This paper examines the issue of whether the UK displays high levels of interregional inequality or only average levels of inequality. The question arises due to major differences in public perceptions. Following on from recent UK public debates, the UK evidence is examined in the context of 28 different indicators and 30 different Organisation for Economic Co-operation and Development (OECD) countries. Answering this question involves a careful consideration of the ways in which we use different spatial units of analysis, different measures of prosperity and different indices of inequality in order to understand interregional inequality, and the issues that arise are common to all countries. In the specific case of the UK, the result is clear. The UK is one of the most regionally unbalanced countries in the industrialized world.
The RIS3 agenda has become a central plank of EU Cohesion policy. It is aimed at spearheading the drive towards research, innovation, and entrepreneurship across all EU regions, especially its weak...
This paper uncovers some little-known properties of the iceberg transport cost functions which are employed in some new economic geography models. In particular the behaviour of the delivered prices generated by the Krugman iceberg specification is very different to those generated by the original Samuelson iceberg model, and these differences require careful interpretation of the outcomes of the explicitly spatial versions of new economic geography models. If the iceberg model is viewed simply as transport costs, then these properties can be shown to be largely implausible when compared with a wide range of empirical evidence. At the same time, even if the iceberg model is viewed as capturing a range of different distance costs, there are still grounds for cautiousness in the way that inferences are made from these models when moving from theory to reality or policy analysis.
Does an individual's housing situation affect community social capital and influence regional development through this channel? A body of literature which followed the seminal work of DiPasquale and Glaeser suggests that homeownership is positively related to social capital formation due to both an increased incentive to engage in the local community and because home ownership creates barriers to mobility. We test the home ownership hypothesis alongside other individual, household and locational determinants of social capital using unique data created by merging the three waves of the New Zealand Quality of Life survey, a local government sponsored survey conducted in 2004, 2006 and 2008. The measures of social capital used in our analysis include trust in others, participation in social networks, attitude towards local governance and sense of community. Because homeownership is not randomly assigned, matching methods are used to control for selection effects.
In this paper we estimate the impact of local authority infrastructure spending in New Zealand using spatial econometric modelling, with the infrastructure spending itself endogenously determined. Utilizing data from the New Zealand Census and Local Authorities Finance data (1991-2008), aggregated to functional labour market areas, we formulate a simultaneous equations growth model of real income, population, land rent and public infrastructure investment. Estimation is conducted using a spatial 3SLS procedure. We find that an increase in local infrastructure spending increases population growth, real income and land values, but is itself endogenous and spatially correlated.
Part I - Analytical approaches to industrial location: classical and neo-classical location-production models, Philip McCann the location of economic activity - central place theory and the wider urban system, John B. Parr geographical grids in new economic geography models, Dirk Stelder firm migration, Piet Pellenbarg, Leo van Wissen, Jouke van Dijk. Part II - Cities and industrial clusters: diversity and specialization in cities - why, where and when does it matter?, Gilles Duranton, Diego Puga global cities, internationalization and urban systems, Ian R. Gordon clusters and networks - institutional settings and strategic perspectives, Michael Steiner theory, methods and a cross-metropolitan comparison of business clustering, Edward Feser, Stuart Sweeney. Part III Multinational firms and location behaviour: the location decision of the multinational enterprise, Ram Mudambi the technological relationships between indigenous firms and foreign-owned MNCs in the European regions, John Cantwell, Simona Iammarino the relationship between the spatial and hierarchical organization of multiplant firms - observations from the global semiconductor industry, Tomokazu Arita, Philip McCann.
This paper aims to sketch rural population decline in Mexico. A comprehensive multi-scale documentation of Mexico's urban and rural fabric and its depopulation dynamics is undertaken. This paper does not identify the causes and effects of rural depopulation, but aims to detect its geographical presence in a manner which other approaches are unable to do. The granularity of rural depopulation is identified as manifested at four different scales: national, state, municipal, and local village scale. We use census data for the period 2000–2010. We relate rural depopulation to spatio-geographical features found in the literature; the relationship of depopulating territories with urban ones and the altitude range in which they are located. We find that proximity to and remoteness from urban areas as well as that of altitude do not follow the typical patterns often observed elsewhere. Unlike that shown in other countries, in Mexico, rural depopulation is manifested mostly in lower-lying areas and despite the fact that most of the country is at altitude. Rural depopulation and disappearance occur mainly in the areas of plains below 1,000 metres above sea level, regardless of the population density of the territories where they were located. Also, our multi-scaling approach has allowed us to identify not just at what scale the phenomenon of rural depopulation is observed, but also, to identify the areas in which the phenomenon is manifested regionally or locally. As we see, much of the phenomenon is concentrated in different areas to what national narratives suggest. El objetivo de este artículo es esbozar el declive de la población rural en México. En él se realizó una exhaustiva documentación de escala múltiple del tejido urbano y rural de México y de su dinámica de despoblación. El artículo no identifica las causas y los efectos de la despoblación rural, sino que intenta detectar su presencia geográfica de una manera que otros enfoques no pueden hacer. Se identificó la granularidad de la despoblación rural, que se manifiesta en cuatro escalas diferentes: nacional, estatal, de municipalidad y por pueblo. Se utilizaron los datos del censo para el período 2000-2010. Se relacionó la despoblación rural con las características espacio-geográficas encontradas en la literatura, a saber, la relación entre los territorios en proceso de despoblación y los urbanos, y el rango de altitud en el que se encuentran. Se comprobó que la proximidad y la lejanía de las zonas urbanas, así como la altitud, no siguen las pautas típicas que suelen observarse en otros lugares. A diferencia de lo que ocurre en otros países, el despoblamiento rural en México se manifiesta sobre todo en las zonas bajas, a pesar de que la mayor parte del país se encuentra en zonas con una altitud elevada. El despoblamiento y la desaparición de las zonas rurales se producen principalmente en las planicies por debajo de los 1.000 metros sobre el nivel del mar, independientemente de la densidad de población de los territorios donde se encontraban. Asimismo, nuestro enfoque multiescalar nos ha permitido identificar no sólo a qué escala se observa el fenómeno de la despoblación rural, sino también identificar las zonas en las que el fenómeno se manifiesta a nivel regional o local. Como vemos, gran parte del fenómeno se concentra en áreas diferentes a lo que sugieren las narrativas nacionales. 本稿では、メキシコの農村の人口減少を概観する。メキシコの都市と農村の構造とその過疎化のダイナミクスの多重スケールで包括的な検討を行った。ここでは農村の人口減少の原因と影響を特定するのではなく、他のアプローチでは不可能な方法で人口の減少を地理的に検出することを目的とする。農村の人口減少の粒度は、4つの異なるスケール、すなわち国、州、市町村、および地方村落の規模で特定する。2000~2010年の国勢調査のデータを用いて、農村の人口減少と文献に見られる空間的・地理的特徴、すなわち人口が減少している地域と都市の地域および人口減少地域の標高の範囲との関連性とを結びつける。都市への近接性や都市からの遠隔性は、標高の高い地域への近接性又は遠隔性と同様に、他の地域で観察される典型的なパターンではないことが分かる。他国の例と異なり、メキシコでは、国土の大部分が高地であるにもかかわらず、低地における人口減少が顕著である。農村の人口減少と消滅は、その農村がかつてあった地域の人口密度に関係なく、主に海抜1,000メートル以下の平原地域で起きている。また、今回用いた多重スケールアプローチは、農村における人口減少の現象がどの程度の規模で観察されるかを特定するだけでなく、現象が地域的又は局所的に発生する地域を特定することも可能であった。このように、人口減少の多くは、国家の言説 (national narratives)による示唆とは異なる地域に集中している。 Depopulation is defined as "a demographic and territorial phenomenon consisting of a decrease in the number of inhabitants in a territory or nucleus relative to a previous period. The fall in absolute terms of the number of inhabitants can result from a negative natural growth (when deaths exceed births), negative net migration (emigration exceeds immigration) or both simultaneously" (Pinilla & Sáez, 2017, p. 2). As of 2008 the majority of the world's population (50.6%) lived in cities, having increased from 30% in 1950, and this share is expected to increase to some 66% by 2050 (United Nations, 2018). These figures indicate the relative increase of urban population of the world with respect to total global population growth. If we focus just on the absolute growth, we also observe that the global urban population has increased from 751 million in 1950 to 4.196 billion in 2018 (United Nations, 2018). The worldwide population living in cities has increased both in relative and in absolute terms along with global population growth. On the other hand, if we consider rural areas, we see that the global rural population has decreased in relative terms from 66.4% of the global population in 1960 to 44.7% in 2018. Nevertheless, in absolute terms, the rural population of the world has also increased from less than 2 billion in 1950 to 3.4 billion in 2018, although it is also projected to decline from just after 2020 onwards (United Nations, 2018). The worldwide population living in rural areas has increased in absolute terms but not in relative terms, while global population growth has continued. That is that although the phenomenon of rural population is not existent worldwide, depopulation is now observable at smaller scales, going from regions of the world such as Asia, whose rural population began to decrease approximately since year 2000, Europe, which has done it so much earlier, or America and the Caribbean region, which has presented an absolute loss of its rural population since the mid-1990s (Figure 1). As the global rural population also begins to decline after 2020, this phenomenon will become even more pervasive across countries (Figure 2) and villages. The villages are the smallest geographic entity of the rurality, and many of them have entirely disappeared or they are in that process. The literature examining rural depopulation considers it to be a complex and multifactor phenomenon, and the literature itself falls broadly into two broad strands, namely, the spatio-geographic and socio-demographic approaches. The first literature strand tends to emphasize more the geographic location and characteristics of depopulating territories as a key influence on population changes, including their remoteness from, or proximity to, urban territories, and also their physiographic and topographical features and the ecological and landscape changes involved in the process (Jacob et al., 2008; Liu, et al., 2010; MacDonald et al., 2000; Montserrat & Villar, 2008; Rizzo, 2016; Yonezawa & Aoyagi-Usuia, 2010). In contrast, the second strand of literature tends to emphasize the dynamics of population growth including issues such as mortality and fertility rates, household composition and formation processes, migration patterns, gender ratios, age and occupational structures, and other social, economic, and cultural features of the remaining populations (Anderlik & Cofer, 2014; Feldhoff, 2013; Noack & Bergmann, 2011; Stratta Fernández & de los Ríos Carmenado, 2010; Traphagan, 2008a, 2008b), as well as discussions of the institutional frames and public policies formulated for depopulating areas (Bielza de Ori, 2003; Cartwright, 2013; ESPON, 2017, 2018; Karcagi-Kováts & Katonáné-Kovács, 2012; López-Ruíz, 2005; Long et al., 2011; Pinilla & Sáez, 2017; Skryzhevska & Karácsonyi, 2012; Thompson, 2008). However, in the particular case of Mexico, there are very few studies about rural depopulation (Bocco Verdinelli & Segundo Métay, 2012; Canales, 2009; García-Barrios et al., 2009; Mojarro & Benítez, 2010; Segundo Métay, et al., 2012; ). Although there is much research on migration, there is relatively very little coherent or a comprehensive research on the remaining rural populations or the characteristics and transformation of the emptying territories, left behind by the out-migrants. While, it is very possible the emergence of literature during the next years when the phenomenon will be at national level (Figure 2), this work coincides with the idea that an approach of any population shrinkage should be done from a multi-scale perspective, even if it is manifested only in one part of a territory, acknowledging that its implications could vary between cities, suburbs, rural villages, or even neighbourhoods (Bontje, 2018; Hollander, 2018). In this work, we approach the phenomenon in a descriptive and explorative manner from a spatio-geographic view. Our thinking regarding this approach is that in topics which have not been explored enough, the descriptive approaches are not just necessary but fundamental in order to think of an agenda for its deeper investigation. As we will demonstrate, because aggregate populations in both Mexican urban and rural areas have increased, as they have across the globe, identifying the spatio-geographical dimensions of Mexican rural depopulation is therefore only possible at smaller spatial scales. Indeed, as we will show in this paper, the extent to which these phenomena can be observed depends crucially on the spatial scales of analysis, and the chosen scale of analysis heavily influences both the observation and the interpretation of the phenomenon. What our multi-scale analysis demonstrates is that although the Mexican population is growing, rural population decline is a phenomenon taking place across the whole country. However, purely national level analyses hide this fact. One level lower, at the state-level, rural depopulation appears to be a regional phenomenon in the north and in the west; whereas, in the centre and south-southeast it is clearly a more local phenomenon. Thus, a multi-scale perspective demonstrates that the phenomenon is widespread in all parts of the country as the scale of observation gets smaller. As for the local scale analysis, we will, also, observe some spatio-geographic features of the depopulating territories identified through this work, such as their relationship with urban territories, which implies a look into different population densities, and the altitude in which they are located. The time period under study is 2000–2010. In order to do this, data from different national data-sources have been merged, coded and categorized using geographical information systems (GIS) techniques. This has been done in order to document and locate every single village in Mexico according to its population, and whether it has depopulated, survived or disappeared during the decade 2000–2010. Each of the nested hierarchical administrative units are also documented, along with all of the urban centres, conurbations and metropolitan zones. This construction provides the requisite multi-level framework for us to uncover the patterns of rural depopulation. The rest of the paper is divided into four sections. Section 2 presents a brief review of the literature concerning rural depopulation and Section 3 outlines the data and methods used in this research. Section 4 reports the results of our analysis in the form of maps, tables and charts, and Section 5 offers some concluding comments. Depopulation territories may exhibit different geographical characteristics. These characteristics can be very complex and often they may be difficult or even impossible to generalize. In such situations, a multi-scale approach which exploits different hierarchical and geographical scales of observation can be a way of making progress in better understanding its key features and ongoing spatial transformations. Rural depopulation has been studied at different levels in some of the European Union member states (Aldrey-Vázquez, 2009; Barrachina, 2007; Bielza de Ori, 2003; Cartwright, 2013; ESPON, 2017 & 2018; Karcagi-Kováts & Katonáné-Kovács, 2012; López-i-Gelats, 2013; López-Ruíz, 2005; MacDonald et al., 2000; Mann, 2005; Montserrat & Villar, 2008; Noack & Bergmann, 2011; Pinilla & Sáez, 2017; Pires de Almeida, 2017; Poinsot, 2007; Rizzo, 2016; Skryzhevska & Karácsonyi, 2012; Viñas, 2019; Westhoek et al., 2006); in Japan (Feldhoff, 2013; Nakagawa & Izuta, 2017; Perez-Barbosa & Zhang, 2017; Thompson, 2003, 2008; Traphagan, 2008a, 2008b; Yonezawa & Aoyagi-Usuia, 2010; Yoshida, 1999); in China (Li, 2015; Liu et al., 2010; Long et al., 2011; Wang et al., 2019), and in some Latin American countries such as Argentina (Lausirica & Sardi, 2008; Steimbreger & Kreiter, 2008; Stratta Fernández & de los Ríos Carmenado, 2010), or Mexico (Bocco Verdinelli & Segundo Métay, 2012; Canales, 2009; García-Barrios et al., 2009; Mojarro & Benítez, 2010). Although an explicit multi-scale approach is absent in these studies, from an overview of these various pieces of research it is possible to identify some key country-related themes in demographic analysis. It is apparent from the literature that countries such as Japan and some European countries, such as Spain, Italy, Greece, Switzerland, France, Baltic countries, and some European Eastern countries, experience rural depopulation primarily in mountainous areas and those which are remote from urban areas. By contrast, China experiences what is called "rural hollowing" manifested as rural depopulation mostly at the core of rural villages located at the urban fringes and in the plains agricultural areas adjacent to cities whose radius is gradually expanded towards surrounding urban centres. The case of the United States is also documented in the Great Plains area (Anderlik & Cofer, 2014; Johnson & Litcher, 2019). In the case of Mexico, however, there is as yet no clear knowledge regarding specific locations and geographic features of the rural depopulating territories. Given that the Mexican population as a whole is still growing, the structural and spatial characteristics of Mexican rural population decline are currently not very evident. Given that the country is predominantly mountainous and experiencing rapid urbanization processes, we would expect that Mexico might share some similarities with other mountainous country cases, and with some other countries experiencing similar rhythms and patterns of urbanization processes. However, our intention here is to demonstrate that the picture on the ground is in reality far more complex and nuanced than either simple country-based comparisons suggest, or simple urban versus rural narratives imply. Rural depopulation in Mexico is taking place in different ways all over the country and a multi-scale approach helps to uncover these various nuances. Our understanding of the diversity, specificity and granularity of rural depopulation is limited by our ability to disaggregate national trends to regional and local trends. Such possibilities are greater in advanced economies with rich, detailed and continually updated datasets, but in many countries such data are not available. In the particular case of Mexico, there are detailed population data which are contained within different national datasets, but in order to understand the Mexican relationships between local, regional and national rural depopulation, and the relationship with the urban territories, in this work it was necessary to build an integrated dataset by combining these different data sources and then disaggregate them at the different spatial scales. The different scales under observation which we employ are the Mexican political-administrative unities (national, state, municipal, and local-village), ordered under a nested hierarchy. As of 2010, there were 196,351 rural villages, contained within 2,456 municipalities, which are grouped together into 32 Mexican states. For each of these spatial-administrative scales we are able to observe changes in population 2000–2010, and we employed GIS techniques to map these changes. The spatial (vector) data for the polygons for states, municipalities, urban territories, and dots for villages, were obtained from the National Geostatistical Framework of the National Institute of Statistics and Geography (INEGI, 2010a). The data sources used at different spatial scales are the National Census for years 2000 and 2010 (INEGI, 2000a, 2010b), the System of Territorial Integration (ITER) for both years from the National Institute of Statistics and Geography (INEGI, 2000b, 2010c), and the National Commission for the Knowledge and Use of the Biodiversity (CONABIO, 2000, 2010). In terms of definitions and categories, if we take the very smallest spatial and administrative unit, according to the last Mexican Census 2010, a village is defined as every place which is occupied by at least one house, even if it is not inhabited. A village is recognized by a name given legally or by tradition of its inhabitants, and a village can be rural or urban. A rural village is a village with less than 2,500 inhabitants (INEGI, 2010d). 11 The definition of what is rural is an important discussion placed in the social scientific table. As this study is devoted to sketch the rural depopulation phenomenon in Mexico, we have taken the official definition of what is rural within this country. We understand that there may be limitations with this approach, given the complexity involved in this concept, which being defined just by population inhabitants and not by density or other factors such as facilities provided, etc., is a very constrained definition. However, for our data purposes this is the most workable definition. In this work we follow the definition of rural depopulation as the absolute loss of population in this kind of villages grouped by the different scales, and this allows us to identify the depopulating and disappearing rural villages during the decade 2000–2010. A rural depopulating village is one with a negative absolute change in its population during that decade, while a rural disappeared village is identified as that one with zero inhabitants in 2010 or the villages that does not appear registered anymore in the database of 2010 neither as a rural nor as an urban settlement. 22 A village could have been registered as a rural village in 2000, but it does not appear as a rural village in 2010. There are two possibilities: (i) The village is not registered as a rural village anymore because it has become an urban village; or (ii) The village has disappeared. In the first case, the code of the village still appears in the database for 2010, but not as a rural village. In the second case, the village is not registered in the database for 2010 anymore. In this study, the rural disappeared villages are the latter cases and those ones which are accounted with zero inhabitants in 2010. The national urban system of Mexico (SUN), is comprised by 384 urban entities, and they are divided by different categories, defined as metropolitan zones, conurbations and urban centres, respectively, in descending order of size (CONAPO, SEDESOL, 2012), (Figure 3). In order to classify the villages and their relationship with the SUN, we have constructed areas around the outside of metropolitan zones, conurbations, and urban centres. These areas are concentric rings of a distance of 10 km, 5 km, and 2.5 km, respectively beyond the edge of the continuously built-up urban boundary (Figure 4). Outside of the metropolitan zone we construct a ring of 10 km, outside the conurbations we construct a ring of 5 km, and outside of urban centres we construct a ring of 2.5 km. Outside of the SUN but within these respective surrounding rings are the areas that we designate as being peri-urban. Even within the SUN there are sometimes standalone/rural villages, because not all city developments are entirely contiguous. Cities have very differentiated and diffuse spatial structures which are often characterized by sprawl, ribbon developments, and fragmented morphologies, and this allows for the presence of standalone/rural villages to exist even within the SUN. In the following subsections we report the findings from our multi-scale GIS-facilitated analysis of the features and nuances of Mexican rural depopulation patterns by breaking the analysis into three elements, namely: a multi-scale analysis, an analysis of the role of proximity versus remoteness, and finally an analysis of the role of altitude. When observing rural depopulation at the national level there is no real evidence of it. This is because the total Mexican rural population has increased from 24,128,494 in 2000 to 24,550,862 in 2010, even though its share of the total population has fallen from 25% to 22%. Yet, changes in the spatial distribution of the rural population have not been homogeneous, and therefore the phenomenon of rural depopulation starts to become visible at smaller spatial scales. As we see in Figure 5, at the level of the 32 states, between 2000 and 2010 rural depopulation appears to be primarily contained within most of the Northern states of the country. These states are generally of low rural population density and in terms of area they are the biggest states of the country, summing all together, an area of 952,541.80 Km2, which represents almost half of the total national area (1,960,670.2 Km2) (INEGI, 2015). The only parts of the north not facing population decline were the states in the Baja California pensinsula, and the state Coahuila, both some of the least densely populated states in the country. If we adjust our observations to a smaller spatial scale, namely the municipal scale, we observe that 43% of all the municipalities of the country showed rural depopulation processes during the decade under study (1,049 out of 2,456 municipalities), as shown in Figure 6. As we also see in Figure 7, in terms of 2010 population densities, 49% of these rural depopulating municipalities lie in the lowest third of municipalities, defined according to population density, with less than 27 inhabitants per Km2. The middle third of municipalities, defined according to their densities, account for 31% of depopulating municipalities, while the top third most densely populated municipalities account for 20% of depopulating municipalities. In other words, as we see in Figure 7b, 51% of depopulating municipalities are actually in the middle and most densely populated areas, most of which, are very small municipalities and lie in the centre or the south of the country. The contrary case, namely that of municipalities without rural depopulation, shows very similar figures but the other way around, from the centre and south to the north of the country. As such, at a municipal scale, rural depopulation is a phenonemon which is equally shared between low density areas and medium or high density areas. This cannot be understood by simply observing national or state-level data. Our understanding and interpretation of the phenomenon of Mexican rural depopulation changes considerably again when we observe it at the very local scale, namely the rural village, where the rural village is the smallest administrative and geographical entity of observation. Around 61% of all rural villages of the country were undergoing a depopulation process during the decade 2000–2010, with 38% of these actually disappearing during this period (Figure 8). The rural villages are the majority type of settlements in Mexico, and they are characterizied by being very dispersed throughout the national territory. Figure 9 shows that rural growing villages are distributed quite evenly throughout the least, middle and most densely populated areas of the country (33%, 35%, and 32%, respectively). Meanwhile, the depopulating and disappeared ones are mainly concentrated at the least densely populated areas (42% and 48% respectively), a situation which is very similar to the municipalities observation. Importantly for our purposes, almost the half of depopulating and disappeared villages are located at the the middle and most densely populated centre and south of the country (Figure 9 and Figure 8). These observations sketch out how, when interpreted as the regional scale, rural depopulation is a phenomenon understood as being mainly present in the least densely populated areas, mostly in the North. However, when interpreted at the smaller municipality and village scales, rural depopulation becomes increasingly understood as being evenly shared by the middle and most densely populated areas of the centre and south of the country. We take a look into the patterns of rural depopulating territories and their spatial relationships with the national urban system (SUN). The location of rural villages with respect to the SUN are shown in Figure 10. What we see is that the patterns of growing, versus declining, versus disappearing villages, are different between villages located within the SUN, versus those located in peri-urban areas, versus those which are remotely located. Figure 10 shows that out of all rural villages located within SUN and peri-urban areas, the majority are growing villages (41%, and 44% respectively), whereas rural villages located at the remote areas are mainly depopulating villages. Regarding disappeared villages they contribute mainly for those rural villages located within SUN. We can also consider how these patterns of village population growth, decline, and disappearance are related to the population density of the surroundings areas. In Figure 11, we can appreciate the great importance of depopulation processes on rural villages located mostly at middle densely populated zones, overall for the remote areas (42% of all remote rural villages in the middle densely populated areas were depopulating), but also within SUN (37% of all rural villages located within SUN in the middle densely populated areas were in a depopulation process). Also, in comparison only with disappeared villages, we can see the important presence of rural depopulating villages at peri-urban zones in general (least, middle and most densely populated areas). In the same line, we can see the great importance of rural disappearing manifestation at the least densely populated areas, but within the SUN. For growing villages, they outstand naturally in the most densely populated areas for all categories. There are three general themes which emerge from Figure 11. First, reading across each group of villages, we see that most disappearance occurred in the least densely populated areas, but within the SUN. Second, rural depopulation is manifested importantly in the least densely populated areas, but mainly away from the SUN, that is in the peri-urban and principally in remote areas. In other words, most demographic declining changes ocurred in low density areas, but depopulation and disappearance relate differently with urban entities. And third, rural depopulation processes expand importantly towards the middle densely populated zones. Figure 12 synthesises schematically our observations to help us to visualize the distributions. Each dot represents each percentage point observed in Figure 11. One other possible issue which might heavily shape the geographical features of Mexican rural depopulation patterns is the question of topography. Just around a third part of the country is flat, and more than the half of the national territory is 1,000 metres above sea level (García-Martínez, 2008). As such, many parts of Mexico are very mountainous and it may be the case, as it is in many other countries that altitude is a key feature of population decline. In this subsection, we categorize the elevation levels of Mexico, and observe seven categories of altitude. We start with the figure that a third of Mexican territory is under 1,000 metres above sea level (masl), which comprises our first category. Given that the capital of the country, Mexico City, is located at 2,240 masl, and most of the SUN is located at the central fringe of the country with the highest altitudes of the country, we have divided the altitude levels above 2,500 masl into shorter interval categories. These altitude intervals are 1,000–1,500 masl, 1,500–2000 masl, 2000–2,500 masl, 2,500–2,800 masl, 2,800–3,000 masl, and above 3,000 masl. These altitude interval categories are used to depict the geography of high altitude rural village depopulation (Figure 13a), disappearance (Figure 13b) and growth (Figure 13c). Figure 13 shows that 50% of the villages that are not facing either depopulation of disappearance are located below 1,000 masl, 12.0% are located between an altitude of 1,000 masl and 1,500 masl, and 19.3% are located between 1,500 masl and 2,000 masl. Figure 13 also shows us that 53.9% of depopulating rural villages are below an altitude of 1,000 masl, 12.4% are located between an altitude of 1,000 masl and 1,500 masl, and 18.1% are located between 1,550 and 2,000 masl. Meanwhile, 62.6% of disappearing rural villages are below an altitude of 1,000 masl, 12.8% are located between an altitude of 1,000 masl and 1,500 masl, and 13.8% are located between 1,550 and 2,000 masl. That is, that the spatial distribution of all rural villages of different categories is similar throughout the whole altitudinal spectrum; but it is clear that rural village population decline is more present on the low lying and plain areas below 1,500 masl, categories in which they overcome the growing rural villages. This characteristic is much more evident for the disappeared villages. That is, that rural village population decline or rural village disappearance in Mexico tends to be relatively more prevalent in lower-lying and flat areas than in mountainous and high altitude areas. This is the opposite of countries such as Japan, and some European members. The last statement can be complemented and ratified by observing the relationship between the (negative) growth rate of rural depopulating villages and altitude (Figure 14). The villages showing figures of population loss roughly between 0% and 40% are mainly concentrated below 200 masl, and also within the range of approximately 1,500 masl <=2,500 masl. Meanwhile, when the figures of population loss are larger than 40%, the phenomenon of depopulation tends to manifest itself mainly at low altitudes, below 200 masl. In other words, much of the most severe depopulation occurs in lower-lying areas. In Figure 15 we repeat the same exercise as in Figure 14 but this time the scatterplot relates to population decline to local population density. What we observe is that much of the population decline is in areas with less than 1,000 inhabitants per square km. For ease of exposition, in Table 1 we synthesize the observations of the spatial relationship between rural depopulation, the SUN and the observations regarding the altitude range in which villages are located. Table 1 is constructed with the information of Figure 10 and Figure 11. The first column states the most relevant information regarding rural depopulating and disappeared villages we found, and their relationship with the SUN. The second column, states the type of density area in which the observations of the first column took place. The third column confirms the altitudinal ranges in which the set of villages referred to in the second column are distributed. For instance, the statement in the first column (a) rural depopulation is an important phenomenon in remote areas (from Figure 10), is broken down in the second column to show that some 40% were located in the least densely populated areas, and 42% were located in middle densely populated areas. The third column of Table 1 shows the altitudinal spectrum in which these (40% and 42%) sets of depopulating villages are distributed. The results reported in this paper are the outcome of the most comprehensive spatial data construction effort yet undertaken regarding the geography of rural depopulation in Mexico. By integrating data from diverse sources and at different spatial scales within a GIS framework we are for the first time able to identify the granularity of rural depopulation proceses at different spatial scales across Mexico. The data construction allows us to observe rural depopulation processes from four scales, namely the national, state, municipal and local village scales. Our multi-scaling approach has allowed us to identify not just at what scale the phenomenon of rural depopulation is observed, but also, to identify the areas in which the phenomenon is manifested regionally or locally. Taking into account just the multi-scaling approach, our research has uncovered two new key observations. First, the observed patterns, and therefore also our understanding of Mexican rural depopulation, are heavily influenced by the choice of the spatial scale we adopt. From a multi-scaling view, rural depopulation is not observed at the national Mexican scale, whereas at the state scale, rural depopulation appears to be primarily a regional phenomenon of the northern and western states. In particular, at the state level rural depopulation appears to be dominated by the least densely populated states, all of which are located in the north of the country. Yet, when we move down through the municipal scale and the local village scales, rural depopulation emerges as a major phenomenon in many parts of the country, including in both the densely populated central and southern parts of the country, the areas which are more rapidly urbanizing. In terms of the specific features of the depopulating territories, namely, their relationship with urban entities and also the altitude in which they are located, our research has uncovered new key observations. In particular, major demographic declining changes in rural villages tended to be observed in the extreme scenarios, namely in conditions of desolate areas of low population density but within the national urban system for the extreme condition of depopulation which the disappearance of villages, or also in the proximity to urban areas for the case of depopulating villages not yet disappeared. It is a much more mixed picture than the literature suggests when points out the low density and remote areas as the areas in which depopulation in prone to occur. Moreover, rural depopulation and disappearance takes place primarily in the plains areas below 1,000 masl. This observation is fundamentally different to the experience of many other countries. There is no systematic altitude aspect of Mexican village decline, and indeed, if anything, the evidence points to the contrary. The results from our multi-scale and multi-altitude analysis can help us reconsider our interpretation of the phenomenon of rural depopulation, along with the effective public policies it may require. Differently from other countries, in Mexico there is currently a vacuum of policy measures oriented explicitly to rural population decline. This work detects the geographical presence of the phenomenon, but not the causes and effects of it, so additional research is therefore needed to be able to diagnose the phenomenon as a negative or positive one at the different scales. These kinds of studies could be really important for advising on public policy-design. The geographical detection of these different patterns undertaken here allows us to begin to identify the target areas in order to study more deeply the spatial impacts and the sectors of population involved.
'%3e%3cpath fill='%23012169' d='M0 0v30h60V0z'/%3e%3cpath stroke='%23fff' stroke-width='6' d='m0 0 60 30m0-30L0 30'/%3e%3cpath stroke='%23C8102E' stroke-width='4' d='m0 0 60 30m0-30L0 30' clip-path='url(%23b)'/%3e%3cpath stroke='%23fff' stroke-width='10' d='M30 0v30M0 15h60'/%3e%3cpath stroke='%23C8102E' stroke-width='6' d='M30 0v30M0 15h60'/%3e%3c/g%3e%3c/svg%3e)
'/%3e%3cuse xlink:href='%23a' transform='translate(288.889 -214.211)'/%3e%3cuse xlink:href='%23a' transform='translate(108 342.749)'/%3e%3cuse xlink:href='%23a' transform='translate(469.189 342.749)'/%3e%3c/svg%3e)
