
Profit before tax (in US$ million) -1.0 34.3
Indirect taxation to government (in US$ million) 32.0 77.8
Dividend (declared/proposed) (in US$ million) 8.8 30.8

Source: Compiled by authors from field Interviews and company annual reports

30
 TABLE FOUR

Performance indicators for Tanzania Cigarette Company Limited

Pre - privatization Post - privatization

Gross turnover (in US$ million) 61.2 88.5

Profit before tax (in US$ million) 7.7 21.5
Number of employees 1,500 730
Number of product brands 3 8
Production capacity (in sticks) 1.5 billion 5 billion
Capacity Utilisation (in per cent) 51.0 80.0
Local market share (in per cent) 60.0 95.0
Excise & VAT (in US$ million) 35.0 39.4
Dividend paid (in US$ million) 4.0 22.7

Source: Compiled by authors from field Interviews and company annual reports

31
NOTES
1
For the examination of MNE affiliate vertical linkages, we use the definition of Lall (1980), who defines
linkages as being the ‘direct relationships established by firms in complementary activities which are external to
‘pure’ market transactions’ and ‘essential to the functioning of any normal industrial market and transactions that
go beyond arm’s length, one-off relations and involve longer-term relations between firms’ (p. 204). The
relationship over time involve sustained exchanges of information, technology, skills and other ownership assets.
2
See for example the strategy document “Tanzania Development Vision, 2025” (GOT 1999). Tanzania’s
economic policy two decades ago involved the outright barring of FDI. The present attitude towards FDI
represents a dramatic turnaround in policy. The privatisation of parastatal companies started in the early 1990s
with a total of 257 companies being divested.
3
These figures also include portfolio investment. Portfolio investment substantially declined in the aftermath of
the Asian crisis to less than 10 per cent of total foreign exchange.
4
In a study on Nigeria, foreign affiliates had a higher propensity to import than their local counterparts (Landi
1986). Similar conclusions have been made in the cases of Ireland, Republic of Korea and India (McAleese and
McDonald 1978, Jo 1980, Kumar 1990).
5
The literature on spillovers tends to ignore another important factor: that not all MNE subsidiaries will provide
spillovers and linkages to the same extent.
6
In the case of Japanese MNEs, Belderbos et al (2001) conclude that acquired affiliates had significantly higher
local content levels than those established through green field investment due to their pre acquisition integration
in the local economy.
7
See for example, Driffield and Mohd Noor 1999, Castellani and Zanfei 1998, Gorg and Ruane, 1998.
8
This has been observed to be the case in East Asia (See e.g., Rasiah 1994), but it is to be noted that firms build
on location advantages that already exist in the host economy and increases in embeddedness are generally in
response to improvements in the domestic technological and absorptive capacity.
9
Questionnaire data was based on three main aspects; information on the affiliate and ownership structure, the
FDI motive and established vertical linkages of the affiliate in the host country. Emphasis was put on the main
changes that have taken place within the firm as a result of privatisation. Multiple visits in certain cases where
undertaken to verify and clarify information sought in the first instance. Interviews were held with company
managers and every effort has been made to interview personnel who could provide an overview of pre and post
privatisation stances. Precautions have been taken to reduce the risks entailed in the case study approach. Taking
advantage of suitability of the case study approach to handle a variety of evidence, the information obtained in
interviews has been complemented by information from official and unofficial documents and by follow-up
interviews to obtain the necessary clarifications.
10
The case study approach can be applied to at least three different situations in evaluation research: to explain
the causal links in real-life interventions that are too complex for other research strategies, to describe the real-
life context in which an intervention has occurred or for illustrative purposes, and to explore those situations
where a single set of outcomes is not clear. Case studies are the preferred approach when 'how' and 'why'
questions are posed, when the investigator has very little control over events and when the focus is on a
contemporary phenomenon within its real-life context (Yin 1984).
11
Precautions have been taken to reduce the risks entailed in the case study approach. To take advantage of the
unique ability of the case study approach to handle a variety of evidence, the information obtained in interviews
has been complemented by information from official and unofficial documents and by follow-up interviews to
obtain the necessary clarifications. In the case of Africa, the extent of industrial development and type of FDI
activity may suggest that technology absorption and diffusion through backward linkages might occur in outlier
cases than occur in an average or typical situation. Even if these firms may be outliers, the approach taken in this
study is that it is also important to remember that outliers are particularly informative.
12
In 1993, SAB miller was invited by the Tanzanian government as its first privatisation partner to become the
strategic shareholder in TB with management control. The shareholding structure of the company stands as
follows; SAB plc 66 per cent, International Finance Corporation (IFC) 8.79 per cent, the Government of
Tanzania 5 per cent, Privatisation Trust 10 per cent and the general public owns 10.17 per cent.
13
TB has also a controlling interest in Tanzania Distilleries and Darbrew. Tanzania Distilleries manufactures and
distributes prominent local spirits in Tanzania as well as imported wines and spirits. Darbrew produces and sells
traditional local beer.
14
Beer consumption per capita levels in Tanzania are 6 litres compared to 13 in Angola, 32 in Botswana.
15
For example, productivity levels in TB’s main brewery in Dar were very low by international standards, 316
hl/capita compared to the industrial standard of 5,400. One of the reasons for the productive inefficiency in the
pre-privatisation period was the operating bottling lines, commissioned in 1991 based on second hand
technology and were never fully utilised to their maximum output, because of extensive breakdown which
frequently resulted in serious delayed production runs. TB’s production efficiency was hindered by inferior

32
quality of intermediate inputs such as barley and glass bottles, obsolete refrigeration system and lack of quality
assurance. Filled bottles often contained too much air, reducing the shelf life of the beer and hence provoking
customer complaints and lost market share.
16
More recently around $4 million have been invested in new bottling and canning facilities aimed to increase
productive efficiency, introduce product innovation and secure quality. Packaging lines have been fitted with
state-of-the-art automated bottle inspectors to avoid infiltration in production. This is the first such facility
commissioned in East Africa and is enabling TB to increase consumer choice in the ‘beer take away as well as
increase export orientation. Exports of Castle in cans has been directed to Uganda.
17
For example, as a result of investing in new plant and equipment, the brewing process now has a more
accurate yeast and oxygen dosing process regardless of production flows, .providing greater uniformity in
product blends and extensive cost reduction. The result is a complete and well documented process line designed
to maintain the characteristics of the beer and meet the quality requirements.
18
Efficiency is a measure of the actual volume produced compared to the theoretical volume over a certain
period of time. In the case of factory efficiency, it applies to the whole brewery and in the case of machine
efficiency to an individual packaging line. TB has achieved 70 per cent factory efficiency (compared to 50 per
cent levels in the early 1990s) which compares to world class industrial standards and the efficiency in ‘sister’
affiliates in the MNE network.
19
TB has to comply with a rigorous, MNE group-wide approach to quality control and product safety. To ensure
that quality is monitored and tracked, new product labelling systems such as the Julian Calendar Codes including
brewing data (brewery name, production line, manufacturing date and time) and ‘best before date’ have been
introduced use. This assists in the management of stocks and helps trace the origin in the event of quality
enquiries.
20
Improvements in the production process have led to a decrease in energy consumption. Effluent is managed
upstream generally by removing a large proportion of the organic pollutants such as yeast and inorganic loading
from the chemical cleaning processes. In addition, the recycling rate of bottle washing water has been increased,
thereby reducing fresh water consumption requirements.
21
For example, the Konyagi brand has been revamped and new brands introduced, permitting the company to
acquire more than 10 per cent of the local market share and link up with international partners. Significant
investment has been undertaken at Dar Brew revamping the quality of local brands such as Kibuku and
Mwamba.
22
The brand portfolio has been widened extensively to include the production and distribution of brands such as
SAB Miller’s flagship brand Castle Lager, Castle Milk Stout, Redds Premium Cold under licence to the parent
company. Safari Lager is the market leader. Kilimanjaro Premium and Safari Lager represent the leading
mainstream clear brands commanding 84 per cent of clear beer volumes. More brands are being produced in the
340ml can which has reduced the reliance on imports and created the opportunity of pursuing exports. TB was
able to embark on the export of its brands to the neighbouring countries on a sustainable basis. Castle Lager is
the leading export brand, with Redds following. No local brands are successfully exported. Following the
acquisition of Kibo Breweries, the Company now also manufactures and distributes certain East African
Breweries (EABL) brands under license.
23
Care for human resources has extended to other economic and social considerations such as provision of good
food from the canteen, better medical services and attractive remuneration. As well as providing in-house
training, selected employees are sent to SAB’s Training Institute in South Africa for advanced managerial
education and technical development. The in-house training department organizes various short courses as part
of departmental on-the-job training. The policy adopted regarding management development in TB is to improve
skills of staff in activities related to the functions of the company. TB prefers to employ already qualified staff
who need only to be introduced to the company's working methods.
24
Due to the political sensitivity of downsizing, the company has undertaken the process in two main stages,
namely, voluntary and forced staff reduction.
25
In some cases departments were reduced to sections under one main department. This represents a continuous
process that aims to ensure that structures contribute to achieving the desired goals.
26
To this end, tailor-specific training courses are given by equipment suppliers in both process and automation
technology and these are designed to extend the skills base and increase the specialist knowledge. Such courses
are offered either on-site or at supplier premises abroad. Time allocated to training is an average of 3.2 days per
employee per annum. Within the MNE parent human resource development categorization, TB has moved from
a lower category of ‘emerging’ to ‘developed’ over a five year period.
27
The main area of concern has been to lead staff to understand the objectives of the company, their internal
systems covering accounts, commercial, human resource and technical areas. The first step in management
development by the new investors has been to expose staff to parent company systems that make reporting and
monitoring easier.

33
28
A market-related salary system has also being introduced as a way of retaining staff.
29
TB cost of beer/hecto litre stands at US$10 compared to 15 for its main international competitors. For TB, low
cost maintenance is important in the context of avoiding unnecessary fluctuations in price so that more
consumers can afford despite the fact that the economic situation in Tanzania changed little if at all between the
share purchase and the time of writing.
30
This is measured according the quality standards and ‘defect’ in the production process of beer and end
product. SABMiller Africa and Asia introduced an inter-brewery competition in October 2001, in which all
production facilities are benchmarked and scored against a set of technical measures to determine a ‘Brewery of
the Year’. In the first round of this competition, TB through its Arusha Brewery attained the second highest
score for producing the best quality beer among all of SAB's international breweries, including those in Europe
and Asia.
31
The 9 units are; leaf, manufacturing, sales and marketing, human resources, finance, legal, corporate affairs,
company services and exports.
32
TCC produces three main types of cigarette blends, namely American, Dark Fired and Virginia.
33
These segments can be broadly categorised as low, value, medium and premium brand cigarettes.
34
This was also the first American blend cigarette to be ever produced in East Africa region.
35
Employment was not drastically cut, since there was an element of natural wastage and re-deployment in
factory activities.
36
There are now manuals, code of conducts, jobs are defined and described systematically. In case of stoppages,
the time, shift, date and personnel involved and which type of machinery will be immediately evaluated.
37
The entire picture of the operations of TCC is known by directors by the first week of every month. If there are
any operational problems, decisions are taken immediately.
38
As a result of this linkage, KIOO Ltd has increased production volumes and invested considerably in its
technological capabilities. Spearheaded by the success in supplying TB, the company stepped up its export sales.
For example, following the supplier agreement in Tanzania, the company entered into a similar supplier
agreement with a SAB affiliate in Uganda.
39
Tanzania boasts a well established tobacco farming industry based mainly in the Tabora, Iringa and Ruvuma
regions. It is the fourth major export crop after cashews, coffee and cotton. Tobacco production in 2001 was
estimated at 30,000 tons. The JT group is jointly the largest buyer of Tanzania tobacco with TCC using around
2,000 tons annually for domestic manufacture of cigarettes and a further 2,200 tons being exported to other
plants worldwide. The tobacco industry accounts for about 10 per cent of the value of agricultural cash crop
exports of Tanzania and employs more than 200,000 families. The cigarette industry generates over US$40
million in tax revenues in 2001, equivalent to about 6 per cent of the Government’s recurrent revenues in 2001.
40
There are certain constraints to these backward linkages. In Tanzania, tobacco yields are low (between 600-
1,500 kg per hectare compared to 1,800 kg/ha in Malawi and 2,000 kg in Zimbabwe), stemming from inadequate
production inputs, poor extension services and research, poor crop husbandry and high level of post harvest
losses.
41
The PSI brings together the major MNE affiliates in Tanzania from a number of important sectors, such as
mining, industry and manufacturing, including TB and TCC.
42
Anecdotal evidence suggests that both TB and TCC have entered into technical collaborative relationships
with their suppliers in a bid to control the quality, reliability and timely supply of inputs. These collaborations
have also led to considerable technological capability development on the part of the suppliers which could draw
on the resources, expertise and support of the foreign affiliates.
43
Although it must be said that in the case of TCC, this linkage is an indirect one since, foreign affiliates as
tobacco processors act as middlemen between TCC and the farmers.
44
For these local economic agents, primarily farmers, the linkages with other foreign affiliates represent notable
development since either these vertical linkages did not exist before privatization or were not so well developed.
45
Lim and Pang (1982) caution on the developmental impact of linkage creation. They argue that affiliate local
sourcing may increase domestic value-added and incomes, but by less than the value and proportion of local
input purchases suggest, since locally procured items themselves are most likely to be imported or embody a
large import content.

34