BIG PUSH OR BIG GRAB? RAILWAYS, GOVERNMENT ACTIVISM AND EXPORT GROWTH IN LATIN AMERICA, Vincent Bignon, Rui Esteves, Alfonso Herranz-Loncán - PDF

Description
BIG PUSH OR BIG GRAB? RAILWAYS, GOVERNMENT ACTIVISM AND EXPORT GROWTH IN LATIN AMERICA, Vincent Bignon, Rui Esteves, Alfonso Herranz-Loncán Abstract Railways were one of the main growth engines

Please download to get full document.

View again

of 19
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Information
Category:

Taxes & Accounting

Publish on:

Views: 17 | Pages: 19

Extension: PDF | Download: 0

Share
Transcript
BIG PUSH OR BIG GRAB? RAILWAYS, GOVERNMENT ACTIVISM AND EXPORT GROWTH IN LATIN AMERICA, Vincent Bignon, Rui Esteves, Alfonso Herranz-Loncán Abstract Railways were one of the main growth engines of Latin American economies during the first globalization boom. At the time economic growth largely depended on export expansion and this, in turn, was closely associated to railway construction. However, railway investment required consistent government support in the form of subsidies, which were highly dependent on government revenues. As government revenues were mostly driven by tariffs on trade, this led to a potential multiple equilibria system. Some Latin American countries seem to have got trapped in a low level equilibrium in which the economy and government revenues did not grow enough due to insufficient railway development, and railway construction was stunted by the scarcity of public resources. The aim of this paper is to test the existence of the bidirectional causality link between government revenues and railway development. Our findings support the hypothesis of a positive two-way relationship between exports and government revenues, on the one hand, and of revenues and railway development, on the other. In an export-led growth context, this positive two-way relationship is consistent with a big push theory of Latin American growth before Our results are also informative of the growing divergence between economies of the region during this period. KEYWORDS Railways; Latin America; export-led growth CONTACT DETAILS Vincent Bignon, Banque de France Rui Esteves, University of Oxford Alfonso Herranz-Loncán, University of Barcelona Big Push or Big Grab? Railways, Government Activism and Export Growth in Latin America, Vincent Bignon 1 Rui Esteves 2 Alfonso Herranz-Loncán 3 1. Introduction State activism has often been identified as either a major force or a necessary condition for economic growth in peripheral countries (Rosenstein-Rodan, 1943, 1961; Rostow, 1960; Murphy et al., 1989). This paper focuses on the role of states in the growth of Latin American economies during the first globalisation boom through their effects on railway expansion. In the half century before the Great War, Latin America as a whole experienced one of the fastest rates of economic growth in the world. According to Maddison s (2001) figures, the economies of the area grew well above the world average between 1870 and 1913, and their growth rate was comparable to that of the Western Offshoots. To a large extent, this growth episode was a consequence of the expansion of exports of primary products during the first globalisation. Growth, however, was not equally shared by all Latin American economies. Although GDP per capita figures are only available for a handful of countries, and their reliability has often been questioned, they show a wide diversity of growth experiences. 4 Moreover, exports per capita, for which information is much more abundant, show even larger variation among countries throughout the period, both in levels and in growth rates. 5 1 Banque de France, DGEI-DEMFI-Pomone ; 2 Department of Economics, University of Oxford; 3 Department of Economic History and Institutions, University of Barcelona; We are grateful to Andrés Álvarez, Leticia Arroyo-Abad, Dan Bogart, Theresa Gutberlet, Silvana Harriet, Nuno Palma, Florian Ploeckl, Maria-Teresa Ramírez, Jaime Reis, Fabbio Sanches, Richard Sicotte, Bill Summerhill, and participants at the ALL-UC Group Conference on Transport, Institutions, and Economic Performance: Historical Perspectives (UC-Irvine, December 2011), the XVI World Economic History Congress ( Stellenbosch, July 2012), the seminar of the Colombian Economic History Association at Los Andes University in Bogotá (October 2012) and the APHES conference in Lisbon (November 2012). We also thank Juan H. Flores and Béatrice Dedinger for sharing unpublished data with us. None of them are responsible for the mistakes of the text. The views expressed in this paper are not necessarily those of the Banque de France or the Eurosystem. 4 The yearly growth rates of the available Latin American GDP per capita estimates between 1870 and 1913 range from 0.2% (Brazil) to 2.3% (Argentina); data from Bértola and Ocampo (2010). 5 In , exports p.c. in the region, in thousand dollars of 1990, ranged from 17.7 (Ecuador) to (Cuba). The yearly growth rate of exports p.c. between 1870 and 1913 went from -1.3% in Brazil to 5.8 in Dominican Republic; see Bértola and Ocampo (2010: 98). 1 Before 1914 differences in exports and income per capita among Latin American economies were closely linked to differences in railway development. The potential capacity of railways to transform the economies they served was much higher in Latin America than in the Western European countries, where the railway technology had been developed, for two reasons. Firstly, due to the almost total absence of alternative transport infrastructure and the scarcity of waterways in the region by the midnineteenth century, railways constituted the only available means to connect the hinterland of those countries with the international markets. Secondly, in the first globalization wave, the growth of Latin American economies was mainly based on exports of natural resources. Consequently, the growth potential of the region crucially depended on the spread of a good transport system throughout the largest possible portion of the territory. Indeed, exports only seem to have taken off in those countries which undertook substantial investment in railways. The direct link between railways and economic expansion in the region allowed Summerhill (2006: 297) to suggest that it seems unlikely that any other technological or institutional innovation was more important in the transition to economic growth in Latin America before However, despite its high potential growth impact, railway investment did not spread evenly throughout the region. In many Latin American countries the development of rail transport was sluggish, the final network mileage was disappointingly low and so the opportunity to take full advantage from the new technology might have been missed. Reasons behind that failure were diverse, but the literature on the topic has often stressed the essential role of governments involvement in railway development. However, owing to the large dependence of Latin American governments on foreign trade taxes, public activism was limited in each country by the degree of external trade success. But, in turn, the growth of exports required, to a large extent, the diffusion of rail transport. In other words, some Latin American economies before 1914 might have been trapped in a double feedback situation wherein railway construction depended on government revenues and these depended on export growth but, at the same time, exports could only increase if the railway network expanded. This multiple equilibria situation created the possibility for some Latin American countries becoming stuck in a non-development trap in which the economy (and government revenues) did not grow enough due to insufficient railway development, but railways did not develop enough either, due to the low level of exports and the subsequent scarcity of public resources. 2 The objective of this paper is to test the existence of such multiple equilibria by analysing the bidirectional causality between government revenues and railway development over the period. Our estimates of a two equation model of government revenues and railway expansion for the Latin American countries between 1865 and 1913 support the existence of a positive two-way relationship between both variables. This is robust to several specifications and is consistent with the division of the region between a group of economies that stagnated and another group of countries with significant economic dynamism. Even though we do not explore here the reasons for why each country fell in one category or the other, we suggest that a combination of structural features and short-term shocks may help to understand differences among economies. Our analysis may therefore be helpful to understand the different ability of the Latin American economies to take full advantage of the opportunities open by the globalisation, and may contribute to explain the significant process of divergence that seems to have taken place within the region before Infrastructure development and the state Our paper touches upon three strands of literature linked to the need for government intervention in triggering infrastructure development. The first one discusses the case for government intervention in the provision of infrastructure or any (other) public good. At least since Walras (1897), economists think of projects exhibiting substantial externalities as a case for state intervention, either to organize or to manage the provision of these goods under a monopoly. Coase (1974) challenged the conventional wisdom in his study of lighthouse provision in England during the 19 th century. He showed that private individuals provided lighthouse services and made profit out of them. This can be seen as an application of the Coase theorem which states that agents can find a way of distributing the gains from trade even when there are externalities. Discussions of Coase s historical example include Van Zandt (1993), Taylor (2001), Bertrand (2006) and Barnett and Block (2007). These authors concurred in showing that the operation of lighthouses also required British taxpayers money, forming a kind of public-private partnership similar to the Latin American railways considered here. Another strand of literature starts from the assumption that government involvement may be needed to coordinate the actions and investments of private agents. Rosenstein- Rodan (1943) pioneered this idea now labeled as the Big Push hypothesis. In the context of the post-wwii period, Rosenstein-Rodan proposed to use government 3 policies to foster growth in some European countries ravaged by the war. Anticipating a shortage of financial capital after the war, he envisaged the specific policies to support industrialization using foreign loans and technical assistance. Murphy et al. (1989) returned to Rosenstein-Rodan s thesis by showing that indivisibilities in the production function were sufficient to render government intervention necessary to coordinate private investment decisions. Coordination failures could be overcome through a Big Push that brought an economy from a low to a high level equilibrium. Infrastructure in general and railways in particular have often been considered as an example of the Big Push (Rostow, 1960; Rosenstein-Rodan, 1961; Murphy et al., 1989). According to Murphy et al. (1989: 1022), one of the reasons why a railroad might not get built, even when it is socially efficient to build it, is that there may remain extrinsic uncertainty over whether or not the potential users of the railroad ( ) become actual users. Several qualifications to the Big Push hypothesis have recently been proposed. Berkovitz and Li (2000) insist that the success of a Big Push is conditional on the proper organization of the tax system and, notably, on the coordination of the public authorities on the power to tax. Essentially, a disaggregated situation with many and possibly competing or unclear layers of authority with the power to tax investment is inferior to a concentrated tax authority. Fontenay (2004) complements this proposal by noting that the success of a Big Push policy may be affected by the market power of the entrepreneur chosen by the government to implement the Big Push. Because entrepreneurs can choose to hold up other investments by charging high prices for their services, the author argues that the success of a Big Push policy directly depended on the strength of institutions particularly on their ability to enforce the entrepreneur s commitment. Bjorvatn and Coniglio (2012), however, show that even if governments are less efficient than market entrepreneurs, a Big Push policy can still promote development. This echoes Rodrik s (1995) argument that governments were helpful in removing coordination failures in investment in Taiwan and South Korea. In Sachs s (2005) version of the hypothesis, the Big Push was reframed as a discussion about the need for significant foreign capital to allow countries escaping a poverty trap. The debate focused on foreign aid (gifts) rather than foreign loans and was challenged by Easterly (2006) who emphasized poor governance over financial constraints as a greater source of entrenched poverty. In all the previous papers, the rationale for government intervention is the inability of the market to coordinate the actions of agents. Indivisibilities in demand, savings or 4 production functions provide authorities with the possibility to implement Paretoimproving policies. However, another case for government intervention in infrastructure development may be connected to the political equilibrium of countries. In this regard, the third strand of literature this paper relates to discusses the political economy of government intervention in the economy. Olson (2000) argued that the choice of the ruler to engage in public goods provision or in extorting activities depended on its time horizon. Dalgic and Van Long (2006) show that this can generate a situation of multiple equilibria, one in which the country is locked in a poverty trap and another unstable in which the country grows without bound. They also show that transparency of the decision process and monitoring by voters make governments more willing to engage in helping policies. Politics may also affect the choice of implementing or not a policy because of the redistribution effects of infrastructure construction. Many noticed that the construction of railways involved significant changes in the price of the surrounding land and in the degree of integration of labor and goods markets. This generates substantial changes in the net gains from trade between regions and industries. It also destroys some rents by making markets more competitive. These redistributive consequences would generate resistance and induce the losers to lobby against or block the implementation of new infrastructure projects. This would be all the more significant in developing countries characterized, as argued by North Wallis and Weingast (2009), by limited access orders in which politics consists in distributing rents to ensure cooperation. Consequently, it is likely that the decision to build infrastructure was dependent on the state redistributing part of the gains to the losers in exchange for their approval. This would then require the state to build up sufficiently large streams of revenues to pay for those compensations. Regardless of the mechanism justifying it public good provision, coordination or redistribution, these different strands of literature have highlighted the importance of the interrelation between infrastructure development and the size of government revenues. In this paper we attempt to measure empirically this relationship in the case of Latin American railways before Moreover, although we cannot differentiate between the relative influence of those three different mechanisms, the available historical evidence suggests that all of them might have been important in the region over the period under study. 5 3. Railway expansion in Latin America before 1914 During the first globalisation boom, railways were a key factor for trade expansion and economic growth in the Latin American economies. Due to the lack of infrastructure and the scarcity of waterways in most of the region, domestic transport costs before the railway were too high to allow a sustained and rapid growth of exports, except for the coastal regions or the products with very high value-to-weight ratios, such as gold or silver. Unlike the US or some European countries, in which the abundance of waterways substantially reduced the cost advantage of the railways, Latin American economies had to rely, with a few exceptions, on overland transport for domestic trade. Only a few feasible navigable routes were available, such as the Amazonas in Brazil, the Magdalena in Colombia or the River Plate system in Uruguay and North-East Argentina, and even there conditions for navigation were not always favourable, as has been stressed in the case of the Magdalena. 6 Latin American pre-railway overland transport was very precarious. Most roads were not accessible for carts, and a huge share of freight transport depended exclusively on pack animals. 7 The primitive character of Latin American road transport can be illustrated by the ratio between the average unit price of pre-railway (largely road) freight transport and the average rates of railway freight. Whereas in England and Wales by 1865, France by 1872 or the US by 1859 this ratio ranged from 2.6 to 3.3, it reached levels of 5 to 13 in Mexico, Brazil and Argentina in the early 20 th century. Only in Uruguay, due to the exceptional natural advantages of the country, was the level of that ratio comparable to the situation in Western Europe (3.7). 8 Under these circumstances, railways became essential for allowing the exploitation of natural resources in the inner areas of each country, and to make long-term export expansion possible. A preliminary illustration of the close association between railway development and export growth is provided in Figure 1, which shows that those 6 Traffic through the Magdalena was slow and highly dependent on weather conditions, and the river was only partially navigable; see Ramírez (2001: 88) or Safford (2010: ). According to Prados de la Escosura (2009: 291), by the early 1840s Colombia had the highest level of transport costs from the port to the capital city (which was placed by the river) in a sample of 9 economies that included countries with harsh relief and inland capitals, such as Bolivia and Mexico. 7 On the state of pre-railway transport infrastructure in Spanish America see Gutiérrez Álvarez (1993), as well as Summerhill (2003:18-33) for Brazil, for Ecuador, Clark (1995: 26-36), for Mexico, Coatsworth (1981: 17-26) and, for Perú, Zegarra (2011: ). Actually, in some cases, such as Mexico, the transport system experienced an involution between the end of the colonial period and the 1870s, with a decrease in wheeled traffic and an increase in the use of pack animals (Riguzzi, 1996: 40-41). 8 Data calculated from Herranz-Loncán (2011a) for Argentina, Coatsworth (1979) for Mexico, Summerhill (2003) for Brazil, Herranz-Loncán (2011b) for Uruguay, Fishlow (1965) for the US, Hawke (1970) for England and Wales, and Caron (1983) for France. 6 countries that reached higher levels of export per capita at the eve of World War I were also those that invested more resources per capita in railways. Indeed, Latin American rail systems were highly specialised in the transportation of freight, which accounted for very high percentages of total railway revenues. By , freight revenues were between 2 and 4 times as large as passenger revenues in Argentina, Brazil, Costa Rica, Mexico, Peru and Uruguay, 9 and a large share of railway freight consisted of exports (e.g. Coatsworth, 1981: 40; Zegarra, 2012: 24-25). 10 Railways were indispensable for the growth of several export sectors, such as saltpetre in Chile (Thomson and Angerstein, 2000: 47), coffee in Brazil (Summerhill, 2003: 140) sugar in Cuba (Zanetti Lecuona and García Álvarez, 1987: 108 and 227), or coffee, bananas and animal skins in Costa Rica (Quesada Monge, 1983: 103), and the lack of railways has been identified as one of the main reasons for the sluggish growth of Colombian coffee exports (Ocampo, 2010: ), or Peruvian exports in general (Zegarra, 2011: ). Figure 1 here Despite the essential role of rail
Related Search
Similar documents
View more...
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks