ArticlePDF Available

Abstract and Figures

This study aims to identify the process by which firms venture into the information, technology and communication (ICT) services global value chain and the obstacles they face. We analyse the prevailing mode of governance of the global value chain (GVC) taking into consideration the degree of complexity of transactions, the ability to codify information, customer-supplier interrelations, and suppliers' capabilities to meet buyers' demands. Our enquiry on the mode of governance could not be as precise as we might have expected. We found a group of firms in transition between captive and relational modes of governance. We identified three clusters with different strategies. We show that complexity of services is associated with the mode of governance. Low added value is associated with captive governance. Outsourcing is especially relevant in firms with captive or hierarchical governance, and a substantial part of exports are from firms with relational governance, characterised by their high capabilities, with more complex services.
Content may be subject to copyright.
nt. J. Value Chain Management, Vol. 8, No. 2, 201
Copyright © 2017 Inderscience Enterprises Ltd.
Mexico: insertion of ICT services in global value
chains, capabilities and public policy
Lilia Domínguez-Villalobos* and
Flor Brown-Grossman
UNAM Av. Universidad 3000,
Ciudad Universitaria, CMX,
Coyoacán, C.P. 04510, México
*Corresponding author
Sergio Carrera-Riva Palacio
San Fernando 37, CDMX 14050, México
Abstract: This study aims to identify the process by which firms venture into
the information, technology and communication (ICT) services global value
chain and the obstacles they face. We analyse the prevailing mode of
governance of the global value chain (GVC) taking into consideration the
degree of complexity of transactions, the ability to codify information,
customer-supplier interrelations, and suppliers’ capabilities to meet buyers’
demands. Our enquiry on the mode of governance could not be as precise as we
might have expected. We found a group of firms in transition between captive
and relational modes of governance. We identified three clusters with different
strategies. We show that complexity of services is associated with the mode of
governance. Low added value is associated with captive governance.
Outsourcing is especially relevant in firms with captive or hierarchical
governance, and a substantial part of exports are from firms with relational
governance, characterised by their high capabilities, with more complex
Keywords: global value chains; globalisation; IT services; international
business; software; Mexico; JEL, L86, F61, F68 and F23.
Reference to this paper should be made as follows: Domínguez-Villalobos, L.,
Brown-Grossman, F. and Carrera-Riva Palacio, S. (2017) ‘Mexico: insertion
of ICT services in global value chains, capabilities and public policy’,
Int. J. Value Chain Management, Vol. 8, No. 2, pp.117–134.
Biographical notes: Lilia Domínguez-Villalobos is a Research Professor at the
Graduate Faculty of Economics in the National Autonomous University of
Mexico. She received her PhD in Economics from the Metropolitan
Autonomous University in 2005. Her research interests consist on industrial
economics, technical change and environment.
118 L. Domíngue
-Villalobos et al.
Flor Brown-Grossman is a Research Professor at the Graduate Faculty of
Economics in the National Autonomous University of Mexico. She received
her PhD in Economics from the Metropolitan Autonomous University in 1997.
Her research interests consist of industrial organization, productivity and
technical change.
Sergio Carrera-Riva Palacio is the CEO at Information technology Institute
(INFOTEC). He received his PhD in Economics from the National
Autonomous University of Mexico in 2014. His research interest is industrial
policy in information technologies.
This paper is a revised and expanded version of a paper entitled ‘Cadenas
Globales de Valor en los Servicios: La Industria de TI en México’ presented at
XVI Congreso Latino-Iberoamericano de Gestión Tecnológica ALTEC 2015,
Porto Alegre, Brasil, 25 October 2015.
1 Introduction
Since the year 2000 information and communication technology and communication
(ICT) services in Mexico have grown substantially, reaching 6.1 billion dollars in 2014
with a 9.2% annual growth rate, well above the growth of the Mexican economy.
According to Select Estrategia, SC, in 2014 subcontracting or outsourcing accounted
for 42% of the market for IT services. Mexico is home to a large number of firms offering
outsourcing services, which include:
1 subsidiaries of large multinational firms
2 firms from emerging countries which serve the domestic market and take advantage
of Mexico’s proximity to the USA to export
3 a group of small and medium-size local firms nationwide, mostly concentrated in
clusters in Monterrey, Guadalajara, Queretaro, Mexico City, and the northwest.
Today, Mexico is the world’s third leading exporter of ICT services (Gartner, 2014). A
study by the Ministry of Economy (PROSOFT, 2015) found that, in terms of operating
costs in the IT sector, Mexico is competitive in areas such as software design and
production of videogames. Thus, compared with the USA, costs in Mexico are lower by
39% in development of digital entertainment and 38% in software design, considering
costs of human capital, infrastructure, transportation, services, and taxes. This has helped
to promote exports and attract investment in information technology, allowing Mexico to
position itself as the third leading exporter of IT services worldwide. However, the
industry produces very little innovation, as the same study confirms; separating spending
in R&D it observes that only large firms allocate 5% of their revenues to such activities,
whereas medium-size and small firms allocate 1% and in micro firms, investment in this
area is practically non-existent.
Although there are different studies which help to understand the characteristics of
development of the industry as a whole, its market structure, regional distribution, and
use of policy instruments (Hualde and Mochi, 2008; Mejia et al., 2006; Schatan, 2012;
Casalet and Buenrostro, 2013; Brown and Domínguez 2012, 2015), as well as excellent
reports prepared by Select Estrategia SC, there are issues of vital interest still to be
exico: insertion of ICT services in global value chains 119
analysed. First, the mode of governance is crucial in understanding global value chains
(GVCs). Given that GVCs are made up by a great number of transactions around the
world, it is important to assess firms’ capabilities to manage complex information and to
assimilate or improve technology, which conditions the distribution of power among
firms. A second question is how governance is related to firms’ strategies or the role of
policy makers in connection with the above-mentioned issues. Finally, there is an area of
concern on the obstacles firms face for growth.
Our enquiry on the mode of governance could not be as precise as we might have
expected. We found a group of firms in transition between captive and relational modes
of governance, which we denominated as captive-relational. Through factor and cluster
analysis we were able to identify three clusters with different strategies in relation to
certification, access to public programs, interrelations with other firms and universities,
and maturity. Attention is frequently drawn to the low level of added value of ICT
services in Mexico; our results show that this is related to captive governance. Greater
complexity of services is associated with relational governance. As regards the position
of firms at nodules in the value chain and their modes of governance, we found that
outsourcing is especially relevant in firms with captive or hierarchical governance. But
our results also show that a substantial part of exports comes from firms with relational
governance characterised by their high capabilities, with more complex services. The
most important obstacle for the growth of firms seems to be the financial, but it was
particularly important for firms under captive governance.
The theory of GVCs has placed emphasis on technological factors such as
fragmentation of the productive process and the global strategies of multinationals to
explain export opportunities for emerging countries (Gereffi, 1994;Cattaneo et al., 2010;
Gereffi et al., 2005). In addition, the theory of capabilities indicates that this approach is
limited, because it is also necessary to consider the accumulation of firm’s capabilities to
successfully escalate to achieve greater added value in the chain (Carlo and Rabellotti,
2010). Also, it should be acknowledged that there are policy factors which have a
significant impact on the presence of the major transnational chains (TNCS) in Mexico
and their optimum functioning (Carrera, 2015).
As we know, in the year 2001 the Mexican government, seeking to take advantage of
a tendency of growth in the global ICT industry, established the strategic importance of
the domestic software industry in terms of its potential for the nation’s economic growth.
In October 2002, it launched the Program for the Software Industry (Spanish acronym
PROSOFT) initiative with the aim of raising competitiveness by strengthening the
Mexican IT sector. Initially it focused on the industry of software and related services
and years later expanded into IT-based creative industries. In its fifteen years of
existence, PROSOFT has evolved, promoting joint efforts between federal and state level
agencies to create economies of agglomeration and form clusters and promote the growth
of firms started in Mexico and, finally, IT-based innovation, and to increase the quality of
the services that persons and organisations offer.
We conducted in-depth interviews with 42 firms, mostly Mexican and in the software
industry. Below we present a review of the literature on the subject which supports our
hypothesis that the development and sophistication of ICT in Mexico is closely related
with the modes of governance of the GVC. Then, we examine the characteristics of
groups of firms with factor analysis, based on which we analyse the firm’s integration in
productive chains, governance, and their exports. In the last section, we present our final
120 L. Domíngue
-Villalobos et al.
considerations and some proposals for public policy drawn from the results of our
2 GVCs, capabilities and public policy
The theory of GVCs gained relevance in the 1990s with authors such as Gereffi (1994)
Gereffi et al. (2005), who observe that, due to the extraterritorial activities of subsidiaries,
GVCs have multiplied in response to the growing possibility of fragmentation of
productive processes, and subcontracting or outsourcing, as various authors recognise in
the literature. Such fragmentation is accompanied by skill in segmenting in modules and
recombining systems of components (Schilling, 2000), all as the result of advances in
codification of knowledge and demand to standardise the interfaces of various stages of
production through better technical standards and design rules (Sanchez and Collins,
2001), as well as progress of information and communication technologies. The result has
opened an opportunity for insertion of emerging countries in such chains. GVCs are made
up by hundreds of thousands of transactions around the world, contingent on firms’
ability to handle complex information, transfer technology, coordinate producers, or
distribute power among themselves.
Analysts of these theories have remarked that insertion in GVCS represents an
opportunity for exports from emerging economies and escalation in the added value of
the chain. However, as Carlo and Rabellotti (2010) observe, it has not been clearly
explained how these chains promote upgrading of firms in developing countries and the
obstacles they face. Despite advances in codification of knowledge, as evolutionist theory
has made clear, it is never complete (Lall, 2006); thus, it is important to examine how
information, knowledge, and materials are transmitted from one nodule to another. This
brings to the forefront of the analysis the importance for firms of accumulating
capabilities and the possibility of asymmetries along the chain, giving rise to different
modes of coordination or governance, which will influence the possibility of upgrading
(Carlo and Rabellotti, 2010).
In these chains, the lead firms are often multinational and can decide rules on what,
how, when, and at what price services are to be produced, which results in a kind of
involvement of participants in the chain. At times, providing firms are able to interact and
collaborate with the client. Less often, the role of lead firm is assumed by exporting firms
in developing countries. Client-provider relationships vary from a market relationship to
building networks, where there is close cooperation (Carlo and Rabellotti, 2010).
According to Sturgeon (2009), three factors determine lead firm’s choice of type of
governance: the complexity of the information involved in transactions, the possibility of
codifying the information, and the intensity of client-provider relations along the chain.
The author identifies five modes (market, modular, relational, captive, and hierarchical).
Subsequently, proponents of theory of capabilities have underscored the need to consider
capacity of absorption and technological capabilities for each of those modes (Carlo and
Rabellotti, 2010). Below we describe those kinds of governance in the terms defined by
these authors.
exico: insertion of ICT services in global value chains 121
Market: Providers and buyers meet in the market, where governance depends on
prices. The degree of coordination and asymmetry between them tends to be the
Modular: The complexity of information is high, but is codifiable in technical
standards, and providers with capabilities predominate. Providers produce in
accordance with client`s specifications and take responsibility for complying with
them. According to the same authors, the need to adhere to such standards induces
learning without being involved in the process.
Relational: Transactions are complex and not easily codifiable, requiring intense
interaction between client and provider. There are capabilities on the providers’ side
and learning can even be mutual. For providers in semi-industrialised countries, it is
important to reinforce their production and linkage capabilities to help them interact
with the lead firm.
Captive: When there are capabilities or not, providers, often small firms, depend on
buying firms which exercise high levels of monitoring and control. Lead firms may
participate in providers learning process giving them technical assistance, but their
support is limited to the needs of the product in question, which usually is not
complex, and therefore the risk of being trapped in niches with added value for
providers is high.
Hierarchical: Corresponds to vertical integration in which the lead firm has the right
of ownership of the operations in the chain and its transactions. Exports occur inside
the firm. Capabilities are transferred through learning by management, spinoffs from
the firm, and training given to the workforce, to the extent that the subsidiary
competes with other subsidiaries worldwide for investment projects; such learning
should not be considered negligible.
In Mexico, it is known that there is a segment of firms with hierarchical governance, and
diagnoses point to another segment of low added value, which suggests captive
governance (Fernandez-Stark et al., 2011). Ideally, Mexican firms should move toward
relational governance, in which there is interaction between client and provider in the
context of a network of collaboration and higher innovation opportunities.
As we mention in the introduction, the theory of GVCs has emphasised technological
factors such as fragmentation of the productive process and the global strategy of
multinationals and the theory of capabilities has unquestionably enriched that
perspective. But we must grant that there are policy-related factors which have a
significant influence on the presence of major TNCs in Mexico and on their optimum
functioning. Many emerging countries have developed programs for the industry of
offshore services which include promotion and attraction of transnationals, support for
investment in services, and economic incentives to influence investors’ perception of
costs and benefits. On the other hand, government agencies can help to improve the
business climate by establishing rules, values, and institutions to reinforce the capabilities
of the private sector, as Carrera (2015) shows in his in-depth study of four countries,
examining industrial policies implemented by Ireland, a country which took advantage of
the diaspora effect and the advantage of English as its native language, implemented an
aggressive fiscal policy to attract ICT multinationals in both hardware and software to
serve the market of the USA and the European Union, stimulated investment in research
122 L. Domíngue
-Villalobos et al.
and development, and strengthened its network of universities and research centres.
These actions have allowed it to have one or two companies among the largest one
hundred and be considered a top-level nation among ICT exporting countries, together
with Israel, India and, naturally, advanced countries like the USA, Germany, and France,
to mention a few (Carmel, 2003). Countries like South Korea, whose educational system
received a substantial boost more than 30 years ago, have succeeded in inserting
themselves in the chain on the second tier of ICT exporting countries. Carrera (2015)
mentions that tax incentives, investment facilities, and attractive financing were provided
to develop an ICT industry focused on the development of software to be used in the
automotive industry, electronics, and medical equipment, in which South Korea has
several successful firms.
In emerging countries, policies had to cover a wider range of action. Mexico, as,
mentioned, designed PROSOFT, which provided support for: upgrading hardware,
training, certification of human resources, English language teaching, adoption of quality
models, internationalisation, and the creation of a ‘country brand’. Also, this policy
scheme sought to create institutions to bolster agglomeration economies at the state level.
Jalisco and Nuevo Leon are two states which have benefitted the most from PROSOFT.
The program has been gradually adapted to new needs, with an eye to expanding the use
of ICT, outreach with universities, and innovation to achieve greater escalation and
improve Mexico’s position as a provider of IT services. However, it is necessary to
mention that PROSOFT’s performance also depends on the local governments treating
the industry as a priority, and while its performance has been laudable in Jalisco, and
particularly in Guadalajara, the same cannot be said of other regions like the Mexico
In addition to PROSOFT, there are other programs to support certification (MEXICO
FIRST) and institutions created under the former FONDO PYME (funding of aid
initiatives for small and medium sized companies) and INADEM (National
Entrepreneurial Institute), such as TECHBA (International Corporate Accelerator and
Consultancy), which supports ICT start-ups, and PROMEXICO to attract investment by
multinational firms.
Aware that there are efforts and programs targeting certain needs of the industry, it is
our view that the promotion of investment in equipment and catch-up programs by the
Ministry of Economy and the National Council for Science and Technology
(CONACYT) is insufficient. Without competitive funding or venture capital, escalation
toward greater added value and innovation for Mexican companies in GVC is
at risk of stagnating in the trap of middle-income countries. For that reason, several
international institutions insist on the need to define a development strategy with
selective measures focused on breaking down the activities in the chain into fine
segments (OECD-WTO-UNCTAD, 2013).
Helping to strengthen relations among stakeholders in the GVC and consolidating an
innovative ecosystem are challenges which require investment to boost the confidence
and organisational capital of the firms that take part in those chains.
exico: insertion of ICT services in global value chains 123
3 Toward a greater understanding of the insertion of firms in the
In this section we present the results of our investigation of interviews with ICT service
companies conducted in the year 2014, with the intention of analysing the chain more
extensively. First, we show the characteristics of the sample and the analytical
methodology to group variables and firms. Next we show the results.
3.1 Characteristics of the sample
The sample sought relevant examples in varying characteristics of size, activity, and
regions. Firms were recommended by the IT Institute of Jalisco (IJALTI), the Monterrey
Software Council, and the IT Innovation and Research Institute (INFOTEC). The sample
does not purport to be representative, but it does allow us to analyse patterns of behaviour
based on differing capabilities and positions in the GVC.
It is true that our sample is small, and therefore the results cannot be generalised for
the whole industry, because we do not include any micro firms with less than five
employees,1 most of which do not outlast the death valley. Nevertheless, the fact that it
includes firms accounting for more than 60% of sales and exports in the industry sheds
light on firms’ strategies to insert in GVCs.
ICT outsourcing services have different levels in the chain’s added value. We use the
Fernandez-Stark et al. (2011) diagram, which classifies outsourcing in three broad
segments: information technologies (ITO), business processes (BPO), and knowledge
intensive processes (KPO). In ITO, the lowest level in added value refers to infrastructure
(administration of applications, networks, and infrastructure); it is followed by software,
where software architecture and development of applications have the highest value in
the market and desktop management and software engineering have the lowest value.
Next are business processes (BPO), in which firms offer outsourcing services in
management of material resources, management of human resources, and management of
client relations. The lowest value in the market corresponds to contact and payroll
services and administration of documents and content; other activities such as accounting,
marketing, and recruitment have medium value. Lastly, knowledge intensive services, or
KPO, have the highest value in the market. The firms interviewed can be located by their
position in the chain in Figure 1.
As a proxy of the added value of services, we used the information on the average
wage reported in Fernandez-Stark et al. (2011) for the different lines of business in ITC
outsourcing with a maximum of 14 points (Table 1). Based on the distribution of scores
in our value variable, we classified the companies in the graph below into three groups:
low (1–3), medium (4–5), and high value (6–14). Note that there are companies with
intersecting activities. These companies declared that they participated in more than one
activity. Very few were classified in the vertical specialisation group, known as market
niches. These niches could represent a significant opportunity for small and medium size
Mexican companies.
124 L. Domíngue
-Villalobos et al.
Figure 1 Firms’ position in the offshore services value chain (see online version for colours)
Source: Own elaboration based in Fernandez-Stark et al. (2011)
3.2 Methodological aspects
With the aim of analysing the relationship between the characteristics of firms in relation
to their maturity, technological effort activities (development of capabilities,
certification), accomplishments and advantages (quality, innovation, linkage, and use of
economies of scale), outsourcing, and access to policies, variables were constructed based
on the answers to the questionnaire. Table 1 shows the points assigned to each variable.
The analysis method was based on multivariate techniques, among them factor
analysis and cluster analysis (Indhumathi and Sathiyabama, 2010; Tabachnick and Fidell,
2001).2 Before presenting our results we devote the next lines to the mode of governance.
We observed that in general there is intense interaction between clients and providers,
which speaks of services tailored to the client’s needs in which, despite standardisation
and certifications, an exchange of tacit knowledge is necessary, which indicates a
potential for learning; consequently, it is essential to enquire about the means of
transmission of knowledge and its limits, which in turn leads us to enquire about
governance in the chain. In principle, it is important to observe that in this sample of
firms we found no evidence of market or modular governance.3
Firms classified as captive firms depend more on their buyers for specifications and
instructions, and report assistance from their technology provider or client.
Among firms with more client-provider interaction, we found one group which can be
considered to use relational governance and another, intermediate group between those
and the captive governance group, which we refer to as ‘captive relational’. The latter
firms includes firms with mid-level capabilities. A low level of capabilities bars them
from interacting with their counterparts on the same level. In contrast, firms with
relational governance firms are characterised by their high capabilities. Finally, the three
firms with hierarchical governance are large firms, with 100% foreign capital, which
export a high percentage of their services to the parent corporation.
exico: insertion of ICT services in global value chains 125
Table 1 Description of variables
Maturity Age Up to one year: 1
1–3 years = 2
3–5 years = 3
5–7 years = 4
7–9 years = 5
+9 years = 6
Spinoff 1 additional point
Maximum = 7
Value of type of
% of sales Systems development = max 5
Systems integration = max 1/2
Software packaging = max 5
Systems support = max 1/2
Business process outsourcing = other
services = max 2
Maximum = 14
Development of
Human resources Own program = 1
Linkage with universities = 1
External training = 1
Development of specific
(methodologies, patents)
Yes = 3
Total = 6
Quality Existence of certification Certification = maximum 5 points
Quality as
Yes = 1
Total = 6
Link ups Universities = 1
Other firms = 1
Strategic alliances Technology partner = 1
Total = 3
Agglomeration The advantage of a region Proximity to universities = 1
Government programs=1
Business environment = 1
Share of exports Exports/sales % Constant variable
Note: *Points were assigned to the variable value in each activity based on the estimate
of salaries made by Fernandez-Stark et al. (2011).
Source: In-depth interviews
126 L. Domíngue
-Villalobos et al.
Table 1 Description of variables (continued)
Exports Value of exports 1–5 MDD = 1
5–10 MDD = 2
10–20 MDD = 3
20–40 MDD = 4
40–60 MDD = 5
50–80 MDD = 6
More than 80 MDD = 7
Innovation Innovation as
Yes = 1
Patent or methodology Yes = 1
Total = 2
Size Sales 2–30 MDD = 1
30–50 MDD = 2
50–100 MDD = 3
More than 100 MDD = 4
Employees Up to 100 = 1
100–500 = 2
More than 500 = 3
Communication with
costumer and suppliers
Two-way interaction Yes = 2
Technical assistance Yes = 1
Licenses Yes =1
Governance Captive = 1 Communication with costumer and
suppliers <3
Development of capabilities <3
Captive-relational = 2 Communication with costumer and
suppliers = 3
Development of capabilities >3 and <4
Relational = 3 Communication with costumer and
suppliers = 4
Hierarchical = 4 Development of capabilities >4
Vertical integration with intrafirm
Note: *Points were assigned to the variable value in each activity based on the estimate
of salaries made by Fernandez-Stark et al. (2011).
Source: In-depth interviews
In summary, half of the firms in our sample have captive and captive relational
governance, in other words, asymmetries between clients and providers in the global
chain of ICT services of Mexican firms, although there are firms which have achieved a
stronger position.
Based on the information from the factor analysis, scores were calculated for each
factor in each of the observations. The factor points represent the relationship of the
exico: insertion of ICT services in global value chains 127
different observations with each factor. Cluster analysis among firms was conducted with
factorial points using the method known as ‘K means’.4
Four factors were identified with a characteristic value greater than one. The rotation
method used was varimax, which seeks to minimise the number of variables that have
high loads in a factor. The highest factor loads are more closely associated with that
variable. The last column corresponds to commonalities, which refer to the part of the
variables explained only by common factors.
3.3 Relationship of variables
The result of the factor analysis5 using the variables described is presented in Table 2. We
observe that the communalities are greater than 70% with one exception; in other words,
the factors explain a high percentage of the variability of the variables.
Table 2 Factor analysis
Variable Capabilities and
Innovation and
economies Communality
Outsourcing –0.04 –0.12 –0.04 0.94
Linkage 0.004 0.24 0.85 0.80
Policy 0.26 0.82 0.18 0.76
Agglomeration 0.23 –0.02 0.82 0.79
Governance 0.92 0.07 0.16 0.80
Certcal 0.85 –0.03 0.14 0.87
Value 0.87 0.12 –0.01 0.76
maturity2 0.47 0.51 0.13 0.53
innovation2 –0.21 0.84 –0.01 0.77
Source: Our estimate using the database for the sample
Factor analysis allows us to appreciate, that there is an association between product
complexity, i.e., value, the intensive presence of human resource programs, certification
efforts, quality, and governance. Overall, factor I explains 30% of the variance and shows
the importance of capabilities in escalating to greater added value. The lack of a
relationship with the perception of innovation as an accomplishment is noteworthy. In
contrast, in the second factor, this variable is related with the use of policies and the
maturity of firms and explains 20% of the variance. The absence of association between
policies and the advantages of agglomeration is unexpected since an effort has been made
to promote clusters. The third factor, called external economies, with 17% of the variance
explained, shows a relationship between economies of agglomeration and linkages or
interaction between firms like strategic alliances. The load coefficient of outsourcing is
noteworthy in this last factor, explaining a smaller percentage of the variance (11%).
Because the first three factors explain 67% of the variance of the original data, the
cluster with the K means method was formed with three initial groups (Indhumathi and
Sathiyabama, 2010). The distribution of firms in those three groups is 21, 12, and 9 firms
(Table 3).
128 L. Domíngue
-Villalobos et al.
Table 3 Cluster analysis of firms
n quality
I (21) 1.6 2.9 9 3.1 0.9 3.9 1.0 1.2 1.6 1.7
II (12) 2.7 4.4 19 4.4 1.8 4.9 1.5 2.8 1.8 2.5
III (9) 3.3 5.2 15 4-9 4.2 5.8 0.5 2.0 1.9 2.5
Note: Based on points in the factorial analysis.
Source: Our estimate using the database for the sample
The clustering of firms gives a good approximation of the importance of the type of
governance of each firm`s chain. Cluster I is characterised by the highest percentage of
firms with captive governance. The youngest firms in the sample with small to medium
sizes, has the lowest values in the different variables. A low level of capabilities bars
them from interacting with their counterparts. As mentioned before, this type of firms
depends more on their buyers for specifications and instructions, and report assistance
from their technology provider or client. For this group external economies from the
specific geographical location are important.
Cluster II, which includes seven firms with relational governance and five with
captive-relational governance, is engaged in activities of medium complexity. Its use of
policies is the most intensive and it has the highest percentage of firms considering
innovation as their most important achievement. Our interviews did not provide evidence
on the depth of these innovations, although we suppose that they are only incremental,
because some firms did mention to be certified in the process to develop software as a
true innovation for them.
Last, cluster III, includes three firms with hierarchical governance and six with
relational governance which show consolidated strategies, mostly very large firms. As
expected, high capabilities in human resources and technology with own programs and
training are combined with high levels of certification and quality and more complex
services, as the variable value indicates. They take advantage of the benefits of different
policy instruments and regional external economies for linkage between firms, or with
In sum we can consider that, on average, the mode of governance for cluster I is
captive, relational captive for the second and hierarchical-relational for the third cluster.
With the aim of analysing the characteristics of insertion of firms in the GVC, we next
present the type of services and exports in our sample.
3.4 Insertion of firms in the GVC
As expected, given the maturity of the market and the overall level of capabilities, ITO is
the most important activity of firms by any standards and KPO is relatively unimportant.
In cluster I (captive firms) these activities are not performed simultaneously; 76% of the
firms’ line of business is ITO, 16% is BPO, and 14% is KPO. On the contrary, in the
other clusters firms have more than one line of business,6 i.e., are more diversified. For
exico: insertion of ICT services in global value chains 129
example, in cluster III (‘hierarchical/relational’), 80% of the firms are in the ITO
segment, 60% in BPO, and 40% in KPO (Figure 2).
Figure 2 Activities in the chain (see online version for colours)
Source: Our estimate using the database for the sample
Not all firms are in the GVC (31%), in the sense that they do not export. Some provide
services to companies in Mexico or to government agencies, either developing
applications to improve management practices from MRP applications, adaptation of SAP
or surveillance systems, or consultancy for certification and other business services.
Some of them (30%) have eventually branched into exporting, although in low
percentages (less than 30% of sales) using the advisory services of the Mexican
technology business accelerator (TECHBA) in the USA or even creating a company there
to raise their profile or partnering with Microsoft or insurance or financial firms in the
USA. Among the largest exporters, we find different patterns of entry to the global chain.
For example, one Mexican firm entered into an early partnership with the financial
outsourcing arm of a multinational, which it then acquired, allowing it to accelerate its
learning of outsourcing services. Another took advantage of the learning of its team of
internal consultants throughout its globalisation process to create its own IT outsourcing
arm, which eventually became a company that provides services to different sectors of
the economy. Firms with foreign capital in the sample vary from small firms which are
created to provide outsourcing services toward the USA to large multinationals which
leverage their local capabilities to scale up to higher added value. Finally, some are
foreign firms from emerging countries, which have opened subsidiaries in Mexico to
serve the domestic market and take advantage of the US market.
Next, Table 5 shows that cluster III exports 32% of its sales. Foreign firms account
for 45% in this cluster with the highest percentage of firms that subcontract and practice
outsourcing. In second place in exports by percentage of sales (19%) is cluster I
characterised by captive governance, smaller firms, lower complexity and 43% engaged
in outsourcing services, although it must be noted that this high percentage is explained
by five large exporters. Finally, cluster II exports only 13% of sales. In sum, there is not a
clear-cut pattern between exports and governance.
130 L. Domíngue
-Villalobos et al.
Table 5 Exports in groups
Cluster % X of sales Foreign firms (number) Value average Outsourcing %
I 19 1 3.9 43
II 13 2 4.9 41
III 32 4 5.8 67
To finish, it is important to refer to firms obstacles for growth. The most important
obstacle for the growth of firms seems to be the financial. It is not surprising that for
cluster III, only one third of the firms mentioned it, whereas for captive firms (cluster I)
competitive credit is a priority. It is important to mention that in spite of government
efforts, there is agreement on the fact that bank credit is scarce for productive activities
and the software industry faces the worst-case scenario. In relation to human resources,
even if an important number of firms reported having specific programs with universities
to attract qualified labour, they consider the lack of English language skills among
qualified engineers an obstacle to export. Under other obstacles, some firms mentioned
marketing abilities to increase demand or the low growth of the economy as a challenge.
4 Final reflections and considerations on public policy
Before turning to the topic of public policy for the ICT service industry in Mexico, it is
helpful to recapitulate the fundamental aspects of our work. In Mexico, as in
other emerging countries, the industry has shown high levels of growth, as the result of
both needs for fragmentation of productive processes in large firms by extending their
supply chains to several countries (outsourcing), and an impulse from Mexican
government policy establishing the strategic importance of the software industry in view
of its potential for the country’s economic growth. This policy has included an effort to
attract major transnational ICT firms with the PROMEXICO program and a program
supporting development of human resources, infrastructure, international certification of
firms, and formation of territorial conglomerates or clusters under the PROSOFT
initiative. A concomitant result has been considerable growth of exports, which today
puts Mexico among the world’s top exporting countries.
Our enquiry on the mode of governance could not be as precise as we might have
expected based on the categories defined by Gereffi et al. (2005) because in some cases
the data did not allow a more refined classification. It was necessary to modify the
classification and subdivide relational governance into captive-relational governance and
hierarchical governance per se. More research is needed on this subject, particularly on
the captive-relational category.
On the basis of our factor analysis, we identified three clusters made up by firms with
different strategies in relation to capabilities, use of policies, use of external economies,
and linkage, as well as their age:
a nine firms, three with hierarchical governance and six with relational governance,
with high capabilities, levels of certification and quality, and more complex
processes exporting the highest ratio of exports to sales
exico: insertion of ICT services in global value chains 131
b 12 firms, seven with relational governance and five with captive-relational
governance, engaged in activities of medium complexity and with the lowest ratio of
exports to sales
c 21 firms predominantly with a captive mode of governance; the youngest firms in the
sample with small to medium sizes and having the lowest values in the different
Therefore, there is no clear-cut pattern between exports and governance. As regards the
position of firms at nodules of the value chain and their modes of governance, we found
that outsourcing is especially relevant in firms with captive or hierarchical governance.
But our results also show that a substantial part of exports comes from firms with
relational governance characterised by their high capabilities, with more complex
We observed that firms have a weakness related to the intrinsic ability to generate
innovation, based on the scores of each cluster: the highest level was obtained by cluster
II. Although our interviews did not provide sound evidence on the depth of these
innovations, firms rarely reported patents or products, suggesting that they are oriented to
incremental innovations, an issue that needs further study. This lack of an innovative
ability of ICT firms has been confirmed by studies in other industries of the Mexican
economy (Dutrénit et al., 2006) and limits the ability of ICT firms to undertake projects
of greater complexity and added value, and thereby escalate within the value chain.
Turning to the issue of public policy, although the three clusters use instruments
designed to support the industry, it should be noted that there are important differences.
Cluster II, which predominantly serves the domestic market and considers itself the most
innovative, is the most intensive user of public policy. This must be considered positively
in as much as it is expected that public support should seek to strengthen that types of
firms, even if there is no evidence that these innovations could claim to be more than
incremental. However, must of the firms in our sample are in low added value niches and
have lesser capabilities, which suggests that there is a long way to go in terms or public
policy efforts that give support not only to certification and equipment but also to
improve bank credit conditions and long-term finance.
Given the importance of the financial obstacle to growth, it is important to stress that
a weak Mexican market in alternatives for venture capital has hindered local
entrepreneurs seeking to implement larger-scale projects due to the scarcity of funds or
the financial cost of available alternatives. This problem should be addressed because the
presence of venture capital in other countries is closely tied to innovative potential and is
a key piece in achieving integration of greater added value in domestic production and in
Mexican exports.
Our results show that there is a group of larger firms with very satisfactory
performance benefiting of public policies. It is recommended that to achieve greater
impact by providing incentives that enable the most advanced companies foster a spill
over of knowledge, business practices, and innovation schemes. Some companies already
do, but it is not a trait that distinguishes the Mexican ecosystem.
We agree with analysts like Carrera (2015) who claim that public support should have
a long-term vision, in order to develop human and organisational capabilities (not only
certifications). The instruments should be consistent with long-term goals while there
should be checkpoints to confirm progress and correct or cancel as needed – it is crucial
132 L. Domíngue
-Villalobos et al.
to assume the risk inherent in the search for innovation and expansion in markets
(development banks should assume a proactive position to boost the accumulation of
intangible capital). If they seek to make inroads in a GVC in leading edge technology,
government agencies need to overcome the challenges of development, which goes
beyond the traditional approach of bridging the gaps created by market slumps.
In the area of talent development, the firms also stressed the need for qualified,
English-speaking engineers, a deficiency which is not addressed by current policies. This
is a key issue that underscores the need for reform in the tertiary public education system.
We also consider it pertinent to align support policies through subsidies with other
instruments of the state such as public purchasing, tax incentives for R&D in ICT, and a
policy of attracting investment to support the development commitments of providers,
which TNCs can usually assume to bolster their competitive position. Likewise, it is
essential to align incentives in government programs that favour modernisation and/or
expansion of other branches of industry, because ICT is a core industry in the digital age
and a modern and strong ICT sector positively impacts the modernisation of the economy
at large.
We thank our colleagues Claudia Schattan and Clemente Ruiz for their observations. We
are also indebted for the valuable comments made by the anonymous reviewers. The
responsibility of mistakes remains our own. Finally, we must recognise the support by
PAPIIT-UNAM project IN305914.
Brown, F. and Dominguez, L. (2012) Can Mexico Set Up in the Aerospace and the Software and it
Global Value Chains as a High-value-added player?, IDB working paper series, pp.1–54.
Brown, F. and Dominguez, L. (2015) ‘Cadenas de valor globales en servicios: el caso de la
industria de TI en México’, Economía:Teoría y Práctica, Nueva Epoca, pp.37–71.
Carlo, P. and Rabellotti, R. (2010) Global Value Chains Meet Innovation Systems: Are there
Learning Opportunities for Developing Countries?, IDB Working Paper Series, pp.1–24.
Carmel, E. (2003) ‘Taxonomy of new software exporting nations’, The Electronic Journal on
Information Systems in Developing Countries, Vol. 13, pp.1–6.
Carrera, S. (2015) La Necesidad de Políticas Industriales En La Economía Moderna: El Caso Del
Fomento a La Industria de Tecnologías de Información en México, UNAM.
Casalet, R. and Buenrostro, E. (2013) ‘La integración regional centroamericana en ciencia,
tecnología e innovación: un nuevo desafío’, Economía: Teoría y Práctica, Vol. 40,
Cattaneo, O., Gereffi, G. and Staritz, C. (2010) Global Value Chians in a Postcrisis World, World
Bank, Washington, DC.
Dutrénit, G., Capdevielle, M., Casas, R., Puchet, M., Unger, K. and Vera- Cruz, A.O. (2006)
Diagnóstico de la Política Científica, Tecnológica y de Fomento a la Innovación en México
(2000–2006), Fondo de Cultura Económica, México.
Fernandez-Stark, Bamber, K. and Gereffi, G. (2011) Skills for Upgrading: Workforce Development
and Global Value Chains in Developing Countries, Duke University, Center on Globalization,
Governance & Competitiveness.
exico: insertion of ICT services in global value chains 133
Gartner, I. (2014) Market Share: IT Services [online] [accessed 18 November
Gereffi, G. (1994) ‘The organization of buyer-driven global commodity chains: how US retailers
shape overseas production networks’, in Gereffi, G.A.M.K. (Ed.): Commodity Chains and
Global Capitalism, Praeger, Westport, CT.
Gereffi, G., Humphrey, J. and Sturgeon, T. (2005) ‘The governance of global value chains’, Review
of International Political Economy, Vol. 12, pp.78–104.
Hualde, A. and Mochi, P. (2008) ‘México: ¿una apuesta estratégica por la industria del software?’,
Comercio Exterior, Vol. 58, pp.335–349.
Indhumathi, R. and Sathiyabama, S. (2010) ‘Reducing and clustering high dimensional data
through principal component analysis’, International Journal of Computer Applications,
Vol. 11, No. 8, pp.975–887.
Lall, S. (2006) ‘Industrial policy in developing countries: what can we learn from East Asia?’, in
Bianchi, P. and Labory, S. (Ed.): International Handbook on Industrial Policy, Edward Edgar
Publishing, Massachusetts.
Mejia, M., Ania, I. and Gamboa, R. (2006) Diagnóstico de la Industria de Servicios de Software en
México [online]
(accessed September 2011).
OECD-WTO-UNCTAD (2013) Implications of Global Value Chains for Trade, Investment,
Development and Jobs, Saint Petersburg [online]
Value-Chains-2013.pdf [accessed May 2015].
PROSOFT (2015) Fuente de Información Estratégica del Sector de TI [online] (accessed May 2016).
Sanchez, R. and Collins, R. (2001) ‘Competing and learning in modular markets’, Long Range
Planning, Vol. 34, pp.645–667.
Schatan, C.Y.L.E. (2012) ‘La producción de bienes y servicios de tecnologías de la información y
la comunicación y las políticas industriales en México’.
Schilling, M.A (2000) ‘Toward a general modular systems theory and its application to interfirm
product modularity’, Academy of Management Review, Vol. 25, pp.312–334.
Select, Estrategia, SC (2014) Reporte.
Sturgeon, T. (2009) ‘From commodity chains to value chains: interdisciplinary theory-building in
an age of globalization’, in Bair, J. (Ed.): Frontiers of Commodity Chain Research, Stanford
University Press, Palo Alto, CA.
Tabachnick, B. and Fidell, L. (2001) Using Multivariate Statistics, Boston, Allyn and Bacon,
1 These firms (around 3,500), account for 85% of firms.
2 The factor analysis model of m common factors considers that the observable p variables
Z1……, Zp depend on m latent variables F1……, Fm, called common factors, and unique p
factors ε1; : : : ; εp, based on Cuadras (2014), with:
ip im mj ip
Z= Fa +e
where a are the coefficients of correlation between Zi and the common factor known as load
134 L. Domíngue
-Villalobos et al.
3 Although this may not hold for the industry as a whole, where there are micro and small
firms which are devoted to meeting specific and basic needs on the basis of prices
(market governance model). Besides, there are other companies which focus on the
development of apps or little systems, on the basis of standards and rules established by other
firms, usually large companies (modular governance model).
5 In the case of our variables, it was possible to use the technique of factorial analysis because
the value of the statistic Kaiser-Meyer-Olkin Measure (KMO) obtained was 0.66; in other
words, because the value of the statistic is greater than 0.5 Tabachnick and Fidell, 2001, the
correlation between variables is adequate to conduct the factorial analysis. Other reasons that
justify using factor analysis are: first, factor loading scores were greater than 0.8, indicating
that the dimensions of the factors are better accounted for by the variables; next, the
correlations between variables are greater than 0.30, suggesting a strong relationship between
them (Tabachnick and Fidell, 2007); and last, our sample variables are heterogeneous.
6 Thus, the percentages do not add up to 100.
México es una de las economías más insertadas en cadenas globales de valor, especialmente con Estados Unidos en la industria automotriz. Este artículo tiene por objeto analizar la integración del sector manufacturero de México en las cadenas globales de valor (CGV), observando la profundidad de las actividades de transformación, los sectores más dinámicos y la importancia para el mercado laboral. Para ello se utilizan las bases de datos Trade in Value Added (TiVA) y Trade in eMployment (TiM) publicadas por la OCDE. Se encuentra que el sector de manufacturas mexicano está fuertemente integrado en CGV, en especial con Estados Unidos, pues abarca una elevada proporción de las exportaciones con valor agregado y de los empleos generados por exportaciones. No obstante, la integración del sector lo expone a dos desafíos principales: los cambios en la política comercial de Estados Unidos y la automatización.
Full-text available
El contexto internacional actual –incluyendo el resurgimiento de gobiernos aislacionistas y proteccionistas– nos obliga forzosamente a pensar de manera intencionada en las diferentes formas, y sus implicaciones, de interrelación entre las diversas economías nacionales: ¿Mayor o menor apertura?, ¿Qué tipo de apertura?, ¿Qué sectores económicos?, ¿Cuál es el papel del Estado respecto al crecimiento económico de una nación? El presente texto hace una revisión histórica del modelo de sustitución de importaciones que predominó en América Latina por buena parte del siglo xx, haciendo un balance entre sus aciertos y sus deficiencias. El artículo pasa revista a su antecedente, el modelo primario-exportador que caracterizó buena parte del siglo xix y principios del xx y su necesaria modificación tras los choques externos de la Gran Depresión y la irrupción de la Segunda Guerra Mundial. Se revisan las contribuciones teóricas de Raúl Prebisch y otros pensadores cepalinos para abordar la razón de ser y las características del modelo de sustitución de importaciones. El trabajo concluye con una reflexión respecto a la viabilidad de repensar dicho modelo en un contexto de lento crecimiento económico de más de 30 años, resaltando los factores de la época contemporánea junto con el regreso de ideologías de menor apertura al comercio internacional.
Full-text available
El surgimiento de las cadenas de valor globales es resultado de un desplazamiento de la integración vertical las transnacionales a favor de las redes de proveedores externos. Le - jos de la noción que identifica la subcontratación de servicios extraterritoriales con la manufactura, la evidencia muestra que aquélla ocupa 80% del total global, dentro del cual están los de tecnologías de la información ( ti ). Mediante estudios de caso se identi - fica el proceso mediante el cual las empresas de software se han podido insertar en la cadena de las ti y los obstáculos a los que enfrentan una vez que ya están inmersas en ella, con la conjetura de que la oportunidad internacional se combinó con la utilización de apoyos de política, como el Prosoft, y la presencia de espíritu empresarial de sus diri - gentes.
Full-text available
Este artículo identifica las capacidades institucionales y productivas adquiridas en varios países de Centroamérica, que fomentan la interacción nacional y crean condiciones favorables para lograr acuerdos que contribuyan a la integración regional en ciencia, tecnología e innovación. Las incipientes redes impulsadas por las políticas nacionales, las instituciones de educación superior, los centros de investigación y los agentes productivos de la región pueden contribuir a una nueva gobernanza que impulse la formación de un sistema regional de innovación.
Full-text available
Many systems migrate towards increasing or decreasing modularity, yet no explicit causal models exist to explain this process. This paper builds a general theory of modular systems, drawing on systems research from many disciplines. This general theory is then used to derive a model of inter-firm product modularity, including testable research propositions. The latter model provides a valuable tool for predicting technological trajectories, and demonstrates how the general theory may be applied to specific systems.
High dimensional data is phenomenon in real-world data mining applications. Developing effective clustering methods for high dimensional dataset is a challenging problem due to the curse of dimensionality. Usually k-means clustering algorithm is used but it results in time consuming, computationally expensive and the quality of the resulting clusters depends on the selection of initial centroid and the dimension of the data. The accuracy of the resultant value perhaps not up to the level of expectation when the dimension of the dataset is high because we cannot say that the dataset chosen are free from noisy and flawless. Hence to improve the efficiency and accuracy of mining task on high dimensional data, the data must be pre-processed by an efficient dimensionality reduction method. This paper proposes a method in which the high dimensional data is reduced through Principal Component Analysis and then bisecting k-means clustering is performed on the reduced data where there is no initialization of the centroids.
This paper explains how the adoption of a modular approach to product and process architectures can greatly improve product development performance and provide a powerful framework for knowledge management and organisational learning. We first consider specific ways in which modular architectures enable both new processes for product development and new product strategies. We then explain how disciplined adherence to the modular way of creating new products helps an organisation to identify its current technological capabilities more clearly, and to define objectives and processes for strategically focused organisational learning. We discuss three seemingly counterintuitive principles in the new strategic logic of managing in the modular way, as well as new management practices involved in the modular way of working.