Content uploaded by Renan Pereira Almeida
Author content
All content in this area was uploaded by Renan Pereira Almeida on Jul 09, 2020
Content may be subject to copyright.
DEPARTAMENTO DE CIÊNCIAS ECONÔMICAS
TEXTO PARA DISCUSSÃO No 4
REAL ESTATE MARKETS IN BRAZIL’S SECOND-, THIRD-
AND FOURTH-TIER CITIES
Renan P. Almeida
Junho de 2020
2
Ficha catalográfica elaborada pelo Setor de Processamento Técnico da Divisão de
Biblioteca da UFSJ
Almeida, Renan P.
Real estate markets in Brazil’s second-, third- and fourth-tier cities / Renan P.
Almeida. – São João del Rei: UFSJ / DCECO, 2020.
19 p. (Textos para discussão; n. 4)
Inclui bibliografia.
1. Setor imobiliário. 2. Rede de cidades. 3. Habitação. 4. Aglomeração. 5.
Urbanização – Brasil. I. Título.
CDU:
330
3
UNIVERSIDADE FEDERAL DE SÃO SOÃO DEL REI
DEPARTAMENTO DE CIÊNCIAS ECONÔMICAS – DCECO
REAL ESTATE MARKETS IN BRAZIL’S SECOND-, THIRD-
AND FOURTH-TIER CITIES
Renan P. Almeida
Professor do DCECO/UFSJ e do
Programa de Desenvolvimento, Planejamento e Território (PGDPLAT)
renan@ufsj.edu.br
DCECO/UFSJ
SÃO JOÃO DEL REI
2020
4
SUMÁRIO
1. INTRODUCTION ..................................................................................................... 07
2. BRAZILIAN CITY NETWORK ............................................................................... 08
3. REAL ESTATE MARKETS IN BRAZIL’S SECOND-, THIRD- AND FOURTH-
TIER CITIES .............................................................................................................. 11
4. CONCLUSONS ........................................................................................................ 17
REFERENCES ............................................................................................................. 17
5
RESUMO
O mercado imobiliário em cidades de rápido crescimento é um aspecto fundamental dos
assentamentos humanos e fundamental para fomentar um sistema urbano nacional
policêntrico. Este trabalho investiga os mercados imobiliários nas cidades de segunda,
terceira e quarta camadas do Brasil. Utilizando um conjunto de dados único, o trabalho
explica a variabilidade e cria tipologias de mercados imobiliários nas cidades
selecionadas através de métodos multivariados. Esta abordagem permite compreender as
variáveis mais relevantes para estudar os mercados imobiliários e mostra semelhanças e
dissemelhanças. Os resultados indicam que os mercados imobiliários são bastante
segregados dentro de cada aglomeração urbana. Dentro de cada uma delas, os mercados
imobiliários variam substancialmente; fora de cada uma delas, há semelhanças
intrigantes. A posição da cidade no sistema urbano não é um critério chave para ser
classificada em um dado cluster. Essas conclusões são relevantes para a política
habitacional, para os debates migratórios e para a relação entre os ciclos
macroeconômicos e os mercados imobiliários.
Palavras-chave: mercado imobiliário; rede de cidades; habitação; aglomeração;
urbanização; Brasil.
ABSTRACT
Real estate markets in fast-growing cities are a fundamental aspect of human settlements
and key to foster a polycentric national urban-system. This paper investigates the real
estate markets in Brazil’s second-, third- and fourth-tier cities. Using a unique dataset, the
paper explains the variability and creates typologies of real estate markets in the selected
cities through multivariate methods. This approach allows for understanding the most
relevant variables to study real estate markets and shows similarities and dissimilarities.
Results indicate that real estate markets are quite segregated within each urban
agglomeration. Inside each of them, real estate markets vary substantially; outside each
of them, there are intriguing similarities. The city rank in the urban-system is not a key
criterion for being classified in a given cluster. These findings hold relevance for housing
policy, migration debates and for the relationship between macroeconomic cycles and
real estate markets
Key words: real estate; city network; housing; clustering; urbanization; Brazil.
Classificação JEL: R3 – R14.
1. INTRODUCTION
Human settlements are essential dimensions of the development process, although these
critical dimensions are sometimes overlooked in favor of macroeconomic
considerations(Storper & Scott, 2003). Simple evidence is the fact that most of the
countries’ GDP is generated in a few urban agglomerations, and they are the privileged
site for innovation and for welfare spillover. After the 1980s, a ‘spatial turn’ occurred in
development studies, with leading economists studying the field (Martin, 1999). Many
development institutions, such as the World Bank and the Inter-American Bank, have
recognized the important role that cities play in the growth rate of a country, even labeling
cities as “growth machines” (Turok, 2014). In this scenario, it is fundamental that policy
makers have a comprehensive understanding of the relationship between cities’ potential
growth and the capacity to spread these benefits across the territory.
There are empirical shreds of evidence that in the early stages of a country’s
development, a single urban agglomeration or a small group of it may grow fast in a
growth pole-like pattern, operating as a driving force to the whole national economy
(Campolina Diniz & Crocco, 2006; Parr, 1999; Perroux, 1967). These poles buy inputs
(raw materials, food and other agricultural inputs) from other locations, which in turn
boost the national growth through backward linkages (Caceres & Seninger, 1980;
Campolina Diniz & Crocco, 2006; Hirschman, 1965)
1
. Negative agglomeration effects,
such as congestion and higher housing costs, create incentives for the economic activity
to flourish in other locations, forming other tiers of urban agglomerations. The dynamics
of this process may lead to a national urban-system(Behrens et al., 2014)
2
.
In this broad discussion, the characteristics and functioning of real estate markets
in fast-growing cities is a fundamental aspect to human settlements and to foster a more
polycentric national urban-system. One peril of the development process is the second-,
third- and fourth-tier cities getting expansive before achieving its full development
potential. It may be an obstacle for firms and families who would like to move from the
primary city to these emerging centralities. Closely related, urban wages premiums and
human capital migration are also key aspects of the process of deconcentration in the
national urban-system(Scherer et al., 2019). Let us assume that a skilled worker desire to
move from first-tier cities due to the high cost of housing and congestion. In her choice,
the second-, third-, and fourth-tier cities are natural sets of options, since they may provide
jobs, a good level of amenities, and a similar way of living. Nonetheless, this migration
heavily depends on these cities’ real estate markets. Obviously, the cost of living is a
wider variable, which includes a number of variables other than housing, but housing is
the most important component on it(Acolin & Green, 2017). If Florida (2002) rhetoric
was right, that is, if the growth of cities in the 21th century relies on the capacity of the
city to attracted skilled and creative workers, so the “creative class” needs to face the real
estate market in the attractive city.
Therefore, this paper aims to describe the real estate market in Brazil’s second-,
third- and fourth-tier cities. I use a new dataset provided by one of the main real estate
brokers companies in the selected cities in the sample. From the best of the author’s
knowledge, it is the first attempt to describe the real estate markets of this group of cities
together in the literature, and the first to use the methodology applied here. The paper
creates a typology of real estate supply in the selected cities through multivariate methods.
1
The growth-pole strategy was one of the most important regional development theory in the 20th century
(Parr, 1999).
2
On the nature of uneven development, see Dunford and Liu (2017).
7
This approach allows for understanding the most relevant variables to describe real estate
markets in these cities. It also shows similarities and dissimilarities in real estate markets
there. As results indicate, real estate markets are quite segregated within each urban
agglomeration, understood as metropolitan regions or city-regions. Inside each of them,
real estate markets vary substantially; outside each of them, there are intriguing
similarities. The city rank in the national urban-system is not a key criterion for being
classified in a given cluster, which indicates that the selected fourth-tier cities may be
considered relatively expensive comparing with their position in the national urban
hierarchy. This pattern may be one obstacle to migration. These findings hold relevance
for housing policy, migration debates and for the relationship between macroeconomic
cycles and real estate markets(Zhang et al., 2016).
The paper is structured in four sections besides this introduction. The next section
presents a short literature review clarifying the Brazilian urban-system and the selected
cities. Section 3 uses principal component analysis to describe the real estate markets and
cluster analysis to create typologies of real estate markets in Brazil’ssecond-, third- and
fourth-tier cities. Section 4 concludes and suggests further steps in research.
2. BRAZILIANCITY NETWORK
The emergence of second-, third- and fourth-tier cities is a phenomenon that has been
taking place in many developing countries. Figure 1 depicts this process in Latin America
and Caribe. In the Brazilian urban-system, the São Paulo Metropolitan Region’s primacy
rate has been falling since the 1970s, with a sprawl of manufacturing activities over other
urban agglomerations in the Southeast, South and Center-West regions. Primacy rate is
commonly understood as a measurement of the demographic preponderance of the largest
urban agglomeration within a country in relation to the rest of its urban network (UN-
HABITAT, p. 30). As Figure 1 shows, São Paulo and Rio de Janeiro (Brazil’s majorurban
agglomerations) grown less than the national average in the last decade, while Belo
Horizonte, Fortaleza and Recife (third-tier) had pronounced growth rates. In this paper,
Brasília is the second-tier urban agglomeration selected in this paper. These
classifications are based on REGIC (IBGE, 2008), Brazil’s official data on urban-system.
Figure1 - GDP per capita growth of the city and the country - LAC (2010)
Source: UN-Habitat (2012)
As the literature has been describing, the drop of São Paulo’s primacy rate after the
1970s is not associated with a continuous concentration, either with a complete
8
dispersion. Indeed, the Brazilian urban deconcentrating process led to the formation of a
“polygon”, where the vertices are formed by the regional metropolises(Campolina Diniz,
1994).The urban agglomerations that compose the sample selected in this paper are some
vertices of this polygon, such as Belo Horizonte, or are within it, such as Florianópolis.
In addition, they are important enclaves or urban agglomerations that are not connected
to the national urban system's main core, such as Salvador, which is also included in the
sample in this paper. Although not making part of the country’s core region, Salvador is
a third-tier urban agglomeration. This pattern has been observed for other urban
agglomerations in Brazil’s Northeast(Scherer et al., 2019).
Moreover, Brazilian ‘march to West’ has been taking place intensely since the 1960s,
with Brasília (the new planned national capital city) playing a key role as regional
centrality and with the agribusiness (soya beans and cattle) invading the Center-West and
the Amazon region (Castriota & Tonucci, 2018; Monte-Mór, 2014). Figure 2 shows this
sort of polygon based on Campolina Diniz(1993). 23 years later, it is easy to suppose that
now Brasília and Vitória are included in that Figure. As Savedoff (1990, 1991) and
Scherer, Amaral, and Folch (2019) concluded, based on evidence from Brazil, the
national labor market is strongly determined by the structure of the demand for labor. It
means that skilled workers are more demanded in second-, third- and fourth-tier cities
than in small cities.
Figure 3 shows the result of polarization analyzes based on economic linkages,
geomorphological characteristics, and political and cultural identities. In this Figure,
Brazilian map is reshaped according to the macro-poles; it highlights Belo Horizonte,
Salvador and Brasília as macro-poles as well as it shows Vitória and Florianópolis as sub-
poles within their main poles.
Figure 2 – Deconcentration axes and the industrial polygon in the 1990s
Source: Campolina Diniz (1993)
9
Figure 3 – Brazilian Macro and Micro-poles (2000s)
Source: Adapted from Brasil (2008)
3. REAL ESTATE MARKETS IN BRAZIL’S SECOND-, THIRD- AND
FOURTH-TIER CITIES
This section shows and discusses the results of principal component analysis and cluster
analysis. Before doing this, I explain the dataset and the methodology.
In the last years, a number of papers have been describing real estate markets in
Brazil(Aguiar et al., 2014; Almeida et al., 2014, 2017; Campos, 2017; Furtado, 2007;
Furtado, 2011; Nadalin, 2010; Paixão, 2010; Paixão & Abramo, 2008; Paixão & Luporini,
2019; Pontes et al., 2011). However, all these works focus on a specific city or urban
agglomeration. Due to the author’s location and data availability, all of them were about
São Paulo or Belo Horizonte. Only very recent attempts for Brasília are known
(Albuquerque et al., 2018).Different from the USA or China, there is no official index for
Brazil’s real estate markets.
Given the lack of consolidated and comprehensive data on real estate markets in
Brazil, I use data from “Netimóveis”, one of the biggest real estate brokers in Brazil. It
provided a unique dataset with more than 30,000 observations for the selected real estate
typologies. It is worth noting that this kind of sample tends to underrepresent informal
markets. Moreover, Netimóveis has an uneven coverage of the national market, and that
is why the selected sample has only five urban agglomerations: Brasília, Belo Horizonte,
Florianópolis, Salvador and Vitória. Certainly, it would be great to have in the sample
other third-tier urban agglomerations such as Porto Alegre, Curitiba, Recife, Fortaleza,
Belém, Manaus and Goiânia, besides São Paulo and Rio. Within each of the selected
10
urban agglomerations, it was necessary to select some of the main cities. It leads to the
sample exposed in Table 1. Table 1 – Selected cities
State
Urban
Agglomeration
Tier
Cities
Population
Annual Per
Capita
Income
(BRL)
Area
(km²)
HDI*
Federal
District**
Brasília
2nd
Brasília
2,570,160
62,859.43
5,780
0.824
Águas Claras
135,000
54,054.12
32
-
Minas Gerais
Belo Horizonte
3rd
Belo Horizonte
Contagem
Nova Lima
2,375,151
603,442
80,998
32,844.41
37,995.25
109,298.94
331
195
429
0.810
0.756
0.813
Bahia
Salvador
3rd
Salvador
Lauro de
Freitas
2,675,656
163,449
18,264.13
28,859.89
693
58
0.759
0.754
Santa Catarina
Florianópolis
4th
Florianópolis
421,240
32,385.04
675
0.847
Espírito Santo
Vitória
4th
Vitória
Vila Velha
327,801
414,586
64,001.91
21,914.19
97
210
0.845
0.80
Serra
409,267
33,039.02
548
0.739
Source: Author and IBGE Cities Database and REGIC (2007). *Human Development Index. **The Federal
District has a different administrative structure, being Águas Claras and Brasília administrative regions/cities.
The sample included not only the capital city but also one or more cities in the
urban agglomeration. This inclusion is crucial since the commuting process is getting
more and more spatially extended, with the distance between the workplace and the home
achieving tens of miles(Almeida et al., 2017; Soja, 2000; Sudjic, 1992). Parr(2005)
defined the city-region exactly based on the commuting capacity of its residents.
11
Furthermore, the sample includes an urban agglomeration from each macro-region in
Brazil, except North region (mostly formed by the Amazon region), which has a very
particular scale and land use pattern(Monte-Mór, 2004).
I used the principal component analysis to summarize the information presented
in the original dataset. I considered the following typologies: apartments, lofts,
“kitchenettes” (small size apartments), houses, gated community houses, terraced houses,
commercial offices, offices, stores, commercial points and industrial hangars. Each
typology has two possible situations: for sale or for rent. The set of variables contains
information such as price per square meter (m²), rent/m², number of rooms, number of
bathrooms, parking space, and percentage of offers that accepts financing. These real
estate typologies combined with the category for sale or for rent led to 27 variables, which
represented the median value for these variables – the average was not used due to
extreme values. Hence, PCA was a useful tool to synthesize so much information.
PCA results show that three components explain around 72% of the total variance.
Figure 4 shows the variables and cities graph for two dimensions. According to Scherer,
Amaral and Folch (2019, p. 13), “it is a useful tool since the angle formed by any two
variables, represented as vectors, reflects their actual pairwise correlation. Also, on the
graph, objects are distributed based on their similarity and attraction to each other.” The
direction and size of arrows represent the loadings for each one of the real estate
characteristics and the position of dots represents the combination of scores of the first
two components for each city. Águas Claras had no offer for renting houses, which made
the software to exclude it automatically.
12
Figure 4 – PCA: variables, cities and two dimensions
Source: Author.
PCA results illustrate the distinguishing features of the selected cities’ real estate
market. Dimension 1 carries information regarding prices and rents as well as real estate’s
physical characteristics, such as the number of bedrooms and bathrooms. Cities with the
highest prices/m² for residential uses have the highest values in this dimension (horizontal
axis). It is the case of Brasília and Nova Lima. The former is the Brazilian capital city,
and its dwellings are mostly described as expensive houses in the so-called “Plano
Piloto”, mainly close to the lake area. Nova Lima became, after the 1980s, a privileged
space for Belo Horizonte’s elites who decided to live in a house in a gated
community(Almeida et al., 2017; H. Costa et al., 2006; S. de A. P. Costa & Perna, 2015).
On the other side, cities such as Contagem and Serra are industrial cities, where dwellings
are produced for middle classes and blue-collar residents. Moreover, these cities have
large poor areas where the human settlements are marked by self-construction and
informal land tenure.
13
The vertical axes (dimension 2 in Figure 4) carries information on apartments and
commercial real estate. It is possible to say that it is an indicator of cities’ verticalization.
Some important variables in this dimension are the number of bedrooms, parking space
and median rent/m² for apartments. Belo Horizonte is well known for its density (more
than 7000 residents/km²), a high level for Brazilian standards. This verticalization sprawl
to Nova Lima’s borders, with a new centrality – “Seis Pistas” and “Vila da Serra” –
having expensive apartments in high-rise buildings. That is why Nova Lima has both
expensive houses (more distant from Belo Horizonte) and expensive apartments (in the
border with Belo Horizonte). On the other hand, real estate markets offer more
predominantly houses in Brasília. This configuration is one of the reasons that led Águas
Claras to be occupied predominately for apartment buildings. Industrial and suburban
cities, such as Contagem, Serra and Lauro de Freitas, are less verticalized.
Moreover, this dataset showed some other interesting characteristics of real estate
markets in these locations. Vitória’s and Florianópolis’ (summed as “Floripa” to ease
visualization) real estate markets are similar. Both are capital cities, islands, located in
the state with the lowest GDP and population of its region (Espírito Santo, in Southeast,
and Santa Catarina, in South). Both have expensive houses, apartments and commercial
real estate.
The variable "percentage of offers that accepts financing" brought another
interesting result. One could expect that high-income suppliers would be more used to
deal with financing, but the result is the opposite: poorer cities have a higher proportion
of suppliers who accepts financing. One possible explanation of is that people have lower
incomes and wealth are not able to buy without financing. Proportionally, housing
subsidies programs (such as Brazil’s My Home My Life) are more relevant for real estate
markets in cities like Contagem and Serra.
Results indicate that real estate markets are quite segregated in these cities. The
variables that explain the variability in the sample do not depend much on proximity
among the cities, but much more on the kind of land use or the profile of its residents.
Real estate markets offer very different products in the same urban agglomeration. For
instance, Belo Horizonte, Nova Lima and Contagem belong to the same urban
agglomeration, but the most important variables for explaining their real estate markets
are quite different. The same happens with Brasília-Água Claras and Vitória-Serra.
Salvador-Lauro de Freitas is the exception in this sample. On the other hand, real estate
markets offer relatively similar products in cities from different urban agglomerations.
From this discussion, a natural further step is to classify the similarities among the
cities through clustering analysis. To do this, I employed a hierarchical cluster, since I did
not have a previous suggestion of how many clusters I would have. Considering the full
information provided by the dataset helped to have a better understanding of the
similarities among these cities – better than just look in the PCA two-dimensional graph.
I used the Ward method and 27 variables, including median calculations of real estate
physical characteristics, prices and rents per city. The agglomeration coefficient was 0.86,
suggesting a strong clustering structure.
14
Figure 5 – Hierarchical Cluster – Selected Cities
Source: Author.
Figure 5 shows the dendrogram or treemap for the hierarchical clustering of the
selected cities. Serra and Contagem form a cluster that one may call as industrial cities,
have similar real estate markets with high levels of low-priced financed real estate, and
humble residential houses with a low number of bedrooms and bathrooms, and large
industrial real estates. On the other side, Nova Lima and Brasília illustrate cities occupied
by the elites, who live in mansion houses close to lakes, and in some cases, expensive
apartments. Vitória and Águas Claras are cities with small areas where the elites and the
middle classes live in relatively expensive apartments.Nonetheless, Vitória is the capital
city of a fourth-tier urban agglomeration while Águas Claras is a suburban administrative
region in a second-tier urban agglomeration (Brasília). Real estate markets offer both
houses and apartments in Vila Velha, Belo Horizonte, Salvador and Florianópolis. As the
most populated city in each respective urban agglomeration, they still have diversified
real estate offers in the same territory. Moreover, they are highly demanded places for
commercial real estate in the respective local market. The capital cities still having
relevant roles as commercial places and still offering a diversity of properties. Lauro de
Freitas, as a suburban extension of Salvador, may be clustered together with these cities.
As PCA results also indicated, there is remarkable segregation in these real estate
markets. Cities within the same urban agglomeration are not classified in the same cluster,
whilst cities in different urban agglomerations are classified in the same cluster. A high-
income homebuyer could pick a similar house in Brasília or Nova Lima, as a low-income
homebuyer could pick similar housing products in Contagem or Serra. Being located in a
second-, third- or fourth-tier urban agglomeration is not a key criterion for a property
being classified in a given cluster. This result suggests that fourth-tier cities such as Vila
Velha, Vitória, Serra and Florianópolis have relatively expensive housing offers
considering their position in the national urban-system. Considering the per capita income
level (Table 1), particularly Salvador (third-tier) and Vila Velha (fourth-tier) raises
15
concerns about their housing affordability, since they are clustered together with much
more rich and developed cities such as Belo Horizonte.
Once cities of different tiers have their real estate markets clustered together, the
macroeconomic policy may have mixed results over them. This result differs from
Zhang’s et. al. (2016) findings, who found similar impacts of macroeconomic exogenous
shocks (e.g. interest rates) in cities of the same tier, in general.
4. CONCLUSIONS
This paper explored Brazil’s real estate markets under the perspective of its urban-
system. PCA and hierarchical cluster show how the selected urban agglomerations have
a much-segmented spatial structure considering its real estate markets. Specific cities are
destined to be the habitat of middle classes and blue-collar workers, such as Serra and
Contagem, while in others the real estate markets produces expensive houses for the
elites, such as Nova Lima and Brasília. It is a clear illustration of the division of labor in
the space and of segregation. These forms of market segmentation take place both in
second-, third- and fourth-tier cities.
A real estate being located in a second-, third- or fourth-tier urban agglomeration is
not a key criterion for being classified in a given cluster of properties. One possible
implication of it is that fourth-tier cities may be relatively less affordable than second-
and third-tier cities, since cities of higher ranks in the urban-system offer higher urban
wage premiums in Brazil, in general (Scherer et al., 2019).
The capital cities still having relevant roles as commercial places and still offering a
diversity of properties. This fact is an evidence of how the Brazilian urbanization differs
from United States, for instance, where the inner cities became decayed areas in many
metropolises and the elites flew to suburban cities (Ehrenhalt, 2012; Harvey, 2014).
Further steps in this research will include more cities in the sample. As recent research
had done, comparisons among different tier of cities may guide to important findings,
both for macroeconomic and housing policy. For instance, the effects of interest rates on
housing prices may varies in different tiers of cities(Zhang et al., 2016). In addition,
spatial explicitly techniques could be employed to capture spatial effects, such as the
neighborhood relation between capital cities and its urban agglomerations.
REFERENCES
Acolin, A., & Green, R. K. (2017). Measuring housing affordability in São Paulo metropolitan
region: Incorporating location. Cities, 62, 41–49.
Aguiar, M. M. de, Simões, R., & Golgher, A. B. (2014). Housing market analysis using a
hierarchical–spatial approach: the case of Belo Horizonte, Minas Gerais, Brazil. Regional
Studies, Regional Science, 1(1), 116–137.
Albuquerque, P. H. M., Nadalin, V. G., Lima Neto, V. C., & Montenegro, M. R. (2018).
Construção de índices de preços de imóveis para o Distrito Federal por meio de vendas
repetidas e GWR. Nova Economia, 28(1), 181–212.
Almeida, R. P., Monte-Mór, R. L. M., & Amaral, P. V. M. (2017). Implosão e explosão na
Exópolis: evidências a partir do mercado imobiliário da RMBH. Nova Economia, 27(2),
323–350.
Almeida, R. P., R. L. M. Monte-Mór, & Amaral, P. V. M. (2014). Implosion and Explosion
through Real Estate Market in the Metropolitan Area of Belo Horizonte, Brazil. 61st
Annual North American Meetings of the Regional Science Association International,
Washington D.C.
Brasil. (2008). Estudo da Dimensão Territorial para o Planejamento. Brasília: MP.
16
Behrens, K., Duranton, G., & Robert-Nicoud, F. (2014). Productive cities: Sorting, Selection and
Agglomeration. Journal of Political Economy, 122(3), 507–553.
Caceres, L. R., & Seninger, S. F. (1980). Endogenous change in an Interregional System of
Growth Centers: An Application to the Central American Common Market. Regional
Studies, 14(1), 1–13.
Campolina Diniz, C. (1993). Desenvolvimento Poligonal no Brasil: nem desconcentração, nem
contínua polarização. Nova Economia, 3(1), 35–64.
Campolina Diniz, C. (1994). Polygonized Development in Brazil: Neither Decentralization nor
Continued Polarization. International Journal of Urban and Regional Research, 18(2),
293–314.
Campolina Diniz, C., & Crocco, M. (2006). Economia Regional e Urbana: contribuições teóricas
recentes. UFMG.
Campos, R. B. A. (2017). O mercado imobiliário residencial no município de São Paulo: uma
abordagem de preços hedônicos espacial. Nova Economia, 27(1), 303–337.
Castriota, R., & Tonucci, J. (2018). Extended urbanization in and from Brazil. Environment and
Planning D: Society and Space, 36(3), 512–528.
Costa, H., Costa, G., Monte-Mór, R. L. M., & Mendonça, J. (2006). Novas periferias
metropolitanas – a expansão metropolitana em Belo Horizonte: dinâmica e
especificidades no Eixo Sul. C/Art.
Costa, S. de A. P., & Perna, S. de A. (2015). The continuous expansion of Brazilian cities: the
case of the city of Belo Horizonte. Journal of Urbanism: International Research on
Placemaking and Urban Sustainability, 8(1), 57–81.
Dunford, M., & Liu, W. (2017). Uneven and combined development. Regional Studies, 51(1),
69–85.
Ehrenhalt, A. (2012). The Great Inversion and the Future of the American City. Vintage.
Florida, R. (2002). The Rise of the Creative Class. And How It’s Transforming Work, Leisure and
Everyday Life. (1st ed.). Basic Books.
Furtado, Bernardo A. (2007). Mercado imobiliário e a importância das características locais: uma
análise quantílico-espacial de preços hedônicos em Belo Horizonte. Revista Análise
Econômica, 25(47), 71–98.
Furtado, Bernardo Alves. (2011). Neighbourhoods in Urban Economics: Incorporating
Cognitively Perceived Urban Space in Economic Models. Urban Studies, 48(13), 2827–
2847.
Harvey, D. (2014). Cities or Urbanization? In Implosions/Explosions: Towards a Study of
Planetary Urbanization. (Eds.) Neil Brenner and Christian Schmidt. Jovis.
Hirschman, A. O. (1965). Política econômica na América Latina. Fundo de Cultura.
IBGE. (2008). Regiões de influência das cidades 2007 - REGIC. IBGE.
Martin, R. (1999). The new “geographical turn” in economics: some critical reflections.
Cambridge Journal of Economics, 23(1), 65–91.
Monte-Mór, R. L. M. (2004). Modernities in the jungle: extended urbanization in the Brazilian
Amazonia [PhD Dissertation]. University of California, Los Angeles.
Monte-Mór, R. L. M. (2014). What is the urban in the contemporary world?
InImplosions/Explosions: Towards a Study of Planetary Urbanization. (Eds.) Neil
Brenner and Christian Schmidt. Jovis.
Nadalin, V. G. (2010). Três Ensaios sobre a Economia Urbana e Mercado de Habitação em São
Paulo [PhD Thesis]. Universidade de São Paulo.
Paixão, L. A. (2010). Externalidades de vizinhança, estruturação do espaço intraurbano e preços
dos imóveis: evidências para o mercado de apartamentos de Belo Horizonte. Ensaios
FEE, 31, 235–258.
Paixão, L. A., & Abramo, P. (2008). Os vetores de expansão da atividade imobiliária em Belo
Horizonte – 1994-2003. Nova Economia, 18, 229–263.
Paixão, L. A., & Luporini, V. (2019). A valorização imobiliária em Belo Horizonte, 1995-2012:
uma análise hedônica-quantílica. Nova Economia, 29(3), 851–880.
Parr, J. B. (1999). Growth-pole Strategies in Regional Economic Planning: A Retrospective View:
Part 2. Urban Studies, 36(8), 1247–1268.
17
Parr, J. B. (2005). Perspectives on the City-Region. Regional Studies, 39(5), 555–566.
Perroux, F. (1967). The Economy of 20th Century. Herder (In Portuguese).
Pontes, E., Paixão, L. A., & Abramo, P. (2011). O mercado imobiliário como revelador das
preferências pelos atributos espaciais: uma análise do impacto da criminalidade urbana
no preço de apartamentos em Belo Horizonte. Revista de Economia Contemporânea,
15(1), 171–197.
Scherer, C., Amaral, P. V. M. do, & Folch, D. (2019).A comparative study of urban occupational
structures: Brazil and United States. Revista Brasileira de Estudos de População, 36, 1–
19.
Soja, E. W. (2000). Postmetropolis: A Critical Study of Cities and Regions. Blackwell Publishers.
Storper, M., & Scott, A. J. (2003). Regions, Globalization, Development. Regional Studies, 37(6–
7), 597–593.
Sudjic, D. (1992). The 100 Mile City. André Deustch.
Turok, I. (2014). Linking urbanisation and development in Africa’s economic revival. In Africa’s
Urban Revolution. Parnell, S. and Pieterse, E. (eds). Zed Books.
Zhang, H., Li, L., Hui, E. C.-M., & Li, V. (2016). Comparisons of the relations between housing
prices and the macroeconomy in China’s first-, second- and third-tier cities. Habitat
International, 57, 24–42.
