¹ See Roberto Cortés Conde, La economía argentina en el largo plazo (siglos XIX y XX) (Buenos Aires, 1997); Guido Di Tella and Manuel Zymelman, Las etapas del desarrollo económico argentino (Buenos Aires, 1967); Carlos F. Díaz Alejandro, Ensayos sobre la historia económica argentina (Buenos Aires, 1970); Aldo Ferrer, La economía argentina (Buenos Aires, 1996); Taylor, Alan M., ‘External Dependence, Demographic Burdens and Argentine Economic Decline after the Belle Epoque’, Journal of Economic History, vol. 52, no. 4 (1992), pp. 907–36CrossRefGoogle Scholar; Taylor, Alan M., ‘Tres fases del crecimiento económico argentino’, Revista de Historia Económica, vol. 12 (1994), pp. 649–83CrossRefGoogle Scholar; Taylor, Alan M., ‘Argentina in the World Capital Market: Saving, Investment and International Capital Mobility in the Twentieth Century’, Journal of Development Economics, vol. 57, no. 1 (1998), pp. 147–84.CrossRefGoogle Scholar

² See Di Tella and Zymelman, Las etapas; Carl E. Solberg, The Prairies and the Pampas: Agrarian Policy in Canada and Argentina, 1880–1930 (Stanford, 1985); Timothy Duncan and John Fogarty, Australia and Argentina: On Parallel Paths (Melbourne, 1984); D. C. M. Platt and Guido di Tella, The Political Economy of Argentina, 1880–1946 (London, 1986); Díaz Alejandro, Ensayos; Taylor, ‘External Dependence’, and ‘Tres Fases’; Cortés Conde, La economía argentina, chap. 1.

³ A more concise and analytical approach which shows that the different economic policies implemented by successive governments provide, in the final analysis, the explanation for Argentina's economic history is to be found in Gerardo della Paolera, Alejandra Irigoin and Guillermo Bózzoli, ‘Passing the Buck: Monetary and Fiscal Policies’, in Gerardo della Paolera and Alan M. Taylor (eds.), A New Economic History of Argentina (Cambridge, 2003), pp. 46–86. See also Yair Mundlak, Domingo Cavallo and Roberto Domenech, ‘Agriculture and Economic Growth in Argentina, 1913–1984’ (International Food Policy Research Institute, Research Report no. 76, 1989), which uses a structural macroeconomic model to simulate what would have happened in Argentina if more appropriate economic policies had been implemented.

⁴ These results are derived from the relative series for GDP per capita using the unit root and structural breaks methodology in line with the suggestions of Perron, and Zivot and Andrews: see Sanz-Villarroya, Isabel, ‘The Convergence Process of Argentina with Australia and Canada, 1875–2000’, Explorations in Economic History, vol. 42, no. 3 (2005), pp. 439–58.CrossRefGoogle Scholar

⁵ Its principal authors are James D. Gwartney and Robert Lawson, although William Easterly also collaborated on the last report published for 2006. The first report, published in 1996, covers the periods 1975, 1980, 1985, 1990 and 1995 for a very broad sample of countries: see the website of the Fraser Institute at <www.fraserinstitute.org>. The Heritage Foundation/Wall Street Journal has also published an annual index of economic freedom since 1995, although the Fraser Institute's figures are more widely used since they cover a longer period of time. In this paper it is termed a ‘reduced index’ since it does not include variables such as the definition of property rights and the regulation of credit, the labour market and business.

⁶ Platt and Di Tella (eds.), The Political Economy, p. 1.

⁷ See, for example, John Fogarty, Ezequiel Gallo and Héctor Dieguez, ‘Australia y Argentina en el periodo de 1914–1933’, in John Fogarty, Ezequiel Gallo and Héctor Dieguez (eds.), Argentina y Australia (Buenos Aires, 1979); Platt and Di Tella (eds.), The Political Economy; Duncan and Fogarty, Australia and Argentina.

⁸ Duncan and Fogarty, Australia and Argentina.

⁹ Platt and Di Tella (eds.), The Political Economy. For a comparison between Argentina and Canada, see also David Sheinin and Carlos Mayo (eds.), Igual pero distinto: Essays in the Histories of Canada and Argentina (Mar del Plata, 1997).

¹⁰ Korol, Juan Carlos, ‘Argentina's Development in a Comparative Perspective’, Latin American Research Review, vol. 26, no. 3 (1991), pp. 201–12.Google Scholar

¹¹ See also Jeremy Adelman, Frontier Development: Land, Labour, and Capital on the Wheatlands of Argentina and Canada, 1890–1914 (Oxford, 1994). In Adelman's view the end of frontier expansion in both Argentina and Canada began in 1914.

¹² Di Tella and Zymelman, Las etapas, p. 123 and Ferrer, La economía argentina.

¹³ Solberg, The Prairies and the Pampas, chaps. 1 and 2. See Adelman, Frontier Development, for a more formal explanation with a sounder economic base. An alternative explanation for the low levels of productivity of Argentine agriculture can be found in Mundlak, Cavallo and Doménech, ‘Agriculture and Economic Growth’. For these authors the lack of incentives in the sector resulting from inappropriate economic policies held up the introduction of new technology, leading to a loss of productivity in comparison with the equivalent sectors in Australia and Canada.

¹⁴ Díaz Alejandro, Ensayos, chaps. 3 and 4. Mundlak, Cavallo and Doménech in ‘Agriculture and Economic Growth’ reach the same conclusions.

¹⁵ Duncan and Fogarty, Australia and Argentina.

¹⁶ Platt and di Tella (eds.), The Political Economy.

¹⁷ Taylor, ‘Tres fases’; Díaz Alejandro, ‘Argentina, Australia and Brazil before 1929’, in Platt and di Tella (eds.), Argentina, Australia and Canada: Studies in Comparative Development, 1870–1965 (London, 1985), pp. 95–109. On the different migration policies adopted in the ‘areas of recent settlement’, see also Timer, Ashley S. and Williamson, Jeffrey G., ‘Immigration Policy Prior to the 1930s: Labor Markets, Policy Interactions and Globalization Backlash’, Population and Development Review, vol. 24, no. 4 (1998), pp. 739–71.CrossRefGoogle Scholar

¹⁸ See Stanley L. Engerman and Kenneth L. Sokoloff, ‘Colonialism, Inequality and Long-Run Paths of Development?’(National Bureau of Economic Research, Working Paper, no. 11057, 2005) for a more detailed analysis of how colonialism affected growth in certain countries.

¹⁹ Taylor, ‘External Dependence’.

²⁰ Ibid., p. 925.

²¹ The theory of demographic dependence has also been criticised. Eduardo Miguez argues that the massive arrival of immigrants had the opposite effect to that noted by Taylor: see Miguez, Eduardo, ‘El fracaso argentino: interpretando la evolución económica en el corto siglo XX’, Desarrollo Económico, vol. 44, no. 176 (2005), p. 492CrossRefGoogle Scholar, note 27.

²² Cortés Conde, Progreso y declinación de la economía argentina (Buenos Aires, 1998).

²³ Cortés Conde, La economía argentina, pp. 204–6.

²⁴ See Taylor, ‘Tres fases’; Taylor, ‘Argentina in the World Capital Market’; and Collins, William J. and Williamson, Jeffrey G., ‘Capital Goods Prices and Investment, 1870–1950’, Journal of Economic History, vol. 61, no. 1 (2001), pp. 59–94.CrossRefGoogle Scholar

²⁵ See Mundlak, Cavallo and Doménech, ‘Agriculture and Economic Growth’.

²⁶ Cortés Conde, Progreso y declinación.

²⁷ None of these authors uses annual data to locate the moment at which the gap Argentine growth began to diverge from Australia and Canada. They use per capita GDP figures for the three countries for specific dates for which data are available. This method involves the loss of large amounts of information and can even lead to rather strange conclusions.

²⁸ Greasley, David and Oxley, Les, ‘A Tale of Two Dominions: Comparing the Macroeconomic Records of Australia and Canada since 1870’, Economic History Review, vol. 51, no. 2 (1998), pp. 294–318.CrossRefGoogle Scholar

²⁹ Leandro Prados de la Escosura, ‘International Comparisons of Real Product, 1820–1990: An Alternative Data Set’, Explorations in Economic History, vol. 37, no. 1 (2000), pp. 1–41.

³⁰ Angus Maddison, The World Economy: Historical Statistics (Paris, 2003); Roberto Cortés Conde and Marcela Harriague, ‘Estimaciones del Producto Interno de la Argentina’ (Universidad de San Andrés, Working Paper, Buenos Aires, 1996).

³¹ Gerardo Della Paolera and Alan Taylor, A New Economic History of Argentina (New York, 2003), Statistical Appendix.

³² The information required to make the adjustment has been kindly provided by Leandro Prados de la Escosura.

³³ The econometric results are included in a previous article: Sanz-Villarroya, Isabel, ‘The Convergence Process of Argentina with Australia and Canada: 1875–2000’, Explorations in Economic History, vol. 42, no. 3 (2005), pp. 439–58.CrossRefGoogle Scholar In this study, using the methodology of structural breaks for time series suggested by Perron, we observe that the first break in the relative series for Argentina's GDP per capita compared with Australia is located in the year 1899, while in comparison with Canada this point occurs in 1896. These dates mark the start of Argentina's period of divergence. Both breaks are statistically significant according to the critical values calculated using a Monte Carlo experiment. Our intention here is not to repeat the whole of the previous analysis, rather to present the results which help us to pursue the objective of this article, which is centred on the causes of Argentina's economic failure.

³⁴ Isabel Sanz-Villarroya, ‘Los procesos de convergencia de Argentina con Australia y Canadá: 1875–2000’ (Universidad Carlos III de Madrid, Departamento de Historia e Instituciones Económicas, Working Paper no. 03-03 (02), 2003). Also see Leandro Prados de la Escosura and Isabel Sanz-Villarroya, ‘Instability and Growth in Argentina: A Long-Run View’ (Universidad Carlos III de Madrid, Departamento de Historia e Instituciones Económicas, Working Paper no. 04-67 (05), 2003), and Leandro Prados de la Escosura and Isabel Sanz-Villarroya, ‘Contract Enforcement and Argentina's Long-Run Decline’ (Universidad Carlos III de Madrid, Working Papers in Economic History, WP 06-03, 2006). Both studies show that the alternative use of Maddison's series and those transformed according to the levels estimated by Prados de la Escosura, ‘International Comparisons’, do not change the results obtained.

³⁵ This is in line with Douglass North's well-known theory which states that the accumulation of physical capital ultimately depends on a structure of incentives that emanate from the institutions in place, and these should constitute the focus for any model of growth: see Douglass C. North, Institutions, Institutional Change and Economic Performance (Cambridge, 1990), p. 137.

³⁶ See, for example, Stephen Knack and Phillips Keefer, ‘Institutions, and Economic Performance: Cross Country Test Using Alternative Institutional Measures’, Economics and Politics, vol. 7, no. 3 (1995), pp. 207–27; Barro, Robert J., ‘Democracy and Growth’, Journal of Economic Growth, vol. 1, no. 1 (1996), pp. 1–27CrossRefGoogle Scholar; Robert J. Barro and Xavier Sala-i-Martín, Economic Growth (New York, 1995).

³⁷ See James Gwartney and Robert Lawson, Economic Freedom of the World: Annual Report (Vancouver, 2001), and Gwartney, James and Lawson, Robert, ‘The Concept and Measurement of Economic Freedom’, European Journal of Political Economy, vol. 19, no. 3 (2003), pp. 405–30.CrossRefGoogle Scholar Other works in this field include Fredrik Carlsson and Susanna Lundström, ‘Economic Freedom and Growth: Decomposing the Effects’, Public Choice, vol. 112, no. 3–4 (2002), pp. 335–44. See also John W. Dawson, ‘Causality in the Freedom-Growth Relationship’, European Journal of Political Economy, vol. 19, no. 3 (2003), pp. 479–95, and Jac C. Heckelman and Michael D. Stroup, ‘A Comparison of Aggregation Methods for Measures of Economic Freedom’, European Journal of Political Economy, vol. 21, no. 4 (2005), pp. 953–66.

³⁸ Gwartney and Lawson, ‘The Concept’, pp. 406–8.

³⁹ This index does not include all the components contained in the index constructed for different periods by the Fraser Institute, but it has the advantage that it is annual and historical. The drawback is that, given that it is not an institutional index which could take into account more stable variables such as regulations or the allocation of property rights, but rather an index of macroeconomic outcomes, its results are more variable over time, giving the impression of a fluctuating institutional framework. The index we present shows that macroeconomic policies were very changeable, and this is what caused such volatile growth. Della Paolera, Irigoin and Bózzoli, ‘Passing the Buck’, carried out a similar study. They constructed an index of macroeconomic and fiscal pressure for Argentina, although the methodology used is different from that used here. However, they produced average calculations for each legislature, making it more difficult to observe continuous changes over time.

⁴⁰ See Prados de la Escosura and Sanz-Villarroya, ‘Contract Enforcement’, for a more detailed study which includes the degree of definition of property rights, the degree of distribution of wealth, and the degree of separation of powers, in addition to the reduced index of economic freedom. The results presented confirm those presented in this paper.

⁴¹ See Jakob de Haan (2003), ‘Economic Freedom: Editor's Introduction’, European Journal of Political Economy, vol. 19, no. 3 (2003), pp. 395–403. The construction of an index with such additional variables, in addition to those representing the macroeconomic outcomes, is complicated by the fact that, as these variables are difficult to measure, dummies are used instead. This makes the results difficult to interpret and complicates the global index. The situation is further complicated when, as is the case in this paper, a lengthy time span is covered.

⁴² Gwartney and Lawson, ‘The Concept’, p. 411. The same position regarding government spending in Argentina is adopted by Mundlak, Cavallo and Domenech, ‘Agriculture and Economic Growth’. According to these authors, an increase in government spending tends to crowd out private consumption and in turn affects the redistribution of the economy's productive resources. They also highlight the perverse effects which different modes of financing government can have on the economy (chap. 2 and pp. 10 and 120).

⁴³ Gwartney and Lawson, ‘The Concept’, p. 414.

⁴⁴ Ibid., p. 415.

⁴⁵ José de Gregorio, ‘Economic Growth in Latin America’, Journal of Development Economics, vol. 39, no. 1 (1992), pp. 59–84; Taylor, Alan M., ‘On the Cost of Inward-Looking Development: Prices Distortions, Growth and Divergence in Latin America’, Journal of Economic History, vol. 58, no. 1 (1998), pp. 1–28.CrossRefGoogle Scholar

⁴⁶ Heckelman and Stroup, ‘A Comparison of Aggregation’.

⁴⁷ This is justified when the long term evolution of the variables is observed. In the corresponding graphs presented in Appendix 1 it can be observed that all the variables have a positive tendency while Tariff falls throughout the period under consideration. For this reason this variable is assigned a different sign and weight compared with the others in the principal components analysis.

⁴⁸ The weightings are: −0.346 for Gi/(Gi+Ci), −0.318 for INFLA/(100+INFLA), 0.198 for Tariff and −0.137 for Black.

⁴⁹ The values of this index and the relative indices that appear later, are normalised given that the original variables have been subjected to the same transformation in order to solve the problem caused by the fact that the principal components method might assign greater weight and impact to some variables than to others because of the effect of the measurement system used, as explained in the text. Consequently, in order to interpret the figures correctly, a higher level of economic freedom should be assigned to positive values and lower levels to negative values.

⁵⁰ See Leslie Bethell (ed.), Argentina since Independence (New York, 1993). It seems that the fiscal policy was so expansionary that, between 1885 and 1893, the levels of public deficit were so persistent and so high that the federal government was forced to seek alternative sources of income in order to cope with the deficit: see Gerardo della Paolera, ‘Experimentos monetarios’, p. 564, and also Pablo Gerchunoff and Luis Llach, El ciclo de la desilusión y el desencanto: un siglo de política económica argentina (Buenos Aires, 2003). These authors refer to the fiscal policy implemented until 1890 as ‘ultraexpansionary’ (p. 49). Moreover, Taylor and Williamson highlighted the fact that the inflow of foreign capital, which came primarily from Britain, increased as a result of the low savings rate caused by a high dependency rate: see Taylor, Alan M. and Williamson, Jeffrey G., ‘Capital Flows to the New Worlds as an Inter-Generational Transfer’, Journal of Political Economy, vol. 102, no. 2 (1994), pp. 348–71.CrossRefGoogle Scholar

⁵¹ Gerardo della Paolera and Alan M. Taylor, Straining at the Anchor: The Argentine Currency Board and the Search for Macroeconomic Stability, 1880–1935 (Chicago and London, 2001).

⁵² Specifically, in 1887 the Ley de Bancos Garantizados was passed which stated that all monetary issues had to be backed by gold reserves and by financial assets issued by the national government valued in gold: see Roberto Cortés Conde, Dinero, deuda y crisis: evolución fiscal y monetaria en la Argentina, 1862–1890 (Buenos Aires, 1989), p. 177.

⁵³ Della Paolera and Taylor, Straining at the Anchor.

⁵⁴ According to della Paolera and Taylor, initially the crisis showed the typical symptoms of a traditional banking crisis, that is an increase in the amount of cash in the hands of the public, an increase in the banks' reserves-deposits ratio, and the elimination of some financial institutions, which meant the destruction of deposits: see della Paolera and Taylor, Straining at the Anchor, p. 68.

⁵⁵ Cortés Conde, Dinero, deuda y crisis, p. 13 and chap. VI. In this chapter Cortés Conde briefly reviews all the studies of the 1890 crisis. Some of these works explain the crisis in terms of external aspects caused by the balance of payments situation. Cortés Conde argues against this idea.

⁵⁶ Gerchunoff and Llach, El ciclo, pp. 71–2 and 97–8.

⁵⁷ The currency board really began to operate in 1899. Between 1891 and 1899 the Baring agreement restricted monetary issues. Between 1900 and 1929 issues by the Caja de Conversión depended on the movement of gold: see Roberto Cortés Conde, La economía política, pp. 43–50 and 58–9. The Gold Standard was reinstated in 1899 and, despite periods outside this system in 1900, 1914 and 1929, the monetary authorities acted as if they were a member country: see Gerardo Della Paolera and Alan M. Taylor, ‘Economic Recovery from Argentina's Great Depression: Institutions, Expectations and the Change of Macroeconomic Regime’ (MS, 1998), p. 12.

⁵⁸ Ibid., p. 3, and della Paolera and Taylor, Straining at the Anchor, p. 31; see also Sidney Homer and Richard Sylla, A History of Interest Rates (3rd edition, New Brunswick NJ, 1996), pp. 626–29.

⁵⁹ According to Arturo O'Connell there were few changes in trade policy, while the rest of the world was returning to protectionism. So, during the 1920s, Argentina continued its free trade policy as a producer of staple goods. The main change was a rise of between 25 per cent and 60 per cent of the official ‘aforo’ values in 1923: see Arturo O'Connell, ‘Free Trade in One Country: The Case of Argentina in the 20s’ in D. C. M. Platt and Guido Di Tella (eds.), The Political Economy of Argentina, 1880–1946 (London, 1986), p. 91; see also Guido di Tella, ‘Economic Controversies in Argentina from the 1920s to the 1940s’ in Platt and di Tella (eds.), The Political Economy of Argentina, pp. 120–32.

⁶⁰ According to della Paolera and Taylor, ‘Economic Recovery’, p. 10, the effects of the fiscal decisions might have been contractionary until 1935, and it cannot be said that New Deal-type policies were implemented.

⁶¹ Peter Alhadeff, ‘Economic Controversies’ in Platt and di Tella (eds.), The Political Economy of Argentina, pp. 96, 107 and 110.

⁶² Cortés Conde, La economía política, pp. 108–11.

⁶³ For di Tella, the 1930 crisis marks the moment of transition from free trade to protectionism in Argentina, although the main change took place alter the Second World War: di Tella, ‘Economic Controversies in Argentina’, p. 128.

⁶⁴ Alhadeff, ‘Economic Controversies’, p. 104.

⁶⁵ Cortés Conde, La economía política, pp. 102–3.

⁶⁶ David Rock, Argentina 1516–1987, desde la colonización hasta Raúl Alfonsín (Madrid, 1988).

⁶⁷ Pablo Gerchunoff, ‘Peronist Economic Policies, 1946–1955’, in Guido di Tella and Rudi Dornbusch (eds.), The Political Economy of Argentina, 1946–1983 (Pittsburgh, 1989), chap. 4

⁶⁸ Roberto Cortés Conde, La economía política de la Argentina en el siglo XX (Buenos Aires, 2005), p. 119.

⁶⁹ During the second phase of Peronism, in 1952 the ‘Economic Plan’ was implemented, an austerity programme which contrasted with the policies of the first phase, and which attempted, above all, to restrain inflation and the foreign trade deficit: see Gerchunoff and Llach, El ciclo, pp. 208–12.

⁷⁰ Juan Carlos Torre and Liliana de Riz, ‘Argentina since 1946’, in Leslie Bethell (ed.), Argentina since Independence (Cambridge, 1993), pp. 271–314; see also Gerchunoff and Llach, El ciclo, p. 295.

⁷¹ Inflation reached 900 per cent in 1975–1976: see Di Tella and Dornbusch, The Political Economy of Argentina.

⁷² Mundlak, Cavallo and Domenech, ‘Agriculture and Economic Growth’, pp. 111–3. The Central Bank had to take control of 60 institutions in 1980: see Gerchunoff and Llach El ciclo, pp. 358–60.

⁷³ Ibid., p. 375.

⁷⁴ The RIEF recovered between 1976 and 1980, coinciding with one of the bloodiest periods of military activity in Argentina's history, a period which came to a close with the restoration of democracy under Raúl Alfonsín in 1983. Although this result may appear contradictory, the explanation lies in the fact that the index used here reflects the results of the policies implemented and, in consequence, the degree of economic freedom has no connection with a country's level of political freedom: see Gwartney and Lawson, ‘The Concept’, pp. 408–9. These authors stress that, although political freedom and economic freedom are usually linked, they are two separate concepts and that there are cases of countries that enjoy high levels of political freedom, but where the government implements policies which restrict economic freedom and vice versa. A similar situation is to be found in Della Paolera, Irigoin and Bózzoli, ‘Passing the Buck’, p. 69. These authors rank successive administrations in Argentina according to the success of their economic results and the period of military government between 1976 and 1983 (Videla/Viola/Galtieri/Bignone) is ranked ninth out of 33.

⁷⁵ The results of the calculation are to be found in Appendix 2 (Table A.2.1).

⁷⁶ Tests applied on the assumption of a model with a constant and a trend and taking the appropriate number of lags into account.

⁷⁷ In the comparison of Argentina with Australia, a three-year lag has been introduced into the relative RIEF as this produces the best fit for the regression. A lag of one year has been introduced into the comparison with Canada for the same reason.

⁷⁸ As can be observed in Appendix 3, the results obtained by considering the alternative GDP per capita series developed by Cortés Conde for the period before 1935 and linked with Maddison's figures from then until 2000 (expressed in 1913 prices for the period 1875–1939 and in 1980 prices for the period 1940–2000), as well as those obtained by della Paolera and Taylor expressed in the same terms, show that cointegration is more intense in the case of comparison with Australia and less intense in comparison with Canada. This suggests that Maddison's figures, expressed in constant 1990 prices, represent the best option. Nevertheless, the results are solid enough in any one of the three scenarios to confirm that relative economic freedom and the results of the policies implemented lie behind the relative performance of Argentina's economy over this long period of time. The fact that the three alternatives lead us to the same conclusion simply confirms that the methodology used in the present study is robust.

⁷⁹ If structural breaks are introduced into the series on which the relative indices are based, a rising trend can be observed until 1899. This year marks a statistically significant break indicating the birth of a new trend which begins to move in the opposite direction. Lack of space makes it impossible to include the results, but they are available upon request. In the case of Canada it would be more accurate to identify 1896 as the key year, but we have decided to use the year 1899 in order to bring the two analyses, Australia and Canada, together.

⁸⁰ Except in the case of the sub-phase 1990–2000, for which a closing of the gap is observed.

⁸¹ In the case of the comparison with Australia, the adjusted R² for the sample which covers the period 1980–2000 is 0.43 and for the period 1875–1980 is 0.16. For the comparison with Canada the corresponding values are 0.62 and 0.42. This shows that the cointegration relationship estimated for the long term does not fade in either case.

⁸² This is also observed in the principal components analysis itself which led us to the construction of the relative indices. We saw that the variable which represents the difference in values for Black and Infla between Argentina and the other two countries has a negative weighting while the (Gi/Gi+Ci) and Tariff variables for Argentina relative to those for Australia and Canada have a positive weighting in both cases (see Appendix 2).

⁸³ The impact of Black in the last ten years of the analysis becomes almost zero after the introduction of the Ley de Convertibilidad and, together with the level of control over inflation, this has a very positive effect on the respective relative indices, which recover significantly between 1990 and 2000. The relative position of Argentina's economy also improves with the result that the relationship between economic freedom and the relative evolution of GDP per capita is maintained until the end of the period under consideration in this study.

⁸⁴ John Coatsworth and Jeffrey G. Williamson, ‘Always Protectionist? Latin American Tariffs from Independence to the Great Depression’, Journal of Latin American Studies, vol. 36, no. 2 (2005), pp. 205–32.

⁸⁵ This is also shown in the work of Coatsworth and Williamson for the region of Latin America; see also Douglas A. Irwin, ‘Did Import Substitution Promote Growth in the Late Nineteenth Century?’ (NBER Working Paper, no. W7851, 2002). Coatsworth and Williamson agree with Irwin that other forms of protection are much more powerful than tariffs, which were often imposed simply as a source of income.

⁸⁶ The impacts of these variables were calculated after estimating a MLS (Minimum Least Squares) regression equation between the respective index as a dependent variable and the variable components of the equation. This produced the coefficients which, when applied to the standard deviation of each variable, allow us to calculate the impact of each of them in the global index. In the case of the Argentine index these coefficients are −0.0080 for BLACK, 4.363 for TARIFF, −1.161 for INFLA and −9.369 for (Gi/Gi+Ci). All the variables are significant with high t-ratios and the model offers a close fit with R²-Adj.=0.986.

⁸⁷ The weights for the variables are: −1.555 for DBlack, 0.814 for DTariff; −2.265 for DInfla and 2.006 for D(Gi/Gi+Ci).

⁸⁸ The weights for the variables are: −3.233 for DBlack, 2.374 for DTariff; −4.509 for DInfla and 4.368 for D(Gi/Gi+Ci).

⁸⁹ The impacts of these variables were calculated after estimating a MLS regression equation between the respective relative index as a dependent variable and the variable components of the equation. This produced the coefficients which, when applied to the standard deviation of each variable, allow us to calculate the impact of each of them in the global index. In the case of the index comparing Argentina with Australia, these coefficients are −0.0079 for DBLACK, 1.078 for DTARIFF, −1.634 for DINFLA and 5.861 for D(Gi/Gi+Ci). All the variables are significant with high t-ratios and the model offers a close fit with R²-Adj.=0.786. In the case of the index comparing Argentina with Canada, these coefficients are −0.0065 for DBLACK, 7.547 for DTARIFF, −1.975 for DINFLA and 4.368 for D(Gi/Gi+Ci). All the variables are significant with high t-ratios and the model offers a close fit with R²-Adj.=0.861.

⁹⁰ The causality analyses are similar in the three cases. The results are not included here due to lack of space but are available upon request. The relative RIEF has been taken with three lags in the case of the comparison of Argentina with Australia and one for the comparison with Canada.