Training Asset 4 How to Work With the Kaltura Media Galler.docx

Training Asset 4: How to Work With the Kaltura Media Gallery
1- In this session, we will learn the functions of your Kaltura
Media Gallery and how to
place the videos created in the Media Gallery into your Sakai
Classroom.
2- Start by selecting Media Gallery on the left hand navigation
menu.
3- The Media Gallery is where you can manage and edit your
uploaded or recorded
multimedia. The tabs Site Library, Collections, and My Media,
are the three visible tabs,
and all have different functions.
4- Site Library is your where you can see videos that have
already been uploaded to the
APUS Course. These videos have already been placed into the
course via the Kaltura
icon within a rich text editor into either the Forums,
Assignments, Announcements, or
Lessons.

5- Collections is where you can make packages of multiple
videos for easy importing,
storage, or grouping. Creating a collection is an effective way
to group multiple videos
relating to the same course / subject together for easy
management.
6- My Media is where you can see all of the uploaded and
recorded files you have made
using Kaltura.
7- Now that the tabs are explained, let’s go into Site Library, so
we can view the various
options present on the toolbar.
8- Downloading, Embedding, Editing the Details, Clipping &
Editing the Media, Removing,
and Adding are all available underneath of the video frame.
Editing clips via the Site
Library will affect all instances of the video in the course.
9- Once again, the Media Gallery is where you can manage all
of your videos. In order to
place the videos you have uploaded or created in your Media
Gallery, you will still have
to go to an instance of the rich text editor and place the video
using the Kaltura button.
Think of the Media Gallery as the location to manage and edit
your videos, while the
Kaltura button in the rich text editor as the way to place them.

Training Asset 3: How to Edit Videos Using Kaltura
1- In this session, we will learn how to edit Kaltura webcam
recordings and videos to
shorten the length of a selected clip.
2- To begin, select “Media Gallery” on the left hand menu.
3- Once within the Media Gallery, you can view multimedia you
have either created or
uploaded. To edit a clip, select the item on the right hand side,
then select “Clip Media”.
4- To begin editing, move the arrow selector to the point where
you want to begin your
new clip, then select “Add New Clip”.
5- This will bring up a grey editing bar, which denotes the
beginning and ending of your

new clip. If you do not want to use the manual slider, you can
also select the start / stop
timers on the top right, and enter in the values for the start and
end times. Remember
to also check “Add to Site Library” to get your video to display
in your media gallery
menu.
6- When you are completed with your edit, select “Save” on the
bottom to create a new
clip of a video. When you return to your “My Media” folder,
your video well display on
the right hand side.
Training Asset 2: How to Upload and Embed Videos into the
Sakai Classroom
Using Kaltura
1- To begin, select the multicolored Kaltura icon located in the
middle row of the rich text
editor.
2- Selecting the Kaltura icon will bring up the following
Browser window, where we will
select “Upload” to begin uploading multimedia from our

computer.
3- For this demonstration, let’s select “Browse” on the “Video”
tab, as this will open your
explorer to select videos present on our computer.
4- Once your explorer opens, you can select a single video, or
multiple videos by holding
down your “CTRL” key. When you have finished your
selection, select “Open”.
5- As you can see, your files will appear in the box, showing
their upload status, name, size,
and the option to remove the file. Select “Upload!” to proceed
placing them into your
Media Gallery, and “Next” to proceed to the next screen.
6- To finalize your video recording, enter the “Title”, “Tags”
which are keywords for search
optimization, and a “Description”, then select “Next”.
7- When completed, you will see the “Done!” screen, where you
can choose to “Add More

Media”, or, “Finish”. Selecting “Finish” will bring you back to
your Media Gallery.
8- Clicking the recording will place the video into your Rich
Text Editor, allowing you to
share your creation with your students.
Training Asset 1: How to Create Videos Using Kaltura
1- In this guide, we will learn how to access Kaltura to create
webcam video content and place it
within the Sakai Classroom.
2- First, access Kaltura from within the rich text editor by
selecting the “Kaltura” icon.
3- Select the “Upload” button.
4- Select “Webcam”, as this will record your personal webcam
input to place into your Sakai
classroom.

5- When you select “Webcam”, you will then need to select
“Allow” in the Adobe Flash Player
Settings to allow video to be captured.
6- To begin recording, select the “Record” button to begin
recording your video and audio input,
and then select “Stop” to cease the recording. When you are
satisfied with your input, select
“Next” to continue.
7- To finalize your video recording, enter the “Title”, “Tags”,
and a “Description”, then select
“Next”.
8- When completed, you will see the “Done!” screen, where you
can choose to “Add More Media”,
or, “Finish”. Selecting “Finish” will bring you back to your
Media Gallery. Selecting the
recording will place the video into your Rich Text Editor,
allowing you to share your creation
with your students.

What drives productivity in Tanzanian manufacturing firms:
technology or business environment?
Micheline Goedhuys
a
*, Norbert Janz
a,b
and Pierre Mohnen
a,c
a
UNU-MERIT, Maastricht, the Netherlands;
b
Aachen University of Applied Sciences, Aachen,
Germany;
c
University of Maastricht, Maastricht, the Netherlands
Using cross-sectional firm-level data, this paper examines the
determinants of productivity
among manufacturing firms in Tanzania. In particular, it seeks
to evaluate the relative
importance of technological advances and the business
environment in which firms operate in
affecting productivity. Of the technological variables, R&D as
well as product and process
innovation, licensing of technology, and training of employees
fail to have any impact; only

foreign ownership, ISO certification and higher education of the
management appear to affect
productivity. Some important influences from the broader
business environment, however,
appear to affect productivity and are robust to different
specifications of the model. Credit
constraints, administrative regulatory burdens and a lack of
business support services depress
productivity; membership of a business association is associated
with higher productivity.
Cet article examine à l’aide de données en coupe transversale
les facteurs qui déterminent la
productivité dans les firmes manufacturières en Tanzanie. Plus
précisément, nous comparons
l’importance relative des avancées technologiques et du
contexte institutionnel comme facteurs
explicatifs de la productivité. Parmi les variables
technologiques, la recherche-développement, les
innovations de produits et de procédés, les licences de
technologie et la formation des employés
n’ont aucun impact. En revanche, la propriété étrangère, la
certification ISO et la formation avancée
des dirigeants d’entreprise semblent influencer la productivité.
Certains facteurs institutionnels,
quant à eux, ont une influence sur la productivité qui se
manifeste de façon systématique dans
plusieurs modèles. Les contraintes de crédit, la lourdeur
administrative de la réglementation et un
manque de services de support aux entreprises sont associés à
une faible productivité, alors que
l’appartenance à des associations de commerce caractérise les
firmes à forte productivité.
Keywords: productivity; technology; R&D; innovation; business
environment; Tanzania

1. Introduction
Innovation is widely regarded as the key to economic growth in
industrialised countries. Firms
invest in R&D to develop new products and/or new processes.
They acquire existing technology
through licensing contracts, cooperation agreements, mergers
and acquisitions. They train their
workers, invest in new technologies, such as in information and
communication technologies
(ICT), or introduce new ways of operating, like selling and
buying on the Internet. By
introducing new products, implementing new technologies, and
reorganising their way of
operating firms remains competitive; by investing in research,
patenting and licensing they stay
at the cutting edge of technologies (Baumol 2002). The
empirical evidence demonstrating the
positive effect of these innovation activities on firm
performance is overwhelming for
industrialised countries (see, for instance, Kleinknecht and
Mohnen 2002).
ISSN 0957-8811 print/ISSN 1743-9728 online
q 2008 European Association of Development Research and

Training Institutes
DOI: 10.1080/09578810802060785
http://www.informaworld.com
*Corresponding author. Email: [email protected]
The European Journal of Development Research
Vol. 20, No. 2, June 2008, 199–218
Since the publication of influential contributions on technical
change in developing
countries (including, among others, Fransman 1985; Katz 1987;
Lall 1992) a rich literature
developed, conceptualising innovation and technological change
in developing economies. In
developing countries, a majority of firms is operating
substantially below the technological
frontier, with lower levels of human capital and older vintage
machinery. In this context, firms’
technological efforts are primarily oriented towards developing
capabilities to absorb, adapt,
master, and eventually improve technologies developed
elsewhere. Several authors (e.g. Enos
1992; Lall 1992; see also UNCTAD 1996 for an overview),
following the evolutionary theory

of economic change (Nelson and Winter 1982), termed the
technological competences of firms
in developing countries by ‘technological capabilities’,
referring to the information and skills –
technical, managerial and institutional – that allow firms to
utilise equipment and technology
efficiently. In a more dynamic setting, firms build up
competences in a process of technological
learning, by engaging in a wide variety of activities, such as
research, training, technology
licensing, investment in new vintage machinery, aimed at
introducing products and production
processes that are new to the firm and reinforce the firm’s
competitive position.
The performance of firms is also found to be strongly and
directly influenced by the wider
business environment, institutional context and socio-economic
framework in which firms’
activities are embedded. High regulatory burdens, low levels of
educational development, weak
industrial inter-firm linkages and poorly functioning financial
markets, that characterise least
developed economies, are likely to hamper firm performance
(Goedhuys 1999).

The main objective of this paper is to investigate whether it is
the technological activities or
the business environment that influence most productivity of
firms in a least developed country,
Tanzania. Tanzania is an interesting representative country as it
shares many of its structural
characteristics with other least developed countries in the Sub-
Saharan African region. It has
undertaken major reforms since the mid-1980s, aimed at
reducing state control over the
economy and increasing the role of private sector firms to
achieve economic growth. Despite
policy reforms, improvements in the business environment
remain a major policy issue, as an
overwhelming majority of private businesses continue to be
small and operating outside the
formal economy. While Tanzanian firms are generally
constrained by finance and managerial
and technical skills, policy documents stress the need to
improve the business environment
further – such as better infrastructure, market information and
reduction of excessive or
conflicting regulation – to unleash productive potential (URT
2006). All this makes the country

an interesting one for our research question on the relative
importance of technological efforts in
the context of a constraining business environment.
Notwithstanding an existing rich literature on several aspects of
technological learning and
innovation in least developed countries, most of the empirical
evidence is based either on case
studies or on small-scale firm surveys, from a particular
industry, location or industrial cluster.
1
Moreover, the link between technological activity or innovation
on the one hand and firm
performance on the other is rarely analysed. Only a few studies
use larger data sets from least
developed countries to explicitly measure the impact of
technological variables on quantitative
firm performance indicators such as productivity, efficiency or
profitability (e.g. Biggs, Shah and
Srivastava 1995; Bigsten et al. 2000; Sleuwaegen and Goedhuys
2003; Biggs and Shah 2006;
Fernandes 2006). We use firm-level data on Tanzania from the
World Bank Investment Climate
Survey (see World Bank 2004a), which contains data on various
technological indicators, input

and output data allowing productivity to be measured, and
information on the business
environment regarding finance, the labour market, infrastructure
and regulations. Our large data
set covers firms from different industries and locations. We use
econometric estimation and
testing techniques to address the issue of the relative
importance of technology and business
environment for productivity.
200 M. Goedhuys et al.
The paper proceeds as follows. Section 2 reviews the literature
on factors affecting firm
productivity in developing countries. Section 3 presents some
background information on
Tanzania and gives an overview of findings from previous
studies on Tanzania relevant to our
analysis. Section 4 presents the econometric specification and
the data that underlie our analysis.
Section 5 discusses the results and section 6 concludes.
2. Technological capabilities and productivity in developing
countries

Following earlier documents by Fransman (1985), Katz (1987),
Lall (1992), a rich literature
developed studying the characteristics of innovation and
technological change in developing
economies. A majority of firms in these countries operates
substantially below the technological
frontier, with lower levels of human capital and older vintage
machinery. Apart from some giant
developing economies such as India or China, where frontier
research is conducted in selected
industries, it is unlikely that a country that has a paucity of
scientists and engineers and that lacks
the institutions propitious to innovation will organise frontier
type of research (Oyelaran-Oyeyinka
2006). This does not mean, however, that less or least
developed countries (LDC) cannot benefit
from technological change. Innovation has to do with adopting
existing technologies rather than
creating new technologies, i.e. reaching the technological
frontier rather than shifting the frontier.
To raise efficiency or establish a better competitive position,
firms’ efforts are oriented
towards developing capabilities to absorb, adapt and master
technologies often developed

elsewhere in a process of technological learning. Cohen and
Levinthal (1989) developed the
concept of ‘absorptive’ capacity to refer to a firm’s ability to
assimilate existing technology and
to adapt it to their own environment. For developing countries a
nascent literature has started to
investigate the link between technological capabilities,
innovation and productivity. So far
results are mixed. Chudnovsky, Lopez and Pupato (2006), using
panel data from Argentina, find
that R&D and technology acquisition raise the probability of
product and process innovation,
which in turn raises productivity. However, using a similar
methodology,
2
Benavente (2006)
found that in Chile firm productivity is not affected by
innovation or research expenditures.
Fernandes (2006) found for Bangladesh that firms’ TFP
improves with higher levels of human
capital, R&D and quality certification, foreign ownership and
exports. For African countries, the
number of firm-level studies is more limited, mostly oriented
towards the analysis of exports,

investment and growth (see Bigsten and Soderbom 2006 for an
overview), and with strong
emphasis on the role of human capital (see e.g. Bigsten et al.
2000). Goedhuys and Sleuwaegen
(1999) find no significant impact of R&D activity or licensing
on labour productivity for
manufacturing firms in Burundi.
This raises the question of what determines productivity in
African firms and draws attention
to the argument of institutional economics, that firm
performance may also be strongly affected
by the institutional and business environment in which firms
operate (Williamson 1987; North
1991; Coase 1998), and which can be particularly constraining
in least developed countries. The
institutional environment consists of formal rules, including
laws, regulations, and property
rights, or informal rules, such as norms, habits and practices,
social conventions. Jointly they
form the basis of the incentive structure in which firms take
decisions, they affect transaction and
production costs and shift firm performance. In developing
countries, several forms of regulation
on the start-up and scope of business activities and labour

regulation still result in severe market
imperfections and create scope for rent-seeking by civil
servants (Djankov, La Porta, Lopez-
De-Silanes and Shleifer 2002). This is often reinforced by a
deficient contract enforcing system.
As a result, in practice, some groups of entrepreneurs and
businessmen have developed business
attitudes by which problems are solved and business deals made
on the basis of trust, reputation
and networking in the framework of unwritten values and norms
of a more traditional society.
The European Journal of Development Research 201
Banerjee and Duflo (2005) discuss how firm productivity is
determined by incentives.
Excessive government intervention, related to a high degree of
formalism or burdensome legal
procedures, may create barriers to entry or growth and protect
inefficient incumbent firms. Credit
constraints in poorly developed financial markets likewise
result in unequal access to finance,
misallocations of capital and productivity differences. In an
overview article on the determinants

of the size structure and productivity performance of
manufacturing firms across developing
countries, Tybout (2000) also mentions the uncertainty about
government policies and demand
conditions, poor rule of law, and corruption as important factors
hampering the operations of
firms. Using firm-level data from 15 countries, including
several African countries, Eifert, Gelb
and Ramachandran (2005), found that high indirect costs – due
to high transportation and utility
costs, bribes, security etc . . . and business environment-related
losses depress productivity in
African firms.
A large literature examines the influence of business practices
on productivity performance,
i.e. the influence of factors such as family ownership, incentive
structures, monitoring activities,
child-care facilities, employee empowerment, flexible working
hours and many others (see for
example Ichniowski, Shaw and Prenushi 1977; Bloom and van
Reenen 2006). Training is often
classified in these practices. We have considered training as a
technological activity. Most of the
other business practices are not recorded in our dataset.

We shall thus examine the importance of technological variables
in explaining productivity
of manufacturing firms in Tanzania, while at the same time
controlling for several possibly
constraining factors originating from the business environment.
Especially in a developing
country firms perceive the institutional framework differently
and are thus differentially affected
by them.
3. Tanzanian industry and technology
Tanzania is a representative country for a larger number of Sub-
Saharan African countries. Its
economy is heavily based on agriculture, which accounts for
46.1% of GDP in 2005 (World
Bank 2007) while industry accounts for only 16.9% of GDP.
After independence in 1960, a
strong socialist centrally planned economy with large state
participation was installed, which
was very hostile to private business. Economic activity was to
be taking place in state-owned
firms, of which 425 were established by the mid-1980s
(Bagachwa 1993, p. 91), about the largest
concentration in the world.

The poor performance of this development policy led to the
implementation of reforms since
1985 towards a more liberal market-based economy (for details
on the industrial experience and
policy reforms over the last three decades see Bagachwa 1993;
Hewitt and Wield 1997; Szirmai
and Lapperre 2001). In the 1990s, a large-scale privatisation
programme was also implemented
(Temu and Due 2000), reducing state participation in industrial
firms, mainly in favour of
foreign participation. In this process, FDI has increased sharply
since 1992,
3
making Tanzania
one of the top African FDI recipient countries (UNCTAD 2002).
An important share of this
foreign investment was concentrated in manufacturing,
especially in the food and beverages
industry. It was expected that FDI would lead to technological
upgrading and transfer of
technology, skills and superior management techniques. A case
study by Portelli and Narula
(2006) on two privatised firms shows that productivity and
technological upgrading increased

sharply after investment by foreign multinational companies.
At the other end of this spectrum, the industrial private sector is
still characterised by a
majority of small local businesses, many of which remain
outside the formal economy – 98%
of all businesses are informal (URT 2006), only 1662
establishments are registered in
manufacturing (National Bureau of Statistics 2004). They are
characterised by weak inter-firm
linkages and a low level of technological capabilities.
Additionally, there is a strong group of
ethnic minority entrepreneurs of Asian (Indian) origin with a
dominant position in light
manufacturing and import/export trade, benefiting from a strong
ethnic network (see Hewitt and
Wield 1997; Biggs and Shah 2006).
This industrial structure and performance is also the historical
result of a broader policy that
did not support the development of private local firms.
Domestic research capability was built in

public research centres, doing research in priority areas
determined by the Tanzania Commission
for Science and Technology. The choice of sectors and research
areas was supply-driven, rather
than based on an analysis of technological needs and problems
of productive private enterprises.
Some state-owned technology-support institutions were
established, but they were hardly aware
of private sector needs and resources and lacked the motivation
to carry out their mandate
successfully (Bongenaar and Szirmai 2001; Utz, 2006). The
linkages between industry,
university and research institutions are weak, as described in
Bangens (2004) and Mwamila and
Katalambula (2004).
Low levels of human capital also hamper technological
upgrading. Although past education
policies made considerable achievements in basic education and
literacy, the educational and
training systems had been insufficiently oriented towards
science and engineering that would
generate managerial and technical skills. This also resulted in
low technology adoption and slow
technological learning from imported technologies, as Wangwe

(1992) demonstrated using four
sector case studies. In addition, the Tanzanian labour force
currently struggles with health
problems,
4
due to high incidence of HIV/aids and other tropical diseases
that lead to high
absenteeism, not only by people infected, but also their
relatives, and reduces the return on
human capital investment.
While insights from the literature stress the importance of local
inter-firm networks and
clusters as a mechanism for technological learning (Bell and
Albu 1999; McCormick 1999) the
evidence of successful network or industrial linkages is patchy
for Tanzania. Comparing the
response of small and medium-sized enterprises (SMEs) to
Structural Adjustment in Tanzania
and Ghana, Dawson (1993) concluded that the stronger
performance of small businesses in
Ghana could be explained mainly by the fact that these firms
had access to modern and
sophisticated technology and to human resources, compared to
little technological enhancement

and few linkages in Tanzania. Murphy (2002) uses data from
manufacturing firms active in the
Tanzanian region of Mwanza and finds that more advanced
social networks are important for
innovation. He also shows that trust in these relations is an
important mechanism to improve
the quality of information exchange and collective knowledge
creation. However, only about
one-quarter of the sampled businesses appeared to be inserted in
a wider social network. Most
entrepreneurs in his sample were socially isolated and the few
relationships they had were
centred on access to capital and short-term market competitive
advantages.
There are also indications that local firms in Tanzania do not
benefit fully from the spillovers
emanating from foreign firms’ backward linkages. Portelli and
Narula (2006) found that
backward linkages with foreign firms based in Tanzania were
primarily related to the sourcing
of medium technology inputs, whereas vertical linkages with
indigenous firms concerned the
sourcing of simple manufacturing inputs, limiting the scope for
technological upgrading.

Similarly, Goedhuys (2007) found that foreign innovative firms
had stronger vertical linkages
with other foreign firms. There was no evidence that backward
linkages of foreign to domestic
firms were strong enough to lead to product innovation in the
local firms.
Hewitt and Wield (1997) and Hewitt, Wangwe and Wield (2002)
have studied the existence
or the lack of formal networks and industrial linkages. They
describe how in recent years more
actors and agents have started to take action in the coordination
of industrial development. Not
only the state, but also industrial associations, the Tanzanian
Chamber of Commerce, Industry
and Agriculture, and the Confederation of Tanzanian Industries
are playing an increasingly
influential role. Supported by the donor community they embark
on negotiations with policy makers
over private sector development issues – such as taxation,
regulation, credit, lack of technical and
managerial education, lack of services to business, and deficient

infrastructure provision. This
dialogue resulted in the ‘Business Environment Strengthening
for Tanzania’ (BEST) Programme
which aims at reducing the cost of doing business by removing
regulatory and administrative
barriers to formal businesses, improving the quality and speed
of government services including
dispute settlement procedures, and empowering private sector
advocacy (URT 2006).
In the presence of weak industrial linkages, low levels of human
capital to absorb external
information, and a changing business environment, the question
arises whether technological
efforts could eventually result in superior individual firm
performance. In what follows we shall
investigate the productivity performance of formal enterprises
in 2002 and explore the relative
influence of their technological efforts and of the constraints
originating from the business
environment in which they operate.
4. Empirical Approach
4.1 Empirical model
To analyse the effects of technological variables and

institutional constraints on firm-level
productivity we use the production function approach. Firms’
value added Yi is a function of the
traditional factors of production, physical capital Ki and labour
Li, as well as other factors
explaining differences in productivity, i.e. technological
variables Z1,i and firm-level constraints
originating from the business environment Z2,i. We assume that
they affect only total factor
productivity, but not the marginal productivity of capital and
labour. Within a Cobb-Douglas
framework allowing for non-constant returns to scale we get the
following specification
Yi ¼ AðZ1;i; Z2;iÞK
a
i L
b
i e
1i ð1Þ
in which a and b denote marginal productivities of physical
capital and labour, respectively.
Constant returns to scale occur if a þ b ¼ 1, which will be tested
empirically. A(Z1,i, Z2,i)
characterises differences in total factor productivity (TFP)

depending on technological variables
and business environment constraints. The stochastic term 1i
summarises other unobservable
factors affecting firms’ output.
As a starting point for our empirical analysis, we get after
taking logarithms
ln Yi ¼ ln AðZ1;i; Z2;iÞ þ a ln Ki þ b ln Li þ 1i ð2Þ
This equation can be rewritten in terms of labour productivity in
the following way:
lnðYi=LiÞ ¼ ln AðZ1;i; Z2;iÞ þ a lnðKi=LiÞ þ ða þ b 2 1Þln Li
þ 1i ð3Þ
The stochastic error term 1i is assumed to be independently and
identically normally distributed.
We further assume that TFP is a linear function of technological
and business constraints
variables. The coefficient of ln Li measures the deviation from
constant returns to scale.
Part of TFP can be attributed to capacity utilisation. When firms
operate at higher capacity,
they can produce more with the same amount of inputs. We
therefore introduce variable ui
measuring the utilisation of actual capacity:
lnðYi=LiÞ ¼ ln AðZ1;i; Z2;iÞ þ a lnðKi=LiÞ þ ða þ b 2 1Þln Li
þ gui þ 1i ð4Þ

We expect parameter g to be positive, i.e. firms are able to
increase labour productivity by using
production capacities more intensively.
To estimate this equation two different estimation techniques
are applied: Ordinary Least
Squares (OLS) regression and quantile regression. If we
summarise the explanatory variables,
including a constant term, to a row vector Xi, the OLS estimator
results from minimising the sum
of squared residuals, i.e. from minimising the criterion function
XN
i ¼1
ðlnðYi=LiÞ 2 XibÞ
2
ð5Þ
where b is the column vector of parameters. Thus, OLS is in
fact estimating the mean effects of
explanatory variables Xi on log value added per employee.
Heterogeneity in firms’

characteristics and abilities that are not reflected in variables Xi
are assumed to be random
and to vanish in the mean. They are not allowed to have an
effect on parameters to be estimated.
Possible differences across firms are thus ruled out.
But, at different levels of productivity firms may face different
conditions and have to cope
with different problems. Technological activities may be
organised differently in high and low
productive firms. High productive firms are likely to have their
own R&D department whereas
low productive firms would rather acquire technology by
licensing. Institutional conditions, such
as rationing on the credit market and overregulation may be a
more severe problem for low than
for high productive firms. Returns to scale may be higher for
high productive firms.
Therefore, in addition to OLS we apply quantile regression
methods (see Koenker and
Bassett 1978; Buchinsky 1998; Koenker and Hallock 2001) to
shed some light on the
heterogeneity of firms and on the technological conditions
creating it.
5

Instead of minimising
the sum of squared residuals, quantile regression coefficients
result from minimising the
criterion function
XN
i¼1
rjlnðYi=LiÞ 2 XibjIðlnðYi=LiÞ . XibÞ þ
XN
i¼1
ð1 2 rÞjlnðYi=LiÞ 2 XibjIðlnðYi=LiÞ # XibÞ ð6Þ
where I(·) is an indicator function taking the value of 1 if the
condition in brackets is met and 0
otherwise, i.e.:
IðlnðYi=LiÞ . XibÞ ¼ 1 if lnðYi=LiÞ . Xib and IðlnðYi=LiÞ .
XibÞ ¼ 0 if lnðYi=LiÞ # Xib:
So, the left term is a weighted sum of all positive residuals, i.e.
the high productive firms, while
the right term is the weighted sum of all negative residuals, i.e.
the low productive firms.
The symbol r is a weighting factor ranging from 0 to 1. In the
special case where r ¼ 0.5,
both terms are equally weighted and minimising the criterion

function leads to the 50% quantile.
This constitutes the well known Least Absolute Deviation
(LAD) or Least Absolute Values
(LAV) estimator. In this case, the procedure will result in the
estimation of median effects in
contrast to the mean effects of the OLS estimator. It is well
known that this LAD estimator is
robust, i.e. less affected by outliers than other estimators like
the OLS estimator. If a few firms,
e.g. foreign-owned firms, behave different from the majority of
local firms, this will influence
the mean results of the OLS estimator but not the median results
of the LAD estimator. In this
case, the median would be a more adequate measure of location
than the mean.
If r ¼ 0.75, the positive residuals in the left term have a higher
weight than the negative
residuals in the right term of the expression. Minimising the
criterion function will then lead to
estimated coefficients whereby 75% of the residuals are
negative. By definition, this is the 75%

quantile, i.e. the upper quartile. The results of the estimation
will show the effect of the
explanatory variables on productivity for the highly productive
firms.
Less productive firms can be examined setting r ¼ 0.25. The
negative residuals in the right
term have higher weight than the positive ones. Minimising the
criterion function will lead to
estimated coefficients where 75% of the residuals are positive,
i.e. the distribution is evaluated at
the 25% quantile, the lower quartile. The lower quartile
represents the less productive firms.
4.2 Data source and construction of variables
Micro-data are needed to analyse differences in firm-level
productivity within a country. While
firm-level data sets are well established for most of the OECD
countries, corresponding data of
good quality were hardly available in the past for most
developing and especially for least
developed countries like Tanzania. Considerable advances have
been made by the World Bank
with the ‘Investment Climate Surveys’ (ICS).
6
They offer harmonised cross-sectional data on

the investment climate, i.e. conditions affecting firm production
and investment behaviour, in
developing countries.
7
In general, firm-level panel data would be the optimal data
source, since
problems of endogeneity resulting from explanatory variables
that are possibly affected by
productivity, could be tackled by using appropriate time-lag
structures. Unfortunately, no panel
data sets are available for most Sub-Saharan African countries
including Tanzania. The
Tanzanian ICS is therefore an interesting alternative source of
recent data, despite its limitations
to interpret causality of relationships in the results.
The Tanzanian ICS, organised and coordinated by the World
Bank, was executed in 2003 by
the ‘Economic and Social Research Foundation’, in
collaboration with the National Bureau of
Statistics. The Tanzanian ICS is a rich data set gathering plant-
level information on the business
environment in which businesses operate, in order to understand
how technological conditions

and business environment constraints affect the operations and
performance of firms, especially
firm-level investment, growth and productivity. The survey
questionnaire includes a series of
questions on firms’ behaviour and their position on financial,
labour and sales markets
accompanied by information on infrastructure, regulation,
international trade, innovation and
learning as perceived by the firm. To benchmark firms’
performance, another set of variables is
included such as sales and material purchases, which can be
used to calculate value added.
The sample in Tanzania includes 275 plants in the
manufacturing sector. These are randomly
selected from a sampling frame constructed from different
official sources and stratified by
branch of industry, size and location.
8
Plants are selected from 11 different locations
representing the major centres of industrial activity in Tanzania:
Dar es Salaam, Arusha,
Morogoro, Mwanza, Kilimanjaro, Tanga, Kagera, Iringa,
Mbeya, Mara on the mainland, and the
island of Zanzibar. The manufacturing sector is divided into

eight industries: food and
beverages, chemicals and paints, construction materials, metal
working, wood working and
furniture, paper and printing/publishing, plastics as well as
textiles, garments and leather
products. With respect to size, the sample is representative for
the formally registered firms. The
median size of the plants in the sample is 30 employees. The
mean size is 125 employees,
showing a highly skewed size distribution with a few very large
firms and a majority of small
firms. The very small firms, with less than 10 employees, and
the informal firms, which are not
registered with any government agency and tend to be small, are
underrepresented in the sample
(World Bank 2004a).
Due to item non-response on variables crucial for the analysis, a
number of observations had
to be excluded from the data set, reducing the number to 187.
9
The distribution of the sample
used for the econometric analysis with respect to sectors and
size classes is shown in Table 1.

The table also presents the number of firms with some share of
foreign ownership. A total of 35
firms are in this category. Foreign ownership is a minority share
in seven firms, a majority share
in 18 firms while ten firms are fully foreign-owned.
The dependent variable is LABOUR PRODUCTIVITY,
measured by the value added per
employee in logarithms. Value added was calculated from the
data as the value of total sales
minus material purchases and fuel and electricity costs. All
values are for the year 2002 and in
logarithmic terms. Value added has two components: prices and
quantity. Thus, not only
efficiency, but also conditions affecting firms’ ability to charge
higher prices, like market power,
result in higher value added. Information on prices is not
contained in ICS datasets.
Labour productivity is a function of the CAPITAL/LABOUR
ratio (in logarithm) and a function
of LABOUR (in logarithm) if there are non-constant returns to
scale. The variable CAPITAL

represents the firm’s capital stock by end of the year 2002,
constructed by the replacement value of
machinery and equipment, plus the net book value of land and
buildings. For a number of firms
replacement values were not available. In these cases,
information on net and gross book value of
machinery and equipment was used to estimate the capital.
Technical details on the construction of
the variable capital are presented in the appendix. Since we
have only cross-sectional data, the
capital stock is measured in nominal terms and therefore part of
its value could be due to higher
mark-ups on the capital goods market without necessarily better
quality equipment. Labour input is
measured by the log value of the total number of employees in
2002, being the sum of permanent
workers and the average number of temporary workers
employed in 2002.
As explained in sections 2 and 3, two additional sets of
variables were constructed. One set
represents information on firms’ technological activity or
sourcing, i.e. ways firms choose to
build up firm-specific skills and increase their knowledge base.
Another set of variables is

referring to the business environment the firm is operating in,
since perception and degree of
being affected may differ from firm to firm even within one
country.
Firms can source technology from abroad through established
ownership linkages that
stimulate transfer of production or organisational capabilities.
This indeed motivated the large
privatisation programme, which resulted in increased foreign
ownership in key industries.
A dummy variable FOREIGN, indicating whether the firm has a
positive share of foreign
ownership, captures the potential effect of foreign ownership
linkages on productivity.
10
Moreover,
firms can directly make use of external technology through
licensing from other firms. The dummy
variable LICENSE marks whether technology has been licensed
from a foreign company.
Firms can also build up a stock of technological knowledge
through a knowledge
accumulation process. From the set of questions related to the
firms’ learning and innovation

Table 1. Composition of sample in terms of sector, foreign
ownership, by size class.
Size class (number of employees)
Sector of activity 1 – 9 10 – 29 30 – 99 100þ Total
Agro-industries 7 16 12 22 57
Chemicals and paints 1 3 6 8 18
Construction materials 1 1 4 2 8
Metal working 0 12 4 4 20
Furniture, wood working 11 22 7 3 43
Paper, printing, publishing 2 9 5 3 19
Plastics 0 0 0 4 4
Textiles, garments, leather products 3 4 6 5 18
Foreign-owned firms 0 5 13 17 35
Total 25 67 44 51 187
activities, variables were constructed to measure the fact of
conducting research and
development (R&D), the intensity of doing it and the incidence
of product and process
innovation. RD is a dummy variable indicating whether a firm
conducts its own R&D. LRDEXP,
the log of the firm’s R&D expenditure, measures the extent of
R&D activities.
11

The dummy
variable PRODUCT indicates whether the firm has introduced a
product that is new to the firm,
while the dummy variable PROCESS points out whether a firm
has implemented a new
production process that substantially changed the way the main
product is produced.
The ability of firms to make use of external technologies and to
efficiently convert research
results in marketable products depends on their absorptive
capacity, especially the educational
level of the labour force and the top manager. This is captured
by the variables AVYEDUC,
measuring the average years of education of the work force, and
EDUCGM, a dummy variable
for managers with higher education. Increasing the educational
level of the labour force through
training, either on the job or through formal training, is
generally regarded to be an important
aspect of competence building. The dummy variable TRAINING
equals one for firms offering
formal training to their employees. TRAININT measures
training intensity by the proportion of

employees that received formal training.
While the use of new information and communication
technologies is fully recognised as an
important instrument in the search for information and
knowledge, with access and use of the
Internet as a major indicator, ICT is still less widespread in
Africa as compared to other developing
regions. Though access to the Internet has increased
substantially in urban areas in Africa, and in
Tanzania in particular, it is still limited to a subset of
businesses (World Economic Forum 2004).
In our data set, INTERNET, a dummy variable measuring
Internet access of firms, captures
this advantage. The technological and organisational level of
firms in developed countries is
sometimes accompanied by certification, such as the well-
known ISO certification. For firms in our
sample this is shown by the dummy variable ISO.
Unfortunately, additional information on
industrial linkages and the quality or intensity of knowledge
flows was not available from the
questionnaire. Table 2 gives an overview of all variables
considered and how they are defined.
A second set of variables deals with the business environment

firms operate in. As in many least
developed countries, many firms – especially small domestic
ones – are financially constrained and
have to rely heavily on trade credit or other forms of informal
credit to finance business operations.
Only a minority of firms has access to more formal forms of
flexible credit. The variable CREDIT
captures the benefit of having access to formal credit, as
reported by the firms.
With respect to firms’ relations with the government, two
related concerns are mainly
reported: firms complain about red tape and high taxes,
combined with poor business
infrastructure and support services (e.g. World Economic Forum
2004; World Bank 2004a).
The extent to which regulation – i.e. the administrative burden
associated with custom and
trade regulation, and bureaucratic business licensing procedures
– is hampering firms’
operations is captured by a dummy variable REGULATION.
The extent to which deficient
business support services hampers operations is taken into
account by another dummy variable
LACKSUPPORT.

Given the increasingly important role played by the industry
association (Hewitt et al. 2002),
both for policy lobbying and as unique formal industrial
network for knowledge and information
exchange, we include a variable BUSASSOC capturing
membership to a business association as
an explanatory variable. Like foreign ownership and education
of management, membership of
business associations might lead to higher value added because
of both higher efficiency and the
ability to charge higher prices for its products and lower prices
on the input markets.
This list is completed by a variable referring to the health
systems. With high HIV/AIDS
infection rates and the high burden of other diseases, including
malaria, absenteeism among the
workers may be depressing firms’ productivity levels. This
effect is measured by the variable
DAYSLOST, the average number of working days lost per
employee due to health-related
problems.

Summary statistics of the variables are presented in Table 3.
Some of the technological
variables have low values. Only 10 – 20% of the firms have a
quality certificate, conduct R&D or
have technology licensed from a foreign company. More
common is the introduction of new-to-
the-firm products (61%) and processes (29%) and the use of
Internet (47%). A majority (66%) of
managers have higher education and firms engage in the
training of their workers (43%).
12
5. Results
The regression results are summarised in Tables 4 and 5. Table
4 reports OLS results for three
different specifications: a simple labour productivity equation
without technological and
institutional variables, the extended model including all
technological and institutional variables
Table 2. Construction and definition of variables.
Dependent variable:
VA/L Value added per employee (in log.)
Total value added is sales minus material purchases, fuel and
electricity expenses

Traditional explanatory variables:
LABOUR (L) Total number of employees, including temporary
workers (in log.)
K/L Capital per employee (in log.)
Capital stock includes machinery, equipment, vehicles, land and
buildings
CAPACITY
UTILISATION
(Actual output produced)/(maximum output that could be
produced with existing
machinery and equipment and regular shifts) [value between 0
and 1]
Technology variables:
FOREIGN Dummy variable equal to 1 if firm has some foreign
ownership
ISO Dummy variable equal to 1 if firm has ISO certification
RD Dummy variable equal to 1 for firms investing in R&D or
design
LRDEXP Expenditures on R&D and design (in log.)
PRODUCT Dummy variable equal to 1 for firms having
developed a major new product line or
upgraded an existing product line in the last three years (2000 –
2002)
PROCESS Dummy variable equal to 1 for firms having
introduced new technology that has
substantially changed the way the main product is produced
LICENCE Dummy variable equal to 1 for firms using
technology licensed from a foreign-owned
company
INTERNET Dummy variable equal to 1 for firms having

internet access
EDUCGM Dummy variable equal to 1 if general manager of the
firm has a graduate or
postgraduate degree or diploma of tertiary college
TRAINING Dummy variable equal to 1 for firms offering
formal training to their employees
TRAININT Training intensity measured as the proportion of
total permanent employees having
received formal training in 2002
AVYEDUC Skills level of the work force, measured as average
number of years of education of the
permanent employees
Business environment variables:
BUSASSOC Dummy variable equal to 1 for firms being member
of a business association
CREDIT Dummy variable equal to 1 for firms reporting not to
be credit constrained
DAYSLOST The number of working days per employee, lost
due to HIV and other diseases
REGULATION Dummy variable equal to 1 if the firm reports
‘Customs and Trade Regulation’ and
‘Business licensing and operating permits’ severely hampering
the operations and
growth of the firm
LACKSUPPORT Dummy variable equal to 1 if the firm reports
lack of business support services as
severely hampering the operations and growth of the firm

listed in Table 3, and a reduced model where only variables
proving to be statistically significant
are included.
13
Starting with the simple specification, we find an elasticity of
output with respect to capital of
0.356 and a scale elasticity of 1.154, significantly different
from one. Thus, increasing returns to scale
cannot be rejected. Productivity also increases with capacity
utilisation. Once we control for
technological and other productivity determinants, the capital
elasticity of output drops to 0.261, and
constant returns to scale can no longer be rejected. Increasing
returns to scale in the simple production
function framework can be attributed to differences in
technology and perceived business
environment. Firms operating on a larger scale are for instance
more technology-intensive and more or
less affected by the business environment. Capacity utilisation
remains significant, but with a slightly
lower coefficient. In the reduced specification the labour and
capital elasticities of output drop even
further, resulting in decreasing returns to scale. The hypothesis

of constant returns to scale has to be
rejected at the 5% level. On the basis of the adjusted R-square,
this specification would be preferred.
Given the cross-sectional nature of our data set, causality
relationships have to be interpreted
carefully. Keeping this in mind, foreign-owned firms have
significantly higher productivity than
local firms. Our results support earlier findings from Portelli
and Narula (2006) on two case
study firms. According to Portelli and Narula (2006),
productivity was raised substantially
following South African and American investment. Also Hewitt
and Wield (1997) mentioned
Asian businesses in Tanzanian industry to have ‘access to
sources of technology, which are not
so easily available to other Tanzanian industrialists’. Biggs and
Shah (2006) equally provide
Table 3. Descriptive statistics on relevant variables.
Mean Standard deviation Lower quartile Median Upper quartile
Dependent variable:
VA/L 14.809 1.484 13.889 14.754 15.714
Traditional variables:
LABOUR 3.618 1.412 2.485 3.401 4.605
K/L 15.826 2.032 14.720 16.030 17.237

CAPACITY UTILISATION 0.587 0.222 0.470 0.600 0.750
FOREIGN 0.187
ISO 0.112
RD 0.187
LRDEXP 14.808 2.099 12.794 15.177 16.118
PRODUCT 0.610
PROCESS 0.289
LICENCE 0.171
INTERNET 0.471
EDUCGM 0.663
TRAINING 0.428
TRAININT 0.120 0.214 0.000 0.024 0.146
AVYEDUC 8.205 2.400 6.800 8.150 10.000
Institutional variables:
BUSASSOC 0.412
CREDIT 0.198
DAYSLOST 0.577 1.223 0.000 0.045 0.727
REGULATION 0.171
LACKSUPPORT 0.086
Note: Number of observations: 187; for binary variables, only
the mean is given. For variable LRDEXP: Values refer to
the sub-sample of 35 R&D performing firms (where RD ¼ 1).
For variable TRAININT: Values refer to the sub-sample
of 80 firms that actually report to offer formal training (where
TRAINING ¼ 1).
evidence that Asian ethnic minority firms have superior
performance and benefit from various

advantages of being in the network, including access to supplier
credit. In a similar way, the
quality of management as reflected in the top manager’s formal
education, and the firms’
technological competence as revealed through ISO certifications
are robust drivers of firm
productivity. An ISO certification opens the access to
international markets, it can act as a signal
of quality, and allows firms to charge higher prices. The
managers with higher education
strongly outperform the minority of managers without formal
schooling beyond secondary level.
However, the most important result from this estimation is that
many of the technology
variables that at least in developed economies are usually found
to be strong productivity
determinants do not have any significant coefficient in the
productivity equation in Tanzania.
Licensing technology from foreign companies (LICENSE) is not
significantly correlated to
higher levels of production.
14
This implies that access to foreign technology mainly runs
through

foreign ownership linkages. In contrast to most findings in the
literature, the skills level and
training activities of the labour force (AVYEDUC, TRAINING,
TRAININT) do not produce
any measurable effect on productivity. Measuring the impact of
the skills level on firm
performance is nevertheless a difficult issue. Sutz (2006, p. 8)
identifies problems related to the
use of indicators based on the qualification of personnel, and
uses the example of ‘proportion of
professionals to total work force’. She explains that for larger
firms, due to a large denominator,
Table 4. Results of OLS regressions.
Dependent variable
OLS regressions
VA/L Simple model Extended model Reduced model
LABOUR 0.154** 20.151 20.184**
K/L 0.356*** 0.261*** 0.246***
CAPACITY UTILISATION 1.514*** 1.413*** 1.353***
FOREIGN 0.448* 0.441**
ISO 0.792*** 0.706**
RD 0.481
LRDEXP 0.045

PRODUCT 20.087
PROCESS 20.124
LICENCE 20.032
INTERNET 20.121
EDUCGM 0.791*** 0.743***
TRAINING 0.031
TRAININT 20.160
AVYEDUC 20.024
BUSASSOC 0.578*** 0.485**
CREDIT 0.552** 0.534**
DAYSLOST 20.134* 20.136**
REGULATION 20.340 20.388*
LACKSUPPORT 20.462 20.524*
Adjusted R-squared 0.320 0.448 0.467
Numbers of observations 187 187 187
Note: Significant at 1% (***), 5% (**) and 10% (*) levels. All
regressions include a constant term and 4 industry
dummies.
the indicator values are depressed; yet the absorptive capacity
can be equally great as in a
smaller firm when a core number of professionals are active in
the firm. The same measurement
problem may relate to our variable AVYEDUC. Sutz further
proposes the indicator – firms

without a single university trainee – that measures
unambiguously difficulties with absorptive
capacity. This effect may indeed be taken up by our variable
EDUCGM, in a sense that firms
managed by persons without higher education are also likely to
lack highly educated engineers
and professionals, and indeed have lower productivity.
Alternatively, it can be that most relevant
skills are learned on-the-job, weakening the real impact of
formal years of schooling and formal
training. Regarding training, our measure does not say anything
about the quality of training, or
of an eventual stock of competences of the work force. In the
same way, the traditional variables
related to R&D and product and process innovation do not
produce any measurable effect on
productivity in Tanzania. Since both variables RD and LRDEXP
are insignificant, knowledge
accumulation through R&D does not improve production
conditions, at least not in the short run.
Even innovations successfully introduced to the market
(PRODUCT) or successfully
implemented in the firm (PROCESS) do not raise productivity.
On the contrary, the variables related to the business

environment capture a fairly large portion
of the variance of value added. This result is reinforced when
the number of explanatory variables
is reduced to those that are significant. First of all, firms that
are members of a business association
(BUSASSOC) have significantly higher productivity. Being a
member of this network is indeed
important for Tanzanian firms. Various reasons could be
invoked to explain the benefit of this
networking effect leading to both higher efficiency and higher
prices: access to information,
increased bargaining power with government and foreign
competitors, exploitation of synergies
(see Hewitt et al. 2002, for a case of private sector influence on
government’s decision to reduce
taxes, through the lobbying of business associations with the
help of independent consultant
institutions).
15
Similarly, firms that have access to external financial funds
(CREDIT) have higher
productivity. This indicates that some projects that would
improve firms’ production technology
Table 5. Results of quantile regressions.

Dependent variable OLS
Quantile regression
VA/L Mean Lower quartile Median Upper quartile
LABOUR 20.184** 20.090 20.165 20.237***
K/L 0.246*** 0.221** 0.288*** 0.301***
CAPACITY UTILISATION 1.352*** 1.422** 1.190** 1.173**
FOREIGN 0.441** 0.293 0.176 0.712*
ISO 0.706** 0.548 1.130*** 0.915***
EDUCGM 0.743*** 0.781** 0.441 0.529
BUSASSOC 0.485** 0.437* 0.522** 0.586**
CREDIT 0.534** 0.521* 0.460* 0.419
DAYSLOST 20.136** 20.302 20.639 20.060
REGULATION 20.388* 20.339 20.124 20.545*
LACKSUPPORT 20.524* 20.434 20.189 20.240
Adjusted R-squared 0.467
Pseudo R-squared 0.269 0.268 0.295
Numbers of observations 187 187 187 187
Note: Significant at 1% (***), 5% (**) and 10% (*) levels.
Regressions include a constant and 4 industry dummies.
are not implemented due to lack of financial resources. But,
alternatively, high productivity firms

could have easier access to the credit market. Thus, problems of
endogenous regressors occur
which cannot be treated straightforwardly with cross-section
data. Overregulation of firms
(REGULATION) and deficient business support services
(LACKSUPPORT) likewise decrease
firms’ productivity, at least at the 10% level of significance.
The same holds for a malfunctioning
health system measured by the number of days lost due to health
problems (DAYSLOST).
Thus, to explain productivity differences in Tanzanian firms
only a limited number of
technology variables turn out to be significant. Some of the
more traditional measures of know-how
and innovation – research and development, product and process
innovation, technology licensing,
skills and training – do not produce any measurable impact on
the productivity of the firm, in
contrast to what could be expected from the mainstream
literature often based on case studies.
In addition, institutional aspects explain a large part of the
variation in firms’ value added, giving
weight to the claims made by the private sector to improve the
business environment further.

These results are valid for the average firm and this picture
seems to be quite homogeneous.
But looking at the quantile regression adds to the information
given by OLS. Results for the 50%
quantile, i.e. the median (LAD or LAV estimator), should more
or less coincide with OLS results
if the conditional distribution of the log value added was nearly
symmetric. But in fact they do
not, since TFP as a rule is skewed to the right. A few highly
productive firms face a majority of
low productivity firms. In this case, higher productivity firms
have a larger influence on the
results of the OLS estimation, implying the risk that factors
affecting these higher productivity
firms are overvalued. Moreover, standard errors are usually
higher in quantile regression since
they have to be bootstrapped (Efron 1981). With respect to
significance, the results of the
quantile regression are thus more conservative.
In contrast to the average firm, the median firm faces constant
returns to scale. Education of
management (EDUCGM) is not a key factor in explaining
productivity, nor is foreign ownership

(FOREIGN) important. Thus, the technological variables reduce
to ISO certification. For the
median firm, being a member of a business association
(BUSASSOC) and having access to
credit (CREDIT) is relevant.
But an interesting picture emerges by comparing the difference
in results for low productivity
and high productivity firms reflected by the results of quantile
regression for the lower and the
upper quartile. Productivity in low productivity firms is mainly
driven by the educational level of
management (EDUCGM). Also access to finance (CREDIT) is a
key factor to increase
productivity in low productivity firms, since a lack of credit is
preventing them from modernising
and installing more advanced technology. Other institutional
aspects like governmental aspects
(REGULATION, LACKSUPPORT) and the health system
(DAYSLOST), do not show up as
significant since these firms have to fight more basic
deficiencies. ISO certification is not an
option at this level of productivity.
High productivity firms do face other problems. The
management is in general well educated

and these firms have access to external finance. For them, ISO
certification is a strategic way of
increasing productivity; foreign ownership linkages (FOREIGN)
offer access to markets and
technology, which they are able to utilise efficiently.
REGULATION is a stumbling block for
higher productivity firms. The only thing which seems to be
beneficial for all firms is to join a
business association (BUSASSOC).
6. Conclusions
This study uses the World Bank Investment Climate Survey
(ICS) data to investigate the relevance
of technological activities and features of the business
environment in explaining productivity
differences among manufacturing firms in Tanzania. ICS
datasets provide rich information on firm
behaviour and the business environment firms are operating in.
Due to the cross-sectional nature of
the data and lack of price information, econometric results have
to be interpreted carefully.

A direct effect of technology and education can hardly be
ascertained. R&D and other
innovation activities – technology variables that are generally
regarded as important in
explaining productivity at least in developed economies – are
not the main drivers of
productivity in Tanzania. Similar results apply for technology
sourcing through licensing from
foreign countries. Even the skills level of the workforce and
activities to improve it do not result
in higher productivity.
Only indirect technological influences that may reflect higher
output and management
quality, more than product innovations per se, show up as
significant determinants of
productivity. Foreign ownership, ISO certification and the
educational level of the general
manager boost productivity in Tanzanian manufacturing firms.
These attempts to signal quality
are accompanied by networking through business associations
that serve as a surrogate for
malfunctioning institutions to reach higher levels of
productivity.
But not all shortcomings of the business environment can be

absorbed by these business
associations. Over-regulation as well as a lack of government
support stand in the way of
efficient production. A deficient health system reduces the
availability of the workforce and
leads to production downtimes. An insufficient financial system
leading to financial constraints
impedes possible expansion of production facilities.
This comprehensive picture of the Tanzanian manufacturing
industry needs to be
differentiated somewhat. Depending on the level of
productivity, firms have different types of
trouble. Low productivity firms face basic needs like a well-
educated management and
appropriate access to financial resources, whereas high
productivity firms are hardest hit by
rudimentary institutions and governmental malfunctioning.
Our econometric study based on a cross-section of firm data
confirms some of the findings of
previous studies for Tanzania and are in line with the observed
industrial characteristics. The
Tanzanian economic structure is characterised by larger foreign-
owned firms and firms belonging

to entrepreneurs of Asian ethnicity, along with a mass of
smaller domestic businesses. Previous
studies have shown that the technology gap is large between
foreign and domestic firms and that
inter-firm linkages for technological upgrading are weak. The
majority of local firms’ linkages are
about reputation and financial issues. On the human capital
side, previous studies identify the lack
of managerial and technical training as constraints to
technological upgrading process.
This study also shows the usefulness of the Investment Climate
Survey data to study
innovation in developing countries, because it sheds light on
many different aspects of relevance
to the innovation system that are not provided in the usual
innovation surveys.
Acknowledgements
The authors would like to thank Samuel Wangwe, David Wield,
Michael Kahn and the participants of the
Globelics Conference 2006, Trivandrum, India and the
participants of the Blue Sky Conference, Ottawa,
2006 for their comments and suggestions. In particular, the
authors wish to thank two anonymous referees
for their constructive comments.
Notes

1. In recent years a rich and insightful literature developed on
the role of industrial clustering for
knowledge flows and firm learning and provides empirical
evidence on this issue (e.g., the work of
Bell and Albu 1999; McCormick 1999; Schmitz and Nadvi
1999).
2. A particular methodology introduced by Crépon, Duguet and
Mairesse (1998) is being applied to a
rising number of developing countries, mainly from Latin
America and Eastern Europe.
3. In 1992 FDI was US$12 million, but it rose to US$193
million in 2002.
4. The life expectancy at birth reduced from 55 in 1985 to 46 in
2005 and 55% of female adults (age
15þ) are HIV positive (World Bank 2006). The impact of the
HIV pandemic on economic life is
devastating.
5. Quantile regressions have been successfully used to analyse a
slightly related problem by Mello and
Perrelli (2003). They examine cross-national differences in
growth and apply quantile regression
techniques to pooled cross-national data, while we use cross-
sectional firm data within a country.
6. These Investment Climate Surveys (ICS) form the basis for
the 2005 World Development Report (see
World Bank 2004b).

7. Firm-level panel data would be an optimal data source since
problems of endogeneity resulting from
explanatory variables that are possibly affected by productivity,
could be tackled by using appropriate
time-lag structures. Unfortunately, no recent panel data sets are
available for Tanzania and are hard to
find for Sub-Saharan African countries more generally.
8. More information on the sampling methodology can be found
in the Investment Climate Assessment
Report on Tanzania (see World Bank 2004a).
9. Some of the variables were imputed using secondary
information to reduce the number of excluded
observations to a minimum. Information on imputation methods
used is delegated to the appendix.
10. Unfortunately, the origin of foreign ownership was not
available for all firms (only for firms owned by
individuals) hence we could not test for differences among these
groups. Also ethnic minorities may
have the Tanzanian nationality.
11. For firms that do not report to have any R&D activities or
expenditures, LRDEXP is set equal to zero.
To correct for this measurement error we include the dummy
variable RD.
12. Yet these values are similar to those of other countries in
Africa. For instance, in Uganda and Zambia,
respectively 16% and 6% of firms have a quality certificate. In
both Kenya and Zambia, 8% of firms
license technology from a foreign company. Training is offered
by 48% of firms in Kenya, 30% in
Uganda, and 34% in Zambia. 47% of firms in Uganda, 50%
report process innovation in Zambia.

13. We control for industry dummies. After eliminating
insignificant dummies, we end up with four
industry dummies: food, chemicals, wood working, and textiles.
14. To control if multicollinearity affects the results, the
variable licence was introduced separately and in
combination with a more limited set of variables, but the
variable remained insignificant in all
different specifications.
15. Other variables on the benefit of being in informal networks
or having informal linkages were tested,
but none of them produced any significant result. One variable
was whether firms produced for
multinational companies located in Tanzania; one variable for
whether the firm acts as subcontractor
for other firms; one variable for whether the firm cooperates
with local producers for borrowing
machinery, product development, market research, training of
workers, purchase of inputs, attracting
investment, exchange of information or subcontracting.
References
Bagachwa, M. (1993), ‘Impact of Adjustment Policies on the
Small-Scale Enterprise Sector in Tanzania,’
in Small Enterprises and Changing Policies, eds. A. Helmsing
and T. Kolstee, London: Intermediate
Technology Publications.
Banerjee, A.V., and Duflo, E. (2005), ‘Growth Theory through
the Lens of Development Economics,’ in
Handbook of Economic Growth Vol. 1A, eds. P. Aghion and S.
Durlauf, North-Holland: Elsevier,
pp. 473 – 552.

Bangens, L. (2004), ‘Clusters of Competence: Forming
Successful Alliances between University and
Industry in East Africa,’ in Innovation Systems and Innovative
Clusters in Africa, Proceedings of the
Regional Conference, 18 – 20 February, eds. B. Mwamila, L.
Trojer, B. Diyamett and A. Temu
Bagamoyo, Dar es Salaam: University of Dar es Salaam, pp. 237
– 256.
Baumol, W.J. (2002), The Free-Market Innovation Machine,
Princeton, NJ: Princeton University Press.
Bell, M., and Albu, M. (1999), ‘Knowledge Systems and
Technological Dynamism in Industrial Clusters in
Developing Countries,’ World Development, 27, 1715 – 1734.
Benavente, J.M. (2006), ‘The Role of Research and Innovation
in Promoting Productivity in Chile,’
Economics of Innovation and New Technology, 15, 301 – 315.
Biggs, T., and Shah, M. (2006), ‘African Small and Medium
Enterprises, Networks and Manufacturing
Performance,’ World Bank Policy Research Working Paper
3855, Washington: World Bank.
Biggs, T., Shah, M., and Srivastava, P. (1995), ‘Technological
Capabilities and Learning in African
Enterprises,’ World Bank Technical Paper 288, Washington:
World Bank.
Bigsten, A., Isaksson, A., Söderbom, M., Collier, P., Zeufack,

A., Dercon, S., Fafchamps, M., Gunning,
J.W., Teal, F., Appleton, S., Gauthier, B., Oduro, A.,
Oostendorp, R., and Pattilo, C. (2000), ‘Rates of
Return on Physical and Human Capital in Africa’s
Manufacturing Sector,’ Economic Development and
Cultural Change, 48, 801 – 827.
Bigsten, A., and Söderbom, M. (2006), ‘What Have We Learned
from a Decade of Manufacturing
Enterprise Surveys in Africa?,’ The World Bank Research
Observer, 21, 241 – 265.
Bloom, N., and van Reenen, J. (2006), ‘Measuring and
Explaining Management Practices across Firms and
Nations,’ Discussion paper #716, Centre for Economic
Performance, London School of Economics and
Political Science, London, UK.
Bongenaar, B., and Szirmai, A. (2001), ‘Development and
Diffusion of Technology: The Case of TIRDO,’
in The Industrial Experience of Tanzania, eds. A. Szirmai and
P. Lapperre, New York: Palgrave,
pp. 171 – 193.
Buchinsky, M. (1998), ‘Recent Advances in Quantile
Regression Models, A Practical Guideline for
Empirical Research,’ Journal of Human Resources, 33, 88 –
125.
Chudnovsky, D., Lopez, A., and Pupato, G. (2006), ‘Innovation
and Productivity in Developing Countries,
a Study of Argentine Manufacturing Firms’ Behaviour (1992 –
2001),’ Research Policy, 35, 266 – 288.
Coase, R. (1998), ‘The New Institutional Economics,’ American
Economic Review, 88, 72 – 74.

Cohen, W., and Levinthal, D. (1989), ‘Innovation and Learning:
The Two Faces of R&D,’ The Economic
Journal, 99, 569 – 596.
Crépon, B., Duguet, E., and Mairesse, J. (1998), ‘Research and
Development, Innovation and Productivity:
An Econometric Analysis at the Firm Level,’ Economics of
Innovation and New Technology, 7,
115 – 158.
Dawson, J. (1993), ‘Impact of Structural Adjustment on the
Small Enterprise Sector: A Comparison of the
Ghanaian and Tanzanian Experiences,’ in Small Enterprises and
Changing Policies, eds. A. Helmsing
and T. Kolstee, London: Intermediate Technology Publications.
Djankov, S., La Porta, R., Lopez-De-Silanes, F., and Shleifer,
A. (2002), ‘The Regulation of Entry,’ The
Quarterly Journal of Economics, 117, 1 – 37.
Efron, B. (1981), ‘Nonparametric Estimates of Standard Error:
The Jackknife, the Bootstrap and other
Methods,’ Biometrika, 68, 589 – 599.
Eifert, B., Gelb, A., and Ramachandran, V. (2005), ‘Business
Environment and Comparative Advantage in
Africa: Evidence from the Investment Climate Data,’ Working
Paper 56, Center for Global Development,
http://ssrn.com/abstract¼997383
Enos, J. (1992), The Creation of Technological Capabilities in
Developing Countries, London: Pinter.
Fernandes, A.M. (2006), ‘Firm Productivity in Bangladesh
Manufacturing Industries,’ World Bank Policy

Research Working Paper 3988, Washington: World Bank.
Fransman, M. (1985), ‘Conceptualising Technical Change in the
Third World in the 1980s: An Interpretive
Survey,’ Journal of Development Studies, 21, 572 – 652.
Goedhuys, M. (1999), ‘Industrial Organisation in Developing
Countries, Evidence from Côte d’Ivoire,’
unpublished Ph.D. dissertation, Catholic University of Leuven.
Goedhuys, M. (2007), ‘Learning, Product Innovation and Firm
Heterogeneity in Developing Countries,
Evidence from Tanzania,’ Industrial and Corporate Change, 16,
269 – 292.
Goedhuys, M., and Sleuwaegen, L. (1999), ‘Barriers to Growth
of Firms in Developing Countries,
Evidence from Burundi,’ in Innovation, Industry Evolution and
Employment, eds. D. Audretsch and
R. Thurik, Cambridge: Cambridge University Press, pp. 297 –
314.
Hewitt, T., Wangwe, S., and Wield, D. (2002), ‘Seeing Eye to
Eye: Organizational Behaviour, Brokering
and Building Trust in Tanzania,’ Public Administration and
Development, 22, 97 – 108.
Hewitt, T., and Wield, D. (1997), ‘Tanzanian Networks,
Networks in Tanzanian Industrialization,’ Science
and Public Policy, 24, 395 – 404.
Ichniowski, C., Shaw, K., and Prenushi, G. (1997), ‘The Effects
of Human Resource Management: A Study
of Steel Finishing Lines,’ American Economic Review, 87, 291
– 313.

Katz, J. (1987), Technology Generation in Latin American
Manufacturing Industries: Theory and Case-
studies concerning its Nature, Magnitude and Consequences,
Basingstoke: Macmillan.
Kleinknecht, A., and Mohnen, P. (2002), Innovation and Firm
Performance. Econometric Explorations of
Survey Data, New York: Palgrave.
Koenker, R., and Bassett, G. (1978), ‘Regression Quantiles,’
Econometrica, 46, 33 – 50.
Koenker, R., and Hallock, K. (2001), ‘Quantile Regression,’
Journal of Economic Perspectives, 15,
143 – 156.
Lall, S. (1992), ‘Technological Capabilities and
Industrialization,’ World Development, 20, 165 – 186.
McCormick, D. (1999), ‘African Enterprise Clusters and
Industrialization: Theory and Reality,’ World
Development, 27, 1531 – 1551.
Mello, M., and Perreli, R. (2003), ‘Growth Equations: A
Quantile Regression Exploration,’ Quarterly
Review of Economics and Finance, 43, 643 – 667.
Murphy, J.T. (2002), ‘Networks, Trust and Innovation in
Tanzania’s Manufacturing Sector,’ World
Development, 30, 591 – 619.

Mwamila, B., and Katalambula, H. (2004), ‘University –
Industry Linkage: The Case of the Prospective
College of Engineering and Technology at the University of Dar
es Salaam,’ in Innovation Systems and
Innovative Clusters in Africa, Proceedings of the Regional
Conference, 18 – 20 February,
eds. B. Mwamila, L. Trojer, B. Diyamett and A. Temu
Bagamoyo, Dar es Salaam: University of
Dar es Salaam, pp. 257 – 295.
National Bureau of Statistics (2004), Central Register of
Establishments, Dar es Salaam.
Nelson, R., and Winter, S. (1982), An Evolutionary Theory of
Economic Change, Cambridge, MA: Harvard
University Press.
North, D. (1991), ‘Institutions,’ Journal of Economic
Perspectives, 5, 97 – 112.
Oyelaran-Oyeyinka, B. (2006), Learning to Compete in African
Industry, Ashgate: Aldershot.
Portelli, B., and Narula, R. (2006), ‘Foreign Direct Investment
through Acquisitions and Implications for
Technological Upgrading: Case Evidence from Tanzania,’
European Journal of Development
Research, 18, 59 – 85.

Schmitz, H., and Nadvi, K. (1999), ‘Clustering and
Industrialization: Introduction,’ World Development,
27, 1503 – 1514.
Sleuwaegen, L., and Goedhuys, M. (2003), ‘Technical
Efficiency, Market Share and Profitability of
Manufacturing Firms in Côte d’Ivoire: A Technology Trap?,’
Cambridge Journal of Economics, 27,
851 – 866.
Sutz, J. (2006), ‘Building Accurate Mirrors: Innovation
Indicators for Better Innovation Policies in
Underdevelopment,’ paper presented at the Catch-up meeting,
Center for Research on Innovation and
Competition, Manchester,
http://www.cric.ac.uk/cric/events/catchup/papers/Sutz.pdf
Szirmai, A., and Lapperre, P. (2001), The Industrial Experience
of Tanzania, New York: Palgrave.
Temu, A., and Due, J. (2000), ‘The Business Environment in
Tanzania after Socialism: Challenges of
Reforming Banks, Parastatals, Taxation and the Civil Service,’
Journal of Modern African Studies, 38,
683 – 712.
Tybout, J. (2000), ‘Manufacturing Firms in Developing
Countries, How Well Do They Do and Why?,’

Journal of Economic Literature, 38, 11 – 44.
UNCTAD (1996), Fostering Technological Dynamism:
Evolution of Thought on Technological Develop-
ment Processes and Competitiveness, A Review of the
Literature, Geneva: United Nations.
UNCTAD (2002), Investment Policy Review, The United
Republic of Tanzania, Geneva: United Nations.
URT (United Republic of Tanzania) (2006), ‘The Economic
Survey 2005,’ www.tanzania.go.tz/economic-
surveyf.html
Utz, A. (2006), ‘Tanzania in the Knowledge Economy:
Fostering Innovation, Productivity and Techno-
logical Change,’ WBI working paper, Washington: World Bank.
http://info.worldbank.org/etools/docs/
library/232302/Tanzania_in_the_Knowledge_Economy.pdf
Wangwe, S. (1992), ‘Building Indigenous Technological
Capacity: A Study of Selected Industries in
Tanzania,’ in Alternative Development Strategies in SubSaharan
Africa, eds. F. Stewart, S. Lall, and
S. Wangwe, London: Macmillan.
Williamson, O.E. (1987), The Economic Institutions of
Capitalism: Firms, Markets, Relational

Contracting, New York: Free Press.
World Bank (2004a), Investment Climate Assessment:
Improving Enterprise Performance and Growth in
Tanzania, Washington: World Bank.
World Bank (2004b), World Development Report 2005: A
Better Investment Climate for Everyone,
Washington: World Bank.
World Bank (2006), World Development Indicators 2006,
Washington: World Bank.
World Bank (2007), Tanzania at a Glance, Washington: World
Bank, http://devdata.worldbank.org/AAG/
tza_aag.pdf
World Economic Forum (2004), The African Competitiveness
Report 2004, Geneva: WEF.
Appendix: Construction of the capital stock
The capital stock used in the estimation is the sum of the value
of machinery and equipment and the value of
land and buildings.
To measure the value of machinery and equipment, we used the
value of replacement cost of machinery
and equipment in 2002. When replacement value was missing,

we used the ‘corrected’ sales value. As a
number of firms provided information on both replacement and
sales value, we used the median ratio of
replacement to sales value, at the industry level, to build the
‘corrected’ sales value. If both replacement and
sales value were missing, we used the ‘corrected’ net book
value of machinery and equipment in 2002.
Similarly, we used the median ratio of replacement value to net
book value, at the industry level, to
construct the ‘corrected’ net book value. If replacement value,
sales value and net book value were missing,
we used the ‘corrected’ gross book value of machinery and
equipment in 2002. To get to the ‘corrected’
gross book value, we used the median ratio of net to gross book
value and additionally the median
replacement to net book value, at the industry level.
To summarise, value of machinery and equipment
¼ Replacement value of machinery and equipment (incl.
vehicles)
if missing:
¼ sales * median (replacement/sales)
if missing
¼ net book * median (replacement/net book)
if missing
¼ gross book * median (net book/gross book) * median
(replacement/net book)
For land and buildings, replacement and sales value are not
available. Hence, we have used the net book
value in 2002, or the ‘corrected’ gross book value, where the
correction factor is the median of net to gross
book value at the industry level.

SELECTION OF A PRESSURE VESSEL MANUFACTURER
On August 1, the engineering department hand-carried a
purchase requisition to Jack Toole,
supply manager, Oceanics, Inc. The requisition covered the
purchase of one pressure vessel to
Oceanics’ specifications as outlined in the requisition.
Immediately, Jack went to work. He
prepared a request for quotations asking twenty major pressure
vessel manufacturers to have their
proposals in his hands no later than Wednesday, August 31. The
response to Jack’s request for
quotations was amazing.
During the month of August, eighteen of the twenty companies
hurriedly prepared their
proposals and submitted them to Jack within the allotted
bidding time. As each proposal was
received on Jack’s desk, copies were forwarded to the engineer

and manufacturing engineer for
preliminary evaluation. By September 5, Jack called a meeting
in his office with the engineer, Mr.
Holpine, and the manufacturing engineer, Mr. Grinn.
During the course of the meeting, proposals were carefully
screened and bidders were
eliminated one by one until two companies remained. It was a
difficult decision for the group to
decide which of the two companies submitted the better
proposal. The advantages and
disadvantages of each supplier appeared to be about equal. Jack
pointed out that Atomic Products
Company submitted a lower estimated price, guaranteed the
equipment, was more suitably
located, and would meet the required delivery date. Jack also
pointed out to Grinn and Holpine
that Nuclear Vessels, Inc., offered Oceanics lower hourly and
overhead rates, a minimum amount
of subcontracting, and excellent past experience in making
similar vessels. Jack stated that a field
trip would be necessary to talk with both suppliers to determine
which one was best qualified. At
this point, the meeting was adjourned and plans were made to

visit both companies the following
week. (See Exhibits 1 and 2.)
In following through with supply management policy, Jack
called the vice president of
Oceanics’ New York sales office and advised him of the
potential trip. Jack learned that Atomic
Products was a potential customer for Oceanics’ products, but
Oceanics’ sales representatives
were unable to get into the plant to meet key people responsible
for procurement of major
equipment. The vice president of sales stated that a sales rep
would be at the airport to meet
Oceanics’ representatives and take them to the Atomic Products
Company first thing Monday
morning. Jack phoned the president, Mr. Wilcox, and advised
him that representatives from
Oceanics would like to be at his plant Monday morning to
review his plant facilities and meet the
responsible people. The president did not appear to be
enthusiastic, but said that he would be
pleased to see them when they arrived.

EXHIBIT 1
Atomic Products Company, New York, N.Y.
We are pleased to submit a proposal in accordance with your
request for the manufacture of one
pressure vessel in accordance with your sketch #835 and all
referenced specifications pointed out
in your letter of August 2.
Price. Because of the potential changes pointed out in your
invitation to bid, and in line with your
request, the work will be performed on a cost-plus-a-fixed-fee
contract detailed as follows:
a. Total price: Estimated cost $1,120,000

Fixed fee 112,000
Total $1,232,000
b. Costing rate: Estimated shop rate $24/hour
Shop overhead 180%
Material Cost + 10% handling charge
Shop facilities. There are adequate facilities at our New York
Plant to manufacture the vessel and
meet the specification to the fullest extent possible. We invite
you and your associates to visit our
facilities.
Past experience. Our company has not made vessels of this size
but does have the equipment and
know-how necessary to perform the work. Our experience has
been in working with vessels up to
60” in length, I.D. 30” and 3” wall.
Subcontracting. We will be able to fabricate the entire vessel
without exception in our shop.
Organization. A total of 2,000 employees is directly associated
with our division.

Delivery. The pressure vessel will be shipped f.o.b. shipping
point via rail in 6 months providing
there are no engineering changes.
Guarantee. We guarantee workmanship and materials to be in
accordance with the specifications
that were supplied to us at the time of this proposal.
EXHIBIT 2
Nuclear Vessels, Inc., Houston, Texas
Reference is made to your invitation to bid dated August 2 to
manufacture the pressure vessel in
accordance with your negative #835, and referenced
specifications and any future changes
necessary.
Price. The work will be performed on a cost-plus-a-fixed-fee
basis, broken down as follows:

a. Total price: Estimated cost $1,560,000
Fixed fee 1
Total $1,560,001
b. Costing rates: Estimated shop rate $16/hour
Shop overhead 160%
Material At cost
Shop facilities. We have adequate shop facilities to manufacture
and deliver the vessel and would
be pleased to have representatives from your company visit our
facilities at any time.
Past experience. The company has had extensive experience in
manufacturing pressure vessels of
heavy plate. Vessels 80” I.D., 40’ long, 5” thick and many
others have been handled by this
company.
Subcontracting. It will not be necessary for the company to
subcontract any of the forming,
welding, machining, or testing for this work. However, forgings
will be purchased from a

competent supplier after he has satisfied the company’s
metallurgist that his forgings will meet
the specifications.
Organization. The Supply, Expediting, Quality Control,
Production and other departments will
each have one man assigned to follow this project from start to
finish. Forms and records are
available for your review. Our organization is familiar with
Oceanics requirements from
knowledge gained as a result of previous work accomplished for
your division.
Delivery. The pressure vessel will be shipped f.o.b. shipping
point, Houston, Texas, to your
Pittsburgh location within your required delivery time of six
months or shortly thereafter.
Guarantee. This company will guarantee only workmanship. The
rigid material specifications
make it difficult for our supplier to furnish plate without any
inclusion of slag deposits. Oceanics
will have to stand the costs of any plate rejected or repaired
after being tested by ultrasonic

methods. Such costs can be negotiated after such defects are
found.
Another call was also made to Nuclear Vessels’ president, Mr.
Winninghoff, who was
quite enthusiastic about the potential visit and asked if he could
meet the group at the airport,
make hotel reservations, or perform other courtesies. Jack
advised Mr. Winninghoff that these
matters were taken care of and that an Oceanics’ sales
representative for the Houston, Texas, area
would accompany the group during the visit.
Monday morning, Messrs. Toole, Grinn, and Holpine took off
from Pittsburgh and
arrived at Kennedy Airport in New York. Mr. Morgan, the sales
manager of Oceanics’ New York
office, met the group and drove them to the main office of the
Atomic Products Company. The
group registered, obtained passes, and went to the conference
room. Shortly thereafter, the
manager of production, Mr. Strickland, entered, introduced
himself, and stated that the president

was tied up but would see them later in the day. Jack Toole
opened the meeting by stating that
Atomic Products’ proposal was among the top contenders for
supplying the pressure vessel and it
was Oceanics’ desire to look over Atomic Products’ facilities
and meet the people responsible for
the job. Jack Toole asked Holpine to explain in greater detail
the use of the vessel in the reactor
system and to give Mr. Strickland some background on the
engineering work relating to the
vessel. Mr. Grinn reviewed the manufacturing aspects of the
vessel as required by the basic
specifications. Near the end of this discussion, Jack Toole asked
Mr. Strickland if Holpine’s and
Grinn’s comments had the same meaning as Atomic Products’
interpretation of the specifications.
Mr. Strickland agreed, but was somewhat concerned over the
rigid cleaning specification. As he
told the group, “It is difficult for a shop our size to construct a
temporary building around the
pressure vessel, make such a building airtight, and compel our
workmen to wear white coveralls
and gloves, and to adhere to surgical cleanliness requirements. I
doubt if we can erect such a

building in our present shop area. Instead, we may add a lean-to
to the outside of our existing
buildings.”
The meeting with the production manager lasted one hour; then
the group commenced to
tour the shop. Grinn noted that most of the machines, such as
the vertical boring mill, horizontal
mill planer, radial drills, and beam press, were comparatively
new and well maintained.
Jack Toole wondered why Atomic Products’ estimated cost was
lower than Nuclear
Vessels’, yet Atomic Products’ costing rates were somewhat
higher. With this thought in mind,
he asked Mr. Strickland, “Do you consider your shop to be
better equipped than your
competitors’?” Mr. Strickland replied that it was their
management’s feeling that this shop was
the best equipped in the United States to handle such vessels,
and that even though the shop rates
were higher than other shops, they would turn out more work in
less time than any competitor.
Holpine asked Mr. Strickland why their past experience was
limited to smaller-sized vessels, to

which Mr. Strickland replied that they could handle any size
vessel up to and beyond the one
required by Oceanics, but had never received a contract for such
vessels.
Atomic Products Company was a union shop that had had
several major strikes during
the past few years. There were 2,000 people employed, and the
plant covered approximately
470,000 square feet of floor area.
The general appearance of the shop was excellent. The group
noticed that the aisles were
clean; that there was ample lighting, adequate ventilation, up-
to-date laboratories, and good
inspection facilities; and that the overall appearance of the
building was extremely neat and well
ordered.
The group pointed out several items in production and asked
Mr. Strickland the ultimate
use of these products. They received a vague reply, such as,
“These are a number of special jobs
we have in the shop that we can handle without any trouble.”

Mr. Strickland interrupted a group of employees standing in a
corner and asked one of
them to show the group the inspection and quality control
departments. Both departments were
well staffed and had up-to-date equipment.
The group asked Mr. Strickland to show them control of
incoming materials vital to
potential Oceanics’ work. Wrong material that might possibly
get into such a pressure vessel
would contaminate the entire nuclear system. Mr. Strickland did
not offer the group any evidence
of materials control, but stated that they had produced hundreds
of smaller vessels and had no
trouble in the segregation of materials.
The metallurgical and chemical laboratories were well staffed
and could provide
Oceanics with adequate test specimens required by the
specifications.

At the end of the tour, the group met with the president, who
asked, “Do you think that
our facilities are adequate to do the job?” Jack Toole replied
that the facilities were impressive,
but that the final selection of the supplier would be determined
by many factors and that facilities
were only part of the total evaluation. The president then
replied, “If you want us to do the work,
let us know and we will commence contract negotiations.”
Several days later, Messrs. Toole, Grinn, and Holpine left New
York and flew to Houston,
Texas, for a visit to Nuclear Vessels, Inc. When the group
registered in the hotel at 5 p.m., they
found a call waiting for them from Mr. Winninghoff, president
of Nuclear Vessels. Mr.
Winninghoff asked the group to meet that evening at the
Houston Country Club for dinner and
business discussions. At 1:30 a.m., the group returned to the
hotel.
The following morning, the Nuclear Vessels’ chauffeur met
Oceanics’ team and the
representative from Oceanics’ sales office at the hotel and took
them to Mr. Winninghoff’s office.

In the office, Mr. Winninghoff was waiting with the vice
president of engineering, vice president
of marketing, vice president of manufacturing, and other key
figures in the organization. Jack
Toole opened the meeting in much the same manner as was done
at Atomic Products Company.
After the Oceanics’ people had gone into detail on the vessel,
Jack Toole asked Mr. Winninghoff
if they had any questions concerning the specifications. There
were no comments, so the entire
group commended to tour the shop.
Mr. Grinn immediately noticed that the company’s machines
were of considerable age
and not of large capacity, but adequate for the job. Some
outside subcontracting work for the
close machining tolerances would be required. Mr. Winninghoff
stated: “True, we may not have
all the necessary machines here, but there are ample machines
available at other divisions, such as
the large vertical boring mill at our El Paso, Texas, subsidiary
plant. The schedule is such that we
can move work into other divisions without delay.” It was noted
that general working conditions

such as heating, lighting, ventilation, and cleanliness were not
as adequate as Atomic Products’.
Jack Toole noted that the higher estimated cost resulted from
more man-hours required to make
the vessel because of less adequate machines.
Mr. Winninghoff stopped by one of the shop foremen and asked,
“Say, Sam, how about
giving these gentlemen from Oceanics an idea of what your
group will be doing in the forming
and rolling of the pressure vessel?” Sam had several of his men
stop work to show the equipment
available and its intended use. Mr. Winninghoff mentioned to
the group that their plant had been
on a profit-sharing plan since it was organized. The employees
never organized a union.
There appeared to be effective control between management and
the shop. For instance,
to carry out the work fully, one member each from supply,
expediting, quality control, and
scheduling was assigned to a task force headed by a project
engineer. It was the responsibility of
this task force to follow the entire project through the shop and

keep the project engineer
informed on a day-to-day basis.
Nuclear Vessels had constructed one vessel considerably larger
than the vessel required
by Oceanics. Mr. Winninghoff claimed that they ran into
numerous problems at the beginning of
manufacturing and that the experience gained in the production
of such a large vessel made them
change their organization for closer follow-up. They also
changed the type of paperwork and
records for better control of material. The group noticed that
each piece of material in the shop
was marked for the project of its intended use. The
metallurgical and chemical laboratories were
very large, but much of their equipment was old. They appeared
to have adequate room for the
location of a cleaning room.
On Friday of the same week, Toole called a meeting of Holpine
and Grinn to evaluate the
two companies being considered. Holpine argued strongly that
Nuclear Vessels should be given a

contract because of their extreme enthusiasm to carry out the
job, their past experience in
manufacturing pressure vessels of equal size, and their previous
Oceanics experience. Said
Holpine, “Atomic Products has not had experience with our
rigid specifications and the price and
delivery will probably slip.” Grinn argued that Atomic Products
should be the company selected
because of their adequate shop and laboratory facilities,
location, ability to meet delivery date,
and ability to guarantee the vessel.
Neither Holpine nor Grinn took into consideration the cost, the
company’s organization,
guarantees, and other business considerations. It was Jack
Toole’s responsibility to evaluate both
of these companies and show which company should be given
the contract.
1. What specific areas and activities should the Oceanics group
have investigated on its
two visits?
2. Evaluate each supplier on each of the above items using
information obtained on the

field visits.
3. Based on the face value of the written proposals, which
company appeared to submit
the better offer?
4. Based on the proposal plus information obtained from the
case history, which
company is likely to be the better supplier?
5. What do you recommend?
Assignment
Payment
Type
Title
Chosen Company
Due Date
Time
CS PowerPoint
$50.00
Case Study 3 Pressure Vessel Manufacturer
12/25/2014
17:00
and Kaltura video
$10.00

Training Asset 4 How to Work With the Kaltura Media Galler.docx

Training Asset 4 How to Work With the Kaltura Media Galler.docx

Recommended

Recommended

More Related Content

Similar to Training Asset 4 How to Work With the Kaltura Media Galler.docx

Similar to Training Asset 4 How to Work With the Kaltura Media Galler.docx (20)

More from MARRY7

More from MARRY7 (20)

Recently uploaded

Recently uploaded (20)

Training Asset 4 How to Work With the Kaltura Media Galler.docx