Esta tesis consta de tres capítulos independientes conectados temáticamente. El primer capítulo analiza el problema de la convergencia en Colombia, en el período 1975-2005. Para ello se propone la estimación de una ecuación de convergencia con efectos fijos individuales, se calculan los estados estacionarios de equilibrio de cada departamento atendiendo al valor de estos efectos fijos |individuales y se definen clubes de convergencia atendiendo a la similitud de los respectivos estados estacionarios. En el capítulo dos se indaga sobre los determinantes del crecimiento de la productividad factorial industrial en los departamentos de Colombia. Descomponemos el cambio en la productividad en el cambio en la eficiencia, el cambio técnico, y los efectos de escala. En el tercer capítulo, se estima la eficiencia técnica de los principales sectores industriales, a través de un modelo de frontera estocástica, y considerando la influencia de la dependencia espacial en dicha estimación. Los resultados de esta tesis sugieren que en Colombia se ha presentado un proceso de polarización en la producción, ya que las disparidades económicas entre los departamentos se están incrementando en los últimos años. / The thesis consists of three independent chapters thematically connected. Chapter one analyzes the problem of the departmental convergence in Colombia between the periods of 1975-2005. For that is proposed the estimation of convergence equation with individual effect fixed, calculating stationary state from equilibrium to each department to pay attention to valor of this individual effect fixed and to define convergence clubs to pay attention the similarity of each stationary state. Chapter 2 investigates the determinants of factor productivity growth of industry in the departments of Colombia. Here we decompose productivity change into efficiency change, technical change, and scale effects. Chapter 3 considers the technical efficiency of the main industrial sectors, through a stochastic frontier model, and considering the influence of the spatial dependence on the estimate. The results of the thesis suggests that in Colombia there has been a polarization process in production, as economic disparities between departments are increasing in recent years.

Treball d’investigació basat en l’àmbit de l’economia de l’educació que tracta d’analitzar els factors determinants en el procés d’elecció de centre universitari, a les comarques lleidatanes, per part dels alumnes que finalitzen el batxillerat , i en concret, com afecta aquesta decisió en la demanda d’estudis a la Universitat de Lleida. En els últims anys s’ha produït uns canvis en l’entorn institucional i acadèmic dels ensenyaments universitaris que justifiquen l’estudi de la demanda universitària en un context lleidatà. En les conclusions del treball, s’hi estableix com a factors determinants de la demanda els relacionats amb la proximitat geogràfica, la qualitat ocupacional, l’itinerari acadèmic previ, la reputació de la universitat, les notes de tall, i el nivell d’estudis de la mare. / Trabajo de investigación basado en el ambito de la economia de la educación que trata de analitzar los factores determinantes del proceso de elección de centro universitario, en las comarcas leridanas, por parte de los alumnos que finalizan el bachillerato, y en concreto, como afecta esta decisión en la demanda de estudios a la Universidad de Lleida. En los ultimos años se han producido unos cambios en el entorno institucional y academico de las enseñanzas universitarias que justifican el estudio de la demanda universitaria en un contexto de Lleida. En las conclusions del trabajo, se establecen como factores determinantes de la demanda los relacionados con la proximidad geográfica, la calidad ocupacional, el itinerario académico previo, la reputación de la universidad, las notes de corte, y el nivel de estudios de la madre. / Research work based on the economics of education that aims to analyze the determinants in the process of choosing a university, by students who have finished secondary education in the province of Lleida, and in particular how this decision affects the demand for studies at the University of Lleida. In recent years there have been some changes in the institutional and academic environment of university programs that justify the study of university demand in a Lleida local context. In the conclusions of the work, the demand determinant factors are those related with the geographical proximity, the quality of employment, the previous academic itinerary, the university reputation, the cut-off mark and the mother’s schooling level.

La tesis analiza los determinantes de la localización de las industrias creativas utilizando microdatos de empresas. Se utiliza un modelo adaptado para distinguir el efecto de las fuerzas económicas generales y de las fuerzas específicas creativas sobre la localización de las industrias creativas. El modelo se aplica a los sistemas locales de trabajo de cinco países europeos: Francia, Italia, Portugal, España y Reino Unido. Los resultados revelan que las economías externas tradicionales (economías de localización y de urbanización) afectan a la localización de las industrias creativas y se complementan con fuerzas específicas creativas. Además, se observan diferencias a nivel nacional y supra-nacional con relación a los principales determinantes de la localización de las industrias creativas. El estudio constituye así pues una base empírica para el diseño de políticas destinadas a estimular la capacidad de los territorios para atraer la creatividad y la innovación, según los objetivos definidos por la Comisión Europea. / This thesis examines the determinants of localisation of creative industries by using plant-level microdata. The thesis proposes a model tailored to differentiate the effect of general-economic and specific-creative forces on the localisation of creative industries. The model is applied to the local labour systems of five European countries, namely, France, Italy, Portugal, Spain and the United Kingdom. On the one hand, the results show that traditional external economies (localisation and urbanisation externalities) affect the location of creative industries, complemented by the effect of specific creative forces. On the other hand, differences are observed at the national level and at the supra-national level with regard to the main drivers fostering the localisation of creative industries. The results offer a novel insight into the determinants of location of creative industries. The work provides thus some empirical basis for the design of policies that may boost the capacity of territories for creativity and innovation, in line with the objectives set out by the European Commission.

El objetivo de este trabajo es analizar el impacto de la liberalización comercial en la distribución espacial de la actividad manufacturera en México de 1981 a 2004. Además, se pretende verificar si la evolución del patrón de localización se comporta de acuerdo con las predicciones de las Teorías del Comercio Internacional y la Nueva Geografía Económica. El propósito es dar explicación a los siguientes cuestionamientos: ¿cómo ha cambiado la especialización de los estados del país?, ¿de qué forma ha evolucionado el patrón de localización de la Industria Manufacturera? y ¿cuáles son los factores que determinan la localización industrial en México?.Los principales resultados del Análisis Descriptivo son por un lado, que los niveles de Especialización Regional para la mayoría de los estados disminuyen, a partir del TLCAN, este indicador aumenta en estados de las regiones Norte, Centro y Periferia, por tanto, se sustenta la hipótesis de Krugman (1991) que las regiones se encontrarán más especializadas cuando se realiza un acuerdo de integración (TLCAN). Por su parte, el grado de Localización Industrial en México presenta una disminución, confirmando la predicción de Krugman y Livas (1996) y la expectativa de Hanson (1997) que ante la apertura comercial las industrias disminuirán sus niveles de concentración. Sin embargo, a partir del TLCAN los niveles de localización permanecen casi sin movimiento.Los resultados del Análisis Empírico indican que las economías de escala y la intensidad en el uso de inputs intermedios nacionales son los factores que de manera más importante explican la Localización Industrial Absoluta en México, como lo predice la Nueva Teoría del Comercio y la Nueva Geografía Económica. Asimismo, se verifica la teoría de Krugman y Livas (1996) que establece que con la liberalización de los mercados las integraciones verticales se vuelven menos importantes para la explicación de la localización de la industria. / "Trade Liberalization and Industrial Localization in Mexico"TEXT:The goal of this work is to analyze the impact of trade liberalization on the space distribution of the manufacturing activity in Mexico in the last too decades of the twentieth century. In addition, we verify whether the evolution of the Localization pattern behaves according to the predictions of the International Trade Theories and the New Economic Geography. Specifically we answer the following questions: How has the specialization of the Mexican States changed? How has the pattern of Localization of the Manufacturing Industry evolved? And, which are the factors that determine the industrial Localization in Mexico.Two main results emerge from the Descriptive Analysis. On the one hand, the level of Regional Specialization for most of the states is reduced. However, since the start of the NAFTA, this indicator increases in states of the North regions, Centre and Periphery, which is consistent with the hypothesis of Krugman (1991) where it is argued that the regions will be more specialized when an integration agreement is undertaken. On the other hand, the degree of Industrial Localization in Mexico shows a reduction, confirming the prediction of Krugman and Livas (1996) and the ones of Hanson (1997) who stated that before trade liberalization the industries will diminish their levels of concentration. Nevertheless, from the beginning of the NAFTA and thereafter the Localization levels remain almost unchanged.The results in our Empirical Analysis indicate that scale economies and the intensity in the use of domestic intermediate inputs are the two main determinants of the Absolute Industrial Localization in Mexico, just as the New Theory of International Trade and the New Economic Geography predict. Finally, the theory of Krugman and Livas (1996) is verified that is, we find that with the liberalization of the markets, vertical integrations became less important as determinants of the Localization levels of the industries.

Aquesta tesi doctoral vol contribuir al camp de la creació d’empreses. En concret, explora com els factors individuals relacionats amb la figura de l’emprenedor influeixen en els resultats de les noves creacions empresarials, tenint en compte els factors de l’entorn. La tesi consta de tres estudis principals, que corresponen als capítols II, III i IV. El primer estudi analitza com la decisió d’esdevenir emprenedor no està condicionada únicament per factors relacionats amb les motivacions, sinó també per la xarxa social de què disposa l’individu. En aquest sentit, aquest capítol explica com l’existència de referents socials a la regió afecta les dimensions de les noves empreses. Concretament, es focalitza en les relaciones que poden existir entre els referents socials potencials (nombre i dimensions dels emprenedors establerts a la província) i les dimensions inicials de les noves creacions. Els resultats palesen que les noves empreses tendeixen a ser més grans en províncies on hi ha empreses també més grans. Contràriament, el nombre d’emprenedors establerts (una mesura que serveix com a símil per capturar la densitat) no té un impacte significatiu en les dimensions inicials. A més, observem que conèixer personalment un emprenedor redueix l’impacte dels referents socials potencials i el conseqüent efecte que tindria en les dimensions de les noves empreses. Resumint, l’entorn més proper de l’emprenedor té un efecte més gran que les consideracions potencials regionals. El segon estudi tracta sobre com el capital humà dels emprenedors modera l’efecte de l’entorn econòmic regional i l’efecte sobre les aspiracions de creixements. Específicament, utilitzant diferents enfocaments, es construeix un model teòric per investigar l’efecte directe de la densitat de població i les diferents dimensions de capital humà dels emprenedors naixents (educació superior, formació en emprenedoria i experiència en propietat i/o direcció) tot afectant les aspiracions de creixement. En el capítol, s’hi argumenta que les aspiracions de creixement dels emprenedors naixents són més altes en aquelles àrees més densament poblades, però la influència de l’entorn és major en aquells individus amb més nivells de capital humà. Podem concloure que les aspiracions de creixement dels emprenedors naixents que tenen educació superior i experiència en propietat i/o direcció són més altes en zones més densament poblades. Finalment, motivat per arguments institucionals i econòmics, el darrer estudi examina els factors que determinen la planificació informal i la influència en el creixement de les noves empreses. Malgrat l’interès d’altres acadèmics per la conveniència d’estudiar els determinants i les conseqüències que té la planificació en el creixement empresarial, aquest aspecte encara avui suscita interès en el camp de la creació d’empreses. D’una banda la planificació pot comportar beneficis a les noves empreses, com ara el coneixement dels projectes i dels objectius organitzats (Delmar and Shane, 2003; Chwolka and Raith, 2012). D’altra banda, pot comportar també aspectes negatius, com la falta de flexibilitat. Aquest capítol intenta aclarir l’actual debat i analitzar com factors institucionals i organitzatius determinen el perfil planificador dels emprenedors al llarg dels anys d’activitat de les seves empreses. Els resultats suggereixen que la planificació incrementa la legitimitat de l’empresa i també influeix positivament en el creixement. Les implicacions proposen que les activitats de planificació informals o bàsiques poden ser una eina útil per a la gestió empresarial i no únicament un requisit formal a complir. / This dissertation contributes to the field of entrepreneurship. Specifically, it explores how individual levels factors related to the entrepreneur influence outcomes in conjunction with factors in the environment. The dissertation has three core studies corresponding to chapters II, III and IV. The first study analyzes that the decision to become entrepreneur is not only driven by motivational factors but also by the social network structure around the individual. Accordingly, this chapter disentangles how the existence of social referents in the region affects the size of new firms. More specifically, we focus on the relationships that may exist between potential social referents (number and size of established entrepreneurs) and the size of the new firms. Also, this influence of potential social referents at the regional model is moderated by the existence of close social ones in the personal network of the entrepreneur. We observe that while new ventures tend to be bigger in provinces where there are bigger firms, the number of established entrepreneurs (a proxy for entrepreneurs density) has not significant impact. Furthermore, we observe that knowing an entrepreneur reduces the impact of potential social referents on the size of new ventures. Overall, and in line with our expectations, the results indicate that the close environment of the entrepreneur has a greater effect versus the impact of the potential regional considerations. The second study looks at how the entrepreneurs’ human capital moderates the effect of the regional economic environment on entrepreneurial growth aspiration. Specifically, we build on different theoretical perspectives to investigate the unique and joint effects of population density and nascent entrepreneurs’ human capital endowments (higher education, entrepreneurship training and owner-manager experience) on entrepreneurial growth aspirations. We argue that growth aspirations of nascent entrepreneurs are higher in more densely populated regions, but that such environmental influence is stronger for individuals with greater human capital. Consistent with our view, we find that the growth aspirations of nascent entrepreneurs with higher education and with owner-manager experience are higher in densely populated provinces. Finally, the last study focuses on institutional and economics arguments to examine the determinants of entrepreneurs’ informal planning behavior influencing the growth of new ventures. Despite the attention paid by academics to the study of the determinants and growth consequences of entrepreneurs’ planning behavior, the convenience of engaging in planning activities is still an open debate in the entrepreneurship literature. Planning may yield benefits for new ventures such as: awareness of the project, goals setting (Delmar and Shane, 2003; Chwolka and Raith, 2012) but may also bring negative aspects such as lack of flexibility. In this chapter, we attempt to contribute to clarify this debate and analyze how institutional and firm level economic factors determine entrepreneurs’ planning behavior over time. Our results indicate that planning, which may be a mean to gain not only legitimacy but also to face business difficulties, increases new firm growth. Implications suggest that informal or basic planning activities can be a really useful tool for management and not simple external requirement.

This dissertation aims to further understand the relationship between innovation and productivity in Spain taking into account the impact of knowledge flows. To this end, Spanish firms from both manufacturing and services sectors are considered. This thesis consists of three empirical studies. The first study analyses the impact that R&D expenditures and intra- and inter-industry externalities have on Spanish firms’ performance considering the firm’s technology level. A Cobb-Douglas production function is employed using the Olley and Pakes (1996) estimator in order to deal with possible selection and simultaneity problems. The results obtained suggest that, unlike previous studies, there is no direct impact of R&D investment on firm performance. However, once the innovation is measured from the output side, a positive effect is found. This finding is in line with Crépon, Duguet and Mairesse (1998) approach which hold that there is a sequential process which drives from R&D to innovation outputs and from these outputs to productivity. As for spillovers, the results point out that Spanish firms are able to benefit from external innovation. In most technology levels, firms increase their sales if the rest of the firms in their sector increase their R&D expenditure, especially low-tech sectors, while in knowledge-intensive services a competitive effect would dominate. Inter-industry externalities, however, present a more ambiguous effect and there appears to be no specific behaviour pattern as for technology level. The second empirical study aims at improving and extending the previous one by analysing the impact of innovation activities and externalities on firms’ productivity using a structural model. The main goal is to determine the extent to which external knowledge may affect both firm behaviour and firm performance. The results from a structural model with three stages show that the greater the number of firms undertaking R&D activities in the same sector the more likely to engage in R&D projects the firm is. Therefore, an external pool of knowledge would encourage firms to carry out R&D activities suggesting the existence of “an incentive effect”. In addition, R&D expenditures incurred by others firms have a positive impact on a firm’s productivity. However, this is only true for manufacturing firms. Finally, regarding the technology level, no clear pattern has been found. All in all, it seems that external knowledge flows play a role along the process of engaging innovation activities and their translation into economic performance. Internal knowledge flows, on the other hand, have faded into the background. Nevertheless, transmission of information and experience within the company can improve its technological performance impacting positively on their innovativeness and leading to higher innovative sales. For that reason, the objective of the final empirical study is to examine the importance of internal knowledge flows on firm’s innovative performance as well as to assess the role of absorptive capacity. In order to deal with selection bias, as only innovative firm are considered in the analysis, the estimator proposed by Wooldridge (1995) is used. The findings point out that internal knowledge flows have a clearly positive impact on innovation performance. In particular, firms increase more innovative sales with voluntary knowledge transfers than with involuntary. More interestingly, the effect of internal knowledge flows is greater than the impact of external knowledge transfers highlighting the importance of internal knowledge flows. Last of all, the effect of absorptive capacity depends on its definition. Crepon, B., Duguet, E., and Mairesse, J. (1998). Research, Innovation and Productivity: An Econometric Analysis at the firm level. Economics of Innovation and New Technology, 7(2), 115–158. Olley, G. S., and Pakes, A. (1996). The dynamics of productivity in the telecommunications equipment industry. Econometrica, 64(6), 1263–1297. Wooldridge, J. M. (1995). Selection correction for panel data models under conditional mean independece assumptions. Journal of Econometrics, 68, 115–132. / Aquesta tesi consisteix en tres estudis empírics sobre l’impacte dels fluxos de coneixement en la relació entre innovació i productivitat a Espanya. En concret, el primer article estudia l’impacte que la despesa en I+D i les externalitats intra- i inter- sectorials tenen sobre la productivitat de les empreses espanyoles. El segon estudi amplia i millora l’anterior utilitzant un model estructural amb tres etapes, analitzant no només l’impacte de les externalitats en la productivitat sinó també en la decisió a dur a terme activitats en I+D. Finalment, el darrer article explora com els fluxos de coneixement interns afecten al rendiment innovador de les empreses. Utilitzant la base de dades PITEC i diferents tècniques economètriques, els principals resultats són: 1) quan major sigui el nombre d’empreses duent a terme activitats en I+D en un sector més gran és la probabilitat que una empresa d’aquest sector iniciï aquest tipus de projectes suggerint l’existència d’un “efecte incentiu”, 2) les empreses manufactureres es beneficien de la I+D feta per altres empreses (del seu mateix sector i d’altres sectors) mentre que les empreses de serveis no obtenen aquest benefici, 3) els fluxos de coneixement interns incrementen les vendes innovadores de l’empresa i el que és més interessant, ho fan en major mesura que els fluxos de coneixement externs.

In the late seventies, after the recovery of democratic institutions in Spain, the planning authority is transferred to the Autonomous Government of Catalonia (Generalitat). Soon after, the incorporation of the Regional policy of 1983 (LPT 1983) allows the development of supra-municipal plans at different scales. During this period the municipal urban planning became a symbol of democratic claim. In contrast, regional planning remains distant from society. Regional plans of the LPT 1983 were postponed until the late nineties. This lack of supra-local planning in Catalonia (and Spain) defines two main problems. First, there is territory broadly managed by fragmented local plans. This is due to the division of the legal system in a law that regulates the regional planning (LPT 1983) and another law that regulates urbanism (TRLUC 2005). The second problem, regarding the lack of supra-local planning, is the opposition to the current processes of urbanization, in which the main vector is the integration of the territory. These processes of metropolization are not new phenomena. However, despite all the theoretical contributions, the plurimunicipal coordination is still very difficult to be executed through operative planning. Here lies the opportunity of the new Supra-local urban plans (PDU). The PDU are instruments intermediate between regional and local plans, implemented as a frame of reference and coordination for the municipalities included in their boundary. Like other regions and countries in the European context, the Government of Catalonia relocates the spatial planning at the regional level, defining a new MODEL of planning and administrative organization to manage its policies. Thus, the Generalitat provides PDU the role of link between the regional and the local levels to overcome the linear hierarchy of planning determined by the LPT 1983. For guiding the planners, graphic and scope guidelines are established for the PDU. However, in practice, the PDU have different APPROACHES to the methodological options and content, showing a potential flexibility to adapt its scope to the uniqueness of each of the territories addressed. The overall objective of this research is to define the innovation and the instrumental specificity of PDU. For this purpose we analyse three pioneers cases: the PDU of Bages, the PDU of Conca d'Òdena and the PDU of Vic. Drafted simultaneously, these plans address the management of three urban areas in the same region, the Central Counties. The first specific objective is, first, to define the uniqueness of the PDU by an individual study of the cases and, secondly, to generalize its scope (in the region of Central Counties) through the comparative analysis of cases. As a second objective we intend to define the role of the PDU within the framework of supra-local planning from two perspectives, one from the planning and one from the project. In particular, we analyse the three cases selected in relation to the Regional Plan for the Central Counties, which together constitute a single territorial project. According to the mentioned objectives, we verified that the PDU are not typeable plans. The instrumental innovation of the PDU lies not only in the flexibility of its contents, but also in the articulation of tested tools from other plans, both urban and regional. In this sense, the efficiency of the system results from the interaction not only of planning but also of the project in two scales, the urban areas and the region. The PDU transcend the rigid segmentation of a legal framework that splits urbanism from the territory. / A finales de los años setenta, con la recuperación de la democracia en España, las competencias urbanísticas son traspasadas al Gobierno Autonómico de Cataluña (Generalitat). Poco después, la incorporació de la Ley de Política Territorial de 1983 (LPT 1983) le permite desarrollar planes supramunicipales en diferentes escalas. En este período, el planeamiento local se convierte un símbolo de reivindicación democrática, supliendo el déficit heredado. En contraste, la ordenación territorial permanece como un propósito distante a la sociedad. Los planes territoriales de la LPT 1983 se postergan durante casi dos décadas. La falta de planeamiento supramunicipal en Cataluña (y en España) determina dos problemas principales. En primer lugar, produce un territorio profusamente ordenado mediante el fragmento municipal. Esto se debe a la división del sistema legal en un texto que regula la ordenación territorial (LPT de 1983) y otro que regula el urbanismo, la Ley de urbanismo de 2005 (TRLUC 2005). El segundo problema es que la falta de planeamiento supramunicipal se opone a unos procesos de urbanización cuyo vector principal es la integración del territorio. Estos procesos de metropolización no son novedosos. Sin embargo, pese a las numerosas aportaciones teóricas, la coordinación urbanística plurimunicipal encuentra históricamente dificultades para implementarse mediante el planeamiento. En esto radica la oportunidad de los Planes directores urbanísticos (PDU). Los PDU son instrumentos intermedios entre el planeamiento territorial y el urbanístico planteados como marco de referencia y coordinación para los municipios que los integran. Al igual que otras comunidades autónomas y países del contexto europeo, la Generalitat relocaliza la ordenación territorial en el nivel regional, definiendo un nuevo MODELO de planeamiento y de organización administrativa para la gestiónde sus políticas. Así, la Administración otorga a los PDU el rol de eslabón entre lo territorial y lo local para superar la jerarquía lineal del planeamiento que plantea la LPT 1983. Para guiar a los redactores, se establecen unas pautas gráficas y de alcance para los PDU. Sin embargo, en la práctica, los PDU presentan diferentes ENFOQUES en cuanto a las opciones metodológicas y de contenido, evidenciando una flexibilidad potencial para adecuar su alcance a la singularidad de cada uno de los territorios abordados. El objetivo general de esta investigación es definir la innovación y la especificidad instrumental de los PDU. Con este fin analizamos tres casos pioneros: el PDU del Bages, el PDU de la Conca d‘Òdena y el PDU de la Plana de Vic. Redactados simultáneamente, estos planes abordan la ordenación de tres áreas urbanas en una misma región, la veguería de las Comarcas Centrales. El primer objetivo específico radica, por un lado, en determinar la singularidad de los PDU mediante un estudio individual de los casos y, por otro, en generalizar su alcance (en las Comarcas Centrales) mediante el análisis comparado de los casos. Como segundo objetivo específico pretendemos definir el rol de los PDU dentro de la estructura de planeamiento territorial desde dos perspectivas, la del plan y la del proyecto. En particular, analizamos los tres casos seleccionados en relación al Plan territorial parcial de las Comarcas Centrales que, en conjunto, constituyen un único proyecto territorial. Según los objetivos planteados, verificamos que los PDU no son planes tipificables. La innovación instrumental de los PDU radica no solo en la flexibilidad de sus contenidos, sino también en la articulación de herramientas ya probadas de otros planes, tanto urbanísticos como territoriales. En este sentido, la eficiencia del sistema resulta de la interrelación no solo del planeamiento sino también del proyecto territorial en dos escalas, la regional y la de las áreas urbanas. Los PDU trascienden así la rígida segmentaciónde un marco legal que disocia el urbanismo del territorio.

A maior integração econômica ocorrida nas últimas décadas, principalmente decorrente da fragmentação do processo produtivo, tornou indispensável a análise da região em um contexto territorialmente mais amplo. Consequentemente, os modelos inter-regionais de insumoproduto, capazes de incorporar à análise aspectos de interdependência regional, ganharam muita relevância. No entanto, a escassez de dados primários para a construção desses modelos impôs a necessidade da estimação de algumas informações que usualmente não estão disponíveis. Desta forma, decorrente das diferentes combinações das técnicas de estimação existentes, sistemas inter-regionais de insumo-produto, para um mesmo conjunto de regiões, em um mesmo ano, podem ser diferentes em termos partitivos e holísticos. Diante do exposto, a presente tese tem como objetivo avaliar como a escolha do método, na estimação dos sistemas inter-regionais, pode influenciar os resultados da análise de insumo-produto, em contextos nacionais e subnacionais. Para isso, inicialmente, são estimados dois sistemas interestaduais de insumo-produto para as 27 UFs brasileiras, utilizando dois dos principais métodos presentes na literatura nacional, o Interregional Input-Output Adjustment System - IIOAS e o Tabela de Usos e Produção Inter-regionais - TUPI. Em seguida, os dois sistemas são comparados. Mesmo apresentando diferenças partitivas significativas, os dois métodos aprestaram bastante acurácia holística, de maneira que a utilização do IIOAS ou o TUPI na estimação do sistema interregional para as 27 UFs brasileiras não compromete, de forma geral, os resultados da análise de insumo-produto. No entanto, para estudos específicos, que envolvam um determinado setor ou uma determinada região, principalmente se esta estiver no Norte do Brasil, é preciso que o analista esteja atento às possíveis variações observadas no presente estudo. Posteriormente, foram descritos e comparados dois dos principais sistemas globais de insumo-produto, o Intercountry Input-Output Model - ICIO - da OCDE e a World Input-Output Table - WIOT - da WIOD. Semelhante ao que ocorreu com os sistemas subnacionais, a baixa acurácia partitiva entre os valores estimados pelo ICIO e a WIOT não se converteu em baixa acurácia holística, isto é, não comprometeu os resultados da análise de insumo-produto para a grande maioria dos países. No entanto, quando se trata de uma análise específica para alguns países como Malta, Chipre, Lituânia e Luxemburgo, e/ou alguns setores, principalmente os de serviços, é preciso ter em conta alguns apontamentos feitos neste estudo. Ao considerarem-se os resultados para modelos globais e subnacionais de forma conjunta, conclui-se que uma política pública orientada por qualquer um dos sistemas aqui estimados não será comprometida. No entanto, esta pesquisa abordou apenas alguns dos principais métodos disponíveis para a construção de modelos inter-regionais de insumo-produto. Um possível desdobramento é a inclusão de outros métodos no processo de comparação, no intuito de corroborar ainda mais a literatura acerca da importância da escolha do método. / The fragmentation of the productive process led to a greater economic integration in the last decades, making it indispensable to analyze the region in a wider territorial context. Consequently, the interregional input-output models have gained relevance due to their incorporating aspects of regional interdependence. However, the scarcity of survey data required for the construction of these models, compelled the use of non-survey techniques to estimate non-available information. Because of various possible combinations of non-survey techniques, interregional input-output systems for the same set of regions in the same year may be different in partitive and holistic terms. Thus, this thesis aims to evaluate how the choice among different interregional systems estimation methods can influence the results of inputoutput analysis in national and subnational approaches. Initially, two interstate input-output systems are estimated for the 27 Brazilian UFs, using two of the main methods in the Brazilian literature, the Interregional Input-Output Adjustment System (IIOAS) and the Supply and Use Interregional Tables (TUPI). When comparing these two systems, even with expressive differences in partitive terms, they presented a high holistic accuracy. Therefore, the use of IIOAS or TUPI to build an interregional system for the 27 Brazilian UFs does not compromise the results of the input-output analysis. Nonetheless, for specific studies that involve a specific sector or region, the analyst must be aware of the possible variations observed in this study, especially if they are in the North of Brazil. Subsequently, two of the most important global input-output systems, Intercountry Input-Output Model (ICIO) from OECD and the World Input-Output Table (WIOT) from WIOD were described and compared. Like IIOAS and TUPI, the low partitive accuracy between systems estimated from ICIO and WIOT did not turn into a low holistic accuracy, i.e., it did not compromise the results of the input-output analysis for most countries. However, when it comes to a specific analysis for some countries such as Malta, Cyprus, Lithuania and Luxembourg and/or some sectors, especially service sectors, it is necessary to consider some remarks pointed out in this study. Considering the results for global and subnational models, it is concluded that, in a general way, the use of any system here estimated would not compromise public policies. Nevertheless, this research addressed just some of the main available interregional input-output estimation methods. A possible future development of this study would be the inclusion of other methods in the comparison process, to further support the literature about the importance of the method.

O objetivo desta tese é analisar o processo de ocupação do Grande Cerrado do Brasil Central no que diz respeito à economia e à geopolítica. Avaliam-se alguns aspectos físico-naturais fundamentais para a interpretação deste processo, procurando mostrar a diversidade do Cerrado brasileiro e as implicações desta diversidade sobre a agricultura, especialmente a aptidão agrícola. Discorre-se sobre aspectos da história da ocupação da região Central do Brasil, procurando chamar a atenção para os pontos históricos que marcaram a formação territorial regional. Resgata-se a influência da geopolítica e traça-se um paralelo entre as diversas políticas públicas e as suas implicações sobre o território. Discute-se a construção de Brasília, resgatando os seus antecedentes históricos e o que sua construção representou para o crescimento da região. Analisam-se alguns dos determinantes da ocupação recente do Cerrado brasileiro - dos anos 1960 até os dias atuais. Para isso, resgata-se o processo de modernização da agricultura brasileira e os principais instrumentos de política utilizados para que a agricultura brasileira entrasse em uma nova fase, dentre os quais a criação do Sistema Nacional de Crédito Rural (SNCR) e os programas de incentivo à incorporação produtiva do Cerrado, como o Polocentro e o Prodecer, além daqueles voltados para a abertura de estradas e a colonização pública e privada. Traça-se um panorama da estrutura produtiva da região, a partir de 1975, com destaque para os principais produtos agropecuários (arroz, feijão, soja, milho, algodão e boi) e a agroindústria, buscando identificar as áreas para onde se dirigiu a produção. Resgata-se o mapa de aptidão agrícola e é analisada a coerência entre as áreas para onde a produção se expandiu e aquelas que têm melhor aptidão agrícola. Analisa-se, ainda, o binômio soja-boi e a conformação do complexo grãos-carne em parte da região. Busca-se caracterizar a agroindústria com ênfase na indústria frigorífica, na indústria processadora de soja e no setor sucroalcooleiro. Em seguida, analisa-se o crescimento populacional e a constituição da malha urbana da região. Discute-se a importância da cidade no processo de desenvolvimento regional e apontam-se os principais trabalhos sobre o tema desenvolvidos no Brasil. Procura-se mostrar que o crescimento da região foi muito diferenciado inter-regionalmente, o que deu origem a uma malha urbana diferenciada segundo sua função. Mostra-se ainda que o crescimento populacional nas últimas três décadas ocorreu ao longo dos principais eixos de transporte da região. Conclui-se que o meio natural influi sobre o processo de localização das atividades agropecuárias e suscita o debate acerca da necessidade de industrialização da região para que a mesma possa manter seu patamar de desenvolvimento e crescimento. Mostra-se que, diferentemente do que o censo comum supõe, o Cerrado é heterogêneo e seu desenvolvimento vem gerando disparidades regionais significativas - áreas muito desenvolvidas e áreas muito atrasadas -, cujo resultado pode ser verificado em alguma medida por meio da análise da rede urbana da região. / This thesis aims at analyzing the occupation of the Grande Cerrado do Brasil Central the great Brazilian woodland-savanna) as to its economic and geopolitical aspects. It analyzes some basic physiconatural aspects for understanding such a process, aiming at showing the Brazilian Cerrado\'s diversity and its implications on agriculture, chiefly on its agricultural ability. It also includes the historical aspects of the occupation of Central Brazil viewing to focalize historical places that have influenced the formation of this regional territory. The influence of geopolitics is recovered and a parallel is made between the several public policies and their implication on the territory. It also discusses the construction of Brasilia by tracing its historical antecedents and by attempting to understand what its role in regional growth. The study analyzes some determinants of the recently occupied Brazilian Cerrado - since the 1960s. For this, it focuses on modernization of the Brazilian agriculture as well as the major specific policy tools employed so as to enable agriculture to enter a new stage. Among such tools, the Sistema Nacional de Crédito Rural - SNCR (national system of rural credit) and programs designed to incorporate the Cerrado into the productive effort, such as Polocentro and Prodecer, were implemented, in addition to those viewing the opening of new roads and public and private colonization. A panorama of the regional productive structure since 1975 is depicted, emphasizing the major agricultural products (rice, beans, soy beans, corn, and bovine livestock) and agroindustry as well, in an attempt to identify those areas where production has been directed to. An agriculture vocational map is traced and the coherence between the areas to where production has been expanded and those better inclined to agriculture is analyzed. Furthermore, the work also analyzes the soy-cattle dual production and the configuration of the grain-meat complex in an area of the region. It also attempts to characterize agroindustry, with emphasis on the slaughter industry and soy processing industry, and on the sugar and alcohol industry as well. Then it analyzes population growth and the constitution of the urban network in the region; the relevance of the urban center in the regional development and finally; mainstream studies on the subject developed in Brazil are shown. It also attempts to show that regional growth has been much differentiated inter-regionally speaking, which has in turn given birth to a differentiated urban network contingent on its function. It also shows that population growth has emerged along the main transport axes within the region during the last three decades. A conclusion has been drawn that natural environment affects localization of agricultural activities and stimulates the debate on the need for regional industrialization so as to enable the region to keep its development and growth levels. Differently from common sense, however, the study shows that the Cerrado is heterogeneous and its development has been provoking significant regional inequalities - some well-developed areas together with lagging ones -, the outcome of which could be verified, in some measure, by means of an analysis of the regional urban network

O objetivo principal desta Tese foi desenvolver um modelo de equilíbrio geral computável (EGC) com especificação detalhada do uso da terra para o Brasil. Primeiramente, pelo lado da demanda, isso foi alcançado com a introdução de uma estrutura aninhada para o fator terra com possibilidades de substituição entre os usos do composto de lavouras, pecuária e floresta. No nível mais inferior desta estrutura, o composto de lavouras definiu as possibilidades de substituição entre diferentes usos de terras de lavouras. Em segundo, pelo lado da oferta, os usos da terra pelas atividades agrícolas foram definidos em termos físicos. Desta forma, modificações no retorno relativo da terra conduzem as variações do uso da terra de acordo com os diferentes usos possíveis. A estratégia de identificação dos usos da terra no modelo foi realizada atrelando os usos às atividades agrícolas correspondentes, mostrando-se distinta das estratégias encontradas na literatura. A capacidade analítica do modelo é ampla, podendo ser aplicado para investigar políticas que afetam direta ou indiretamente o uso da terra. Duas aplicações foram realizadas no contexto de análise de desenvolvimento do modelo. A primeira foi a mensuração dos efeitos econômicos de mudanças climáticas sobre o Brasil. Para a construção dos choques foi adotada uma estratégia de integração entre os resultados de um modelo econométrico do uso da terra, que forneceu a sensibilidade da alocação de terra entre os usos possíveis a mudanças climáticas, e o modelo de equilíbrio geral. A integração foi definida de forma que as mudanças climáticas produzissem deslocamentos na função de demanda por terra neste modelo. Para a construção dos choques foram consideradas projeções para temperatura e precipitação com base em dois cenários do IPCC, A2 e B2. Os resultados desta análise indicaram uma redução do PIB real nacional em ambos os cenários e nos intervalos de projeção. As atividades agrícolas foram as mais afetadas negativamente. A economia da maioria dos estados apresentou variação negativa do PIB real. A outra análise implementada foi a avaliação dos impactos regionais da mudança na produtividade da terra agrícola das lavouras entre 1996 e 2006. A mudança na produtividade agrícola produziu efeitos positivos sobre o PIB real nacional e da maioria dos estados. / The main objective of the dissertation was to develop a computable general equilibrium (CGE) model with detailed specification of land use for Brazil. First, in the demand side, it was achieved with the introduction of a nested structure for land with substitution possibilities among the uses of crops composite, livestock and forestry. At the lowest level of this structure, the crops composite defined the substitution possibilities among different uses of crop land. Second, in the supply side, the land uses by agricultural activities have been defined in physical terms. Thus, changes on relative return of land drive the changes in land use according to the different possible uses. The strategy of identification of land uses in the model was peculiar wherein the land uses were linked to the corresponding agricultural activities, being distinct from the strategies found in the literature. The analytical capability of this model is broad and it can be applied to investigate policies that directly or indirectly affect land use. Two applications were made using the model. The first one was to measure the economic effects of climate change on Brazil. For the construction of the shocks was adopted a strategy of integration between the results of an econometric model of land use, which provided the sensitivity of the allocation of land among the possible uses to climate change, and the general equilibrium model. The integration was set so that the climate changes produce shifts in the demand function for land in this model. For the construction of the shocks were considered projections for temperature and precipitation based on two IPCC scenarios, A2 and B2. The results of this analysis indicated a reduction in national real GDP in both scenarios and projection intervals. Agricultural products were the most adversely affected. The economies of most states showed negative change in real GDP. A further analysis was implemented to evaluate the impact of the regional change in productivity of agricultural crops between 1996 and 2006. The change in agricultural productivity had positive effects on national real GDP and the majority of states.