Cluster-based MSE development: the role of Kaizen training.
Sonobe, Tetsushi ; Otsuka, Keijiro
This paper attempts to address the issues of why firms in
developing countries are less able to innovate and manage business than
their counterparts in developed countries and how their entrepreneurship
can be nurtured. For this purpose, this paper reviews our case studies
of industrial clusters in Asia and sub-Saharan Africa, including
randomised controlled trials of Kaizen management training, and
discusses what the government can do to support training, awareness
campaigns, and other efforts to nurture entrepreneurship.
JEL Classification: L60, M11,012, 014
Keywords: Industrial Development, Industrial Clusters,
Entrepreneurship, Management Training, Randomised Controlled Trial
I. INTRODUCTION
It has been increasingly recognized that entrepreneurship holds the
key to industrial development in developing countries [World Bank
(2012)]. Indeed, a significant number of studies find that productivity
and profitability vary greatly across enterprises even in the same
industry in the same country, and that a large part of the variation can
be accounted for by the difference in management practices (1)
Identifying and supporting high-potential entrepreneurs may be the key
to the success of industrial development.
Entrepreneurship is the capacity to introduce new ideas into
practice and to manage enterprise operations efficiently. Innovation
here does not necessarily mean scientific discovery or engineering
invention but the Schumpeterian creation of a new combination of
production resources and new ideas to increase profits. In the context
of developing economies, innovation includes borrowing technology or
learning from abroad. The first introduction of products and production
processes from developed countries into a developing country and the
first adoption of management practices that may be common in developed
countries but are novel in developing countries are considered to be
innovations.
Despite its importance, we know little about the entrepreneurship
of business owners and managers in developing countries. (2) Why are
firms there less able to innovate and manage than their counterparts in
developed countries? How can their entrepreneurship be nurtured? The
purpose of this paper is to explore these questions by reviewing our
case studies of industrial clusters in Asia and sub-Saharan Africa
(SSA). These studies include randomized controlled trials (RCTs) of
management training for micro and small enterprises (MSEs). While RCTs
of management or business training are not new, a major feature of our
training programmes is to teach basics of production management called
Kaizen as well as basics of are financial management and marketing which
are usually taught in other programmes. Kaizen is based on industrial
engineering developed in the US, assimilated and modified into a
bottom-up approach to management in Japan, and imported by the US
automobile industry under the name of Lean Manufacturing, which does not
emphasises bottom-up flows of information as much as Kaizen.
We highlight cluster-based industrial development because
low-income countries should have a comparative advantage in
labour-intensive manufacturing industries, which are characterised by
the dominance of MSEs located in industrial clusters. (3) In other
words, we are interested in cultivating entrepreneurship that will
foster cluster-based MSE development since such development will be
conducive to reducing poverty and income equality and crucial for
inclusive growth. (4) In Asia and Latin America, there are a large
number of industrial clusters. (5) Clusters are not uncommon in SSA.
Altenburg and Meyer-Stamer (1999) find that a number of industrial
clusters in Latin America are "survival clusters" producing
generally low-quality products and selling them predominantly to
domestic markets. Yoshino (2011) finds that most industrial clusters in
SSA are also survival clusters, where firms are barely breaking even.
Why do these firms fail to innovate? Possibly because of unfavourable
business environments: such as bad governance, poor infrastructure, and
inadequate financing. Or, they may not possess much entrepreneurship.
The former is important but, we believe the latter is crucial. There are
cases in which clusters have successfully grown despite unfavourable
business climates. This paper explores how that happens. We begin by
asking how business owners and managers who are not entrepreneurs dare
to start their businesses.
To answer this question, Section II describes how cluster-based
MSEs in various industries developed in different countries in Asia and
Africa. We find that many owners and managers come to feel keenly their
lack of managerial and innovative capacities, as their clusters become
survival clusters. Section III begins by discussing the possible reasons
why they cannot improve these skills and what the government can do to
support training, awareness campaigns, and other efforts to nurture
entrepreneurship. We then review the results of management training
experiments, including our own, to see whether entrepreneurship can be
taught and how management training programmes can be improved in future.
Section IV contains concluding remarks.
II. MANAGERIAL AND INNOVATIVE CAPACITIES OF MSEs
In this section, we use the results of our observational studies to
discuss why managerial and innovative capacities are not important
initially but assume importance later in the process of industrial
development.
Our arguments in this paper are based mainly on our 19 case studies
of cluster-based industrial development in Asia and Africa, which are
listed in Table 1. Of these case studies, 8 cases (cases 6, 8, 10, 14,
15, 16, 17, and 19) involved experiments, of which 5 experiments (cases
6, 10, 14, 16, and 19) were randomized controlled trials. Discussion of
the results of these experiments is deferred to the next section.
Selection of the Cases
We conducted the 19 case studies over nearly 15 years without a
long-run study plan specifying which industrial clusters should be
studied based on random selection from a list of clusters. This is
because there was and is no such a list. The number of clusters is
unknown, especially in SSA. It is likely that SSA has many more clusters
and entrepreneurs than governments and researchers recognise. Most would
be survival clusters. (6) The 19 clusters that we studied are not
representative of industrial clusters in Asia and Africa, and they are
more likely to be focused on dynamically growing rather than survival
clusters. Nonetheless, they cover a variety of MSE clusters in Asia and
Africa.
All these clusters have been voluntarily and spontaneously formed
by private firms without much assistance from the government, including
the four cases in China. Thus, the China case studies are comparable to
the others. The studies of four industrial clusters in Japan and Taiwan,
China are also comparable because they were developed before these
countries became highly industrialised.
The clusters in the first group (i.e., cases 1 to 10) were
initiated by pioneering businesses that produced imitations of imported
goods without any help from the government or foreign firms. The
development of the second group of clusters (cases 11 to 17) was based
on foreign technologies brought in by foreign direct investments (FDIs),
state-owned firms (SOEs), or foreign artisans who migrated from Southern
Europe and ran home-based businesses. (7) In the third group (cases 18
and 19), the clusters came into existence thanks to training programmes
provided by foreign firms and international organisations. Consistent
with the Heckscher-Ohlin theorem, the initial products of these
clusters, as shown in the parentheses in the last column, tended to be
labor-intensive items. This tendency is clearer in the first group than
in the second and third groups (cases 18 and 19).
Indeed in the first group, new industries emerged as main products
or activities shifted to take advantage of existing institutions,
infrastructures, and technologies. In Kumasi, Ghana (case 10), for
example, metalworking firms, such as lathe turners, welders, and casting
foundries, are clustered where thousands of car repair garages are
concentrated. These garages need a lot of metal processing services and
also provide an abundant supply of scrap metal from disabled cars that
cannot be repaired. The metalworking firms provide repair services but
also manufacture a variety of metal products, such as com mill machines,
wheelbarrows, and cash safes. Among tailors in Addis Ababa and Nairobi
(cases 8 and 9), those who made money and won the buyers' trust
expanded their businesses from tailoring into the factory or workshop
production of ready-made garments. In cases 1, 4, 5, 7, 13, and 15, a
number of owners and managers started as traders and then diversified
their businesses into manufacturing. Such former traders tend to be
high-performers, which may reflect the importance of management and
marketing skills to manufacturing firms in developing countries.
Why Business Owners anil Managers who are not Entrepreneurial Set
up Firms
New indigenous industries in developing countries begin with the
imitation of imported products. In cases 1 to 10 in Table 1, an industry
pioneer exerted great efforts to find new production methods, sources of
materials, and marketing channels for imperfect imitations of imported
products. Once these difficulties were overcome, the pioneer made
sizable profits, despite the low quality of his or her products, because
poor domestic consumers created a high demand for cheap substitutes of
imported products and because competitors were few. This profit
attracted a swarm of imitators to the industry. Many were spin-offs,
i.e., the former employees of the pioneer, who produced the same
low-quality products by using the same low-quality inputs and sold their
products on the same local markets as the pioneer. In this way, an
industrial cluster was formed, as illustrated in Figure 1. Industries
that began with technology brought by FDIs, SOEs, and international
organisations followed almost the same path (cases 11 to 17 and 19).
[FIGURE 1 OMITTED]
Earning high profits for the time being, owners and managers in the
early phase are not particularly interested in introducing new products
or new production processes. Moreover, they are rarely capable of
innovation. The pioneers are imitators, even though they are innovative
in starting new business activities by imitating. Other firms in the
cluster are imitators of imitators and even less likely to possess
innovative capacity. Interestingly, setting up a firm in the cluster
becomes increasingly easy over time for such entrepreneurs in the early
phase of cluster development. As the cluster expands in terms of the
number of firms and hence their output, an increasing number of traders
buying products and supplying materials come to the cluster, which makes
production and transactions in the cluster more convenient. Some firms
specialise in the production of parts. In other words, the division of
labour between assemblers and part-suppliers and between manufacturers
and traders increases with the market size [Ruan and Zhang (2009)]. The
problem of imperfect contract enforcement and asymmetric information is
reduced in the industrial cluster by the "community mechanism"
an environment where people know each other and spread information
regarding unscrupulous behaviour, as in rural communities [Hayami
(2009); Arif and Sonobe (2012)]. (8)
The number of firms and the total output in the cluster increase
over time, but productivity and product quality are hardly improved.
This phase of cluster development may be termed the "quantity
expansion" phase. In clusters in their quantity expansion phase,
even casual observers can see that very poorly managed firms are not
forced to exit. Possible reasons include the low-level of competition
within the cluster, the high demand for low-quality products, and the
availability of low-wage labour. Thus, spontaneously formed and
quantitatively expanded industrial clusters abound in firms with very
limited managerial and innovative capacities. Since such firms make
positive profits, more of the same kinds of firms are attracted to the
cluster. They, however, will face difficulty in the next phase.
Increasing Importance of Managerial and Innovative Capacities
As the cluster expands quantitatively, the product price and
profitability will eventually decline because increased output is
supplied to the limited domestic market for the cluster's
low-quality product. This process, illustrated by Figure 1, took place
in the subsequent quantity expansion phase of industrial development in
all cases except for case 18 listed in Table 1. As the profitability
declines, the entry of new firms slows down and eventually ceases.
Declining profitability will induce business owners and managers to
seek more profitable products [Aghion, et al. (2005)], which should be
of higher quality and differentiated from the inferior products of other
firms. Such endeavour, however, do not necessarily bear fruit. While a
firm may successfully improve product quality by using high-quality
materials and parts and by employing skilled workers, consumers may not
immediately perceive the improved quality and the new product may fail
to command a high price in market. As Akerlof (1970) points out,
branding and quality guarantees would be effective countermeasures to
this problem, but they may be known by only a few owners and managers,
if any, in industrial clusters. Moreover, since brands may be stolen,
branding may have to be supplemented with the use of exclusive sales
agencies and other distribution methods [Sonobe, Hu, and Otsuka (2004)].
Since the improved products contain differentiated parts and components,
and since such intermediate goods embody new ideas, it is also important
to establish trust-based, long-term subcontracting relationships with
parts suppliers. In addition, to enter the high-quality segment of the
market, product quality must be strictly controlled. If these reforms
are successfully implemented, production can be expanded profitably.
As the firm size is enlarged accordingly, the management of cash
flows, inventory, and labour will assume greater importance. Good
management is not an easy task for owners and managers in the cluster,
as they have operated only small organisations and may be unaware of how
to improve management. In this sense, they are not likely to be
efficient managers. They now need to study management or hire competent
managers or probably both.
The restoration of high profitability involves these
multidimensional improvements in production and management. If one
dimension of the improvements is missing, other improvements may not
produce positive results. For example, even a firm that has successfully
launched a popular product may suffer from losses if its supply chain is
inadequate. Moreover, even a firm making profits may go into bankruptcy
if its cash management is poor. If a firm's attempt to improve
production and management miscarries, the news will spread quickly in
the cluster, and others may recoil from any new attempt at similar
improvements. This will lead the cluster to a long-run stalemate in
which profitability is so low that existing firms struggle for survival
and there are no new entrants (see the dotted curves in Figure 1). Most
industrial clusters in SSA have been trapped in this equilibrium. A
possible exception is the leather shoes industry (case 15) in Addis
Ababa, which seems to be going through multidimensional improvements in
production and management by incorporating production and management
knowledge from Italy.
Many clusters in Asia, including cases 1 to 5 and 11 to 14,
succeeded in multidimensional improvements and followed a development
path leading to the dramatic growth of the industry with a smaller
number of much larger firms, as illustrated by the solid curves in
Figure 1. Such a dynamic development phase may be termed the
"quality-improvement" phase, in which managerial human capital
plays a decisive role. The growth of the industry in this phase entails
creative destruction. An increasing number of firms attempt to imitate
the multidimensional improvements in the cluster, and some of them
further improve production and management practices. Those firms
undertaking continual improvements in production and management will
grow, whereas firms that fail to keep pace will be forced to exit the
industry or will be merged with growing firms. While the number of firms
in the cluster will decrease as the result, the total production value
and employment of the cluster will continue to grow, and the products
will be sold in larger markets, including export markets.
Continual improvements in production and management will be
achieved by learning from outside of the cluster, especially from
abroad. More concretely, innovative entrepreneurs acquire new knowledge
of technology and management by visiting foreign countries frequently to
participate in trade fairs and training programmes. They also send
workers abroad for training, and invite in foreign experts. They may
also be able to learn from foreign firms operating within their
countries. Entrepreneurs in Asia have been keen on learning from the
successful experiences of advanced firms in neighbouring countries.
Taiwan and Korea learned from Japan, while China learned from Taiwan and
Korea as well as from Japan. At present, South Asian countries are
learning from China, among other countries. In SSA, only a few owners
and managers seem to clearly recognise the importance of learning from
abroad. (9)
Employment Size of Industrial Clusters
To illustrate how significant industrial clusters are in terms of
job creation and how they differ in size, with and without successful
multi-dimensional improvements in production and management, Table 2
shows the number of manufacturing jobs and the total number of jobs
(including those in garages in case 10 and subcontractors in case 14).
It also shows the gender and schooling composition of workers in the
selected industrial clusters. Because we often failed to collect the
relevant job data in case studies conducted earlier, we focus on
recently surveyed industrial clusters. Except for cases 10, 14, 15, and
18, the clusters listed in this table are in the final stage of the
quantity expansion phase, and their total employment size ranges from
600 to thirteen thousand and has been relatively stable for the last
several years. The metalwork cluster in Ghana (case 10) was expanding in
terms of the number of enterprises at the time of our baseline survey in
2003.However according to the results of our follow-up surveys it
probably has been in the final stage of the quantity expansion phase
since 2005. The number of jobs is significantly higher in the knitwear
cluster in Vietnam (case 14) and leather shoe cluster in Ethiopia (case
15), which have entered the quality improvement phase by newly
introducing multidimensional improvements [Sonobe and Otsuka (2011)].
III. TRYING TO MAKE MANAGEMENT TRAINING MORE EFFECTIVE
In recent years, an increasing number of randomized controlled
experiments have been carried out to test the effectiveness of
management training and consulting services provided to MSEs in various
parts of the developing world. (10) A common finding obtained by these
studies, including our own randomised experiments (cases 6, 10, 14, 16,
and 19) and as non-randomized ones (cases 8, 15, and 17), is that the
owners and managers of 0 MSEs had very limited management knowledge
before receiving management training or consulting service. While the
training contents were rather rudimentary in many of these experiments,
many participants in the training programmes adopted the management
practices taught in their programme. This indicates they had not been
aware of those management practices before.
Reasons for and Countermeasures to Low Awareness
Most likely, the majority of the MSEs in the samples of these
studies operate in survival clusters, and they know they need something
new in order to increase profitability. Why then are these owners and
managers ignorant of even basic management practices? There are two
possible explanations. One attributes this ignorance to market failures.
One market failure occurs because owners and managers do not know who
possesses the knowledge they want to acquire, and when they do know,
they are not able to verify whether or not the person has passed that
knowledge to them. Moreover, once the seller shares his or her
knowledge, the buyers may quickly grasp and become unwilling to pay for
it. Because of this asymmetric information problem, the transaction of
knowledge is difficult unless the seller has established a good
reputation. The second type of relevant market failure arises from the
difficulty in keeping the purchased knowledge secret. If imitation or
spillover is expected to be widespread, businesses will be reluctant to
pay, preferring to get a free ride.
The second explanation may be down to ignorance. MSE owners and
managers may value of learning as well as overestimating their own
abilities." Many owners and managers in our study sites maintain
their management is better than average. When asked how they know what
the average is, they simply smile. Moreover, some owners and managers
may have a tendency to put off paying for expensive activities such as
learning and investing. (12)
The asymmetric information problem may be mitigated if
management-training programmes are provided by organisations with good
reputations, such as governmental orienteer national organisations,
foreign aid agencies, well-known companies and NGOs. The gap between the
social and private benefits of owners and managers' acquiring
management skills may be bridged by government's support for
management training or the provision of financial incentives to either
owners or managers who learn management skills or to organisations that
train them. Public awareness campaigns may help address ignorance about
the value of management skills. Owners and managers who tend to
procrastinate may be nudged into action if microfinance or some other
favorable treatment is linked with participation in a training
programme. (13) Implementing these programmes to assist MSEs is unlikely
to lead to serious corruption, compared with infrastructure projects
[Shleifer and Vishny (1993)].
Estimating the Benefits of Management Training
The above arguments about countermeasures are based on the premise
that it is possible to design and implement a management-training
programme with social benefits exceeding costs. There has, so far, been
little empirical evidence that clearly supports this premise because the
social benefits of management training programmes are not easy to
estimate even if sample owners and managers are randomly assigned to
control and treatment groups. Few studies have attempted cost-benefit
analyses of management training. Many papers on management experiments
do not even mention costs.
There are probably two major difficulties in measuring or
estimating the social benefits of management training. (14) One arises
from the possible external effects of the training. The social benefit
is the sum of private benefits to the participants and those to
non-participants, minus any harmful effects, if any. Non-participants
may benefit from the training programme through knowledge spillovers.
However, if participants increase sales revenues by taking customers
from non-participants, someone will lose. Few attempts have been made to
measure or estimate these positive and negative externalities, with
Calderon, et al. (2012) being an exception.
The second major difficulty in estimating the effects of training
is that no one knows how long they take to become fully effective or how
long they last. Existing studies measure how training programmes change
participants' knowledge levels, their adoption of management
practices taught in the programme, their longevity or survival in the
market, and their accounting-based indicators of business performance,
such as revenues, value added, and profits. Most of the studies find
that the training effects on the participants' level of management
skills and adoption of useful management practices are positive and
statistically significant. Mano, et al. (2012) find that management
training improves a participant business's longevity significantly.
However, the training effects on accounting-based performance indicators
are significant in only a small number of studies. Presumably, this is
partly because business performance is affected by a multitude of
factors beyond the participants' control, and partly because the
training effect on business performance takes a longer time to make its
full impact. It is likely this effect was measured too early in most
studies.
Tentative Results of Cost-Benefit Analysis
Since 2007 and in collaboration with the World Bank and the Japan
International Cooperation Agency (JICA), we have been conducting pilot
projects in eight clusters where management training programmes are
provided for small businesses free of charge.. Cases 6, 10, 14, 16, and
19 can be regarded as randomized controlled experiments because owners
and managers were assigned to treatment and control groups at random.
Three cases (8, 15, and 17) are not randomized controlled experiments.
In each cluster, local business consultants provided management
training in local languages in a classroom setting. An interpreter
accompanied any international consultants. In the clusters in Vietnam
(cases 6 and 14), Ethiopia (case 16), and Tanzania (case 19), we also
offered on-site training several months after the classroom training was
completed. In the on-site training programme, instructors visited
participants' firms to teach them how to adopt useful management
practices. They later visited the firms again to check if the
assimilation was going well and to give further advice. The assignment
to the on-site training was random and independent of the random
assignment to the classroom training.
To measure management practices, we use a simpler version of the
management score developed by Bloom and Van Reenen (2007, 2010) for
medium-sized firms in developed and emerging economies. Our version is
suitable to MSEs in low-income countries. More than half of the elements
of our management score are based on visual inspection by our
enumerators. They base the remainder on the interviews with the owners
or managers of the firm. In each experiment, we observed the management
score and the accounting-based performance indicators of each owner or
manager in the treatment group and the control group before and after
the training programme. In cases 6, 14, 16, and 19, we also observed
management scores between the classroom training and the on-site
training (Survey 2).Another round of surveys has been conducted very
recently in one study site and is planned in the others.
In our randomized experiments, we randomly invited a number of
owners and managers to take part in a training programme. It was up to
them to decide whether or not to attend. Since some did not, we
estimated the local average treatment effects (LATE). The estimated LATE
on the participants was positive and highly significant for the
management score in all the randomized experiments that we conducted
(cases 6, 10, 14, 16, and 19). As mentioned earlier, the LATE on firm
survival is also positive and significant in case 10, even though it
could not be estimated in the other cases because the incidence of exit
was negligibly low. Since owners and managers may not be fully aware of
the value of management skills, it is interesting to examine how
training affects the willingness to pay (WTP for management training. We
asked our sample owners and managers about their willingness to pay
about 400 US dollars before and after four of our experiments (cases 6,
14, 16 and 19). In three cases, the WTP for the participants increased
sizably after the training. In two of these three cases, the WTP also
increased for non-participants in the control group. These WTP results
en suggest that many owners and managers were unaware of the value of
training before taking part, and that the vast majority of training
participants attach high value after taking part. In addition, some
non-participants also become willing to pay, based solely on what they
hear about the programme.
The LATEs on the participants' accounting-based performance
indicators, however, are not statistically significant in case 16. This
is most likely because the follow-up survey was conducted too soon, only
two months after the on-site training was completed and five months
after the classroom training. In cases 6 and 14, a number of MSEs had
not collected receivables from their customers in the short period
before the follow-up survey--so accounting-based performance data are
not available until the next round of survey.
In case 10, where the follow-up survey was conducted a year after
the classroom training programme, the estimated LATE on annual value
added is 13,890 US dollars and is marginally significant. This indicates
that participation in the classroom training for three weeks increased a
participant firm's value added by 13,890 dollars on average
relative to the non-participant firms' value added for the single
year immediately after the training programme. The training effect may
persist for several years, and the estimate of the effect may be diluted
by the improved performance of non-participants due to knowledge
spillovers. Thus, the estimated LATE of 13,890 dollars is likely to
underestimate the social benefit of the training. The cost of the
training per participant is 740 dollars. For the experiment in this
study site, we did not hire international consultants, and the venue was
provided by a nearby vocational school free of charge. The cost per
participant in this study site was relatively low. It could have been
even lower if the same training had been rolled out for a large number
of participants because some costs, such as preparing teaching material
are fixed.
In case 19, the follow-up survey was conducted just three months
after the on-site training had been completed. Since almost all the
invited owners and managers participated in the training in this site,
we estimated the average treatment effect on the participants. It is
estimated that the classroom training increased the annual value added
of firms taking part by 4,181 US dollars. On-site training increased
that by 4,038 dollars on average relative to non-participants. (15) The
classroom training cost per invited firm was 2,905 US dollars, which
includes the fee for an international consultant from Japan and banquet
hall rental paid to in one of the largest hotels in Dar es Salaam. The
on-site training cost per participating firm was 2,043 dollars, which
again includes the fee for an international consultant from Japan.
Despite cost being over-estimated and benefits under-estimated, it is
remarkable and encouraging that the estimated effect tends to exceed the
cost.
Issues for Future Research
It may be interesting to note that training effects depend on the
education level of the training participant, as our studies in Ghana and
Ethiopia, Bjorvatn and Tungodden (2010), and some other studies attest.
For example, educated owners and managers tend to benefit more from
training in record keeping practices probably because record keeping
requires math skills. Berge, et al. (2011) measure preference variables
and argue that female owners and managers perform less well because they
are more risk-averse and less willing to compete than their male
counterparts. In our sample of garment producers in Tanzania (case 19),
female owners and managers comprise the majority, and many of them
operate businesses not to make a living but to have disposable income
that they can spend as they wish, according to our unstructured
interviews with them. Yet, we find no evidence for underperformance of
female owners and managers in terms of business performance. In terms of
management skills, they were somewhat superior to their male
counterparts because many of them had started their businesses after
receiving garment business training provided by some organisations
exclusively for females. In our experimental training programme, male
owners and managers participated and benefited from the training a
little more than female participants. This could be a catch-up rather
than the gender difference in the effect of training.
More generally, there are those training participants who can apply
new ideas to their businesses successfully, and those who cannot acquire
useful knowledge. Indeed different effects in terms of education have
already been reported by several studies, including Mano, et al. (2012).
To the extent that these different effects are more a rule than
exception, the cost-benefit analysis of the training would need to take
them into account. We will come back to this point shortly after the
discussion of spillovers below.
Knowledge spillovers are a vexing problem for researchers who want
to estimate training effects, but they are socially beneficial. To
design better training programmes, we should pay more attention and
devote more efforts to explore what type of knowledge is easy or
difficult to spread, what the major channels of spillovers are, and how
fast and how accurately knowledge is conveyed through spillovers. The
data generated by our experiments in Ethiopia and Tanzania suggests that
knowledge spills overs not only from participants to non-participants
but also from participants to participants. Also, participants benefit
from talking about the training contents to others. In other words, you
can better yourself by observing others and the best way to learn
something is to teach it. To the extent that these tentative results are
robust, it is useful to increase opportunities for participants to
discuss management and to visit each other's firms, for example, by
encouraging them to organise alumni associations.
The methodology featuring randomised controlled trials and the
estimation of the average treatment effect on the treated (ATT) has been
extensively used in medical studies of the effects of medicines.
Spillovers are not a major concern in such studies. By contrast, the
effect of teaching knowledge and skills is likely to be accompanied by
spillovers. If management training generates considerable external
effects, it may not necessarily be a good idea to estimate ATT or LATE
in order to grasp the training impacts. If training contents spill over
widely, we should focus on analysing what types of owners and managers
assimilate training contents better and disseminate them to a greater
number of others. If training contents do not spill over widely, but the
success story of participants spurs many others to actively try to
acquire useful knowledge and skills, it is desirable to concentrate the
limited training resources on nurturing promising owners and managers
and developing success stories or role models. In either case, it is
meaningless to be obsessed by the idea of estimating the impact on
average and to evaluate training programmes through the estimated impact
on average.
To design better training programmes, what to teach and how to
teach are critical issues to be explored. These issues have not received
due attention in existing studies of management training. A major
exception is the study by Drexler, Fischer, and Schoar (2010), which
compares different ways of teaching bookkeeping. Traditionally, Japanese
businessmen, business consultants, and aid workers are fond of teaching
the Japanese style of production and quality management, Kaizen. Itis
"a common-sense, low-cost approach to management" according to
Imai (1997). It is now adopted by a number of large firms in the
developed and emerging economies, often under the name of lean
management. Business development service (BDS) providers who received,
say, the ILO's entrepreneurship and business management training
must have heard of Kaizen, but they do not allocate much time to it when
they serve as instructors in training programmes. However, in our
experience of hiring them as instructors in our pilot projects, they are
excellent at teaching entrepreneurship, business strategy, marketing,
and record keeping. The tentative results of our training experiments
indicate that training participants appreciate both Kaizen-type training
and BDS-type training. Therefore we have designed our programmes to
include both.
While the discussion in this section has focused on management
training, we believe that training programmes for MSE owners and
managers in developing countries should include innovation as a major
subject. As we saw in the previous section, it is vital for these owners
and managers to succeed in product quality improvement, branding,
improvements in marketing, strengthening relationships with suppliers,
and improvements in management of labour, inventory, and finance.
Moreover, since a number of clusters in developed and emerging economies
have achieved these multidimensional improvements in production and
management, a number of them should be able to share their experiences
with owners and managers in survival clusters in the developing world,
especially in SSA. Thus, these issues should be incorporated in training
programmes to cultivate both the innovative and managerial capacities of
owners and managers.
If MSE owners and managers are likely to benefit from learning
diverse subjects, should all subjects are to be taught in the same way.
Given the heterogeneity of owners and managers, the question arises also
as to whether to include them all in the same class. According to our
tentative results classroom training and on-site training have similar
effects in BDS-type training but differ in Kaizen-type training. It is
difficult for training participants to understand how to apply the
knowledge of Kaizen without looking at concrete examples. So for
Kaizenon-site training and visiting other participants' workshops
seem to be more effective. Another related finding is that while the
effect of on-site training on the adoption of Kaizen-type practices does
not depend on the participant's education level, the effect of
classroom training on the adoption of BDS-type practices does. These
preliminary results warrant further studies on how to efficiently teach
management to MSE owners and managers in developing countries.
IV. CONCLUSIONS
The pace of industrial development is largely determined by the
dissemination and adoption of new technologies and good management
practices. The market for management consulting and training services,
however, is likely to fail. This is partly because knowledge spillovers
create a gap between the social and private benefits of acquiring
managerial capacity, and partly because many owners and managers are not
aware of the importance of management knowledge. In this paper we argue
that to help MSEs in developing countries prosper management training
should be supported by government and foreign aid. In Taiwan, the
Industrial Technology Research Institute has facilitated importing
foreign technologies and disseminating them by means of adaptive
research and training [Hong and Gee (1993)]. It is hoped that
Sub-Saharan Africa will have such effective institutions in charge of
adaptive research and training toward improvements in production and
management in the not-too-distant future.
Although this paper focuses on cultivating managerial and
innovative capacities, financial development and infrastructure
development are as important. Management training can have an additional
impact if it is linked with finance and infrastructure development.
High-performing firms require finances not only to invest in capital
goods but also to move from their original industrial clusters to more
spacious and better-equipped industrial zones. If a training programme
is provided before offering low-interest credits, it is relatively easy
to distinguish innovative, promising entrepreneurs. It is desirable to
link management training programmes with the provision of credit and
space in the industrial zone. This way firms with improved managerial
and innovative capacities and, hence, higher ability to repay loans,
will be more likely to receive loans and relocate their production bases
with expanded scale of operations.
Much remains to be studied about the design of management training
programmes for owners and managers, public awareness campaigns,
trainers' training systems, trainers' quality assurance
systems, and the overall strategy of dissemination and adoption. In this
paper we argue, based on our case studies, that management training
programmes are likely to have significant impact. They should facilitate
learning from the experience of multidimensional improvements in
production and management to and cultivate innovative capacities.
Considerably many more empirical studies are warranted.
Tetsushi Sonobe <sonobe te@grips.ac.jp> and Keijiro Otsuka
are affiliated with the National Graduate Institute for Policy Studies,
Tokyo, Japan.
Authors' Note: We thank Prof. Abdul Jalil of Quaid-e-Azam
University for his extremely thoughtful comments. This study was
supported by JSPS Kakenhi Grant Number 25 101002.
REFERENCES
Aghion, P., N. Bloom, R. Blundell, R. Grifitt, and P. Howitt (2005)
Competition and Innovation: An Inverted-U Relationship. Quarterly
Journal of Economics 120:2, 701-28.
Akerlof, G. A. (1970) The Market for 'Lemons': Quality
Uncertainty and the Market Mechanism. Quarterly Journal of Economics
84:3, 488-500.
Altenburg, T. and J. Meyer-Stamer (1999) How to Promote Clusters:
Policy Experiences from Latin America. World Development 27:9,
1693-1713.
Arif, B. W. and T. Sonobe (2012) Virtual Incubation in Industrial
Clusters: A Case Study in Pakistan. Journal of Development Studies 48:3,
377-392.
Berge, L. L, K. Bjorvatn, and B. Tungodden (2011) Human and
Financial Capital for Microenterprise Development: Evidence from a Field
and Lab Experiment. (CM1 Working Paper).
Bjorvatn, K. and B. Tungodden (2010) Teaching Business in Tanzania:
Evaluating Participation and Performance. Journal of the European
Economic Association, 8 (23), 561-70.
Bloom, N. and J. Van Reenen (2007) Measuring and Explaining
Management Practices across Firms and Countries. Quarterly Journal of
Economics 122:4, 1351-409.
Bloom, N. and J. Van Reenen (2010) Why Do Management Practices
Differ across Firms and Countries? Journal of Economic Perspectives
24:1, 203-24.
Bloom, N., A. Mahajan, D. McKenzie, and J. Roberts (2010) Why do
Firms in Developing Countries Have Low Productivity? American Economic
Review 100:2, 619-23.
Bloom, N., B. Eifert, A. Mahajan, D. McKenzie, and J. Roberts
(2011) Does Management Matter? Evidence from India. World Bank. (Policy
Research Working Paper No. 5573).
Bruhn, M., D. Karlan, and A. Schoar (2010) What Capital is Missing
in Developing Countries? American Economic Review 100:2, 629-33.
Calderon, G., J. Cunha, and G. de Giorgi (2012) Business Literacy
and Development: Evidence from a Randomized Trial in Rural Mexico.
Stanford University. (Mimeo).
Drexler, A., G. Fischer, and A. Schoar (2010) Financial Literacy
Training and Rule of Thumbs: Evidence from Field Experiment. (CEPR
Discussion Papers, No. 7994).
Hayami, Y. (2009) Social Capital, Human Capital, and the Community
Mechanism:
Towards a Conceptual Framework for Economists. Journal of
Development Studies 45:1,96-123.
Hong, C.-M. and S. Gee (1993) National Systems Supporting Technical
Advance in Industry: The Case of Taiwan. In R.R. Nelson (ed.) National
Innovation Systems: A Comparative Analysis. Oxford: Oxford University
Press.
Huang, Y. and A. M. Bocchi (eds.) (2008) Reshaping Economic
Geography in East Asia, Washington, DC: World Bank.
Iddrisu, A., Y. Mano, and T. Sonobe Entrepreneurial Skills and
Industrial Development: The Case of a Car Repair and Metalworking
Cluster in Ghana. Journal of the Knowledge Economy 3:3, 302-326.
Imai, M. (1977) Gemba Kaizen: A Commonsense: Low-cost Approach To
Management. New York: McGrow-Hill.
Karlan, D. and M. Valdivia (2011) Teaching Entrepreneurship: Impact
of Business Training on Microfinance Clients and Institutions. Review of
Economics and Statistics 93:2,510-27.
Long, C. and X. Zhang (2011) Cluster-Based Industrialization in
china: Financing and Performance. Journal of International Economics
84:1, 112-23.
Malmendier, U. and G. Tate (2005) CEO Over confidence and Corporate
Investment. Journal of Finance 60:6, 2661-700.
Mano, Y., A. Iddrisu, Y. Yoshino, and T. Sonobe (2012) How Can
Micro and Small Enterprises in Sub-Saharan Africa Become More
Productive? The Impacts of Experimental Basic Managerial Training. World
Development 40:3, 458-68.
McKenzie, D. and C. Woodruff (2014) What Are We Learning from
Business Training and Entrepreneurship Evaluations around the Developing
World. World Bank Research Observer 29:1, 48-82.
O'Donoghue, T. and M. Rabin (1999) Doing It Now or Later?
American Economic Review 89:1, 103-24.
Rodriguez-Clare, A. (2007) Clusters and Comparative Advantage:
Implications for Industrial Policy. Journal of Development Economics
82:1, 43-57.
Ruan, J. and X. Zhang (2009) Finance and Cluster Based Industrial
Development in China. Economic Development and Cultural Change 58:1,
143-64.
Shleifer, A. and R. W. Vishny (1993) Corruption. Quarterly Journal
of Economics 108:3,599-617.
Sonobe, T. and K. Otsuka (2006) Cluster-Based Industrial
Development: An East Asian Mode. Basingstoke, UK: Palgrave Macmillan.
Sonobe, T. and K. Otsuka (2011) Cluster-Based Industrial
Development: A Comparative Study of Asia and Africa. Basingstoke, UK:
Palgrave Macmillan.
Sonobe, T., J. E. Akoten, and K. Otsuka (2011) The Growth Process
of Informal Enterprises in Sub-Saharan Africa: A Case Study of a
Metalworking Cluster in Nairobi. Small Business Economics 36: 3,
323-335.
Sonobe, T., A. Suzuki, and K. Otsuka (2011) KAIZEN for Managerial
Skills Improvement in Small and Medium Firms: An Impact Evaluation
Study, Japan PHRD Trust Fund No. TF096317, Washington, DC: World Bank.
Sonobe, T., D. Hu, and K. Otsuka (2004) From Inferior to Superior
Products: An Inquiry into the Wenzhou Model of Industrial Development in
China. Journal of Comparative Economics 32:3, 542-63.
Sonobe, T., J. E. Akoten, and K. Otsuka (2009) An Exploration into
the Successful Development of the Leather-Shoe Industry in Ethiopia.
Review of Development Economics 13:4,719-736.
Syverson, C. (2011) What Determines Productivity? Journal of
Economic Literature 49:2, 326-65.
World Bank (2012) World Development Report 2013--Jobs. Washington,
DC: World Bank.
Yoshino, Y. (2011) Industrial Clusters and Micro and Small
Enterprises in Africa: From Survival to Growth. Washington, DC: World
Bank.
Comments
Thank you, Dr Sonobe for a really fruitful presentation. It is
really an honour for me to be a discussant on the work of Dr Sonobe.
But, unfortunately, I did not have your paper for a detailed discussion.
However, I went through the abstract of your paper and come up with some
brief comments and questions.
Dr Sonobe your presentation tries to put many things together. For
example, the cluster-based industries, Kaizen Management and MSE growth.
Surely your presentation/paper is a coherent explanation of title and
subtitles. However, I have just few comments and question for more
understanding the subject.
(1) The Clusters must be expected to display the negativities of
asymmetric knowledge. How do you think how the clusters may work in this
situations?
(2) I just wonder how the concept of KAIZEN management is different
than Lean Manufacturing. In my understanding, lean manufacturing also
reduces costs, lead times, and inventory requirements, ensures greater
productivity, and brings about overall efficiency. Similarly, what is
differences between 5s approach and Six sigma approach.
(3) The KAIZEN management has a constant focus on improvement
becomes an obsession and causes stress in the workforce. How do you
think about the stress management in your KAIZEN management model?
(4) No Margins for Error.
(5) There is no margin of error in KAIZEN management arrangements.
Apparently, it requires a favourable external climate. For instance is
not possible in places with unreliable energy supply, inadequate
transportation infrastructure, and or poor work culture in the society.
Similarly, in KAIZEN management arrangements.
(6) Is management training enough? Surely, Technology is also
important, Credit/finance is also important, Infrastructure is also
important
(7) The KAIZEN arrangement over-focus on elimination of waste, and
it overrides other concerns. For example, it ignores other crucial
parameters such as employee wellness, and corporate social
responsibility. Please comment.
(8) KAIZEN is more a culture than a method, and there is no
standard KAIZEN production model.
Abdul Jalil
Quaid-i-Azam University, Islamabad.
(1) See, for example, Bloom and van Reenen (2007, 2010), Bloom,
Eifert, et al. (2011), Bloom, Mahajan, et al. (2010), Bruhn, Karlan, and
Schoar (2010), and Syverson (2011).
(2) In this paper, we refer to owners or other ultimate decision
makers of enterprises as business owners and managers to avoid confusion
between entrepreneurs and entrepreneurship. The term entrepreneur is
used to denote owners and managers with varying entrepreneurial skills.
(3) We define an industrial cluster simply as the localization of
firms producing similar and related products (e.g., final products and
parts).
(4) Our approach is consistent with the theoretical proposition of
Rodriguez-Clare (2007) that the best industrial policy entails the
direct promotion of industrial clusters in the sector in which the
country has a comparative advantage.
(5) As reported by Huang and Bocchi (2008). Long and Zhang (2011),
and Sonobe and Otsuka (2006).
(6) We have found that the number of business people operating in
our study sites in SSA has not been captured by statistical and tax
authorities, and that few clusters in SSA were known to researchers.
(7) Although printed circuit boards (PCBs) are core devices for
electronics items and IT equipment, the PCBs produced in the clusters
listed in Table 1 in their early days were technologically simple and
produced without using expensive machinery.
(8) Hayami (2009) argues that the community mechanism is effective
not only for the enforcement of contract-based transactions but also for
the management of local commons and the provision of local public goods
in rural communities. In this paper, however, we focus on the
enforcement of contract-based transactions.
(9) Some enterprise managers in East African countries, however,
are learning from other countries by employing technicians and
managerial advisers from Asia and by visiting European countries.
(10) See, for example, Karlan and Valdivia (2011), Drexler,
Fischer, and Schoar (2010), Bruhn, Karlan, and Schoar (2010), Bjorvatn
and Tungodden (2010), and Mano, et al. (2012).
(11) In the panel of Forbes 500 CEOs, Malmendier and Tate (2005)
find that overconfident CEOs over estimate investment returns.
(12) There is growing literature on procrastination [e.g.,
O'Donoghue and Rabin (1999)]. A number of randomized controlled
trials have confirmed the existence of procrastination.
(13) There are already a number of management or business training
programmes linked with microfinance provision, including the training
programmes discussed by Karlan and Valdivia (2011) and Bjorvatn and
Tungodden (2010).
(14) For a more comprehensive and rigorous discussion of the
technical problems of the management experiments, see an excellent
literature survey by McKenzie and Woodruff (2012).
(15) Note that the effect of the on-site training would be more
difficult to be realised by the time of the follow-up survey because the
on-site training was implemented after the classroom training.
Table 1
Origin of Development of Selected Industrial Clusters in Asia and
Africa
Location Main Product
1 Bingo, Japan Working clothes
2 Hamamatsu, Japan Motorcycles
3 Taichung, Taiwan, China Machine tool
4 Zhili, Zhejiang, China Baby clothes
5 Wenzhou, Zhejiang, Electrical fittings
China
6 Bac Ninh, Vietnam Rolled steel bars
7 Sargodha, Pakistan Electric fittings
8 Addis Ababa, Garment
Ethiopia
9 Nairobi, Kenya Garment
10 Kumasi, Ghana Metalwork
11 Northern Taiwan, China Printed circuit
board
12 Chongqing, China Motorcycles
13 Three cities in Jiangsu, Printed circuit
China board
14 Hatay, Vietnam Knitwear
15 Addis Ababa, Leather shoes
Ethiopia
16 Addis Ababa Metalwork
Ethiopia
17 Nairobi, Kenya Metalwork
18 Dhaka, Bangladesh Garment
19 Dar es Salaam, Garment
Tanzania
Location Initial Firm Type and Initial Product
1 Bingo, Japan Farm households (traditional clothes)
2 Hamamatsu, Japan Small- and medium-scale (woodwork)
3 Taichung, Taiwan, China Small- and medium-scale (simple
machinery)
4 Zhili, Zhejiang, China Farm households (hand-made silk
products)
5 Wenzhou, Zhejiang, Farm households (repair parts for
China electrical fittings)
6 Bac Ninh, Vietnam Farm households (agricultural
implements)
7 Sargodha, Pakistan Small- and medium-scale (electrical
fittings)
8 Addis Ababa, Taylors (tailored suits)
Ethiopia
9 Nairobi, Kenya Taylors (tailored dresses)
10 Kumasi, Ghana Small- and medium-scale (car repair)
11 Northern Taiwan, China FDIs (printed circuit boards)
12 Chongqing, China Small- and medium-scale and FDIs
(motorcycle)
13 Three cities in Jiangsu, SOEs and TVREs (printed circuit
China boards)
14 Hatay, Vietnam Cooperative (knitwear for SOEs)
15 Addis Ababa, Migrant artisans (leather shoes)
Ethiopia
16 Addis Ababa Migrant artisans (metal work)
Ethiopia
17 Nairobi, Kenya FDIs and SOEs (metalwork)
18 Dhaka, Bangladesh Training in Korea (garment)
19 Dar es Salaam, UNIDO training (garment)
Tanzania
No. of Entrepreneur's
Firms in Schooling
Location Cluster Years
1 Bingo, Japan 200 10.6
2 Hamamatsu, Japan 150 Na
3 Taichung, Taiwan, China 100 13.0
4 Zhili, Zhejiang, China 5000 7.5
(inch
subcons)
5 Wenzhou, Zhejiang, 200 10.6
China
6 Bac Ninh, Vietnam 133 6.7
7 Sargodha, Pakistan 1200 9.4
8 Addis Ababa, 700 10.1
Ethiopia
9 Nairobi, Kenya 640 8.6
10 Kumasi, Ghana 500 11.0
11 Northern Taiwan, China 60 13.2
12 Chongqing, China 50 15.1
13 Three cities in Jiangsu, 200 12.1
China
14 Hatay, Vietnam 160 7.9
15 Addis Ababa, 1000 9.2
Ethiopia
16 Addis Ababa 130 10.8
Ethiopia
17 Nairobi, Kenya 150 11.0
18 Dhaka, Bangladesh 4100 15.0
19 Dar es Salaam, 700 10.7
Tanzania
Notes: FDIs, SOEs, and TVREs stand for foreign direct investments,
state-owned enterprises, and township-village-run enterprises,
respectively. Sources: Cases 1, 2, 3, 4, 5, 11, 12, and 13 are from
Sonobe and Otsuka (2006). Cases 6, 7, 8, 9, 14, 15, 16, and 17 are
from Sonobe and Otsuka (2011). Case 10 is from Iddrisu, el at. (2012)
and Mano, el at. (2012). Cases 16 and 19 are from Sonobe, Suzuki, and
Otsuka (2011). Case 18 is from Mottaleb and Sonobe (2012).
Table 2
The Number of Jobs, Monthly Wages, and Sex and Schooling Compositions
of Workers in Selected Industrial Clusters
No. of
Manufacturing
No. (1) Cluster Year Jobs (1000)
6 Rolled steel, Vietnam 2010 3.3
7 Elec, fittings, Pakistan 2008 13.2
8 Garment, Ethiopia 2007 7.2
10 Metalwork, Ghana 2003 6.7
14 Knitwear, Vietnam 2010 3.2
15 Leather shoes, Ethiopia 2008 22.2 (5)
16 Metalwork, Ethiopia 2009 9.1
17 Metalwork, Kenya 2008 1.2
18 Garment, Bangladesh 2004 2,100 (9)
19 Garment, Tanzania 2010 0.6
% of High
School
Total No. of Jobs Monthly %of Education
No. (1) (1000) (2) Wage(US$) (3) Women and above
6 n.a. 161 5 0
7 n.a. 105 10 18
8 n.a. 36-45 (4) 59 17 (5)
10 99.2 (6) 32 0 25
14 20-30 (7) 59 70 25
15 n.a. 35-53 (4) 23 15 (8)
16 n.a. 63-88 (4) 9 44 (8)
17 1.7 117-208 (9) 0 43
18 n.a. 61-97 (11) 40 50
19 n.a. 57 83 (12) 11
Notes:
(1.) The numbers are the same as in Tables 1 and 2.
(2.) Including related jobs in the cluster (e g., car repairing in
the metal work cluster).
(3.) Converted to US dollars by the prevailing exchange rate.
(4.) The lower number refers to female wage and the higher one to
male wage.
(5.) The number refers to 2009.
(6.) Including 80 thousand jobs generated by about 8 thousand
garages.
(7.) Including more than 5,000 subcontractors in nearby villages.
(8.) Including those who have completed vocational training.
(9.) The lower number refers to casual worker wage and the higher
one to regular worker wage.
(10.) The data include the employment in the country as a whole
including the Greater Dhaka and Chittagong.
(11.) The lower number refers to casual worker wage and the higher
one to regular worker wage.
(12.) Proportion of female owners, not workers.
