€¦  · Web viewTaxonomy of Portuguese regional competitiveness . Abstact: This study proposes a conceptual framework made up of six pillars of competitiveness, considered inputs,

Taxonomy of Portuguese regional competitiveness

Abstact:

This study proposes a conceptual framework made up of six pillars of competitiveness, considered inputs,

which impact on GDP per capita, monthly income and the unemployment rate.

The objective of this work involves determining a competitive index for the Portuguese regions (NUTS

III) through means of advancing a methodology based on Data Envelopment Analysis (DEA) as well as

establishing an index of factors constituting this regional competitiveness and determining the respective

taxonomy. The results convey how the Lisbon Metropolitan Area, Alentejo Litoral and the Coimbra

Region emerge at the top of the competitiveness ranking.

Key words: Regional competitiveness; Competitive index, Pillars of competitiveness, Portugal, Data

Envelopment Analysis

1. Introduction

The competitiveness of territories, such as at the regional level, has proven an area of substantial

theoretical controversy, in particular due to the argument that the companies, and not the territories,

compete for resources and markets (Huggins, Izushi, & Thompson, 2013). However, the research work

undertaken in recent years has sought to theorise and empirically measure the competitiveness of regions

(Annoni & Dijkstra, 2013; Charles & Zegarra, 2014; Dijkstra, Annoni, & Kozovska, 2011; Elissalde &

Santamaria, 2014; Huggins et al., 2013; Snieška & Bruneckiene, 2009; Titze, Brachert, & Kubis, 2011;

Ženka, Novotný, & Csank, 2014), and the attraction of political decision makers to this concept since the

1990s has been notable (Boschma, 2004; Bristow, 2005).

The concept of the competitiveness of regions within a country proves similar to the concept of

competitiveness at the national level and the main conclusions of the literature on competitiveness may be

applied to the regional level within a country (Aiginger, Bärenthaler-Sieber, & Vogel., 2013; Aiginger &

Vogel, 2015; Donaldson, 2001; Institute for Management Development, 2014; Porter, 1990; Sölvell,

2015). However, this should nevertheless take into account the different processes and dynamics inherent

to each particular reality (Boschma, 2004). The concept of regional competitiveness encapsulates far

more than its export capacity or the existence of a positive trade balance as this reaches beyond its

capacity to produce goods to embrace a broad diversity of factors and indicators portraying both tangible

and intangible resources (Kitson, Martin, & Tyler, 2004).

Evaluating regional competitiveness proves dependent on the level of region selected and especially in

terms of the European Union through the NUTS – the National Unit of Territorial Statistics. The

reference period, the availability and regularity of data as well as the selection of specific factors play an

equally substantial role.

1

The calculation of regional competitiveness for Portugal proves crucial to reducing regional inequalities

and boosting national competitiveness. Since the launch of the single European currency, the Portuguese

economy has practically stagnated with regional inequalities remaining fairly acute, especially between

regions located on the coast and those inland (Silva & Ferreira-Lopes, 2014; Soukiazis & Antunes, 2011).

Hence, guaranteeing sustainable economic growth and reducing regional inequalities constitutes an

important means of raising the competitiveness of all regions in the country. Regional competitiveness

indexes thus represent a crucial step given their aid in not only identifying those regions and areas with

particular shortcomings but also in policy decision making tailored towards each regional level (Berger,

2011; Charles & Zegarra, 2014; Huggins et al., 2013).

Despite the growing number of scientific studies on regional competitiveness related issues, there is

practically no academic research determining the competitiveness of regions in the Portuguese context

and hence we seek here to meet that gap. Despite composite indices based on univariate and multivariate

statistical analysis representing the conventional method for analysing the problem in a complex form,

other quantitative approaches are eligible for application as is the case with factor efficiency analysis

through Data Envelopment Analysis (DEA). This study deploys DEA to determine the competitiveness of

Portuguese regions (NUTS III) seeking also to serve as a catalyst for the utilisation of this methodology

to measure competitiveness, in particular the competitiveness of regions.

This current study also holds the objective of determining a competitive index for Portuguese regions

through applying a DEA data based methodology alongside stipulating the indices for the factors making

up this competitiveness and determining its taxonomy.

This takes the following organisational structure: after this introduction (Section 1), in Section 2, we

define regional competitiveness as well as discriminating some of the factors to regional competitiveness

and some regional competitiveness indices before finally setting out a regional competitive index. In

Section 3, we describe the methodology applied in this study before Section 4 details the results of the

empirical analysis. Section 5 analyses the main results and discusses their theoretical and practical

implications as well as identifying study limitations and future lines of research.

2. Regional competitiveness

Porter (1990) was the first to define national competitiveness as stemming from a nation’s capacity to

innovate with the objective of attaining or maintaining a competitive position in relation to other nations

in specific key industrial sectors and switching his unit of analysis from companies and industries to the

spatial borders occupied by companies and industries. Following on from this initial approach by Porter

(1990), the notion of clusters emerged, with the transfer of knowledge, innovation and cooperation

between the companies associated representing constructed phenomena operating at a regional level.

Furthermore, this regional level is where many of the policy levers associated with the microeconomic

2

determinants of economic development are located. Correspondingly, the idea that microeconomic

determinants generate prosperity and wealth, in opposition to factors related to monetary and exchange

rates and similar, lie at the core of the regional competitiveness concept (Porter, 2000, 2003a). The focus

on the regions reflects a growing consensus that these represent the primary spatial units, competing to

attract investments and the appropriate level for the circulation and transfer of knowledge and resulting in

agglomerations, that is clusters, of companies, industries and services (Huggins & Izushi, 2015).

2.1. The notion of regional competitiveness

Despite the developments in the regional competitiveness literature, there remains no single framework or

unquestionably accepted definitions nor even any agreement over how this concept best gets measured

(Huggins et al., 2013). Regional competitiveness proves determined by the productivity returned by just

how each region deploys its resources, both human and natural, and its capital (Porter and van der Linde

1995; Porter 1990). According to the European Commission (European Commission, 1999),

competitiveness derives from the capacity to produce goods and services that satisfy the test of

international markets, and, simultaneously maintain high and sustainable levels of income or, in general

terms, the capacity of regions to generate high income and employment whilst exposed to external

competition. According to Huovari, Kangasharju, & Alanen (2002) regional competitiveness encapsulates

the capacity of regions to nurture productive environments, highly accessible and that perpetuate and

attract factors of production and thereby fostering economic growth. Kitson et al. (2004) propose regional

competitiveness as a complex concept that should focus essentially on the long term regional indicators

and dynamics and not on restrictive notions of competition such as market shares and resources and that,

in the final instance, competitive regions are those where companies and persons wish to settle and invest.

According to Borozan (2008), regional competitiveness encapsulates the sustained capacity of a region to

compete with other regions whilst simultaneously guaranteeing its sustainable growth and economic

development and thus able to attract productive capital that enables the strengthening of its innovation

abilities. Dijkstra et al. (2011) define regional competitiveness as the capacity to provide an attractive and

sustainable environment to companies and inhabitants to live and work in. According to Huggins et al.

(2014), regional competitiveness stems from the differences in the economic growth rates between

regions and their respective capacities for leveraging future economic growth in relation to other regions

in similar stages of economic development.

2.2. Factors of regional competitiveness

Regional competitiveness proves a hermetic concept and thus not susceptible to any simplistic or singular

definition. This situation derives from the complex dynamics intrinsic to the regional reality and the

absence of any consensual definition for this concept. The competitive position of a region reflects a set

of factors that determine its competitive capacity and based on the efficient utilisation of the resources

available and hence regional competitiveness takes place through the continuous consolidation of the

3

respective competitive position based on increasing the competitive advantages of the existing industries

and on fostering new highly efficient segments of the economy (Porter 1990).

The notion of comparative advantage, with its roots reaching back to Ricardo, initially served as an

alternative to the concept of competitive advantage or competitiveness (Budd and Hirmis 2004; Kitson et

al. 2004; Porter 1998). This comparative advantage concept affirms that trade between countries and

regions incorporates the differences in their endowment of factors such as land (Budd & Hirmis, 2004;

Huggins & Williams, 2011; Maskell & Malmberg, 1999), labour (Borozan, 2008; Boschma, 2004;

Bristow, 2005; Budd & Hirmis, 2004; Maskell & Malmberg, 1999), natural resources (Annoni &

Dijkstra, 2013; Borozan, 2008; Camagni & Capello, 2010; Dijkstra et al., 2011; Huggins et al., 2013;

Kitson et al., 2004; Maskell & Malmberg, 1999; Porter, 1998, 2003b; Ström & Nelson, 2010) and capital

(Borozan, 2008; Boschma, 2004; Bristow, 2005; Budd & Hirmis, 2004; Charles & Zegarra, 2014;

Huggins & Williams, 2011; Kitson et al., 2004; Porter, 1998), which are hard to impact on through direct

public intervention (Dijkstra et al., 2011).

According to Porter (1998, 2003b), the productive advantages deriving from the existence of clusters,

such as access to specialised inputs, institutions and infrastructures, serve as the incentives for companies

to agglomerate and mutually reinforce the clusters over the course of time. This strengthening and the

agglomeration then attracts new companies given that there are already competitive advantages emerging

out of the already existing concentration.

The clusters include those governmental and non-governmental institutions fostering the creation of

knowledge (Huggins et al., 2013; Huggins & Williams, 2011; Maskell & Malmberg, 1999; Ženka et al.,

2014), diffusing processes (Huggins et al., 2013), exchanging information and ideas and regulating the

interactions between actors (Boschma, 2004), as well as institutions that have as their objective the

attracting of national and international direct investment and talent (Bristow, 2005; Huggins & Williams,

2011; Turok, 2004), and encouraging entrepreneurship (Huggins & Williams, 2011).

The global level of development of regional infrastructures directly interrelates with the respective level

of socioeconomic development (Komarova, 2014; Snieška & Simkunaite, 2015). A region may obtain

competitive advantages whenever in possession of quality infrastructures able to benefit individual

companies (Bristow, 2005; Kitson et al., 2004; Rozmahel, Grochová, & Litzman, 2016). A regional

competitiveness strategy to expand the areas potentially eligible to national and international investment

involves targeting the provision of efficient transport and communications networks (Camagni & Capello,

2010) as well as the existence of efficient public transportation systems able to reduce congestion and

boost mobility levels (Turok, 2004).

The prevailing economic framework, especially the relative cost of inputs and social duties incurred by

companies (Bristow, 2005; Camagni & Capello, 2010; Huggins & Williams, 2011; Kitson et al., 2004;

Porter & Ketels, 2003; Rozmahel et al., 2016; Snieska & Simkunaite, 2015; Ženka et al., 2014) and the

fiscal burden (Audretsch, Hülsbeck, & Lehmann, 2011; Boschma, 2004; Huggins et al., 2013; Porter,

1998; Ström & Nelson, 2010; Turok, 2004) also represent factors impacting on regional competitiveness.

4

Furthermore, their influence also reflects on the terms, conditions and efficiency in the deployment of

factors of production, in particular their respective productivity levels (Boschma, 2004; Bristow, 2005;

Budd & Hirmis, 2004; Huggins et al., 2013; Huggins & Williams, 2011; Porter & Ketels, 2003; Porter,

2003b; Turok, 2004), the intensity of competition (Kitson et al., 2004; Snieška & Bruneckiene, 2009),

levels of business sophistication (Budd & Hirmis, 2004; Huggins et al., 2013; Porter, 2003b) and

framework of demand (Bristow, 2005; Budd & Hirmis, 2004; Porter & Ketels, 2003; Porter, 1990).

Quality, competences and the level of education of human resources in a region serve to support and

sustain improvements to a region’s competitive performance (Charles & Zegarra, 2014; Kitson et al.,

2004), shaping the typology of industries and companies located in a region as the greater the level of

labour qualifications the greater the probability of industries with high rates of added value taking up

location there (Charles & Zegarra, 2014; Sepic, 2005). Hence, knowledge has become a crucial asset

determining current systems of production and generating knowledge also constitutes a fundamental

process to maintaining or raising competitiveness (Maskell & Malmberg, 1999). This knowledge may

also flow from regional institutions or complementarily between companies within the same space

through the founding of business networks that return mutual advantages (Bair & Gereffi, 2001; Dahl &

Pedersen, 2004; Keeble & Wilkinson, 1999; Lawson & Lorenz, 1999; Porter, 1998; Smith, 2003), given

their role in interlinking the different regional entities and enabling a more cohesive and harmonious

evolution of the region (Brykova, 2007). These networks of cooperation form the foundations for the

emergence and development of local clusters contributing: i) to the creation and diffusion of knowledge

(Boschma, 2004; Huggins & Williams, 2011; Maskell & Malmberg, 1999; Turok, 2004), ii) to the rapid

spread of innovative dynamics (Beugelsdijk, 2007; Calia, Guerrini, & Mourac, 2007; Dasgupta & David,

1994; Lai & Shyu, 2005; Markman, Siegel, & Wright, 2008; Miyazaki & Islam, 2007; Porter & Ketels,

2003), and iii) to the development of regional innovation systems (Brykova, 2007; Huggins & Williams,

2011; Ženka et al., 2014). This positioning also attributes innovation with a leading role in boosting

regional competitiveness (Boschma, 2005; Keeble & Wilkinson, 1999; Martin & Sunley, 2003; Porter,

2003a; Tallman & Phene, 2007; Yeung et al., 2006). The openness to knowledge and creativity also

enhances the capacity to attract human capital and boost the dynamics driving the entrepreneurial spirit

and capacity (Ács, Szerb, Ortega-Argilés, AIdis, & Coduras, 2014; Delgado, Porter, & Stern, 2010;

Huggins & Williams, 2011).

2.3. Indices of regional competitiveness

The motivation underlying the calculation of complex indices targets establishing a tool with the

objective of comparing the competitiveness of different regions. In recent years, efforts have focused on

building composite indicators of regional competitiveness in accordance with identical trends in the

advance of national competitiveness indicators (Berger, 2011). Due to the complexity of the concept and

the diverse existing definitions, various indices have been put forward. Berger (2011) identified the

existence of 46 regional indices aggregating a diversified set of potentially measures.

5

Given the complexity inherent to setting out this range of existing indices, we decided here to analyse

those stemming from similar perceptions as to the future composition of a competitive index for the 23

NUTS III (Version NUTS 2013) on mainland Portugal, the version implemented by the National and

European Statistical Systems on 1st January 2015 (INE, 2015a). Therefore, we identified and selected six

regional competitiveness indices: (1) Regional competitive index by Huovari et al. (2002); (2)

Competitive index for the United Kingdom (Huggins, 2003); (3) Objective competitiveness – Ranking the

Regions of the European Union (y Oliva & Calvo, 2005); (4) Regional competitive index for Lithuania

(Snieška & Bruneckiene, 2009); (5) Regional competitive index of the European Union (Annoni &

Kozovska, 2010); and (6) Synthetic index of regional development - Competitive index (INE, 2014).

Table 1 presents a summary of these competitive indices and identifying the factors and indicators they

incorporate.

Table 1 – Summary of the factors, subfactors and indicators contained in competitive indices

Index Factors Subfactors

Regional competitive index by Huovari et al. (2002)

Human capitalCapacity for innovationLevel of agglomeration

Accessibility

Competitive index for the United Kingdom (Huggins, 2003)

InputsOutputsResults

Objective competitiveness (y Oliva & Calvo, 2005)

Ageing of the populationEmployment market participation

Regional dynamicsFactors determinant to regional

competitivenessBasic development factorsEconomic and residential attractiveness to familiesPublic R&DDevelopment potential

EducationLevel of urbanisation

Per capita demographic pressures

Regional competitive index for Lithuania (Snieška & Bruneckiene, 2009)

Production

Human resourcesPhysical infrastructures and

geographic positioningKnowledge

Capital

Demand

Dimensions and structure of internal demand

Quality and prices of internal client demand for regional products

Quality and prices of external client demand for regional products and their international market profile

Factors of company competitivenessRegional competitive index for the European

Union (Annoni & Kozovska, 2010) Basic

InstitutionsMacroeconomic stability

InfrastructuresHealthcare

Quality of primary and secondary teaching

EfficiencyHigher education lifelong learning

Employment market efficiencyMarket scale

Innovation Technological availabilityBusiness sophistication

6

Index Factors SubfactorsCapacity for innovation

Synthetic Regional Development Index - Competitive index (INE, 2014)

Capacity to generate earnings and penetrate external markets

Factors favourable to competitiveness

Human resourcesInfrastructures

Economies of scale

Efficient transformation potential of regional economies

Production specialisationEducation and professional mobility

Entrepreneurial initiative and robustness

Exposure to external competitionInvestment in producing knowledge

As regards calculating the competitiveness indices for the Portuguese regions, there are only a few

academic studies attempting this benchmarking. Of these, we would highlight the analytical studies by

Annoni & Dijkstra (2013) and de Dijkstra et al. (2011) even while this spans the overall context for

European Union NUTS II regions. In another research project, Oliveira (2014) analyses the

competitiveness of NUTS III regions (version 2002) deploying 22 indicators to portray the

competitiveness of Portuguese regions across three major components: (1) structural (human capital,

economic base and territorial system of organisation), (2) dynamics (human capital and economic base),

and (3) institutional (density of the regional innovation system, entrepreneurial spirit). Oliveira (2014)

concludes that the most competitive region is that of Greater Lisbon, followed by Greater Oporto, the

Setúbal Peninsula, Baixo Vouga and Baixo Mondego, Pinhal Litoral and Cávado with the regions

returning the greatest lags in competitive development being Pinhal Interior Norte, Pinhal Interior Sul,

Alto Alentejo, Beira Interior Norte and Serra da Estrela. This author identifies four stages of

competitiveness in Portugal (competitive regions, regions with competitive shortcomings, non-

competitive regions and peripheral regions). In terms of the factors of regional competitiveness, Oliveira

(2014) reports that the Greater Lisbon region attains a strong performance in practically all the subfactors

and the exact opposite to the Serra da Estrela regions that display a poor level of performance in almost

all subfactors.

3. Methodology

3.1. Defining a regional competitive index

Based on the description set out above, we observe the existence of a great diversity as regards the choice

of the respective factors and indicators and that on occasion result from subjective decisions taken by the

researchers themselves. Nevertheless, the indicators set out convey ideas as to the type of factors needing

consideration whenever studying regional competitiveness.

The main concern in developing the index was its incorporation of available and comparable data at the

regional and national level and that in some way reflect the interconnection between the factors of

competitiveness driving the tangible and intangible regional performance levels.

7

The factors, in this study referred to as pillars, are: (1) human capital, (2) business dynamics, (3)

employment market, (4) market scale, (5) technological availability, and (6) innovation. In relation to

regional performance, the measures adopted are GDP per capita, average monthly earnings per capita

and the unemployment rate. Figure 1 depicts the model serving to measure regional competitiveness.

Figure 1 –Regional competitiveness model

3.2. Data

The regional analysis focuses on the NUTS III level and not on the NUTS II level as is the case with

different competitiveness indices. This fact derives from the existence of just 5 NUTS II regions in

continental Portugal. Thus, the study object features the 23 NUTS III regions of mainland Portugal and

thus excluding from the analysis the Autonomous Region of Madeira and the Autonomous Region of the

Azores on the one hand due to difficulties in accessing statistical data on these two regions and on the

other hand because they span completely different realities given they are two archipelagos located out in

the Atlantic at a considerable distance from the continent. The data derived from the National Institute of

Statistics Territorial Statistics (INE, 2015b).

3.3. Variables

The regional competitive index structure designed in this study rests upon six pillars of competitiveness

and three components to the results of competitiveness.

3.3.1. Pillars of competitiveness

i) Human capital (Pillar 1) – Human capital and knowledge rank as essential factors in determining the

competitive differences between regions (Camagni & Capello, 2010; Huggins et al., 2013; Huggins,

2003), with the utilisation of indicators portraying education (Huggins, 2003; Huovari et al., 2002; Porter,

1998, 2003b; Snieška & Bruneckiene, 2009), the illiteracy and school dropout rates, the proportion of the

8

population aged over 15 with a higher education decree (Annoni & Kozovska, 2010) and the number of

higher education students (y Oliva & Calvo, 2005).

ii) Business dynamics (Pillar 2) – The dynamics in effect at companies are an essential determinant in

regional competitiveness strategies (Brykova, 2007). Evaluating such dynamics made recourse to

variables such as GVA per employee (Annoni & Kozovska, 2010), export volumes per worker and the

proportion of high technology exports (INE, 2014), the number of companies per thousand inhabitants

(Huggins, 2003), the proportion of company turnover invested back into the business (Snieška &

Bruneckiene, 2009), the company start-up rate (Audretsch et al., 2011; Delgado et al., 2010; INE, 2014)

and the proportion of employees working in knowledge intensive companies (Snieška & Bruneckiene,

2009).

iii) Employment market (Pilar 3) – An important component of regional competitiveness stems from the

employment market (Annoni & Dijkstra, 2013; Ženka et al., 2014), given that the presence of efficient

and flexible markets contribute towards the efficient allocation of resources and thus correspondingly

boosting competitiveness (Schwab & Porter, 2008). This pillar includes statistics detailing the active

population (Huovari et al., 2002), the employment rate (Snieška & Bruneckiene, 2009; Annoni &

Kozovska, 2010), the percentage difference between the male and female employment rates (Annoni &

Kozovska, 2010), the female unemployment rate (Annoni & Kozovska, 2010), the number of full time

employees holding higher education qualifications (Snieška & Bruneckiene, 2009) and the migratory

balance (Snieška & Bruneckiene, 2009).

iv) Market scale (Pilar 4) – Another determining factor of competitiveness stems from the effects of the

regional market scale (Sepic, 2005) given that larger markets enable companies to develop and benefit

from economies of scale and stimulating entrepreneurship and innovation (Dijkstra et al., 2011). The

indicators included in this pillar portray GDP (Annoni & Kozovska, 2010; INE, 2014), the population

totals (Annoni & Kozovska, 2010), population density (Huovari et al., 2002; Snieška & Bruneckiene,

2009; INE, 2014), business export volumes (Snieška & Bruneckiene, 2009; INE, 2014), number of crimes

registered per thousand inhabitants (Annoni & Dijkstra, 2013; Charles & Zegarra, 2014), the total amount

of accommodation and the average occupancy rate (y Oliva & Calvo, 2005; Snieška & Bruneckiene,

2009; INE, 2014).

v) Technological availability (Pilar 5) – The information and communication technologies (ICTs) have

profoundly changed the organisational structures of companies, facilitating the adoption of new and more

efficient working practices and styles, improving on commercial processes and increasing productivity

having become an essential factor to competitiveness (Dijkstra et al., 2011). To evaluate technological

availability, the study adopts indicators such as the level of Internet access per thousand inhabitants

(Snieška & Bruneckiene, 2009; Annoni & Kozovska, 2010; INE, 2014), the number of residential

telephones per thousand inhabitants (Charles & Zegarra, 2014; Snieška & Simkunaite, 2015), number of

9

inhabitants per ATM (Institute for Management Development, 2014) and the proportion of school

computers with Internet access (Snieška & Bruneckiene, 2009).

vi) Innovation (Pilar 6) – Innovation is attributed one of the lead roles in regional competitiveness

catalysts (Boschma, 2005; Keeble & Wilkinson, 1999; Martin & Sunley, 2003; Porter, 2003a; Tallman &

Phene, 2007; Yeung et al., 2006) alongside openness to knowledge and creativity, the capacity to attract

human capital and the dynamic entrepreneurial capacity (Ács et al., 2014; Delgado et al., 2010; Huggins

& Williams, 2011). This pillar contains indicators spanning the proportion of GDP on R&D expenditure

carried out by the state, companies and higher education institutions (Huovari et al., 2002; y Oliva &

Calvo, 2005), the proportion of company employees with doctoral degrees, the number of employees

working full time on R&D in companies and the number of patent requests per million inhabitants

(Huovari et al., 2002; Snieška & Bruneckiene, 2009; Annoni & Kozovska, 2010; y Oliva & Calvo, 2005).

3.3.2. Variables for the resulting effects

The aforementioned pillar impact on regional competitiveness. Thus, in terms of tangible results, the

study furthermore evaluated indicators on the average level of remuneration per capita and the

unemployment rate and, regarding the intangible results, this also measured GDP per capita (Huggins,

2003).

3.4. Data analysis methods

The data analysis methods targeting the determining of the regional competitiveness indices primarily

applied Principal Component Analysis (PCA) (Annoni & Dijkstra, 2013; Dijkstra et al., 2011; y Oliva &

Calvo, 2005), Regression Analysis (Huggins, 2003; Snieška & Bruneckiene, 2009) and through numerical

indices (Huggins, 2003, Huovari et al., 2002; INE, 2014). Despite being less frequent, the Data

Envelopment Analysis (DEA) methodology has also served as a methodology for evaluating regional

competitiveness (Charles & Zegarra, 2014; Staníčková & Skokan, 2012).

This study applies a mixed methodology based on three methods: DEA, PCA and numerical indices. The

determining of the regional competitive index thus follows three procedures.

Initially, this applies DEA to establish a regional index for each one of the pillars under study, with the

objective of determining regional competitiveness by the respective efficiency level, thus, the quantity of

inputs necessary to obtaining a determined output. The DEA is a non-parametric mathematical technique

developed in order to determine the efficiency of productive units where considering the financial aspect

is not relevant in which each individual observation is optimised with the objective of calculating the

efficiency frontier that converts multiple inputs into multiple outputs. Despite a range of means of

determining this frontier, two models are most commonly applied, the CCR (Charnes, Cooper, & Rhodes,

10

1978) and the BCC (Banker, Charnes, & Cooper, 1984). The originally presented CCR model works with

constant returns to scale, therefore, any variation in the inputs produces proportional variations in the

outputs. The BCC model considers variable returns to scale, hence enabling the productive units

operating with low input values to return rising scales and those operating with high values to return

falling scales. Both models orient towards either inputs (by optimising the division between the weighted

sum of the outputs and the weighted sum of the inputs) or the outputs (maximising the outputs while

maintaining the inputs unchanged). The BCC model supplies a more realistic expression of economic

reality and the relationships ongoing between countries and regions (Staníčková & Skokan, 2012) and

thus adopting this modelling approach and with outputs oriented by each pillar of competitiveness. The

variables incorporating the pillars of competitiveness are the inputs and the variables resulting are the

outputs. The second phase of analysis applies PCA, which represents a multivariate statistical technique

with the objective of synthesising a set of variables within a smaller quantity of factors (Hair, Black,

Babin, Anderson, & Tatham, 2010) for the set of indicators making up each one of the pillars, retaining

the standardised component score that explains the greatest proportion of the variation in these indicators

with this process involving running PCAs. A further phase calculates the numerical index of between 0

and 100 by the minimax method (subtracting the minimum reference from the observed value and

dividing this difference by the difference between the maximum reference value and the minimum

reference value) for each pillar and each territorial unit. Finally, this once again makes recourse to DEA

in which the inputs are the numerical indices for each pillar and the variables are the output results.

In order to attain a taxonomic description of the competitiveness of regions, we applied cluster analysis¸ a

multivariate technique with the objective of grouping cases with similar profiles across a defined set of

characteristics and to this end deploying the average distance to the squared Euclidian distance and then

the Ward method to the grouping within the scope of gaining a similar number of regions for each group

(Hair et al., 2010). The variables applied to determine the groups of regions reporting similar

competitiveness profiles were the regional indices stemming from each of the six pillars.

All the DEA calculations made recourse to the Frontier Analyst software version 4.2.0 (Banxia Holdings

Ltd, Kendal, UK) and for estimating ACP, the numerical indices and the cluster analysis, we applied IBM

SPSS software version 22.0 (IBM Corporation, New York, USA).

4. Results

The results span three different levels. In a first phase, we present results stemming from the regional

competitiveness index and the respective ranking, followed by analysis of competitiveness in terms of the

pillars. Finally, we set out the groups of regions according to the profile of the competitive pillars.

4.1. Regional competitiveness index

11

Set out in Table 2, we find the most competitive group of regions is led by the Lisbon Metropolitan Area

followed by the coastal Alentejo Litoral and Coimbra, the Algarve and Beira Baixa regions. In turn, the

NUTS III reporting the weakest competitive positions contain Viseu Dão Lafões, Beiras e Serra da

Estrela, Ave, Alto Minho and Alto Alentejo. Taking into consideration these results, we identify a certain

discrepancy between these results and the index published by the INE in which the most competitive

regions are the Lisbon Metropolitan Area, trailed by Aveiro region, the Oporto Metropolitan Area,

Cávado and Alto Minho regions, whilst the least competitive regions are Alto Alentejo, Douro, Alto

Tâmega, Beiras e Serra da Estrela, Médio Tejo and Terras de Trás-os-Montes.

As regards the noticeable difference between the index calculated by this study and that of the INE, the

Oporto Metropolitan Area stands out and ranked third in the INE table whilst this study’s calculations

place the area in 17th place. In relation to the NUTS Aveiro and Cávado regions, the INE ranking places

them among the leading regions whilst the index ranks these regions in intermediate positions, in 11th and

12th position respectively. On the contrary, the NUTS Alto Minho (5th) and Ave (7th) regions attain

upper places in the INE ranking but come in among the bottom ranked regions here (20th and 21st placed

respectively). In terms of the results similar to both of the respective indices, we may highlight the high

competitive performance of the Lisbon Metropolitan Area and the low levels of competitiveness of the

NUTS Beiras e Serra da Estrela and Alto Alentejo regions.

Despite the NUT III regions proving different, this study also contains equal similarities and differences

to the results of the study by Oliveira (2014). Both these research findings and those returned by Oliveira

(2014) confirm the Lisbon Metropolitan Area (NUTS 2013)/Greater Lisbon (NUTS 2002) as the most

competitive region whilst the Beiras e Serra da Estrela (NUTS 2013)/ Serra da Estrela (NUTS 2002)

regions propped up the rankings for regional competitiveness. The main divergence between this study

and that of Oliveira (2014) interrelates with the ranking attributed to the NUTS Oporto Metropolitan Area

(NUTS 2013)/Greater Oporto (NUTS 2002) taking one of the upper positions in the Oliveira (2014)

ranking and 17th position in that returned here. The differences, between this study and that of the INE,

may stem from the fact that the methodologies applied by the INE and by Oliveira (2014) do not take into

account the results of competitiveness in accordance with the factors of competitiveness, thus, their

efficiency in terms of GDP per capita, average monthly income and the unemployment rate given they

apply numerical indices. One example of this fact incorporates how the Oporto Metropolitan Region

contains high levels of human capital, market scale, technological availability and innovation but that

these do not come in for efficient application as the region contains the highest rate of unemployment and

a GDP per capita rate below the national average. In contrast, Alentejo Litoral, which returns low levels

of human capital, employment market, market scale and innovation proves to be the NUT III region with

the highest GDP per capita in national terms and even outpacing the Lisbon Metropolitan Area in this

indicator of results.

Table 2 – Regional competitiveness ranking Ranking NUTS 3 Score Ranking

INE

12

1 Lisbon Metropolitan Area 100.0 1

2 Alentejo Litoral 99.0 6

3 Região de Coimbra 98.6 9

4 Algarve 98.1 13

5 Beira Baixa 96.3 17

6 Alentejo Central 96.3 14

7 Alto Tâmega 95.6 21

8 Lezíria do Tejo 94.3 15

9 Douro 93.9 22

10 Baixo Alentejo 92.9 12

11 Região de Aveiro 92.9 2

12 Cávado 91.9 4

13 Terras de Trás-os-Montes 91.6 18

14 Região de Leiria 90.3 8

15 Tâmega e Sousa 89.4 16

16 Médio Tejo 89.2 19

17 Oporto Metropolitan Area 89.1 3

18 Oeste 87.8 11

19 Alto Alentejo 86.5 23

20 Alto Minho 83.6 5

21 Ave 82.6 7

22 Beiras e Serra da Estrela 80.1 20

23 Viseu Dão Lafões 75.1 10

4.2. Pillars of regional competitiveness

Due to its complexity, this theme requires thorough analysis of each pillar of competitiveness and

representing a further added value arising from this project. Table 3 displays the rankings for the six

pillars defined in the study. We should from the outset stress that the Lisbon Metropolitan Area leads in

competitiveness across all pillars and that the coastal Alentejo Litoral region takes second place in six of

the rankings whilst located in 8th place in the innovative capacity pillar. The Coimbra region features as a

region with high levels of competitiveness in terms of the pillars detailing human capital, business

dynamics, employment market, technological availability and innovation, and taking 2nd position in the

innovation ranking. On the contrary, the NUTS Viseu Dão Lafões and Beiras e Serra da Estrela regions

display significant competitive shortcomings across all of the pillars subject to analysis. Both these

findings and those of Oliveira (2014) confirm the Lisbon Metropolitan Area (NUTS 2013)/Greater

Lisbon (NUTS 2002) as the most competitive region in terms of the defined determinants of

competitiveness. Another similar result between this study and that of Oliveira (2014) derives from the

Beiras e Serra da Estrela (NUTS 2013)/ Serra da Estrela (NUTS 2002) region occupying the lowest

positions in the factors of regional competitiveness rankings.

13

Table 3 – Regional competitiveness pillar rankings Human capital

(Pillar 1)Business dynamics

(Pillar 2)Employment market

(Pillar 3)Market scale

(Pillar 4)Technological availability

(Pillar 5)Innovation(Pillar 6)

1 - Lisbon Metropolitan Area (100,0) 1 - Lisbon Metropolitan Area (100,0) 1 - Lisbon Metropolitan Area (100,0) 1 - Lisbon Metropolitan Area (100,0) 1 - Lisbon Metropolitan Area (100,0) 1 - Lisbon Metropolitan Area (100,0)

2 - Alentejo Litoral (98,9) 2 - Alentejo Litoral (99,2) 2 - Alentejo Litoral (98,8) 2 - Alentejo Litoral (98,1) 2 - Alentejo Litoral (97,6) 2 - Região de Coimbra (99,9)

3 - Região de Aveiro (98,9) 3 - Região de Coimbra (99,1) 3 - Oporto Metropolitan Area (98,5) 3 - Lezíria do Tejo (97,9) 3 - Região de Aveiro (97,3) 3 - Região de Aveiro (99,5)

4 - Região de Coimbra (98,7) 4 - Algarve (98) 4 - Região de Coimbra (97,8) 4 - Baixo Alentejo (95,9) 4 - Região de Coimbra (96,8) 4 - Cávado (99)

5 - Região de Leiria (98,6) 5 - Douro (96,9) 5 - Região de Leiria (97,2) 5 - Algarve (93,9) 5 - Lezíria do Tejo (95,7) 5 - Algarve (98,9)

6 - Lezíria do Tejo (98) 6 - Alto Tâmega (94,8) 6 - Alentejo Central (97,1) 6 - Cávado (93,4) 6 - Beira Baixa (94,9) 6 - Oeste (97,8)

7 - Alentejo Central (97,8) 7 - Tâmega e Sousa (93,6) 7 - Algarve (96,7) 7 - Douro (92,5) 7 - Terras de Trás-os-Montes (94) 7 - Alentejo Central (97)

8 - Baixo Alentejo (97,1) 8 - Beira Baixa (92) 8 - Cávado (96,5) 8 - Beira Baixa (92,2) 8 - Alto Tâmega (93,2) 8 - Alentejo Litoral (96,8)

9 - Algarve (96,7) 9 - Alentejo Central (91,4) 9 - Alto Alentejo (96,1) 9 - Terras de Trás-os-Montes (92,2) 9 - Tâmega e Sousa (92,7) 9 - Alto Tâmega (96,6)

10 - Médio Tejo (94,6) 10 - Oporto Metropolitan Area (90,2) 10 - Douro (94) 10 - Alto Tâmega (90,9) 10 - Douro (92,3) 10 - Baixo Alentejo (96,6)

11 - Cávado (94) 11 - Terras de Trás-os-Montes (89,3) 11 - Beira Baixa (93,7) 11 - Alentejo Central (89,5) 11 - Médio Tejo (92,3) 11 - Beira Baixa (95,1)

12 - Beira Baixa (93,7) 12 - Beiras e Serra da Estrela (88,1) 12 - Terras de Trás-os-Montes (91,9) 12 - Oporto Metropolitan Area (89,3) 12 - Alto Minho (91) 12 - Douro (95)

13 - Terras de Trás-os-Montes (91,4) 13 - Região de Leiria (87,4) 13 - Alto Tâmega (91,4) 13 - Médio Tejo (89,3) 13 - Baixo Alentejo (90,2) 13 - Lezíria do Tejo (93,9)

14 - Alto Tâmega (91,2) 14 - Médio Tejo (85,5) 14 - Tâmega e Sousa (90,3) 14 - Região de Coimbra (88,6) 14 - Região de Leiria (90,2) 14 - Médio Tejo (93,9)

15 - Tâmega e Sousa (91) 15 - Baixo Alentejo (85,2) 15 - Baixo Alentejo (89,7) 15 - Alto Alentejo (87) 15 - Ave (89,8) 15 - Região de Leiria (93,7)

16 - Oeste (89,5) 16 - Lezíria do Tejo (83,7) 16 - Ave (89,2) 16 - Região de Aveiro (86,3) 16 - Viseu Dão Lafões (88,9) 16 - Terras de Trás-os-Montes (92,1)

17 - Alto Minho (85,2) 17 - Oeste (83,5) 17 - Lezíria do Tejo (87,2) 17 - Tâmega e Sousa (84,5) 17 - Cávado (87,9) 17 - Tâmega e Sousa (90,7)

18 - Ave (84,4) 18 - Região de Aveiro (82,2) 18 - Beiras e Serra da Estrela (86,7) 18 - Alto Minho (82) 18 - Oporto Metropolitan Area (87,2) 18 - Alto Alentejo (92,1)

19 - Douro (82,6) 19 - Alto Alentejo (80,8) 19 - Alto Minho (85,3) 19 - Oeste (79,6) 19 - Alentejo Central (86) 19 - Oporto Metropolitan Area (89,2)

20 - Viseu Dão Lafões (81,4) 20 - Alto Minho (79,4) 20 - Oeste (84,7) 20 - Região de Leiria (75,2) 20 - Alto Alentejo (84,6) 20 - Alto Minho (88,5)

21 - Alto Alentejo (81,1) 21 - Cávado (78,1) 21 - Médio Tejo (84,2) 21 - Beiras e Serra da Estrela (70,7) 21 - Oeste (82,7) 21 - Viseu Dão Lafões (85,8)

22 - Oporto Metropolitan Area (79,2) 22 - Viseu Dão Lafões (77,3) 22 - Região de Aveiro (83,6) 22 - Viseu Dão Lafões (69,7) 22 - Beiras e Serra da Estrela (81,6) 22 - Beiras e Serra da Estrela (81,7)

23 - Beiras e Serra da Estrela (74,6) 23 - Ave (76) 23 - Viseu Dão Lafões (82,5) 23 - Ave (69,4) 23 - Algarve (70,7) 23 - Ave (80,7)

14

4.3. Regional competitiveness profiles

Cluster analysis with the objective of returning a taxonomic description of Portuguese regions

discriminated between four regional competitiveness profiles. Table 4 presents the areas making up the

regional taxonomy with the map of Portugal below detailing the 23 NUTS III and their respective states

of competitiveness in order to put forward a better interpretation of the results (Figure 2).

Table 4 – Regional groupings according to the competitiveness taxonomy

Highly competitive regions Averagely competitive regions in pillars 1,2,3 and 6

Averagely competitive regions in pillars 4 and 5

Regions with low competitive levels

Lisbon Metropolitan Area Algarve Beira Baixa Oeste

Alentejo Litoral Alentejo Central Alto Tâmega Alto Alentejo

Região de Coimbra Região de Leiria Lezíria do Tejo Alto Minho

Oporto Metropolitan Área Douro Beiras e Serra da Estrela

Região de Aveiro Ave

Baixo Alentejo Viseu Dão Lafões

Cávado

Terras de Trás-os-Montes

Tâmega e Sousa

Médio Tejo

Figure 2 – Regional grouping according to the six pillars of regional competitiveness

15

The taxonomic profiles of the four regional groups are: (1) highly competitive regions; (2) regions with

average competitive levels in pillars 1,2,3 and 6; (3) regions with average competitive levels in pillars 4

and 5; and (4) regions with low levels of competitiveness.

Lisbon Metropolitan Area, Alentejo Litoral and Coimbra Region constitute one group of regions

displaying high levels of competitiveness across all of the six defined pillars. The Algarve, Alentejo

Central, Leiria Region and Oporto Metropolitan Area establish a second group with average positions in

terms of the competitiveness of their human capital, business dynamics, employment market and

innovation and low competitive performance levels in the pillars for market scale and technological

availability. The third group, made up of Beira Baixa, Alto Tâmega, Lezíria do Tejo, Douro, Aveiro

Region, Baixo Alentejo, Cávado, Terras de Trás-os-Montes, Tâmega and Sousa and Médio Tejo display

average levels of competitiveness in the pillars for market scale and technological availability and low

levels in terms of the other pillars. Finally, the group of lesser competitive territories across all pillars

contains the Oeste, Alto Alentejo, Alto Minho, Beiras e Serra da Estrela, Ave and Viseu Dão Lafões

regions. The regions making up the regional competitiveness profiles defined by this study do not align

with those groups reported by Oliveira (2014).

5. Conclusions

Regional competitiveness emerges out of diverse factors and hence is not susceptible to any full definition

by any one or various social and economic indicators and hence the complexity encountered in attempting

its measurement. This study proposes a conceptual framework made up of six pillars (Human capital,

Business dynamics, Employment market, Market scale, Technological availability and Innovation) as

well as variables measuring different facets of regional competitiveness and bearing an impact on GDP

per capita, monthly income levels and the unemployment rate.

The study objective targeted the determining of a competitive index for Portuguese regions (NUTS III),

establishing indices of the factors composing this regional competitiveness and determining its respective

taxonomy.

The results demonstrate that the Lisbon Metropolitan Area, the Alentejo Litoral and Coimbra regions

occupy the top three places in the rankings. The Lisbon Metropolitan Area, hosting the Portuguese

capital, takes first place whether in the global ranking or that for each respective pillar while the Alentejo

Litoral region, hosting the main Portuguese maritime port and the country’s main petrochemicals cluster,

holds second place both in overall terms and in five of the six pillars. Coimbra Region, in keeping with its

poles in healthcare and higher education, attains third place in the regional competitiveness ranking. In

contrast, the NUTS Beiras e Serra da Estrela and Viseu Dão Lafões regions, two adjoining inland

territories, display the lowest levels of competitiveness.

16

Implications

These results hold fundamental implications. As regards their theoretical consequences, this study sets out

a concept of regional competitiveness that displays on the one hand factors impacting on such

competitiveness and tangible and intangible results for competitiveness on the other hand. Analysis of the

efficiency of the diverse factors of competitiveness provides a different approach to the conventional

means and leading to another vision on competitiveness.

As regards the practical implications, Portugal needs to still further strengthen and deepen the regional

focus of its economic policies. Should the Portuguese government wish to reduce regional asymmetries

and inequalities, the competitiveness of regions requires evaluation as despite the national economy

having developed in recent decades, there remain major differences in the competitiveness of its regions.

In order to return better balanced growth and reduce regional asymmetries, the government clearly needs

to invest in education and the technological capacities of the less competitive regions as well as endow

incentives to fostering regional NUTS III systems of innovation. Portugal should furthermore found and

develop centres of industrial competences in less competitive regions as well as backing the launch of

strategic alliances and sharing both infrastructures and specialist knowledge. In these less competitive

regions, subsidies should serve to foster cooperation between businesses and universities. Portugal should

also advance with efforts to revitalise its less favoured regions through making market based approaches.

Portuguese firms should also play a leading role in certain regions given that, despite this lesser

developed regional status, they also report greater levels of efficiency in the utilisation of factors of

competitiveness than regions otherwise displaying higher levels of development.

Limitations and future lines of research

This study makes a range of contributions but also incorporates certain limitations. Another means of

evaluating the efficiency of territorial units would incorporate stochastic frontier models with the

objective of modelling the regional competitive index and the efficiency of regions with a certain degree

of probability and then comparing both results.

Another study limitation stems from the chronology of the data collected as despite this index undergoing

calculation based on the year 2014, some of the variables relate to earlier years (in particular to 2011 and

2013) given that for many of the statistical indicators applied there were no figures available for 2014.

Longitudinal studies would serve to ascertain the chronological evolution of regional competitiveness.

References

17

Ács, Z. J., Szerb, L., Ortega-Argilés, R., AIdis, R., & Coduras, A. (2014). The Regional Application of the Global Entrepreneurship and Development Index (GEDI): The Case of Spain. Regional Studies, ahead-of-p, 1–18. http://doi.org/10.1080/00343404.2014.888712

Aiginger, K., Bärenthaler-Sieber, S., & Vogel., J. (2013). Competitiveness under New Perspectives.

Aiginger, K., & Vogel, J. (2015). Competitiveness : from a misleading concept to a strategy supporting Beyond GDP goals. Competitiveness Review, 25(5).

Annoni, P., & Dijkstra, L. (2013). EU Regional Competitiveness Index - RCI 2013. Luxembourg: Publications Office of the European Union. http://doi.org/10.2788/61698

Annoni, P., & Kozovska, K. (2010). EU Regional competitiveness Index.

Audretsch, D. B., Hülsbeck, M., & Lehmann, E. E. (2011). Regional competitiveness, university spillovers, and entrepreneurial activity. Small Business Economics, 39(3), 587–601. http://doi.org/10.1007/s11187-011-9332-9

Banker, R. D., Charnes, A., & Cooper, W. W. (1984). Some models for estimating technical scale inefficiencies in data envelopment analysis. Management Science, 30(9), 1078–1092.

Békés, G. (2015). Measuring regional competitiveness : A survey of approaches, measurement and data.

Berger, T. (2011). An Overview and Analysis on Indices of Regional Competitiveness. Review of Economics & Finance, 1, 17–33.

Beugelsdijk, S. (2007). The regional environment and a firm’s innovative performance: A plea for a multilevel interactionist approach. Economic Geography, 83(2), 181–199.

Borozan, D. (2008). Regional competitiveness: Some conceptual issues and policy implications. Interdisciplinary Management Research, 4(1), 50–63.

Boschma, R. (2004). Competitiveness of regions from an evolutionary perspective. Regional Studies, 38(9), 1001–1014. http://doi.org/10.1080/0034340042000292601

Boschma, R. (2005). Proximity and innovation: A critical assessment. Regional Studies, 39(1), 61–74. http://doi.org/10.1080/0034340052000320887

Bristow, G. (2005). Everyone’s a `winner': problematising the discourse of regional competitiveness. Journal of Economic Geography, 5(3), 285–304. http://doi.org/10.1093/jeg/lbh063

Brykova, I. (2007). Determinants of the International Competitiveness of National Regions in the Global Economic Environment *. International Economic Policy, 2, 5–25.

Budd, L., & Hirmis, A. K. (2004). Conceptual framework for regional competitiveness. Regional Studies, 38(9), 1015–1028. http://doi.org/10.1080/0034340042000292610

Calia, R. C., Guerrini, F. M., & Mourac, G. L. (2007). Innovation networks: From technological development to business model reconfiguration. Technovation, 27(8), 426–432. http://doi.org/10.1016/j.technovation.2006.08.003

Camagni, R., & Capello, R. (2010). Macroeconomic and territorial policies for regional competitiveness: an EU perspective. Regional Science Policy & Practice, 2(1), 1–19. http://doi.org/10.1111/j.1757-7802.2010.01016.x

18

Charles, V., & Zegarra, L. F. (2014). Measuring regional competitiveness through Data Envelopment Analysis: A Peruvian case. Expert Systems with Applications, 41(11), 5371–5381. http://doi.org/10.1016/j.eswa.2014.03.003

Charnes, A., Cooper, W. W., & Rhodes, E. (1978). Measuring the efficiency of decision-making units. European Journal of Operational Research, 2, 429–444.

Dasgupta, P., & David, P. A. (1994). Toward a new economics of science. Research Policy, 23(5), 487–521.

Delgado, M., Porter, M. E., & Stern, S. (2010). Clusters and entrepreneurship. Journal of Economic Geography, 10(4), 495–518. http://doi.org/10.1093/jeg/lbq010

Dijkstra, L., Annoni, P., & Kozovska, K. (2011). Competitiveness Index : Theory , Methods and Findings.

Donaldson, T. (2001). The ethical wealth of nations. JOURNAL OF BUSINESS ETHICS, 31(1), 25–36. http://doi.org/10.1023/A:1010776922597

Elissalde, B., & Santamaria, F. (2014). Territorial cohesion and regional competitiveness : Defining key-notions in the EU ’ s regional policy. Social Sciences, 1(27-35). http://doi.org/10.11648/j.ss.s.20140101.14

European Commission. (1999). Sixth Periodic Report on the Social and Economic Situation of Regions in the EU.

Hair, J. F., Black, B., Babin, B., Anderson, R. E., & Tatham, R. L. (2010). Multivariate Data Analysis (7th Editio). London, United Kingdom: Pearson Prentice Hall.

Huggins, R. (2003). Creating a UK competitiveness index: regional and local benchmarking. Regional Studies, 37(1), 89–96. http://doi.org/10.1080/0034340022000033420

Huggins, R., & Izushi, H. (2015). > The Competitive Advantage of Nations>: origins and journey. Competitiveness Review, 25(5).

Huggins, R., Izushi, H., Prokop, D., & Thompson, P. (2014). The Global Competitiveness of Regions.

Huggins, R., Izushi, H., & Thompson, P. (2013). Regional Competitiveness: Theories and Methodologies for Empirical Analysis. The Business and Economics Research Journal, 6(2), 155–172. http://doi.org/10.7835/jcc-berj-2013-0086

Huggins, R., & Williams, N. (2011). Entrepreneurship and regional competitiveness: the role and progression of policy. Entrepreneurship & Regional Development, 23(9-10), 907–932. http://doi.org/10.1080/08985626.2011.577818

Huovari, J., Kangasharju, A., & Alanen, A. (2002). Constructing an Index for Regional Competitiveness. In Z. J. Acs, H. L. F. de Groot, & P. Nijkamp (Eds.), The Emergence of the Knowledge Economy (pp. 121–138). Springer Berlin Heidelberg. Retrieved from http://books.google.com/books?hl=en&lr=&id=33qYWb_sJ3oC&oi=fnd&pg=PA121&dq=CONSTRUCTING+AN+INDEX+FOR+REGIONAL+COMPETITIVENESS+*&ots=XeHNM5UeQt&sig=BVTr-tFEJldSwxUZney0qK4g2FM

INE. (2014). DOCUMENTO METODOLÓGICO: Índice Sintético de Desenvolvimento Regional (Versão 1.4).

INE. (2015a). As novas unidades territoriais para fins estatísticos - NUTS 2013.

19

INE. (2015b). Estatísticas Territoriais. Retrieved 1 June 2015, from https://www.ine.pt/xportal/xmain?xpid=INE&xpgid=ine_unid_territorial&menuBOUI=13707095&contexto=ut&selTab=tab3

Institute for Management Development. (2014). IMD World Competitiveness Yearbook 2014. Geneve: IMD.

Keeble, D., & Wilkinson, F. (1999). Collective learning and knowledge development in the evolution of regional clusters of high technology SMEs in Europe. Regional Studies, 33(4), 295–303.

Kitson, M., Martin, R., & Tyler, P. (2004). Regional Competitiveness: An Elusive yet Key Concept? Regional Studies, 38(9), 991–999. http://doi.org/10.1080/0034340042000320816

Komarova, V. N. (2014). An Infrastructure Factor in Regional Competitiveness. Mediterranean Journal of Social Sciences, 5(18), 355–360. http://doi.org/10.5901/mjss.2014.v5n18p355

Lai, H. C., & Shyu, J. Z. (2005). A comparisson of innovation capacity at science parks across the Taiwan strait: the case of Zhangjiang High-Tech Park and Hsinchu Science-based Industrial Park. Technovation, 25(7), 805–813. http://doi.org/10.1016/j.technovation.2003.11.004

Markman, G. D., Siegel, D. S., & Wright, M. (2008). Research and Technology Commercialization. Journal of Management Studies, 45(8), 1401–1423. http://doi.org/10.1111/j.1467-6486.2008.00803.x

Martin, R., & Sunley, P. (2003). Deconstructing clusters: chaotic concept or policy panacea? Journal of Economic Geography, 3(1), 5. Retrieved from http://search.proquest.com/docview/195069078?accountid=26401 LA  - English

Maskell, P., & Malmberg, A. (1999). Localised learning and industrial competitiveness. Cambridge Journal of Economics, 23(2), 167–185. http://doi.org/10.1093/cje/23.2.167

Miyazaki, K., & Islam, N. (2007). Nanotechnology systems of innovation - An analysis of industry and academia research activities. Technovation, 27(11), 661–675. http://doi.org/10.1016/j.technovation.2007.05.009

Oliveira, A. F. R. de. (2014). A Problemática da Competitividade Regional - Entendimento Conceptual e Formas de Medição: O Caso Português (Master’s thesis). Instituto Universitário de Lisboa. Retrieved from http://hdl.handle.net/10071/8832

Porter, M. E. (1990). The Competitive Advantage of Nations. New York: The Free Press.

Porter, M. E. (1998). Clusters and the new economics of competition. Harvard Business Review, (November-December), 77–90.

Porter, M. E. (2000). Location, competition, and economic development: Local clusters in a global economy. Economic Development Quarterly, 14(1), 15–34. Retrieved from http://search.proquest.com/docview/215400421?accountid=26401 LA  - English

Porter, M. E. (2003a). Building the Microeconomic Foundations of Prosperity: Findings from the Business Competitiveness Index. In M. E. Porter, K. Schwab, X. Sala-i-Martín, & A. Lopes-Claros (Eds.), The Global Competitiveness Report 2012-2013 (pp. 29–56). Geneva, Switzerland: World Economic Forum.

Porter, M. E. (2003b). The economic performance of regions. Regional Studies, 37(6-7), 549–578. http://doi.org/10.1080/0034340032000108688

20

Porter, M. E., & Ketels, C. H. M. (2003). UK Competitiveness : moving to the next stage. DTI Economics Paper, (3). Retrieved from http://178.79.154.241/media/asset_publics/resources/000/000/794/original/paper3-porter-ketels.pdf

Porter, M. E., & van der Linde, C. (1995). Toward a new conception of the environment-competitiveness relationship. The Journal of Economic Perspectives, 9(4), 97–118.

Rozmahel, P., Grochová, L. I., & Litzman, M. (2016). The Competitiveness of the EU Countries: A Multi-dimensional Cluster Analysis Approach. In P. Huber, D. Nerudova, & P. Rozmahel (Eds.), Competitiveness, Social Inclusion and Sustainability in a Diverse European Union (pp. 13–36). Springer International Publishing. http://doi.org/10.1007/978-3-319-17299-6

Schwab, K., & Porter, M. E. (2008). The Global Competitiveness Report 2008 – 2009. Geneva, Switzerland: World Economic Forum.

Sepic, D. (2005). The regional competitiveness : some notions. Retrieved from http://www.recep.ru/files/documents/regional_competitiveness_en.pdf

Silva, R., & Ferreira-Lopes, A. (2014). A Regional Development Index for Portugal. Social Indicators Research, 118(3), 1055–1085. http://doi.org/10.1007/s11205-013-0455-z

Smith, A. (2003). Power relations, industrial clusters, and regional transformations: Pan-European integration and outward processing in the Slovak clothing industry. Economic Geography, 79(1), 17–40.

Snieška, V., & Bruneckiene, J. (2009). Measurement of Lithuanian regions by regional competitiveness index. Engineering Economics, 1(61), 45–57. Retrieved from http://www.scopus.com/inward/record.url?eid=2-s2.0-70349119917&partnerID=40&md5=3f0a9e5c4da8dda58f570e3d89183fd4

Snieška, V., & Simkunaite, I. (2015). Socio-Economic Impact of Infrastructure Investments. Engineering Economics, 63(4), 16–25. Retrieved from http://www.scopus.com/inward/record.url?eid=2-s2.0-70349792199&partnerID=40&md5=1c3e75effe48eac9df1d65c0272a4b0d

Sölvell, Ö. (2015). The Competitive Advantage of Nations Book 25 Years – Opening Up New Perspectives on Competitiveness. Competitiveness Review, 25(5).

Soukiazis, E., & Antunes, M. (2011). Is foreign trade important for regional growth? Empirical evidence from Portugal. Economic Modelling, 28(3), 1363–1373. http://doi.org/10.1016/j.econmod.2011.02.023

Staníčková, M., & Skokan, K. (2012). Evaluation of the EU Member States Efficiency by Data Envelopment Analysis Method. International Journal of Mathematical Models and Methods in Applied Sciences, 6(2), 349–357.

Ström, P., & Nelson, R. (2010). Dynamic regional competitiveness in the creative economy: can peripheral communities have a place? The Service Industries Journal, 30(4), 497–511. http://doi.org/10.1080/02642060903067555

Tallman, S., & Phene, A. (2007). Leveraging knowledge across geographic boundaries. Organization Science, 18(2), 252–260. http://doi.org/10.1287/orsc.1060.0227

Titze, M., Brachert, M., & Kubis, A. (2011). The Identification of Regional Industrial Clusters Using Qualitative Input-Output Analysis (QIOA). Regional Studies, 45(1), 89–102. http://doi.org/10.1080/00343400903234688

21

Turok, I. (2004). Cities, Regions and Competitiveness. Regional Studies, 38(9), 1069–1083. http://doi.org/10.1080/0034340042000292647

y Oliva, M. de V., & Calvo, M. Á. M. (2005). Objective Competitiveness Ranking amongst EU Regions. In ERSA conference papers. (No. ersa05p625). European Regional Science Association.

Yeung, H. W. C., Liu, W. D., & Dicken, P. (2006). Transnational corporations and network effects of a local manufacturing cluster in mobile telecommunications equipment in China. World Development, 34(3), 520–540. http://doi.org/10.1016/j.worlddev.2005.07.019

Ženka, J., Novotný, J., & Csank, P. (2014). Regional Competitiveness in Central European Countries: In Search of a Useful Conceptual Framework. European Planning Studies, 22(1), 164–183. http://doi.org/10.1080/09654313.2012.731042

22