Ownership, technology and buyers:

explaining exporting in
China and Sri Lanka

Ganeshan Wignaraja *

This paper examines several characteristics besides foreign ownership

that influence the decision of clothing firms in China and Sri Lanka
whether or not to export – namely, the acquisition of technological
capabilities and learning from buyers. As a by-product of the exercise,
the model also describes the effect of other explanatory variables
(capital, skill adjusted wages and age). The findings indicate that foreign
ownership, the acquisition of technological capabilities and learning from
buyers are positive and significantly correlated with the probability of
exporting in Chinese and Sri Lankan clothing firms. Skill adjusted wages
are also significant and with the expected negative sign. Comparative
econometric analysis is a powerful tool to verify and extend the findings
of detailed enterprise case studies on innovation and learning processes
in developing countries.

Key words: foreign investment, technological capabilities, buyers of

output, exports, China, Sri Lanka
JEL Classification: F14, O31, L67

1. Introduction
There is a large literature on the determinants of international trade
across countries and industries. With the increased availability of firm-level
surveys, there has been growing attention to firms’ export behaviour using
econometric analysis (for surveys see Bleaney and Wakelin, 2002; Rasiah,
2004; and Greenaway and Keller, 2007). Drawing on the literature on applied
international trade and investment as well as that on innovation and learning,
attempts have been made to explain why some firms are better exporters than

* Principal Economist, Asian Development Bank, Manila. Email address: gwignaraja@

adb.org. The views expressed here are solely mine and are not to be attributed to the Asian
Development Bank. I am indebted to the late Professor Sanjaya Lall for introducing me to
LQGXVWULDOWHFKQRORJLFDOGHYHORSPHQW7KDQNVDUHGXHWR5RVHFKLQ2O¿QGRIRUHI¿FLHQWUHVHDUFK
assistance and to two reviewers for comments. I am also grateful to suggestions from participants
at seminars at UNUMERIT, Maastricht in June 2006, the University of Colombo in December
2006 and Renmin University, Beijing in April 2008.
others. A positive relationship between foreign ownership and firm-level
export behaviour emerges from several studies (Lall, 1986; Wilmore,
1992; Rasiah, 2003; Correa et al., 2007; Du and Girma, 2007). The
superior export behaviour of foreign firms relative to domestic firms is
typically attributed to access to the marketing connections and know-how
of their parent companies coupled with accumulated learning experience
of producing for export. Research and development (R&D) intensity
(and innovation more generally) has also been found to have a positive
effect on export behaviour at firm-level (Kumar and Siddharthan, 1994;
Ito and Pucik, 1993; Bleaney and Wakelin, 2002).
Case studies of firms have long indicated that exporters in
developing countries rarely undertake formal R&D activities at frontiers
of technology. Instead, they focus on the difficult process of acquiring
technological capabilities to use imported technologies efficiently and
learning from buyers of output (e.g. Lall, 1987, 1992; Rhee, 1990; Ernst
et al., 1998; and Keesing and Lall, 1992; Wignaraja, 1998; Mathews and
Cho, 2002). However, there has been limited econometric analysis to
date to verify the findings of case studies. Further econometric study of
innovation and learning processes will provide statistical confirmation
of case study findings and significantly improve our understanding of
firm-level exporting behaviour in developing countries.
This paper examines a variety of characteristics besides foreign
ownership that influence a firm’s decision of whether or not to export –
namely, the acquisition of technological capabilities and learning from
buyers. As a by-product of the exercise, the model also describes the
effect of other explanatory variables such as capital, wages and age in
production. Background studies and hypotheses are reviewed in section
2. The results of Probit estimates carried out on samples of 353 clothing
firms in China (surveyed in 2003) and 205 clothing firms in Sri Lanka
(surveyed in 2004) are presented in section 3. Both economies sought
to promote exports and attract foreign investment by adopting outward-
oriented policies in the late-1970s. An improved incentive regime and
inward investment has facilitated China’s rapid emergence as one of the
world’s largest clothing exporters and allowed Sri Lanka to achieve the
highest clothing exports per capita in South Asia. This paper suggests
that technological and marketing factors also underlie export success at
firm-level in both countries. The econometric results indicate that foreign
ownership, acquisition of technological capabilities and learning from
foreign buyers are positively associated with the probability of exporting
in Chinese and Sri Lankan clothing firms. Skill adjusted wages are also
significant and with the expected negative sign. Section 4 concludes.

2 Transnational Corporations, Vol. 17, No. 2 (August 2008)

2. Background and hypotheses
2.1 Literature
The analysis of firm-level export performance has attracted
the attention of two related schools of applied economics. Relatively
recently, applied international trade and investment specialists have
explored the effects of the theoretical determinants of comparative
advantage on firm-level export performance. This literature (which has
roots in the neo-Heckscher-Ohlin Model and the neotechnology theories)
suggests that the theoretical determinants of comparative advantage,
which are traditionally recognized as industry-level factors,1 can also
operate at firm-level (see, for instance, Lall, 1986; Dunning, 1993;
Kumar and Siddharthan, 1994; Bleaney and Wakelin, 2002). Conditions
of imperfect markets with widespread oligopoly as well as differences
in technologies, learning and tastes underlie the notion of firm-specific
advantages. It follows that almost all theories of comparative advantage
can be firm-specific determining not only which countries will enjoy a
comparative advantage in international markets but also which firms can
exploit that comparative advantage better than others. Incorporating the
notion of firm-specific advantages somewhat modifies the predictions
of the theories of international trade as follows: (1) there are country-
specific and industry-specific advantages which apply to all firms
equally; and (2) within this, some advantages will be firm-specific since
certain managerial, organizational, marketing and other skills will be
peculiar to each firm as will production methods, technologies and
experience based know-how.
The other group with an interest in firm-level export behaviour is
the literature on technological capabilities. Focusing on innovation and
learning processes in developing countries, this literature emphasizes
the acquisition of technological capabilities as a major source of
export advantage at firm-level (see Lall, 1987, 1992; Ernst et al., 1998;
Mathews and Cho, 2002; Rasiah, 2004; Nelson, 2008). Drawing on
the evolutionary theory of technical change, the capability literature
underlies the difficult firm-specific processes involved in building
technological capabilities to use imported technology efficiently. The
1
The major trade theories (the Heckscher-Ohlin Model, theories of economies of
scale and oligopolistic competition, the neo-technology theories and theories of economic
geography) attribute the export performance of an open developing economy to its
comparative advantage over another in terms of access to certain factor inputs – capital,
labour, economies of scale, technology and geography (for a survey see Deardorff,
2005). Empirical applications to developing countries have sought to explain the export
performance of each industry/product in terms of their various characteristics.

Transnational Corporations, Vol. 17, No. 2 (August 2008) 3

central argument is that firms have to undertake conscious investments
in search, training, engineering and, even research and development,
to put imported technologies to productive use. Furthermore, capability
building rarely occurs in isolation and involves active cooperation
between firms, buyers of output and support institutions for technology
and export marketing. Buyers of output have been especially helpful
in supporting the firm-level learning in consumer goods industries
like textiles and clothing by providing marketing advice and technical
knowledge (Rhee, 1990; Keesing and Lall, 1992). Hence, differences
in the efficiency with which firm-level capabilities are created are
themselves a major source of competitive advantage.
It is challenging, however, to measure inter-firm differences in
technological capabilities in developing countries. In the last decade
or so, studies have begun to develop a simple summary measure of
technological capabilities by ranking the technical functions performed
by enterprises (see the pioneering work on Thailand by Westphal et
al., 1990).2 The ranking procedure integrates objective and subjective
information into measures of a firm’s capacity to set up, operate and
transfer technology. The typical approach is to highlight the various
technical functions performed by enterprises and to award a score for
each activity based on the assessed level of competence in that activity.
An overall capability score for a firm is obtained by taking an average
of the scores for the different technical functions. As discussed below,
the overall capability score (often referred to as a technology index or
TI) has proved robust in statistical analysis of export and technological
performance.
The increasing availability of large micro-level datasets,
particularly for developing countries, has stimulated econometric
research at firm-level rather than country or industry-level. This research
has sought to test the importance of the theoretical determinants of
comparative advantage as well as technological capabilities at firm-
level. Multiple regressions (OLS, Tobit, Probit and Heckman selection
models) were run relating export behaviour to various enterprise
characteristics (including foreign ownership, R&D and technological
capabilities, advertising, firm size, skill intensity and capital intensity).
The results from selected studies on China and other developing
countries can be highlighted.3 A study by Zhao and Li (1997) tested the
2
More recent applications include Pakistan by Romijn (1999), Mauritius by
Wignaraja (2002), and China by Guan and Ma (2003).
3
Similar studies include Wilmore (1992) on Brazil, Kumar and Siddharthan (1994)
on India, Wignaraja (2002) on Mauritius, Bhaduri and Ray (2004) on India, and Rasiah
(2006) on South Africa.

4 Transnational Corporations, Vol. 17, No. 2 (August 2008)

relationship between R&D and export propensity in manufacturing firms
in China and found R&D and firm size to be positive and significant
determinants. Capital intensity was also significant but with a negative
sign. A study of Chinese firms by Guan and Ma (2003) reported that
firm-level export performance is positively associated with an index of
innovative capability and firm size. More recently, Du and Girma (2007)
report that foreign ownership, access to finance and product innovation
were found to be positively associated with the propensity to export in
firms in China.
In an early study of Indian engineering and chemicals firms, Lall
(1986) found evidence for technological determinants of enterprise
exporting. Foreign equity was found to be significant in chemicals,
licences were highly significant in engineering, and R&D was significant
in both industries (but with opposite signs). Rasiah (2003) examined the
influence of ownership, R&D expenditure, age and skills in determining
exports in electronics firms in Malaysia and Thailand. All four variables
had positive signs and were significant. Correa et al. (2007) report that
R&D, firm size and foreign ownership are positively associated with
exporting behaviour in firms in Ecuador. Finally, Wignaraja (2008)
found that geographical location, human capital, size, ownership and a
technology index were significant and positively associated with firm-
level export performance in Sri Lanka. Thus, econometric studies have
generally confirmed the importance of the theoretical determinants of
comparative advantage as well as technological capabilities at firm-
level in developing countries.
2.2 Hypotheses
Building on the econometric literature on firm-level exporting
discussed above, this paper estimates separate functions on the probability
of exporting for clothing firms in China and Sri Lanka:
Y = BX
BX + E , (1)
where Y is the vector denoting the probability of exporting at firm-level,
X is the matrix of explanatory variables, B is the matrix of coefficients,
and E is the matrix of error terms. The dependent variable of the model,
Y, is a binary variable taking the value of one if the firm is an exporter
and zero if the firm is a non-exporter.
The hypotheses and explanatory variables in X in equation (1) are
described below.

Transnational Corporations, Vol. 17, No. 2 (August 2008) 5

Foreign ownership
From existing empirical studies, the share of foreign equity
(FOR) is expected to have a positive influence on the probability of
exporting (Lall, 1986; Wilmore, 1992; Rasiah, 2003; Correa et al., 2007;
Du and Girma, 2007). There are two a priori reasons. First, access to
the marketing connections and know-how of their parent companies as
well as accumulated learning experience of producing for export make
foreign affiliates better placed to tap international markets than domestic
firms.4 Second, foreign firms tend to be larger than domestic firms and
therefore better placed to reap economies of scale in production, R&D
and marketing. A large firm will be better able to exploit such scale
economies and enjoy greater efficiency in production, enabling it to
export more.

Technological capabilities
We expect technological capabilities to be positively associated
with the probability of exporting. Case studies and econometric work
indicates that the learning process in enterprises is not just a simple
function of years of production experience but of more conscious
investments in creating skills and information to operate imported
technological efficiently (see Westphal et al., 1990; Ernst et al., 1998;
Rasiah, 2003, 2006; Wignaraja, 2002, 2008; Guan and Ma, 2003). Such
investments would include search, training and engineering activities.
In the tradition of Westphal et al. (1990), a firm-level technology index
(TI) has been developed to represent technological capabilities. The
TI used here is a simple production capability based variant of indices
based on the Lall (1992) taxonomy of technological capabilities. It was
constructed by ranking a clothing firm’s competence across a series
of technical functions and the results were normalized to give a value
between 0 and 1 (see appendix 1 for details of the TI).

Foreign buyers
Marketing and information links, and associated learning
processes are an under-studied area in the econometric literature on
firm-level exporting. New developing country export firms in consumer
goods industries rarely engage in independent export marketing efforts
including advertising. Instead, case studies suggest that they typically

4
See Dunning (1993) for a discussion of the ownership advantages of
transnationals.

6 Transnational Corporations, Vol. 17, No. 2 (August 2008)

manufacture to orders from buyers from industrial countries (see
Rhee, 1990; Keesing and Lall, 1992). Buyers’ help (or that of technical
consultants) is indispensable in showing new and potential exporters how
to meet the price, quality and delivery requirements of demanding export
markets. Equipment and technical assistance are frequently provided
by buyers to purchase new equipment and improve technological
capabilities (including quality management, control inventory and
product designs). Accordingly, the presence of a marketing relationship
with a leading buyer of output is considered to be positively associated
with the probability of exporting. A dummy variable (BUYER)  which
takes a value of 1 when a marketing relationship with a buyer is present
 is used to represent such a relationship.

Age
As firms with experience are regarded as enjoying greater
experimental and tacit knowledge, age is considered to be positively
associated with the probability of exporting and the building capabilities
(Rasiah, 2003). Age is represented by the absolute age of the firm in
number of years (AGE).

Capital
For capital-poor developing countries, the Heckscher-Ohlin trade
theory predicts a negative relationship between capital intensity and
exports and a positive relationship between capital intensity and imports.
Some econometric studies (e.g. Wilmore, 1992; Zhao and Li, 1997) have
confirmed the predicted negative relationship between capital intensity
and the probability of exporting at firm-level. Accordingly, trade theory
may be useful in predicting whether or not a firm will export. Capital
is difficult to measure and the proxy used by empirical studies depends
on data availability. Capital is represented by fixed assets capital per
employee (CAP).

Skill adjusted wages

In skill-poor developing countries, the Heckscher-Ohlin trade
theory predicts a negative sign for skill intensity in export functions
and empirical evidence at firm-level verifies this prediction (Wilmore,
1992; Bhavani and Tendulkar, 2001). Skill intensity is represented by

Transnational Corporations, Vol. 17, No. 2 (August 2008) 7

skill-adjusted wages (WAGE).5 For a given level of material intensity,
the lower the wage share, the lower is the (skill adjusted) wage rate in
relation to labour productivity, the more likely a firm has a comparative
advantage in exporting. Thus, skill-adjusted wages are expected to have,
ceteris paribus, a negative association with the probability of exporting.

3. Data and empirical findings

3.1 Data and t-test
The data used come from the World Bank Investment Climate
Surveys, conducted in 2003 for China and 2004 for Sri Lanka.6 These
surveys provide a representative sample of the population of clothing
firms in both countries by selecting firms on a largely random basis
using a stratified simple random sample design. Summary descriptive
statistics for 353 clothing firms in China and 205 firms in Sri Lanka are
provided in table A1. The Chinese sample includes 59 foreign-owned
firms while the Sri Lankan sample includes 47 foreign-owned firms.
Apart from ownership, these samples cover a wide range of market-
orientation, size, age groups and technology levels.
Table 1 reports t-test results on mean values for a variety of firm
characteristics. The comparison considers clothing exporters and non-
exporters. Exporters are defined as continuing and new exporters in
China in 2003 and Sri Lanka in 2004. Non-exporters are defined as the
rest of the firms.
There is a significant difference in foreign equity between
exporters and non-exporters in China and Sri Lanka. This is probably
the most striking difference between exporters and non-exporters. On
average, exporters in China have 4.8 times more foreign equity than
non-exporters while exporters in Sri Lanka have 6.1 times more foreign
equity.

5
Bhavani and Tendulkar (2001), among others, argue that it is not just cheap
labour (a low wage rate per worker) that results in a comparative cost advantage but
a low wage in relation to productivity of that labour. The skill adjusted wage rate in
UHODWLRQWRSURGXFWLYLW\DW¿UPOHYHOLVGH¿QHGDVIROORZV:6 :(6(ZKHUH:
WKHZDJHELOO6 YDOXHRIVDOHVDQG( QXPEHURIHPSOR\HHV
6
Private contractors conduct these surveys on behalf of the World Bank. The Sri
Lanka survey was conducted in collaboration with the Asian Development Bank. See
www.enterprisesurveys.org for details of the China and Sri Lanka surveys.

8 Transnational Corporations, Vol. 17, No. 2 (August 2008)

 Technology differences (as indicated by the technology index,
TI) between exporters and non-exporters are also significant in both
countries. Interestingly, the TI gap between Chinese exporters and
non-exporters is somewhat narrower than that between Sri Lankan
exporters and non-exporters. The value of TI for Chinese exporters is
0.47 compared with 0.43 for non-exporters. Meanwhile, the TI value for
Sri Lankan exporters is 0.50 compared with 0.40 for non-exporters.
Table 1. Mean characteristics of clothing exporters and non-exporters in
China in 2003 and Sri Lanka in 2004

China Sri Lanka

Non- Non-
Characteristics
Exporters Exporters Exporters Exporters
(n=130) (n=223) t-values (n=119) (n=86) t-values
Foreign equity, % 22.91 4.80 6.02*** 28.77 4.72 4.85***
Technology Index (0 to 1) 0.47 0.43 2.13** 0.50 0.40 3.17***
$JHRI¿UPQXPEHURI\HDUV 12.83 16.79 -2.64*** 16.99 23.18 -2.80***
Capacity utilization, % 82.12 69.43 4.81*** 80.40 70.41 3.91***
Fixed assets per employee, US$ 4.63 8.83 -0.84 2.91 3.10 -0.14
Wage bill, % sales 12.22 23.60 -4.00*** 25.34 50.65 -4.67***
No. of permanent employees
No. 471 245 3.38*** 683 96 5.81***
Source: author’s analysis.
t-values for two-sample t test with equal variance: mean(exporter) – mean(non-exporters); ***
significant at 1% level, ** at 5% level, and * at 10% level.

Exporters are younger than non-exporters in both countries. The

average age for an exporter in China is just under 13 years while that for
a Sri Lankan exporter is 17 years. Non-exporters are 16.8 years and 23.2
years, respectively.
Looking at capacity utilization, once again we observe significant
differences between exporters and non-exporters. Capacity utilization
levels in exporters are at least 10 percentage points higher than non-
exporters in both countries.
The wage bill as a percentage of sales and firm size are two
additional characteristics that differ between exporters and non-
exporters.
The Chinese and Sri Lankan samples reproduce some of the
stylized facts reported by the literature on exporting. The stylized facts are
consistent with the studies reported in section 2. By applying the t-tests,
which are a useful descriptive device, we can establish that exporters
have higher foreign ownership, are technologically more sophisticated
and have higher capacity utilization levels. These differences alone do

Transnational Corporations, Vol. 17, No. 2 (August 2008) 9

not shed much light on causation. Hence, we develop a regression model
below.
3.2 Econometric analysis
A general to specific modelling approach was adopted for
econometric testing. Initially, the general model (with all the explanatory
variables mentioned in section 2) was estimated. Then a specific model
or reduced form was estimated with only the significant variables.
Table 2 shows the estimated Probit models. Estimated equations (1) and
(3) report the general models for Chinese and Sri Lankan firms while
equations (2) and (4) show the reduced form models with only the
significant variables.
The results of equation (2) for Chinese clothing firms can be
considered following diagnostic testing.7 The pseudo R2 in equation (2)
is acceptable for a cross-section model. Of the six original independent
variables in equation (1), five are significant (three at the 1% level) and
have the expected sign.
The findings underline the critical link between three
complementary factors and the probability of exporting in clothing firms
in China. First, FOR is positive and significant (1% level) which indicates
that foreign ownership is associated with the probability of exporting in
Chinese firms. The explanation seems to lie in a combination of access
to marketing connections and know-how of their parent companies,
accumulated learning experience of producing for export, and economies
of scale linked to firm size. Second, TI is significant (10% level) and
positive. This emphasizes that investments in creating the requisite
technological capabilities to operate imported technology efficiently
is linked to the probably of exporting. Third, BUYER is significant
(1% level) and with the correct sign. This suggests that a marketing
relationship with a foreign buyer of output increases the probability
of exporting at firm-level. Finally, the control variables suggested
by trade theory  CAP (at the 10% level) and WAGE (1% level) are
also significant and with the expected negative sign. Accordingly, the
predictions of the Heckscher-Ohlin trade theory receive support from
firm level analysis in the case of China.

7
6LQFHWKHGDWDVHWFRQWDLQV¿UPVRIDOOVL]HVGLIIHUHQWRZQHUVKLSVWUXFWXUHH[SRUW
orientation, among others, the Probit estimation used the robust standard errors to
account for mild heteroskedascity that is expected in the dataset. Furthermore, correlation
analysis indicated no large correlations between any of the independent variables.

10 Transnational Corporations, Vol. 17, No. 2 (August 2008)

 The results for the Sri Lankan clothing firms are similar,
indicating that certain factors are closely associated with the probability
of exporting at firm level. The pseudo R2 in equation (4) is better than
equation (2) and four of the independent variables are significant in the
reduced form equation. The three complementary factors are significant
and with the correct sign. While TI is significant at the 10% level, FOR
and BUYER are significant at the 1% level. Furthermore, WAGE has
the expected negative sign and is also significant (5% level).
Table 2. Probit estimates of export behavior of garments firm
Binary Variable: Exporter (1) and Non-exporter (0)

China Sri Lanka

Independent Variables
(1) (2) (3) (4)
FOR 0.0158 0.0158 0.0103 0.0103
(5.29)*** (5.37)*** (2.75)*** (2.71)***
TI 0.7942 0.7958 1.0175 0.9572
(1.82)* (1.82)* (1.92)* (1.91)*

BUYER 0.5224 0.5223 1.4547 1.5533

(2.61)*** (2.61)*** (5.81)*** (6.48)***
AGE -0.0010 -0.0057
(-0.15) (-0.93)

CAP -0.0070 -0.0070 -0.0000

(-1.92)* (-1.92)* (-0.17)

WAGE -0.0263 -0.0265 -0.0127 -1.3140

(-4.76)*** (-4.90)*** (-2.25)** (-2.40)**
Constant -0.7169 -0.7297 -0.5417 -0.6436
(-2.50)** (-2.62)*** (-1.56) (-2.17)**

n 314 314 171 180

Wald Ȥ2 62.86*** 62.88*** 54.59*** 57.28***
Pseudo R2
Pseudo 0.17 0.17 0.38 0.39
Logg pseudo
Log p likelihood -172.09 -172.11 -70.93 -73.57
Source: Author’s analysis.
z values are in parentheses; *** significant at 1% level, ** at 5% level, and * at 10% level.
Coefficients were estimated using robust standard errors.

4. Conclusion
The paper uses a rich microeconomic dataset to explore the
determinants of a firm’s decision of whether or not to export in clothing

Transnational Corporations, Vol. 17, No. 2 (August 2008) 11

firms in China and Sri Lanka. It emphasizes that several factors must be
taken into account to explain the decision to export at firm level. Firm-
level export functions were estimated using a Probit model for Chinese
and Sri Lankan clothing firms with proxies for foreign ownership,
technological capabilities, learning from buyers and standard control
variables (capital intensity, skill adjusted wages and firm age). As a part
of the exploratory data analysis, t-tests were also conducted on exporters
and non-exporters. Another interesting aspect of the research was the
inclusion of a technology index to represent technological capabilities
and a dummy variable to capture learning from buyers. To the best of
our knowledge, this is one of the first econometric studies to test the
influence of these two variables along with foreign ownership and other
control variables.
The econometric results indicate that foreign ownership, the
acquisition of technological capabilities and learning from buyers are
positive and significantly correlated with the probability of exporting
in clothing firms in China and Sri Lanka. The role of technological and
marketing factors in the decision to export at firm-level is thus underlined
by the econometric results. First, access to the marketing connections
and know-how of parent companies as well as accumulated experience
of production makes foreign affiliates better placed to tap international
markets than local firms. Second, conscious investments in skills
and information to use imported technologies efficiently give firms a
competitive advantage in exporting. Third, buyers help is indispensable
in showing potential exporters how to meet the demanding requirements
of export markets.
Furthermore, skill adjusted wages  a control variable suggested
by trade theory  is significant and with the expected negative sign
in both Chinese and Sri Lankan firms. Meanwhile, fixed assets per
employee (a proxy for capital intensity) has a negative correlation with
the probability of exporting in China but not in Sri Lanka. These last two
results indicate that the predictions of the Heckscher-Ohlin trade theory
also receive some support at firm-level in China and Sri Lanka.
Comparative econometric research on firm-level exporting
behaviour using large samples is a relatively new development in the
literature, stimulated by the availability of large enterprise survey
datasets and methodological developments (e.g. the technology index).
Nonetheless, as this paper and others highlight, it provides a powerful
means to verify and extend the findings of detailed enterprise case
studies on innovation and learning processes in developing countries.

12 Transnational Corporations, Vol. 17, No. 2 (August 2008)

Appendix 1. The Technology Index (TI) for Chinese and
Sri Lankan firms
The Lall (1992) taxonomy of technological capabilities provides
a comprehensive matrix of technical functions required for a developing
country firm to set up, operate and transfer imported technology
efficiently. Lall groups these functions under the three sets of capabilities -
investment, production and linkages. The Lall taxonomy of technological
capabilities has been successfully used by case study research to assess
levels of firm-level technological development in developing countries
(for a selection see Lall, 1987; Lall and Wignaraja, 1998; Wignaraja,
1998; Romijn, 1999). Subsequently, a technology index based on the
Lall taxonomy (or its variants) has been developed for econometric
testing of the relationship between technological capabilities and exports
in several developing countries (see, for instance, Westphal et al., 1990;
Romijn, 1999; Wignaraja 1998, 2002, 2008).
The application of the Lall (1992) taxonomy in this study was
influenced by data availability on technical firms performed by firms
in the 2003 Investment Climate Surveys of China and Sri Lanka. Five
technical functions were common to both the Chinese and Sri Lankan
samples. Hence, the TI used here was based on firms’ competence in
the following  (i) search for technology, (ii) ISO quality certification,
(iii) process adaptation, (iv) minor adaptation of products, and (v)
introduction of new products. A firm is given a score of 1 for each
technical function it undertakes and the result is normalized to give a
value between 0 and 1. This figure can be interpreted as the overall
capability score for a firm.
Table A1. Summary descriptive statistics for clothing firms
in China and Sri Lanka

China Sri Lanka

Characteristics
obs Mean Std. Dev. obs Mean Std. Dev.
Exports to sales ratio, % 350 27.89 41.89 205 49.51 46.89
Fixed assets per employee, US$ 351 7.27 45.07 195 2.98 9.00
Wage bill, % sales 315 19.30 25.01 194 35.22 38.59
No. of years since establishment
No. 253 15.33 13.68 218 19.38 15.63
Foreign equity, % 353 11.47 28.55 218 18.89 36.89
Technology Index (0 to 1) 353 0.45 0.17 204 0.45 0.22
No. of permanent employees
No. 352 328 614 204 435 764

Source: Author’s analysis.

Transnational Corporations, Vol. 17, No. 2 (August 2008) 13

 References
%OHDQH\ 0 DQG . :DNHOLQ  ³(I¿FLHQF\ LQQRYDWLRQ DQG H[SRUWV´ Oxford
Bulletin of Economics and Statistics, 64(3), pp. 315.
Bhaduri, S. and A. Ray (2004). “Exporting through technological capability: econometric
HYLGHQFH IURP ,QGLD¶V SKDUPDFHXWLFDO DQG HOHFWULFDOHOHFWURQLFV ¿UPV´ Oxford
Development Studies, 32(1), pp. 87100.
%KDYDQL 7$ DQG 6XUHVK ' 7HQGXONDU  ³'HWHUPLQDQWV RI ¿UPOHYHO H[SRUW
SHUIRUPDQFHDFDVHVWXG\RI,QGLDQWH[WLOHJDUPHQWVDQGDSSDUHOLQGXVWU\´ Journal
of International Trade & Economic Development, 10(1), pp. 65–92.
Correa, P., M. Dayoub and M. Francisco (2007). “Identifying supply-side constraints to
H[SRUWSHUIRUPDQFHLQ(FXDGRUDQH[HUFLVHZLWKLQYHVWPHQWFOLPDWHVXUYH\GDWD´
World Bank Policy Research Working Paper, 4179, March.
'HDUGRUII$³+RZUREXVWLVFRPSDUDWLYHDGYDQWDJH´ Review of International
Economics, 13, pp. 10041016.
'X - DQG 6 *LUPD  ³)LQDQFH DQG ¿UP H[SRUW LQ &KLQD´ Kyklos, 60(1), pp.
3754.
Dunning, J.H. (1993). Multinational Enterprises and the Global Economy, Wokingham,
United Kingdom: Addison-Wesley.
Ernst, D, T. Ganiatsos and L. Mytelka (eds.) (1998). Technological Capabilities and
Export Success in Asia, London: Routledge.
Greenaway, D. and R. Kneller (2007). “Firm heterogeneity, exporting and foreign direct
LQYHVWPHQW´ Economic Journal, 117, 134161.
Guan, J. and N. Ma (2003). “Innovative capability and export performance of Chinese
¿UPV´ Technovation, 23(9), pp. 737747.
Ito, K. and V. Pucik (1993). “R&D spending, domestic competition and export
SHUIRUPDQFHRI-DSDQHVHPDQXIDFWXULQJ¿UPV´ Strategic Management Journal, 14,
pp. 6175.
Keesing, D.B. and S. Lall. (1992). “Marketing manufactured exports from developing
FRXQWULHV OHDUQLQJ VHTXHQFHV DQG SXEOLF VXSSRUW´ LQ *. +HOOHLQHU HG Trade
Policy, Industrialization and Development: New Perspectives, Oxford: Clarendon
Press.
.XPDU1DQG166LGGKDUWKDQ³7HFKQRORJ\¿UPVL]HDQGH[SRUWEHKDYLRXU
LQ GHYHORSLQJ FRXQWULHV WKH FDVH RI ,QGLDQ HQWHUSULVHV´ Journal of Development
Studies, 31(2), pp. 289309.
Lall, S. (1986). “Technological development and export performance in LDCs: leading
HQJLQHHULQJDQGFKHPLFDO¿UPVLQ,QGLD´ Weltwirtschaftliches Archiv, 122(1), pp.
8091.
Lall, S. (1987). Learning to Industrialize, Basingstoke, United Kingdom: Macmillan.

14 Transnational Corporations, Vol. 17, No. 2 (August 2008)

/DOO6³7HFKQRORJLFDOFDSDELOLWLHVDQGLQGXVWULDOL]DWLRQ´World Development,
20, pp. 16586
/DOO 6 DQG * :LJQDUDMD  ³0DXULWLXV G\QDPLVLQJ H[SRUW FRPSHWLWLYHQHVV´
Commonwealth Economic Paper, No 33, London: Commonwealth Secretariat.
Mathews, J.A and D. Cho (2002). Tiger Technology: The Creation of a Semiconductor
Industry in East Asia, Cambridge: Cambridge University Press.
Nelson, R.R (2008). “Economic development from the perspective of evolutionary
HFRQRPLFWKHRU\´Oxford Development Studies, 36(1), pp. 921.
Rasiah, R. (2003). “Foreign ownership, technology and electronics exports from
0DOD\VLDDQG7KDLODQG´Journal of Asian Economics, 14, pp. 785811.
Rasiah, R. (2004). Foreign Firms, Technological Intensities and Economic Performance:
Evidence from Africa, Asia and Latin America, Cheltenham: Edward Elgar.
Rasiah, R. (2006). “Ownership, technological intensities and economic performance
LQ6RXWK$IULFD´International Journal of Technology Management, 36(1/2/3), pp.
166189.
Rhee, Y.W. (1990). “The catalyst model of development: lessons from Bangladesh’s
VXFFHVVZLWKJDUPHQWH[SRUWV´World Development, 18(2), pp.333346.
Romijn, H. (1999). Acquisition of Technological Capability in Small Firms in Developing
Countries, Basingstoke, United Kingdom: Macmillan Press.
Westphal, L.E., K. Kritayakirana, K. Petchsuwan, H. Sutabutr and Y. Yuthavong (1990).
“The development of technological capability in manufacturing: a macroscopic
DSSURDFK WR SROLF\ UHVHDUFK´ LQ 5( (YHQVRQ DQG * 5DQLV HGV Science and
Technology: Lessons for Development Policy, London: Intermediate Technology
Publications.
Wignaraja, G. (1998). Trade Liberalization in Sri Lanka: Exports, Technology and
Industrial Policy, London: Macmillan Press, and New York: St. Martins Press.
Wignaraja, G. (2002). “Firm size, technological capabilities and market-oriented policies
LQ0DXULWLXV´Oxford Development Studies, 30(1), pp. 87104.
Wignaraja, G. (2008). “Foreign ownership, technological capabilities and clothing
H[SRUWVLQ6UL/DQND´Journal of Asian Economics, 19, 2939.
:LOPRUH/³7UDQVQDWLRQDOVDQGIRUHLJQWUDGHHYLGHQFHIURP%UD]LO´Journal
of Development Studies, 28(2), pp. 314335.
World Bank (2008). Enterprise Surveys, www.enterprisesurveys.org
Zhao, H. and H. Li (1997). “R&D and export: an empirical analysis of Chinese
PDQXIDFWXULQJ ¿UPV´ Journal of High Technology Management Research, 8(1),
pp. 89105.

Transnational Corporations, Vol. 17, No. 2 (August 2008) 15

16 Transnational Corporations, Vol. 17, No. 2 (August 2008)