Faculty of Economics
Cambridge Working Paper Economics
Cambridge Working Paper Economics: 1636
TRACING THE EVOLUTION OF AGGLOMERATION ECONOMIES:
SPAIN, 1860-1991
Francisco J. Beltrán Tapia
University of Cambridge
Alfonso Díez-Minguela
Universitat de València
Julio MartinezGalarraga
Universitat de
València
3 June 2016
This article attempts to quantify how the effect of agglomeration economies on population growth has
evolved over time. Using district population in Spain between 1860 and 1991, recorded approximately
every decade, this article examines whether initial population affects subsequent population growth. Our
results show that, while the relationship between these two variables hardly existed during the second
half of the 19th century, this link increased significantly between 1910 and 1970, although this trend
was abruptly interrupted by the Civil War and the autarkic period that followed. The intensity of this
relationship debilitated in the 1970s, a process that continued during the 1980s as rural out-migration
diminished and de-industrialisation hit traditional manufacturing sectors. Our findings also stress that
agglomeration economies were stronger in medium-size districts, especially from 1960 onwards, thus
suggesting that congestion costs began to mitigate the benefits arising from agglomeration economies in
the largest locations.
Tracing the Evolution of Agglomeration Economies: Spain, 1860-1991
Francisco J. Beltrán Tapia
Alfonso Díez-Minguela
Julio Martinez-Galarraga
University of Cambridge
Universitat de València
Universitat de València
(fjb38@cam.ac.uk)
(alfonso.diez@uv.es)
(julio.martinez-galarraga@uv.es)
May 30, 2016
Abstract
This article attempts to quantify how the effect of agglomeration economies on
population growth has evolved over time. Using district population in Spain
between 1860 and 1991, recorded approximately every decade, this article
examines whether initial population affects subsequent population growth. Our
results show that, while the relationship between these two variables hardly
existed during the second half of the 19th century, this link increased
significantly between 1910 and 1970, although this trend was abruptly
interrupted by the Civil War and the autarkic period that followed. The
intensity of this relationship debilitated in the 1970s, a process that continued
during the 1980s as rural out-migration diminished and de-industrialisation hit
traditional manufacturing sectors. Our findings also stress that agglomeration
economies were stronger in medium-size districts, especially from 1960
onwards, thus suggesting that congestion costs began to mitigate the benefits
arising from agglomeration economies in the largest locations.
Keywords: Agglomeration economies, Regional growth, Spain
JEL classification: N93, N94, O18, R11, R12
Acknowledgements: This paper was presented at the 2016 Economic History Society Annual Conference at
Cambridge. We would like to thank the comments received there, as well as those by Javier Silvestre and
Rafael González-Val. This research received financial support from the Spanish Ministry of Science and
Innovation (projects HAR2012-30732, ECO2012-33286 and ECO2015-65049-C12-1-P).
1
1. Introduction
Agglomeration economies refer to the benefits that arise when firms and people locate near
one another (Glaeser, 2010). These might appear from the scale and/or the diversity of the
local economy (Marshall, 1890; Jacobs, 1969; Henderson, 2003). In particular, the spatial
concentration of economic activity increases market access, thus resulting in cheaper and
more varied inputs, as well as allowing the sharing of risk and indivisible facilities (i.e.
airports, universities, hospitals). Besides, denser locations enable a more efficient matching
between firms and workers or buyers and sellers, both in terms of quantity and quality of
matches; and facilitate knowledge spillovers within and across industries (Duranton and
Puga, 2004)1. Yet, agglomeration is also associated with expensive housing, long commutes
and pollution, among other costs. There appears then to be a trade-off between increasing
returns and congestion costs (Fujita and Thisse, 2002)2.
In a seminal contribution, Krugman (1991) suggests that, through a process of
circular causation, the interaction between economies of scale and transport costs might
lead to the emergence of an industrialised «core» and agricultural «periphery»3. In fact, this
economic geography model also entails that initial conditions (i.e. population density) are
self-reinforcing, thereby stressing the role of history on the spatial concentration of
industry. In this vein, Puga (1999) stresses the relevance of workers mobility to income
differentials. If the agglomeration of economic activity raises wages and workers are
perfectly mobile, these will tend to relocate near industrial clusters. Consequently, structural
change should be one of the major driving forces of urbanisation, as Michaels et al. (2012)
show. However, urban agglomeration continues, even in countries where employment
shifted away from agriculture long ago, thereby stimulating the need for further research4.
The relationship between economic density and productivity has been at the core of
most debates on agglomeration economies. Ciccone and Hall (1996) pioneered this strand
of the literature studying productivity differentials within the United States. On the whole,
these authors found that variation in output per worker across states partly reflects
differences in the density of economic activity. Since then, several studies have attempted
to quantify the effect of economic density on productivity. Although results may vary
according to the level of aggregation, period of study and/or estimation method, it is
1
See Combes and Gobillon (2015) for a recent review.
Using a novel methodology, Combes et al. (2012) estimate the elasticity of urban costs with respect to
population, and provide evidence in support of the trade-off between agglomeration economies and urban
costs.
3 See also Krugman and Venables (1995), Venables (1996), Fujita et al. (1999) and Combes et al. (2008).
4 As the World Urbanisation Prospects: The 2014 Revision indicates, all regions are expected to urbanise further in
the next decades.
2
2
somewhat accepted that density increases the productivity of firms and workers (Combes
and Gobillon, 2015)5. Yet, most empirical studies within this regional economics literature
have a static or short-term perspective, thereby ignoring long-run dynamics6. Data
availability has usually constrained research on how agglomeration economies vary along
the process of economic development. In fact, long-run studies have usually replaced
employment for population, and use population growth as a proxy for economic dynamism
(Beeson et al., 2001; Dobkins and Ioannides, 2001; Michaels et al., 2012; Desmet and
Rappaport, 2016).
In addition, the literature in the field of urban economics has focused on the study of
one of the most peculiar empirical regularities, the Gibrat’s law, which suggests that city
growth is independent of its initial size (Clark and Stabler, 1991; Gabaix, 1999; Eeckhout,
2004). In short, if the spatial concentration of population follows a random process,
agglomeration economies would be irrelevant. But, if this is not the case, and hence
Gibrat’s law is violated, agglomeration economies might play a critical role. Similarly,
location fundamentals or «first nature» advantages (i.e. climate, geography) are expected to
be essential in the location of preindustrial populations (Beeson et al., 2001; Dobkins and
Ioannides, 2001; Davis and Weinstein, 2002; Bosker and Buringh, 2015)7. Yet,
industrialisation brought about a new scenario where agglomeration economies overcame
other major forces (Gabaix and Ioannides, 2004; Duranton, 2007; Rossi-Hansberg and
Wright, 2007; Michaels et al., 2012). In this regard, it appears that the link between initial
population and subsequent population growth only started to be visible in the late 19th
century, accelerated from then onwards and slowed down in the late 20th century. Michaels
et al. (2012) argue that this pattern is related to reallocation away from agriculture, a
process that weakened during the latter period. However, congestion costs, especially in the
largest locations, may also contribute to explaining the mitigation of agglomeration
economies in recent decades (Puga, 1999; Graham, 2007; Combes et al., 2012)8.
In this paper, we analyse how the relationship between agglomeration economies and
the spatial distribution of population has evolved over time. Our contribution is twofold.
5 Ciccone and Hall (1996) find that doubling employment density increases average labour productivity by 6
per cent. Melo et al. (2009) use a meta-analysis, which includes 729 estimates from 34 studies, to analyse how
study characteristics might affect results. Combes and Gobillon (2015) survey the existing literature and
report that the elasticity of productivity with respect to density usually ranges between 0.04 and 0.07.
Furthermore, recent studies have shown that agglomeration economies appear to have a greater impact in
developing economies, such as China and India (Combes et al., 2013; Chauvin et al., 2014).
6 Ciccone and Hall (1996); Desmet and Fafchamps (2005).
7 Cronon (1991) introduced the terms «first nature» and «second nature» referring to “original, prehumen nature”
and “artificial nature that people erect atop first nature” respectively.
8 See also Krugman and Venables (1995) and Venables (1996).
3
First, we introduce a novel dataset that traces the evolution of the Spanish population at
the district level during the period 1860-1991. The data, which comprises 464 districts and
is recorded on a decadal basis, allow us to capture the transition from a pre-industrial
society to a modern economy. Second, our empirical analysis examines whether initial size
affects subsequent growth. If agglomeration economies play a role, orthogonal growth
would not hold and large districts would grow more rapidly than small ones, thus violating
Gibrat’s law and increasing spatial concentration. In order to isolate the effect of initial size
from other potential determinants of population growth, we have considered climatic and
geographic information to capture the «first nature» advantages of each district. We also
control for other issues, such as the «capital effect» and the economic dynamism of
neighbouring locations. Potential endogeneity is further addressed by instrumenting the
size of the local economy using historical urban population.
Our results show that the relationship between district size and population growth
hardly existed during the second half of the 19th century. Although interrupted by the
Spanish Civil War and the autarkic period that followed, the link between these two
variables increased significantly between 1910 and 1970. These findings, in line with
previous studies, illustrate the relevance of structural change and agglomeration economies
in the shaping of a modern economy. The intensity of this relation slightly weakened in the
1970s, a process that continued during the 1980s as rural out-migration slowed down and
de-industrialisation hit traditional manufacturing sectors (i.e. metallurgy, extractive). Lastly,
we also find that agglomeration economies appear to have a greater impact on medium-size
districts, especially from 1960 onwards, thus suggesting that congestion costs might have
started to mitigate the benefits arising from economic density in the largest locations.
The rest of the paper is structured as follows. Section 2 provides a brief historical
account of the evolution of the Spanish economy between 1860 and 1991. Data and
descriptive analyses are introduced in Section 3, whilst Section 4 presents the empirical
analysis. Section 5 discusses our main findings and Section 6 concludes.
2. Historical background: Spain 1860-1991
Our period of study, 1860-1991, spans one and a half centuries, thus covering most of the
process of economic development. During this time, the Spanish economy undertook a
profound structural transformation that turned a predominantly agricultural society into a
modern economy by the late 20th century: labour shifted away from agriculture to industry
and services, and income per capita increased accordingly (see table 1 and figure 1).
4
TABLE 1. Real GDP, population and per capita GDP growth, 1850-2000
GDP
Population
Per capita GDP
1850-1883
1.8
0.4
1.4
1884-1920
1.3
0.6
0.7
1921-1929
3.8
1.0
2.8
1930-1952
0.8
0.9
0.0
1953-1958
4.7
0.8
3.9
1959-1974
6.9
1.1
5.8
1975-1986
2.5
0.7
1.8
1986-2000
3.5
0.2
3.3
Source: Prados de la Escosura (2008, 288). Annual average logarithmic rates.
20
Figure 1. Labour force by main economic activity in Spain, 1860-2000 (unit: millions)
Agriculture
Industry
0
5
10
15
Services
1860
1877
1887
1900
1910
1920
1930
1940
1950
1960
1970
1981
1991
2000
Source: Population censuses
The Spanish economy initiated the early stages of modern economic growth in mid19th century. Economic growth was initially fostered by the integration of the national
market and the adoption of industrial innovations. However, up to the First World War,
economic growth rates progressed at a slow pace, industrialisation advanced with
difficulties and unevenly distributed across space, and structural change was limited. By
1910, as figure 1 illustrates, more than two thirds of the labour force were still enrolled in
agriculture.
5
The integration of Spain’s domestic market received a strong impulse in the middle
of the 19th century9. Before that, as a consequence of the persistence of barriers and
limitations to internal trade, the national market was fragmented into various local and
regional markets that were largely unconnected. Local tariffs and regulations restricting
trade were widespread and weights and measures differed across regions. In addition,
transport costs were very high due to low public investment in transport infrastructures,
the use of traditional means of transport and the particular geography of Spain, which was
highly rugged and lacked an extensive water transport system. As a result, regional
commodity markets were scarcely integrated and prices markedly differed from one region
to another. It is true though that some interdependence in commodity prices had existed
since the 18th century (Ringrose, 1998).
The successive political reforms of the 19th century promoted market liberalisation.
Laws were unified, legal support was given to property rights, and tariffs and local
restrictions on home commerce were eliminated (Tedde, 1994). In addition, the expansion
of the rail network brought with it major changes that favoured the progressive
development of the domestic market. The first line was finished in 1848, covering the 28
kilometres that separated Barcelona and Mataró. By 1866 the railway linked up Spain’s
main economic centres and by 1901 all the provincial capitals were connected to the
railway (Wais, 1987); at that time, the railway network covered a distance of 10,827
kilometres (Herranz, 2005)10 and the country’s infrastructure stock as a share of GDP rose
from 4,27% in 1850 to 27,2% in 1900 (Herranz, 2007). Transport improvements,
particularly the completion of Spain’s railways network, favoured the fall in transport costs
and the creation of a national market for most important commodities during the second
half of the 19thcentury11. Indeed, regional wheat prices converged in this period (Peña and
Sánchez-Albornoz, 1984)
In parallel to the integration of the domestic market, manufacturing industries
became increasingly concentrated in space (figure 2). While inland regions experienced a
substantial process of deindustrialisation (with the exception of Madrid), Spanish
industrialisation was mainly led by Catalonia and the Basque Country (Nadal, 1987)12. By
1910, the contribution of these regions to Spanish industrial output was 31.6% and 6.9%,
9
See Rosés et al. (2010) for a detailed description.
By 1990, the length of the railway network was 12,560 km (Gómez Mendoza and San Román, 2005).
11 According to Herranz (2005), the introduction of the railway represented a reduction of up to 86% in
haulage costs in 1878.
12 In Catalonia, the cotton industry, with a tradition that stretched back to the 18th century, gradually became
mechanised in the 19th century. In the Basque Country, the iron and steel industry underwent rapid growth
in the last quarter of the century.
10
6
while their population only represented 10.5% and 3.4% respectively (Tirado and MartinezGalarraga, 2008). In addition, internal migratory flows were relatively low throughout most
of the 19th century (Silvestre, 2005). Due to the predominance of agrarian activities and
their subsequent seasonality, an important part of these movements was temporary and
occurred over short distances (Silvestre, 2007). Indeed, up to the 1920s the number of
permanent internal migrations remained small (see figure 3)13. International migration, on
the other hand, experienced a notable increase in the late 19th century and the first decades
of the 20th century, mainly heading to Latin America (Sánchez-Alonso, 2000).
0
.2
Gini index
.4
.6
.8
Figure 2. Concentration in manufacturing, Spain 1856-1995
1856
1893
1913
1929
1955
1975
1995
Source: Paluzie et al. (2004)
The integration of the Spanish market further progressed throughout the interwar
years, especially with a substantial increase of paved roads, which complemented the
previous development of the railway network (Herranz, 2005)14. In addition, although the
notable progress of electrification mitigated previous energy restrictions traditionally faced
by Spain’s industry and the number of industrial locations expanded, mainly to Madrid and
Zaragoza, the spatial concentration of manufacturing continued. The increasing market
13 The number of Spaniards residing outside their province of origin was relatively small, about 9.3 per cent
by 1910 (Silvestre, 2005).
14 The road network increased from 36,300 km in 1900 to 109,176 in 1935. By 1960 the distance covered was
130,644 and reached 162,298 km in 1990 (Gómez Mendoza and San Román, 2005).
7
integration was accompanied by large inter-regional migrations: Spaniards left declining
regions, which were mainly rural and agrarian, to reallocate in the richest regions, which
were more urban and specialised in industry and services. Yet, these migratory flows did
not originated in the poor areas of the south of the country because these provinces were
far away from the industrial centres (Silvestre, 2005)15. In parallel to these developments,
structural change accelerated and the share of agrarian employment decreased substantially
while economic growth rates significantly increased.
10
Figure 3. Emigration, returns and internal migration in Spain by period (per 1,000 people)
Emigra2on
Returns
Internal migra2on (1)
0
2
4
6
8
Internal migra2on (2)
1887-1900
1901-10
1911-20
1921-30
1931-40
1941-50
1951-60
1961-70
1971-80
1981-91
Source: (1) Emigration and returns: Sánchez-Alonso (1995) and Bover and Velilla (1999)16.
(2) Internal migrations: Paluzie et al. (2009b: Table 1, p. 255)
Note: Internal migrations refer to the annual rate (per 1,000) of born in another province during the
inter-census period from 1877 to 1940. Yet, between 1961 and 1991, it is the annual rate (1,000) of
residents in another province (at previous census date) during the inter-census period (García-Coll
and Stiwell, 1999). These two indicators are dissimilar hence they are not strictly comparable. See
Paluzie et al. (2009b) for more information.
The Spanish Civil War (1936-1939) and the first years of Franco’s regime supposed a
major setback, not only to economic growth, but also to economic integration. The
autarkic policy that followed the Civil War came hand in hand with a tight regulation of
commodity and input markets, including state control of prices and quantities in most
15
This process favoured the convergence registered in regional wages that had begun in the mid-19th century
(Rosés and Sánchez-Alonso, 2004).
16 We would like to thank Olympia Bover for kindly sharing the data with us.
8
goods. Although these policies created a false impression of price convergence, internal
trade hardly increased. In addition, due to the lack of investment in infrastructure,
transport costs remained unaltered during the 1940s and early 1950s. Economic growth
and structural change therefore came to a halt: agrarian employment actually increased
during the 1940s and it took twenty years to return to the pre-Civil War per capita income
levels.
The economic liberalisation and stabilisation measures introduced at the end of the
1950s, together with foreign assistance, favoured the transition of the Spanish economy
toward a new phase of economic development (Prados de la Escosura et al., 2012). This
period was characterised by high economic growth rates and by the lead taken by the
industrial sector in the country’s economic activity. New investments in infrastructures
such as roads, railways and communication networks led to further reductions in internal
transport costs. Spanish economic growth in the 1960s was also characterised by the
growing mobility of the labour force that was becoming increasingly concentrated in the
big cities. Rural exodus towards cities, as well as to more developed European countries
(figure 3), resulted in a substantial decline in agrarian employment and an increase of the
share of manufacturing, construction and services sectors (Ródenas, 1994; Bover and
Velilla, 1999; Bentolila, 2001). Contrary to the previous phase of the 1920s, migrants from
the southern provinces now played a key role and massively participated on migratory
flows. However, although to a lower extent, migrants’ destinations were still limited to a
relatively small number of large cities, mainly Madrid and Barcelona17. Regarding
international migration, a new wave of out-migration flows took place in 1960-73, with
more than 100,000 workers migrating per year to the core European countries.
However, the crisis of the 1970s, which in the case of Spain stretched well into the
1980s, put a brake on these upward trends and GDP growth rates were lower than in the
previous decades. The concentration of manufacturing industries somewhat receded during
these years, thus causing the spatial distribution to present a bell-shaped evolution in the
long-term (figure 2). Furthermore, traditional industries (i.e. mining, metallurgy) underwent
severe reconversion processes in the 1980s. Importantly, inter-regional migration rates fell
in the 1970s and early 1980s, arguably as a result of the high unemployment registered
during those years (Bentolila and Blanchard, 1990; Bentolila and Dolado, 1991).
17
In 1930, almost half of the population living in a province different from the birth province was residing in
Madrid (22.9%) and Barcelona (22.9%). By 1970, the percentages were similar, 23.8 and 23.7% respectively
(Paluzie et al., 2009b, 248).
9
The new phase in Spanish economic growth, which started after the entry of the
country into the European Union in 1986, was no longer linked to the leadership of
industrial production, but rather to that of the services and construction sectors. Internal
migrations were now characterised by an increase in the dispersion of migratory flows due
to the growing importance of services, an economic sector that is much less spatially
concentrated than industry. Increasing congestion costs, such as the rise in housing prices,
the higher weight of amenities and other aspects related to the quality of life or the effect
of redistributive policies would also account for the lower intensity of migratory flows in
the last decades (Bover and Velilla, 1999; Bentolila, 2001). Yet, these declining interprovince migrations, shown in figure 3, were partially counterbalanced by growing intraprovince migrations (Paluzie et al., 2009b). In addition, a new wave of investment in
infrastructure helped to further reduce transport cost across Spanish regions and also
across national borders. Huge investments in freeways, high-speed railway and
telecommunications were developed during these years and led to major advances in the
integration of the internal Spanish market and its connection to international markets. In
this respect, the accession of Spain to the EU in 1986 bolstered the Spanish economy, thus
further promoting the catching-up process with the most developed countries during the
1990s. The tertiarisation of the economy was completed at the same time that substantial
GDP per capita growth rates were reached (Prados de la Escosura and Rosés, 2009).
Different studies have examined the relevance of agglomeration economies and the
spatial distribution of population in Spain. On the one hand, using provinces as unit of
analysis, Martinez-Galarraga et al. (2008) find that, in line with Ciccone and Hall (1996),
since the mid-19th century, doubling employment density increases industrial labour
productivity by around 3-5%. It appears though that this relationship declines over time. In
addition, Pons and Tirado (2008) analyse the inter-provincial disparities in the distribution
of economic activity throughout the 20th century and conclude that the evolution of
regional income inequality has eventually been shaped by agglomeration economies, while
the impact of the initial geographic conditions has decreased over time18.
On the other hand, exploring the evolution of the 100 largest cities during the 20th
century, Lanaspa et al. (2003) find that a convergent pattern of growth dominated between
1900 and 1970 and divergence followed thereafter. While this study is just concerned with
18 Other studies have shown the relevance of agglomeration economies in the industrial sector during the
early stages of economic growth in Spain (Tirado et al., 2002; Martinez-Galarraga, 2012); examined and
verified the existence of the wage equation in industrial wages (Paluzie et al., 2009a; Tirado et al., 2013); and
proved the direct relationship between internal migration decisions and market access in the 20th century
(Paluzie et al., 2009b).
10
the upper tail of the distribution, Goerlich and Mas (2008; 2009) employ all municipalities
to illustrate a tendency towards spatial concentration over the whole 20th century, especially
between 1950 and 198119. Besides, they also stressed the relevance of the “capital status”:
provincial capitals attracted a significant bulk of the population from neighbouring areas.
Analysing provinces, Ayuda et al. (2010) find that location fundamentals or «first nature»
explain the spatial distribution of population before industrialisation (up to 1900).
Disparities in population density increased from then onwards once «second nature»
factors grew in relevance. The initial effect of «first nature» was thus reinforced by «second
nature», which is closely related to increasing returns and thus the presence of
agglomeration economies. More recently, González-Val et al. (2015) explore the
relationship between market potential and city growth during the period 1860-1960. Urban
growth was first the result of location fundamentals, whereas market potential appears to
be significant over the 20th century. Although these studies suggest the increasing
importance of agglomeration economies over time, further analysis is required.
3. Data
In order to better understand the long-run evolution of agglomeration economies in Spain,
we have built a panel data set that traces the population of 464 districts from 1860 to
199120. Based on the Population Censuses that were carried out approximately every
decade, our dataset thus covers 13 periods. Within the framework of an integrated
economy, the use of population rather than income data in measuring economic activity
has the advantage of taking into account that migration flows respond to income
differences and tend to mitigate them. Regional differences in productivity might then be
better reflected in population figures (Glaeser et al., 1995; Beeson et al., 2001)21.
19
Goerlich and Mas (2008; 2009) are based on a database that covers all Spanish censuses from 1901 to 2001,
with municipalities homogenised to their structure in 2001 (Goerlich et al., 2006).
20 Due to data availability the Canary Islands districts -Arrecife, Guía, La Laguna, Las Palmas, Orotava, Santa
Cruz de la Palma, and Santa Cruz de Tenerife-, have been excluded from our sample. The period of study
finishes in 1991 because the population category that had been employed since 1860 by the Spanish Statistical
Agency (Población de hecho) disappears from later censuses.
21 Our study is somewhat constrained by data availability. Using population, instead of working-age
population or employment, might not capture relevant changes in population structure. While migration
flows increase the number of workers in the receiving districts, the working-age population decreases in
sending regions. Although population growth may initially reflect these flows, it may also result from
differential demographic patterns present in younger populations. Employing population data could thus lead
to overestimate the effect of agglomeration economies. We acknowledge this issue, though we think it would
be more evident from 1991 to 2011.
11
The unit of analysis is the district or Partido Judicial, an administrative category
composed of several municipalities22. Employing districts as units of analysis presents
several advantages (Beeson et al., 2001; Desmet and Fafschamps, 2005; Partridge et al.,
2008). On the one hand, districts better capture the potential effects of agglomeration
economies than cities because they allow taking into account their hinterland, as well as
avoiding the comparability problems generated by the rise of metropolitan areas. On the
other hand, given that we cover the whole Spanish territory, district-level data avoids the
sample selection bias usually present in the literature focusing on cities. These studies only
consider settlements above a certain threshold, thus only focusing on those that have been
relatively successful and missing those that did not grow enough to reach that limit or
those that declined and fell below that figure. These two features of the data are crucial not
only because most population has traditionally lived in rural areas, but also because rural
out-migration was an essential dimension of how the spatial distribution of the population
evolved. Michaels et al. (2012, 536) show that examining both rural and urban areas
significantly enhances our understanding of the urbanisation process. As these authors
point out, the unit of analysis should be stable over time (548). Given that during the
period under analysis, legislative changes have somewhat affected these administrative
boundaries, we have homogenised our panel dataset using the administrative boundaries
existing in 1860. We therefore rely on district boundaries that are consistent over the whole
sample period. Overall, the average surface area is 1,075 squared kilometres, which allow us
to capture metropolitan areas (i.e. Madrid, Barcelona), but at a lower level of aggregation
than provinces (NUTS3), thus reducing potential distortions arising from the Modifiable
Areal Unit Problem (MAUP) (Briant et al., 2010)23. By means of comparison, the average
size of a US county is around 1,500 squared kilometres (Michaels et al., 2012, 551).
The Spanish population increased significantly during our period of study: from
around 15.6 million people in 1860, to 18,5 in 1900, 23.6 in 1930, 30.4 in 1960, reaching
forty million in the 1990s. Its spatial concentration also underwent major changes
throughout this period. Map 1 depicts how district population evolved between 1860 and
1991. Generally speaking, districts grew in size, yet unevenly, showing a tendency to
concentrate along the coast and around Madrid, the capital city.
22 The spatial unit chosen to study agglomeration economies is still an open issue. Theoretically, it depends on
the type of agglomeration effect but, in practice, the effect of choosing different spatial units is not that
important (Briant et al., 2010; Combes and Gobillon, 2015, 294).
23 The Modifiable Areal Unit Problem (MAUP) refers to the sensitivity of results to the zoning system
employed in the empirical analysis. Figure A.1 in the appendix shows the surface area distribution of districts.
12
Map 1. Population in Spain by district, 1860-2000
Source: Population censuses
Figure 4 illustrates structural change, proxied with the share of employment in
industry and services, and the average and median size of districts by year. In the early
stages, there appears to be a steady, though timid, increase in the districts population. This
is expected given that structural change was rather modest. Yet, from 1920 to 1970, the
Spanish population rapidly concentrated. The mechanisation of agriculture and the spread
13
of industrialisation, especially since the 1950s, fuelled rural-urban migration24. Districts,
where modern industries located, grew rapidly. Backward and agrarian ones shrunk. This
diverging pattern between the average and median sizes continued, but at a slower pace, in
the late decades of the twentieth century, thus coinciding with the growth of the service
sector and the rise of information and communication technologies (ICT).
.6
Figure 4. Structural change and spatial concentration of population in Spain, 1860-1991
80
Services
Industry
Mean size (Popula4on district)
60
50
.1
30
.2
40
District popula4on size ('000)
Employment share in Industry and Services
.3
.4
70
.5
Median size (Popula4on district)
1860
1890
Civil War
1936-39
1920
1950
1980
Source: Population censuses
Agglomeration economies take place in densely populated areas. Therefore, the
evolution of the spatial concentration of the population can also be examined focusing on
the relationship between the initial level of the district population size and subsequent
population growth. Figure 5 fits kernel regressions showing the link between these two
variables in each period (approximately 10-year intervals). Broadly speaking, while the
second half of the 19th century appears to be characterised by quasi-orthogonal growth, the
early 20th century witnessed how large districts tended to grow faster than small ones, a
feature that intensified during the second half. Interestingly, the positive link between initial
population and subsequent growth reverses for the largest locations from 1970 onwards,
24 As mentioned in the previous section, emigration abroad was also important during the 1950s and 1960s. Although of
smaller magnitude, international emigration flows, especially to Latin America, occurred between 1880 and 1920
(Sánchez-Alonso, 2000).
14
suggesting that congestion costs began to exert a significant effect during this later period.
Next section examines these issues in more detail with the aim of quantifying how the
effect of agglomeration economies evolved over time.
Figure 5. District population growth vs. initial population
1877-87
1887-1900
1900-10
1910-20
1920-30
1930-40
1940-50
1950-60
1960-70
1970-81
1981-91
-.05
0
.05
.1
.15
-.05
0
.05
.1
.15
-.05
0
.05
.1
.15
1860-77
8
10
12
14
16
8
10
12
14
16
8
10
12
14
16
8
10
12
14
16
Source: Population censuses; each period roughly corresponds to (approximately) 10-year intervals.
4. Empirical Specification
In order to examine how the size of the local economy affects subsequent growth over
time, we first estimate the following model for each period using OLS:
Δyi = α + β yi + x’i γ + εi
(1)
where Δyi is the population growth rate of each district between two censuses (Δyi = ln yt+1i –
ln yti) while yi refers to the log of the population level at the beginning of each period. Given
that both variables are measured in logs, the estimated parameters can be interpreted as
elasticities. In addition, x’i is a vector of control variables taking into account geographic,
climatic and geological features of each district.
15
On the one hand, given that we attempt to isolate the effect of agglomeration
economies from other determinants of population growth, we have included different
geographic characteristics that capture the locational fundamentals of each district. Firstly,
using the WorldClim 1 kilometre digital data, we have computed average annual
temperature and average annual rainfall. Secondly, the SRTM 90-meter resolution digital
elevation data allows obtaining a measure of the median altitude of each district, as well as a
ruggedness index that measures the standard deviation of altitude. Thirdly, drawing on the
European Environment Agency WISE Large rivers dataset, a dummy variable takes the
value of 1 if a district has access to a large river. Similarly, we have also computed distance
to the coast. Fourthly, to further proxy for the potential agricultural productivity, we have
relied on certain soil quality parameters provided by the European Soil Database (ESDB 1kilometer resolution): top soil available water capacity, base saturation of the top soil,
volume of stones and top soil organic content. Following Combes et al. (2010), we have
computed which is the most common category in each district and then assigned the
corresponding dummy variables to control for that. Given the large heterogeneity of the
districts’ geographic size, the specification also controls for district area. Moreover, given
the central location of Madrid, the country’s capital, a dummy variable has also been
created to account for this25.
On the other hand, the growth of a particular district not only depends on its own
economic dynamism, but also on that of competing neighbouring population, so our model
incorporates the existence of important neighbouring locations using GIS techniques26.
More specifically, we have computed for each period the population living in towns larger
than 10,000 inhabitants within a certain radius from the district centroid: less than 50, 50100, 100-250 and 250-500 kilometres respectively. Lastly, to further capture potentially
unobserved characteristics, we include regional dummies27. Tables A1 and A2 in the
Appendix explain how the variables employed have been obtained and report summary
statistics.
Figure 6 visually reports the results of regressing population growth on initial
population, as well as the full set of controls, for each period (see table A3 in the Appendix
25 In order to capture the potential capital status effect at the province level, we have also added to our controls a set of
dummy variables for the districts where the province capitals are located. Although the results become less precise
because initial size and capital status are highly correlated, they remain qualitatively unchanged and are available upon
request.
26 Rappaport and Sachs (2003) and Desmet and Fafchamps (2005) follow a similar approach.
27 We followed the classification employed by Simpson (1996), which divides the country into ten macro regions: Galicia,
Biscay, Castile-Leon, Upper Ebro, Lower Ebro, Levant, Centre, Extremadura, Eastern Andalusia and Western Andalusia.
16
for the actual results). Standard errors are clustered at the provincial level to take into
account that same-province districts may share unobserved characteristics. The estimated
coefficients evidence that the effect of size was negligible in the second half of the 19th
century and the first decade of the 20th century28. Centripetal forces only started to
consistently induce population concentration after 1910. The estimated coefficients kept
growing during the first third of the 20th century, from 0.055 in the 1910s to 0.077 in the
1930s. The Civil War and its aftermath, however, induced a set back on the intensity of
agglomeration economies which were reduced to figures around 0.040. While the 1950s
witnessed how the estimated coefficients returned to the levels existent prior to the war,
the 1960s experienced a major boost that situated them around 0.156. Although the 1970s
still enjoyed relatively high figures (around 0.131), the effect of initial size on subsequent
population growth was significantly decreased in the 1980s (0.056).
Figure 6. Evolution of agglomeration effects (OLS)
The previous model, however, may suffer from endogeneity problems. It is plausible
that larger locations are the result of some local characteristics, so their growth may not be
the result of agglomeration economies but of some underlying feature, such as better
agricultural potential or the presence of certain amenities (administrative or transportation
28 Although quite low, the estimated coefficient is statistically significant for the period 1887-1900. The end-of-thecentury crisis, however, temporarily suppressed these incipient developments.
17
infrastructures), that promotes their future growth. Although we have included a
comprehensive set of variables that attempt to control for this issue, we have further
attempted to mitigate this concern by instrumenting the size of the local economy using
historical urban population. In particular, we employ population living in cities larger than
5,000 individuals in 150029. By doing so, we exploit the long-term persistence of the spatial
distribution of population due to the inertias that local population and economic activity
generate. Given the long lag employed, this instrument is plausibly exogenous because the
modern sources of local productivity differ from those existing in such a distant past.
However, there might be local characteristics that affected population growth in the
past and still continue to influence it in more recent times: suitable agro-climatic
conditions, the presence of a large river or another geographical feature that increases the
location’s potential such as the centrality of the location in the country or having access to
the sea (Combes and Gobillon, 2015, 287). This is especially important because we start
measuring agglomeration economies in the mid-19th century when structural change
remained limited and agriculture was still a significant source of local wealth. Crucially, as
discussed above, our model directly controls for other geographical, climate and geological
variables that may have still influenced each district potential productivity during the period
under study.
As the first stage evidences, our instrument is highly correlated with the instrumented
variables (see Table A.4 in the Appendix)30. Fig. 7 compares the IV results to those
obtained using OLS. Broadly speaking, both sets of estimated coefficients depict similar
trends, the IV results tend to be higher, especially between 1910 and the Civil War and
between 1960 and 1980, thus reinforcing the image of rapid structural change and the
increasing spatial concentration of the population that these two periods witnessed. By the
1930s, when initial size was twice as large, population growth was around 16.1 per cent
higher. The magnitude of agglomeration economies was even higher during the 1960s. In
comparison, these forces paled during both the aftermath of the Civil War and the 1980s.
29 The historical population is taken from Bosker et al. (2013). Using the population figures for the years 1600, 1700 or
1800 does not alter the results reported here.
30 Also, the Sanderson-Windmeijer (SW) test always rejects that the endogenous regressor is unidentified.
18
Figure 7. Evolution of agglomeration effects (OLS vs IV)
5. Discussion
Our results are consistent with Krugman’s (1991) model: agglomeration economies
strengthen as transport costs are reduced over time. The structural transformation away
from agriculture also helps explaining the increasing effect of agglomeration over time,
especially for intermediate densities. The increasing relationship between initial size and
subsequent population growth at intermediate U.S. locations was stronger from 1880 to
1960 than during the late 20th century (Michaels et al., 2012, 537). This pattern is arguably
related to reallocation away from agriculture, a process that was significantly more intense
during the former period. Michaels et al. (2012, 536) argue that, in more populated areas,
where non-agricultural activities already dominate, further population growth is not
necessarily correlated with initial population density. Taking into account the obvious
differences between the history of the U.S. and Spain, the evolution of the effect of initial
size on subsequent population growth is also similar. In Spain, the relationship between
these two variables when the largest districts are excluded increased significantly between
1910 and 1970, although this trend was abruptly interrupted by the Civil War and the
autarchic period. Using provinces as unit of analysis, Ayuda et al. (2010) also find that
increasing returns only started to play a role in the geographical concentration of the
Spanish population from 1900 onwards as the share of increasing-returns sectors in the
Spanish economy grew.
19
The relatively high coefficients found around 1960/70s are consistent with the
comparatively higher incidence of agglomeration economies found in developing countries
nowadays31. The intensity of the link between initial population and subsequent population
growth debilitated in the 1970s, a process that continued during the 1980s. Desmet and
Fafschamps (2005, 262) argue that these recent developments can be attached to deindustrialisation. While services, mostly a non-tradable activity, had traditionally been
spread out, declining transportation and communication costs have recently facilitated
services to concentrate. In contrast, the same processes have weakened the benefits from
agglomeration in manufacturing, thus promoting their geographical dispersion32. The
depression that followed the 1970s crisis indeed meant that the service and construction
sectors replaced industrial production as the main engines of economic growth in Spain
(Bentolilla and Blanchard, 1990; Bentolilla and Dolado, 1991). Similarly, as land prices
increase, more land intensive activities, such as manufacturing, are replaced by less land
intensive activities, such as services (Desmet and Fafschamps, 2005, 262). Consistent with
our results, Paluzie et al. (2004) show that the spatial distribution of manufacturing in Spain
presented a bell-shaped evolution, with an initial phase characterised by a significant
increase in industrial concentration followed by a trend reversal since the 1970s when a
growing dispersion of industry is observed33. In this regard, Paluzie et al. (2009b) show that
the geography and intensity of internal migrations mirrored the patterns of industrial
concentration.
Increasing congestion costs, arising from rising housing prices and the higher weight
of amenities and other aspects related to the quality of life (Bover and Velilla, 1999;
Bentolila, 2001), may also explain why the coefficients on initial population got smaller
during the 70s and the 80s. Focusing on the evolution of the largest cities, Lanaspa et al.
(2003) find that while differences in city size got amplified between 1900 and 1970, small
and intermediate cities grew faster than large ones from that moment onwards34. Due to
congestion and pollution costs, agglomeration economies are subject to diminishing
returns, so we now explore how agglomeration economies depend on district size. The size
of the coefficients we have estimated reflect the total net impact of the concentration of
The coefficients are 0.10-0.12 and 0.09-0.12 for China and India, respectively (Combes et al., 2013; Chauvin et al.,
2014).
32 In addition, “the splitting up of the production process into different stages has allowed manufacturing firms to
relocate certain activities to less dense areas” (262). On these issues, see also Desmet and Rossi-Hansberg (2009).
33 Similarly, Martinez-Galarraga et al. (2008) find that, although employment density is related with inter-regional
differences in industrial labour productivity during the early stages of industrialisation, this association becomes less
important over time and actually disappears in recent decades (1985-1999).
34
See also González-Val et al. (2014) where the authors test the validity of Gibrat’s Law for the complete set of Spanish
municipalities during the 20th century. Additionally, García-López et al. (2015) show that the extensive construction of
highways after the late 1960s, and especially from the 1980s onwards, fostered suburbanization in Spain.
31
20
economic activity. Given that congestion costs increase as population density grows, it is
interesting to examine their importance by estimating the same model but excluding the
largest districts.
Figure 8 compares the 2SLS results of estimating equation 1 using the whole sample
to those of replicating the same exercise but sequentially excluding the largest locations:
those with a population above one million, 500,000 and 250,000 inhabitants respectively
(see also Table A5 in the Appendix). This exercise evidences that our results are not just
driven by the upper tail of the distribution. The estimated coefficients of initial size on
subsequent population growth actually remain qualitatively unchanged up to the 1930s,
when they slightly increase, once the largest districts are excluded from the sample.
Furthermore, once the disruptions caused by the Civil War and its aftermath waned, this
occurs again, though the difference between the coefficients is only statistically significant
in the 1980s. These results could be driven by the dynamism of middle-sized districts and
the presence of congestion costs in the largest districts, especially during the more recent
decades. Yet, it is worth noting that, despite congestion costs, the gains from
agglomeration prevailed.
Figure 8. Evolution of agglomeration effects. IVs. Excluding largest districts
21
6. Conclusion
Agglomeration economies play a fundamental role in the location of economic activity.
However, the impact of these forces is likely to have varied over time as economies evolve.
We have shown how economic density only started to significantly influence the spatial
concentration of the Spanish population in the early 20th century, a process closely related
to the structural transformation of the economy. The effect of initial population on
subsequent population growth increased between 1910 and 1970, although this trend was
briefly interrupted by the Spanish Civil War and the autarkic period that followed. The
intensity of this relationship, however, receded in the 1970s, and especially during the
1980s, as rural out-migration slowed down and several traditional industries fell apart.
Furthermore, the largest locations did not benefit as much as medium-size ones from the
presence of agglomeration economies after the 1960s. This raises several questions, which
are not discussed in the paper, but it points to the fundamental trade-off between
increasing returns and congestion costs.
22
References cited:
Ayuda, M.I., Collantes, F., and Pinilla, V. (2010), ‘From locational fundamentals to
increasing returns: The spatial concentration of population in Spain, 1787-2000’,
Journal of Geographical Systems 12 (1): 25-50
Beeson, P.E., Dejong, D.N., and Troesken, W. (2001), ‘Population growth in US counties,
1840-1990’, Regional Science and Urban Economics 31: 669-699
Bentolila, S. (2001), ‘Las migraciones interiores en España’, FEDEA Working Papers 200107
Bentolila, S., and Blanchard, O. (1990), ‘Spanish unemployment’, Economic Policy 10: 233281
Bentolila, S., and Dolado, J.J. (1991), ‘Mismatch and internal migration in Spain, 19621986’, in F. Padoa-Schioppa (ed.), Mismatch and labour mobility, (Cambridge:
Cambridge University Press): 231-288
Bosker, M., Buringh, E., and Van Zanden, J.L. (2013), ‘From Baghdad to London:
Unravelling urban development in Europe and the Arab world, 800-1800’, Review of
Economics and Statistics 95 (4): 1418-1437
Bosker, M., and Buringh, E. (2015), ‘City seeds: Geography and the origins of the
European city system’, Journal of Urban Economics (In Press).
Bover, O., and Velilla, P. (1999), ‘Migrations in Spain: Historical background and current
trends’, Banco de España Working Paper 9909
Briant, A., Combes, P.P., and Lafourcade, M. (2010), ‘Dots to boxes: Do the size and shape
of spatial units jeopardize economic geography estimations?’, Journal of Urban
Economics 67 (3): 287-302
Chauvin, J.P., Glaeser, E., and Tobio, K. (2014), ‘Urban economics in the US and India’,
Harvard University
Ciccone, A., and Hall, R.E. (1996), ‘Productivity and the density of economic activity’,
American Economic Review 86 (1): 54-70
Clark, J.S., and Stabler, J.C. (1991), ‘Gibrat’s law and the growth of Canadian cities’, Urban
Studies 28 (4): 635-639
Combes, P.P., Duranton, G., Gobillon, L., and Roux, S. (2010), ‘Estimating agglomeration
economies with history, geology, and worker effects’, in E.L. Glaeser (ed.),
Agglomeration economies (Chicago: The University of Chicago Press): 15-65
23
Combes, P.P., and Gobillon, L. (2015), ‘The empirics of agglomeration economies’, in G.
Duranton, J.V. Henderson and W. Strange (eds.), Handbook of urban and regional
economics, vol. 5 (North-Holland: Elsevier): 247-348
Combes, P.P., De Murger, S., and Li, S. (2013), ‘Urbanisation and migration externalities in
China’, Discussion Paper 9352, Centre for Economic Policy Research
Combes, P.P, Mayer, T., and Thisse, J.F. (2008), Economic geography. The integration of regions
and nations (Princeton: Princeton University Press)
Combes, P.P., Duranton, G., and Gobillon, L. (2012), ‘The costs of agglomeration: Land
prices in French cities’, Discussion Paper 9240, Centre for Economic Policy Research
Cronon, W. (1991), Nature’s metropolis: Chicago and the great west (New York: W.W. Norton &
Company)
Davis, D.R., and Weinstein, D.E. (2002), ‘Bones, bombs, and break points: The geography
of economic activity’, American Economic Review 92 (5): 1269-1289
Desmet, K., and Fafchamps, M. (2005), ‘Changes in the spatial concentration of
employment across US counties: A sectoral analysis 1972-2000’, Journal of Economic
Geography 5 (3): 261-284
Desmet, K., and Rappaport, J. (2015), ‘The settlement of the United States, 1800-2000: The
long transition towards Gibrat’s law’, Journal of Urban Economics (In Press)
Desmet, K., and Rossi-Hansberg, E. (2009), ‘Spatial growth and industry age’, Journal of
Economic Theory 144: 2477–2502
Dobkins, L.H., and Ioannides, Y.Y. (2001), ‘Spatial interactions between U.S. cities: 19001990’, Regional Science and Urban Economics 31: 701-731
Duranton, G. (2007), ‘Urban evolutions: The fast, the slow, and the still’, American Economic
Review 97: 197–221
Duranton, G., and Puga, D. (2004), ‘Micro-foundations of urban agglomeration
economies’, in J.V. Henderson and J.F. Thisse (eds.), Handbook of Regional and Urban
Economics (Amsterdam: North-Holland): 2063-2117
Eeckhout, J. (2004), ‘Gibrat’s law for (all) cities’, American Economic Review 94 (5): 1429-1451
Fujita, M., Krugman, P., and Venables, A.J. (1999), The spatial economy: Cities, regions and
international trade (Cambridge: MIT Press)
Fujita, M., and Thisse, J.F. (2002), Economics of agglomeration: Cities, industrial location, and
regional growth (Cambridge: Cambridge University Press)
Gabaix, X. (1999), ‘Zipf’s law for cities: An explanation’, Quarterly Journal of Economics 114
(3): 739-767
24
Gabaix, X., and Ioannides, Y. (2004), ‘The evolution of city size distributions’, in J.V.
Henderson and J.F. Thisse (eds.), Handbook of Regional and Urban Economics, vol. 4
(North-Holland: Elsevier): 2341-2378
García-Coll, A., and Stillwell, J. (1999), ‘Inter-provincial migration in Spain: Temporal
trends and age-specific patterns’, International Journal of Population Geography 5: 97-115.
García-López, M.A., Holl, A., and Viladecans-Marsal, E. (2015), ‘Suburbanization and
highways in Spain when the Romans and the Bourbons still shape its cities’, Journal of
Urban Economics 85: 52-67
Glaeser, E.L., Scheinkman, J., and Shleifer, A. (1995), ‘Economic growth in a cross-section
of cities’, Journal of Monetary Economics 36: 117-143
Glaeser, E.L. (2010), ‘Introduction’, in E.L. Glaeser (ed.), Agglomeration economics (Chicago:
University of Chicago): 1-14
Goerlich, F.J., Mas, M., Azagra, J., and Chorén, P. (2006), La localización de la población sobre el
territorio. Un siglo de cambios: un estudio basado en series homogéneas (1900-2001) (Bilbao:
Fundación BBVA)
Goerlich, F.J., and Mas, M. (2008), ‘Algunas pautas de localización de la población española
a lo largo del siglo XX’, Investigaciones Regionales 12: 5-33
Goerlich, F.J., and Mas, M. (2009), ‘Drivers of agglomeration: Geography vs history’, The
Open Urban Studies Journal 2: 28-42
Gómez Mendoza, A., and San Román, E. (2005), ‘Transportes y comunicaciones’, in A.
Carreras and X. Tafunell (eds.), Estadísticas históricas de España, siglos XIX-XX (Bilbao:
Fundación BBVA): 509-572
González-Val, R., Lanaspa, L., and Sanz-Gracia, F. (2014), ‘New evidence on Gibrat’s
Law for cities’, Urban Studies 51 (1): 93-115
González-Val, R., Tirado, D.A., and Viladecans-Marsal, E. (2015), ‘Market potential
and city growth: Spain 1860–1960’, Cliometrica (online first)
Graham, D.J. (2007), ‘Variable returns to agglomeration and the effect of road traffic
congestion’, Journal of Urban Economics 62: 103–120
Henderson, J.V. (2003), ‘Marshalls’ scale economies’, Journal of Urban Economics 53 (1): 1-28
Herranz, A. (2005), ‘La reducción de los costes de transporte en España (1800-1936)’,
Cuadernos Económicos ICE 70: 183-203
Herranz, A. (2007), ‘Infrastructure investment and Spanish economic growth, 1850-1935’,
Explorations in Economic History 44: 452-468
Jacobs, J. (1969), The economy of cities (New York: Vintage)
25
Kim, S. (1995), ‘Expansion of markets and the geographic distribution of economic
activities: The trends in US regional manufacturing structure, 1860-1987’, Quarterly
Journal of Economics 110 (4): 881-908.
Krugman, P. (1991), ‘Increasing returns and economic geography’, Journal of Political
Economy 99 (3): 483-499
Krugman, P., and Venables, A.J. (1995), ‘Globalization and the inequality of nations’,
Quarterly Journal of Economics 110 (4): 857-880
Lanaspa, L., Pueyo, F., and Sanz, F. (2003), ‘The evolution of Spanish urban structure
during the twentieth century’, Urban Studies 40 (3): 567-580
Marshall, A. (1890), Principles of economics (London: Macmillan)
Martinez-Galarraga, J. (2012), ‘The determinants of industrial location in Spain, 1856-1929’,
Explorations in Economics History 49: 255-275
Martinez-Galarraga, J., Paluzie, E., Pons, J., and Tirado, D.A. (2008), ‘Agglomeration and
labour productivity in Spain over the long term’, Cliometrica 2 (3): 195-212
Melo, P.C., Graham, D.J., and Nolan, R.B. (2009), ‘A meta-analysis of estimates or urban
agglomeration economies’, Regional Science and Urban Economics 39: 332-342
Michaels, G., Rauch, F., and Redding, S.J. (2012), ‘Urbanization and structural
transformation’, Quarterly Journal of Economics 127 (2): 535-586
Nadal, J. (1987). ‘La industria fabril Española en 1900. Una aproximación’, in J. Nadal, A.
Carreras, and C. Sudrià (eds), La economía española en el siglo XX: Una perspectiva histórica
(Barcelona: Ariel): 23-61
Paluzie, E., Pons, J., and Tirado, D.A. (2004), ‘The geographical concentration of industry
across Spanish regions, 1856-1995’, Jahrbuch für Regionalwissenschaft 24: 143-160.
Paluzie, E., Pons, J., and Tirado, D.A. (2009a), ‘A test of the market potential equation in
Spain’, Applied Economics 41 (12): 1487-1493
Paluzie, E., Pons, J., Silvestre, J., and Tirado, D.A. (2009b), ‘Migration and market potential
in Spain over the twentieth century: A test of the new economic geography’, Spanish
Economic Review 11: 243-265
Partridge, M.D., Rickman, D.S., Ali, K., and Olfert, M.R. (2008), ‘Lost in space: Population
growth in the American hinterlands and small cities’, Journal of Economic Geography 8:
727-757
Peña, D., and Sánchez-Albornoz, N. (1984), ‘Wheat prices in Spain, 1857-1890: An
application of the Box-Jenkins methodology’, Journal of European Economic History 2:
353-373
26
Pons, J., and Tirado, D.A. (2008), ‘Los determinantes de la desigualdad económica regional
en España’, Información Comercial Española, ICE: Revista de Economía 842: 195-216
Prados de la Escosura, L. (2008), ‘Inequality, poverty and the Kuznets curve in Spain,
1850-2000’, European Review of Economic History 12 (3): 287-324
Prados de la Escosura, L., and Rosés, J.R. (2009), ‘The sources of long-run growth in Spain,
1850-2000’, The Journal of Economic History 69 (4): 1063-1091
Prados de la Escosura, L., Rosés, J.R., and Sanz-Villarroya, I. (2012), ‘Economic reforms
and growth in Franco’s Spain’, Revista de Historia Económica / Journal of Iberian and Latin
American Economic History 30 (1): 45-89
Puga, D. (1999), ‘The rise and fall of regional inequalities’, European Economic Review 43: 303334
Rappaport, J., and Sachs, J.D. (2003), ‘The United States as a coastal nation’, Journal of
Economic Growth 8 (1): 5-46
Ringrose, D.R. (1998), Spain and Europe, 1700-1900: The myth of the failure (Madrid: Instituto
Universitario Ortega y Gasset)
Ródenas, C. (1994), ‘Migraciones interregionales en España, 1960-1989’, Revista de Economía
Aplicada 2 (4): 5-36
Rosés, J.R., Martinez-Galarraga, J., and Tirado, D.A. (2010), ‘The upswing of regional
income inequality in Spain (1860-1930)’, Explorations in Economic History 47: 244-257
Rosés, J.R., and Sánchez-Alonso, B. (2004), ‘Regional wage convergence in Spain 18501930’, Explorations in Economic History 41: 404-425
Rossi-Hansberg, E., and Wright, M.L. (2007), ‘Urban structure and growth’, Review of
Economic Studies 74: 597–624
Sánchez-Alonso, B. (1995), Las causas de la emigración española, 1880-1930 (Madrid: Alianza
Editorial)
Sánchez-Alonso, B. (2000), ‘Those who left and those who stayed behind: Explaining
emigration from the regions of Spain, 1880-1914’, Journal of Economic History 60 (3):
730-755
Silvestre, J. (2002), ‘Las emigraciones interiores en España durante los siglos XIX y XX:
Una revisión bibliográfica’, AGER-Revista de Estudios sobre Despoblación y Desarrollo
Rural 2: 227-248
Silvestre, J. (2005), ‘Internal migrations in Spain, 1877-1930’, European Review of Economic
History 9 (2): 233-265
27
Silvestre, J. (2007), ‘Temporary internal migrations in Spain, 1860-1930’, Social Science History
31: 539-574
Simpson, J. (1996), Spanish agriculture: The long siesta: 1765-1965 (Cambridge: Cambridge
University Press)
Tedde, P. (1994), ‘Cambio institucional y cambio económico en la España del siglo XX’,
Revista de Historia Económica 3: 525-538
Tirado, D.A., Paluzie, E., and Pons, J. (2002), ‘Economic integration and industrial
location: The case of Spain before World War I’, Journal of Economic Geography 2: 343363
Tirado, D.A., and Martinez-Galarraga, J. (2008), ‘Una nova estimació retrospectiva del
VAB regional industrial. Espanya (1860-1930)’, Documents de Treball de la Facultat de
Ciències Econòmiques i Empresarials E08/192
Tirado, D.A., Pons, J., Paluzie, E., and Martinez-Galarraga, J. (2013), ‘Trade policy and
wage gradients: Evidence from a protectionist turn’, Cliometrica 7: 295-318
Venables, A.J. (1996), ‘Equilibrium locations of vertically linked industries’, International
Economic Review 37: 341-359
Wais, F. (1987), Historia de los ferrocarriles españoles (Madrid: Fundación de los Ferrocarriles
Españoles)
28
Appendix
300
200
0
100
Frequency of districts (N=464)
400
500
Figure A.1. Surface area distribution of districts
0
1000
2000
3000
Surface area (sq. km)
Source: European Soil Data base (ESDB). District area computed using ArcGIS.
29
4000
Table A.1. Description of the variables employed
District population. Municipalities have been grouped into Partidos Judiciales
using the 1860 boundaries. Data taken from the Population Censuses
(available at http://www.ine.es/intercensal/).
Sum of population living in cities larger than 10,000 inhabitants within 50,
Population living in
50-100, 100-250 and 250-500 kilometers from the district geographical
neighbouring cities
centre.
Annual average temperature taken from WorldClim 1 kilometre digital data
Temperature
(Hijmans et al., 2005) which can be found at http://www.worldclim.org/.
The climate information refers to the average during the period 1950-2000.
Average annual rainfall. As with temperature, this is taken from
Rainfall
WorldClim 1 kilometre digital data (Hijmans et al. 2005) which averages
pluviosity over the period 1950-2000.
Median altitude in each district using the SRTM 90-meter resolution digital
Altitude
elevation data (http://srtm.csi.cgiar.org).
Population
Ruggedness
Large rivers
Capital city
Standard deviation of altitude.
Dummy variable that takes the value of 1 if a district has access to a large
river, namely Rivers Ebro, Tagus, Duero, Guadiana and Guadalquivir.
Georeferenced data can be found at http://www.eea.europa.eu/data-andmaps/data/wise-large-rivers-and-large-lakes
Dummy variable that takes the value of 1 for the Madrid district, which
hosts the country’s capital.
Distance to coast
Distance from the district centroid to the nearest coastline.
Soil quality
A number of dummy variables have been generated relying on the
following features provided by the European Soil Database (ESDB 1kilometer resolution): Top soil available water capacity, Base saturation of
the top soil, Topsoil organic content, Volume of stones and Distance to
rock. After computing which is the most common category in each
district, dummy variables have been assigned to the corresponding
categories.
Raw
raster
data
can
be
found
at
http://esdac.jrc.ec.europa.eu/resource-type/european-soil-database-soilproperties
District area
District area (km2).
Note: The geographical data have been computed using ArcGIS.
30
Table A2. Summary statistics
Population (in thousands)
Annual population growth (%)
Pre-modern population, 1500 (in thousands)
Urban pop. within 50 km radius (000)
Urban pop. within 50-100 km radius (000)
Urban pop. within 100-250 km radius (000)
Urban pop. within 250-500 km radius (000)
Distance to coast (kms)
Temperature (°C)
Rainfall (mm)
Altitude (m)
Ruggedness
Area (km2)
Capital
Large rivers
Top soil average water capacity, d2
--, d3
Base saturation of the top soil, d2
--, d3
Top soil organic content, d2
--, d3
--, d4
Volume of stones, d2
--, d3
--, d4
Distance to rock, d2
--, d3
--, d4
Source: See table A1.
Obs.
6,032
5,568
6,032
6,032
6,032
6,032
6,032
6,032
6,032
6,032
6,032
6,032
6,032
6,032
6,032
6,032
6,032
6,032
6,032
6,032
6,032
6,032
6,032
6,032
6,032
6,032
6,032
6,032
31
Mean
52.9
0.18
1.8
192
453
1,981
4,936
103.2
13.7
642.1
571.1
178.4
1,075.2
0.00216
0.235
0.907
0.0216
0.0194
0.741
0.373
0.218
0
0.0603
0.213
0.205
0.224
0.330
0.0280
St. dev.
118.8
1.22
6.6
387
613
1,814
3,814
89.8
2.5
263.2
342.9
115.0
664.5
0.0464
0.424
0.290
0.145
0.138
0.438
0.484
0.413
0
0.238
0.410
0.404
0.417
0.470
0.165
Min.
2.0
-6.34
0
0
0
51
239
0.3
4.1
269
2
7
12.2
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
Max.
3,291.7
13.24
70
4,553
5,125
10,925
18,796
356.2
18.2
1,517
1,915
707
4,154.8
1
1
1
1
1
1
1
1
0
1
1
1
1
1
1
Table A3. Evolution of agglomeration economies. OLS results
Dependent variable: Population growth (ln yt+1 - ln yt)
1860
1877
1887
1900
1910
1920
1930
1940
1950
1960
1970
1981
(1)
(2)
(3)
(4)
(5)
(6)
(7)
(8)
(9)
(10)
(11)
(12)
Initial
pop.
0.013
(0.010)
0.003
(0.009)
0.020*
(0.008)
0.008
(0.007)
0.055**
(0.008)
0.068**
(0.011)
0.077**
(0.013)
0.040**
(0.005)
0.083**
(0.009)
0.156**
(0.023)
0.131**
(0.021)
0.056**
(0.010)
Obs.
464
464
464
464
464
464
464
464
464
464
464
464
R2
0.359
0.309
0.309
0.343
0.440
0.355
0.486
0.256
0.499
0.636
0.631
0.510
Note: Robust standard errors clustered at the provincial level in parentheses; ** p<0.01, * p<0.05. All specifications include the full set
of controls discussed in the text. Both the dependent and the independent variables are expressed in natural logs, so the coefficients can
be interpreted as elasticities.
32
Table A4. Evolution of agglomeration economies. IV results
Panel A:
Second
Stage
Dependent variable: Population growth (ln yt+1 - ln yt)
1860
1877
1887
1900
1910
1920
1930
1940
1950
1960
1970
1981
(1)
(2)
(3)
(4)
(5)
(6)
(7)
(8)
(9)
(10)
(11)
(12)
Initial pop.
0.030*
(0.014)
-0.007
(0.021)
0.037*
(0.014)
-0.004
(0.016)
0.064**
(0.017)
0.091**
(0.022)
0.161**
(0.028)
0.043**
(0.013)
0.074**
(0.019)
0.184**
(0.040)
0.144**
(0.026)
0.072**
(0.013)
R2
0.349
0.306
0.299
0.339
0.438
0.344
0.342
0.256
0.497
0.633
0.629
0.502
S-W F test
66.94
65.49
60.22
57.05
53.48
53.08
55.82
70.49
72.39
85.34
112.2
129.9
p-value
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
1860
1877
1887
1900
1950
1960
1970
1981
Panel B:
First Stage
Dependent variable: Initial population (ln)
Pre-modern
pop., 1500
1910
1920
1930
1940
(1)
(2)
(3)
(4)
(5)
(6)
(7)
(8)
(9)
(10)
(11)
(12)
0.192**
(0.023)
0.199**
(0.025)
0.195**
(0.025)
0.208**
(0.028)
0.207**
(0.028)
0.221**
(0.030)
0.240**
(0.032)
0.280**
(0.033)
0.292**
(0.034)
0.310**
(0.034)
0.369**
(0.035)
0.428**
(0.038)
Obs.
464
464
464
464
464
464
464
464
464
464
464
464
Note: Robust standard errors clustered at the provincial level in parentheses; ** p<0.01, * p<0.05. All specifications include the full set
of controls discussed in the text. Both the dependent and the independent variables are expressed in natural logs, so the coefficients can
be interpreted as elasticities.
33
Table A5. Evolution of agglomeration economies. IV results. Different samples
Panel A: Whole Sample
Dependent variable: Population growth (ln yt+1 - ln yt)
1860
1877
1887
1900
1910
1920
1930
1940
1950
1960
1970
(1)
(2)
(3)
(4)
(5)
(6)
(7)
(8)
(9)
(10)
(11)
(12)
0.030**
(0.014)
-0.007
(0.021)
0.037**
(0.014)
-0.004
(0.016)
0.064***
(0.017)
0.091***
(0.022)
0.161***
(0.028)
0.043***
(0.013)
0.074***
(0.019)
0.184***
(0.040)
0.144***
(0.026)
0.072***
(0.013)
464
464
464
464
464
464
464
464
464
464
464
464
R-squared
0.349
0.306
0.299
0.339
0.438
0.344
0.342
0.256
0.497
0.633
0.629
0.502
S-W F test
66.94
65.49
60.22
57.05
53.48
53.08
55.82
70.49
72.39
85.34
112.2
129.9
p-value
0,0000
0,0000
0,0000
0,0000
0,0000
0,0000
0,0000
0,0000
0,0000
0,0000
0,0000
0,0000
1981
Initial pop.
Observations
1981
Panel B: Excluding districts with population over 1,000,000 inhabitants
Dependent variable: Population growth (ln yt+1 - ln yt)
Initial pop.
Observations
1860
1877
1887
1900
1910
1920
1930
1940
1950
1960
1970
(1)
(2)
(3)
(4)
(5)
(6)
(7)
(8)
(9)
(10)
(11)
(12)
0.030**
(0.014)
-0.007
(0.021)
0.037**
(0.014)
-0.004
(0.016)
0.064***
(0.017)
0.091***
(0.022)
0.184***
(0.027)
0.044***
(0.015)
0.084***
(0.019)
0.214***
(0.034)
0.164***
(0.023)
0.083***
(0.011)
464
464
464
464
464
464
462
462
462
462
462
462
R-squared
0.349
0.306
0.299
0.339
0.437
0.344
0.326
0.240
0.499
0.652
0.653
0.516
S-W F test
67.14
65.64
60.28
57.14
53.54
53.16
67.50
82.10
86.02
100.2
117.5
123.6
p-value
0,0000
0,0000
0,0000
0,0000
0,0000
0,0000
0,0000
0,0000
0,0000
0,0000
0,0000
0,0000
1981
Panel C: Excluding districts with population over 500,000 inhabitants
Dependent variable: Population growth (ln yt+1 - ln yt)
Initial pop.
Observations
R-squared
1860
1877
1887
1900
1910
1920
1930
1940
1950
1960
1970
(1)
(2)
(3)
(4)
(5)
(6)
(7)
(8)
(9)
(10)
(11)
(12)
0.030**
(0.014)
-0.007
(0.021)
0.037**
(0.014)
-0.008
(0.018)
0.065***
(0.019)
0.093***
(0.025)
0.184***
(0.027)
0.044***
(0.015)
0.101***
(0.017)
0.237***
(0.035)
0.194***
(0.021)
0.105***
(0.012)
464
464
464
462
462
462
462
462
461
459
457
454
0.349
0.306
0.299
0.337
0.422
0.316
0.326
0.240
0.522
0.647
0.658
0.518
S-W F test
67.14
65.64
60.28
79.67
73.53
67.87
67.50
82.10
86.42
73.25
95.15
78.69
p-value
0,0000
0,0000
0,0000
0,0000
0,0000
0,0000
0,0000
0,0000
0,0000
0,0000
0,0000
0,0000
Panel D: Excluding districts with population over 250,000 inhabitants
Dependent variable: Population growth (ln yt+1 - ln yt)
Initial pop.
Observations
1860
1877
1887
1900
1910
1920
1930
1940
1950
1960
1970
1981
(1)
(2)
(3)
(4)
(5)
(6)
(7)
(8)
(9)
(10)
(11)
(12)
0.022
(0.014)
-0.013
(0.023)
0.030**
(0.015)
-0.008
(0.018)
0.070***
(0.020)
0.097***
(0.027)
0.205***
(0.032)
0.049**
(0.020)
0.122***
(0.021)
0.285***
(0.028)
0.232***
(0.025)
0.133***
(0.016)
462
462
462
462
461
459
459
457
456
454
450
441
R-squared
0.347
0.301
0.295
0.337
0.421
0.299
0.269
0.222
0.509
0.612
0.630
0.470
S-W F test
77.83
82
69.08
79.67
79.53
47.80
47.30
57.18
57.49
63.81
48.70
42.69
p-value
0,0000 0,0000 0,0000 0,0000
0,0000
0,0000
0,0000
0,0000
0,0000
0,0000
0,0000
0,0000
Robust standard errors clustered at the provincial level in parentheses; *** p<0.01, ** p<0.05, * p<0.1; All specifications include the full
set of controls discussed in the text.
34