¹ Brading, D. A., Miners and Merchants in Bourbon Mexico, 1763–1810 (Cambridge: Cambridge University Press, 1971), pp. 36–42Google Scholar; Socolow, Susan M., The Merchants of Buenos Aires, 1778–1810: Family and Commerce (Cambridge: Cambridge University Press, 1978), pp. 16–20Google Scholar; Sebastián, Jesús Turiso, Comerciantes españoles en la Lima borbónica: anatomía de una élite de poder, 1701–1761 (Valladolid: Universidad de Valladolid, 2002)Google Scholar; Twinam, Ann, Miners, Merchants, and Farmers in Colonial Colombia (Austin, TX: University of Texas Press, 1982)Google Scholar; Zaldívar, Trinidad et al. , Los vascos en Chile, 1680–1820 (Santiago: Editorial Los Andes, 1998)Google Scholar.

² Brading, D. A., ‘Government and Elite in Late Colonial Mexico’, Hispanic American Historical Review, 53: 3 (1973), pp. 389–414CrossRefGoogle Scholar.

³ Pérez, Paloma Fernández and Sola-Corbacho, Juan Carlos, ‘Regional Identity, Family, and Trade in Cadiz and Mexico City in the Eighteenth Century’, Journal of Early Modern History, 8: 4 (2004), pp. 358–85CrossRefGoogle Scholar.

⁴ Montañeses were natives of the mountains of Santander, roughly corresponding to present-day Cantabria. While politically this geographical area formed only a small part of the province of Burgos, itself part of Old Castile, its inhabitants have maintained a distinct identity up to the present.

⁵ In addition to the works already cited, also see Shaw, Carlos Martínez, La emigración española a América, 1492–1824 (Colombres, Asturias: Archivo de Indianos, 1994)Google Scholar.

⁶ Following the distinction between independent immigrants and officials, the term ‘economic elite’, as used here, refers to the upper tiers of the wealthy population whose wealth and status derived from accumulation.

⁷ Figures of transatlantic migration are based on my analysis of passenger licences from the Archivo General de Indias, Contratación, vols. 5492–5507 (years 1751–1764); as well as Domínguez, Isabelo Macías, La llamada del nuevo mundo: la emigración española a América, 1701–1750 (Sevilla: Universidad de Sevilla, 1999)Google Scholar; Macías, Rosario Márquez, La emigración española a América, 1765–1824 (Oviedo: Universidad de Oviedo, 1995)Google Scholar. For the population of Mexico City, the most nuanced estimate is a total of over 150,000 in 1790, offered by Grijalva, Manuel Miño, ‘La población de la Ciudad de México en 1790: variables económicas y demográficas de una controversia’, in Grijalva, Manuel Miño and Toledo, Sonia Pérez (eds.), La población de la Ciudad de México en 1790 (México, DF: Colegio de México, 2004)Google Scholar. Estimates of Spanish immigrants in Mexico City are based on my own analysis of the census of 1811 (see Tables 1 and 2) and Aguirre, Jorge González Angulo, ‘Los inmigrantes de la ciudad de México en 1811’, in Anaya, Delia Salazar (ed.) Imágenes de los inmigrantes en la ciudad de México, 1753–1910 (Mexico City: Plaza y Valdeés Editores: CONACULTA/INAH, 2002), pp. 114–18Google Scholar. According to John Kicza, 90 per cent of New Spain's wealthiest families lived in the capital: Kicza, John E., Colonial Entrepreneurs: Families and Business in Bourbon Mexico City (Albuquerque: University of New Mexico Press, 1983), pp. 15–17Google Scholar.

⁸ By the late eighteenth century, Spanish population reached about 11 million. The most frequently quoted estimate of Spanish emigration to the New World is Mörner, Magnus, ‘Spanish Migration to the New World Prior to 1810: A Report on the State of Research’, in Chiapelli, Fredi (ed.), First Images of America: The Impact of the New World on the Old (Berkeley, CA: University of California Press, 1976), pp. 737–82Google Scholar. For the eighteenth century, a more refined estimate of 100,000 emigrants is offered by Martínez Shaw, La emigración española a América, pp. 164–7. Emigration was constrained by transatlantic shipping traffic, which was in turn constrained by commercial policy. An Annales-informed seminal study of the Cadiz monopoly and transatlantic commerce is González, Antonio García-Baquero, Cádiz y el Atlántico, 1717–1778 (Sevilla: Escuela de Estudios Hispano-Americanos, 1976)Google Scholar.

⁹ The Casa de Contratación was transferred from Seville to Cadiz in 1717 and abolished in 1790: see Haring, C. H., The Spanish Empire in America (New York: Oxford University Press, 1963 [1947]), pp. 293–309Google Scholar; on policy see also Márquez Macías, La emigración española, pp. 34–8.

¹⁰ On the costs of emigration, see Jacobs, Auke P., Los movimientos migratorios entre Castilla e Hispanoamerica durante el reinado de Felipe III, 1598–1621 (Amsterdam: Rodolpi, 1995), pp. 33–47, 67–71Google Scholar.

¹¹ Illegal immigration, partly comprising members of ships' crews, made possible the participation of the poor, who could not afford the journey: see Bueno, Pablo Emilio Pérez-Mallaína, Los hombres del océano: vida cotidiana de los tripulantes de las flotas de Indias, siglo XVI (Sevilla: Diputación de Sevilla, 1992)Google Scholar; Jacobs, Auke Pieter, ‘Pasajeros y polizones: algunas consideraciones sobre la emigración española a las Indias durante el siglo XVI’, Revista de Indias, 43: 172 (1983), pp. 439–79Google Scholar.

¹² Because of the highly skewed gender imbalance, this analysis refers only to male immigrants. The single exception is the data in Table 1, which includes both men and women.

¹³ Boyd-Bowman, Peter, ‘Patterns of Spanish Emigration to the Indies until 1600’, Hispanic American Historical Review, 56: 4 (1976), p. 589Google Scholar; Macías Domínguez. La llamada del Nuevo Mundo, pp. 87–90.

¹⁴ Notable studies that highlight transatlantic immigrant networks are: Altman, Ida, Emigrants and Society: Extremadura and America in the Sixteenth Century (Berkeley: University of California Press, 1989)Google Scholar; Altman, Ida, Transatlantic Ties in the Spanish Empire: Brihuega, Spain, and Puebla, Mexico, 1560–1620 (Stanford, CA: Stanford University Press, 2000)Google Scholar; Pescador, Juan Javier, The New World inside a Basque Village: The Oiartzun Valley and its Atlantic Emigrants, 1550–1800 (Reno: University of Nevada Press, 2003)Google Scholar.

¹⁵ Immigrants' wealth was accumulated only after arrival in the New World. Pro- and anti-peninsular writers commented abundantly on immigrants' initial poverty. See Brading, Miners and Merchants, pp. 109–10. All the works on merchants cited here claim that immigrants started off from scratch, usually as apprentices. Institutional obstacles made the transfer of capital or property virtually impossible, as illustrated by the rudimentary ship baggage inventories of a few immigrants presented in Jacobs, Los movimientos migratorios, pp. 183–7. Even if we assume that immigrants arrived with initial wealth, it would have been negligible in comparison to the fortune eventually attained by the more successful among them.

¹⁶ Although we only consider immigrants here, it should be noted that creoles were present in all tiers of wealth. However, their participation in the upper classes was slim.

¹⁷ In many types of records, import-export merchants, lesser ones, and commercial employees were all designated as comerciantes. Brading noted the same problem with mining: Brading, ‘Government and Elite,’ pp. 394. He also commented that extant records inform us about success stories but are silent about the rather frequent cases of bankruptcy: Brading, Miners and Merchants, p. 169.

¹⁸ The Appendix describes the Marriage Applications dataset in detail.

¹⁹ Since many immigrants clustered in centrally located commercial houses, where they lived and worked, residential patterns may contain unwarranted biases, especially if we seek to single out the truly wealthy.

²⁰ The gap between Old and New Castile – Old Castilians were better off – was observed with regards to all indicators discussed below.

²¹ Mörner, Magnus, ‘Economic Factors and Stratification in Colonial Spanish America with Special Regard to Elites’, Hispanic American Historical Review, 63: 2 (1983), pp. 335–69Google Scholar.

²² My discussion of the concept of networks addresses its broad sociological meaning and use in historical studies. Despite their sophistication, Social Network Analysis studies using computerised mathematical modelling frequently suffer from similar flaws to less systematic historical studies, notably the confusion between description and explanation.

²³ The concept of embeddedness goes back to Karl Polanyi who argued that under modern capitalism economic relations were no longer embedded in social relations. An influential work on trade networks and diasporas applying the concept of ‘embeddedness’ is Curtin, Philip, Cross-Cultural Trade in World History (Cambridge: Cambridge University Press, 1984), pp. 12–14Google Scholar.

²⁴ A key work concerning trust-based networks in the Spanish empire is Lamikiz, Xabier, Trade and Trust in the Eighteenth-Century Atlantic World: Spanish Merchants and Their Networks (London: Royal Historical Society, 2010)Google Scholar. An extensive literature in economic history follows a similar line of reasoning: see, for example, Landa, Janet T., Trust, Ethnicity, and Identity: Beyond the New Institutional Economics of Ethnic Trading Networks, Contract Law, and Gift-Exchange (Ann Arbor: University of Michigan Press, 1994)Google Scholar; North, Douglass C., Institutions, Institutional Change, and Economic Performance (Cambridge: Cambridge University Press, 1990), pp. 36–45Google Scholar. A critique of the concept of trust, arguing for its redundancy, is Guinanne, Timothy W., ‘Trust: A Concept Too Many’, Jahrbuch für Wirtschaftsgeschichte, 1 (2005), pp. 77–92Google Scholar.

²⁵ Lockhart, James, ‘The Merchants of Early Spanish America,’ in Lockhart, James, Things of the Indies: Essays Old and New in Early Latin American History (Stanford, CA: Stanford University Press, 1999), pp. 158–82Google Scholar, provides a synthesis on this theme. Although the term ‘networks’ is not always used, the idea is the same: commerce was grounded in personal ties. While part of my discussion focuses on commerce, the argument applies to other sectors that immigrants entered, notably silver mining. Both were lucrative yet burdened by heavy risk. Baskes argued that merchants’ monopolisitic practices emerged as a response to a risky environment, which may highlight the role of networks: see Baskes, Jeremy, ‘Risky Ventures: Reconsidering Mexico's Colonial Trade System’, Colonial Latin American Review, 14: 1 (2005), pp. 27–54Google Scholar.

²⁶ In this sense my position differs from studies that view networks, implicitly or explicitly, as a causal factor of economic mobility. Lamikiz argues that Spanish merchants' trust networks embedded in ethnic identity formed a prerequisite for trade, and an independent factor in explaining participation in international trade: see Lamikiz, Trade and Trust, pp. 182–5.

²⁷ Examples of this prevalent line of argument include Brading, Miners and Merchants, pp. 111–12; Socolow, Susan, ‘Marriage, Birth and Inheritance: The Merchants of Eighteenth Century Buenos Aires’, Hispanic American Historical Review, 60: 3 (1980), pp. 387–406Google Scholar; Hoberman, Louisa Schell, Mexico's Merchant Elite, 1590–1660 (Durham, NC: Duke University Press, 1991), pp. 64–8Google Scholar; Fernández Pérez and Sola-Corbacho, ‘Regional Identity, Family, and Trade’, pp. 369–76.

²⁸ Some historians have qualified the role of marriage by stressing merchants' advanced age at the altar, although they did not dismiss it altogether as a channel of social mobility or offer an alternative explanation: Kicza, Colonial Entrepreneurs, pp. 164–6. On late marriage among Buenos Aires merchants, see Socolow, ‘Marriage, Birth and Inheritance’, pp. 390–1.

²⁹ Fernández Pérez and Sola-Corbacho, ‘Regional Identity, Family, and Trade’, pp. 370–1. Using evidence from testaments, they show that 40 per cent (!) of Mexico City's merchants died as bachelors. My own findings, based on the 1811 census, show that about a third of the pool of wealthy peninsulares aged 45 and over were bachelors.

³⁰ This calculation, based on the marriage application dataset, only includes first marriage. In eighteenth-century Spain, males' average age at first marriage was 26.5 for Basques, 25.0 for Asturians, 25.5 for Galicians, 24.9 for New Castilians, and 24.3 for Andalusians: see Rowland, Robert, ‘Sistemas matrimoniales en la Península Ibérica (siglos XVI–XIX): una perspectiva regional’, in Moreda, Vicente Pérez and Reher, David S. (eds.), Demografía histórica en España (Madrid: El Arquero, 1988)Google Scholar, p. 95. The overall gap between immigrants and those who stayed behind is remarkable. Furthermore, although regional differences in Spain were modest, they reversed in the New World, suggesting that marriage was responsive to circumstances rather than fixed in cultural tradition.

³¹ Although this type of social capital had a transatlantic dimension I discuss it here because its dynamics in regards to wealth accumulation took place after arrival in the New World: see Brading, Miners and Merchants, pp. 104–8, and also Socolow, The Merchants of Buenos Aires, pp. 21–2, and James Lockhart, ‘The Merchants of Early Spanish America’, pp. 164–6.

³² Kicza, Colonial Entrepreneurs, pp. 140–1.

³³ Based on the enumeration of the central wards (1, 5, and 9) in the 1811 census. Only peninsular employees were counted.

³⁴ Sola-Corbacho, Juan Carlos, ‘Los comerciantes mexicanos frente a la muerte (1765–1800)’, Revista Complutense de Historia de América, 25 (1999), p. 191Google Scholar. Although Brading did not argue that inheritance was transferred directly to nephews, his argument, based largely on literary sources, suggested that young relatives from the peninsula ended up taking over commercial ventures, while creole sons received landed estates, an enterprise less risky but also less profitable than commerce, thus leading to cross-generational downward mobility: Brading, Miners and Merchants, pp. 208–19.

³⁵ Socolow argued that ‘a combination of business conditions, legal restrictions and fertility patterns seriously impaired the merchants' ability to transfer their estates intact to the next generation’: Socolow, ‘Marriage, Birth and Inheritance’, pp. 404–5. In a detailed study of emigration from the Oyarzun valley in the Basque province of Guipúzcoa, Pescador argued that the extant evidence belies the prevalent notion of kinship-based recruitment and favouritism: Pescador, The New World Inside a Basque Village, pp. 82–4.

³⁶ A study highlighting conflicting group interests is Stein, Stanley J. and Stein, Barbara H., Apogee of Empire: Spain and New Spain in the Age of Charles III, 1759–1789 (Baltimore, MD: Johns Hopkins University Press, 2003)Google Scholar. On the shift from creole to peninsular dominance in the audiencias, see Burkholder, Mark A., From Impotence to Authority: The Spanish Crown and the American Audiencias, 1687–1808 (Columbia, MO: University of Missouri Press, 1977)Google Scholar.

³⁷ For bureaucrats crossing the Atlantic whose origin is stated in their passenger licence, only 14 per cent came from the north, while Andalusians and Castilians constituted 53 per cent: see Márquez Macías, La emigración española, p. 202. Among high-ranking peninsular bureaucrats in late colonial Buenos Aires, Andalusians predominated (6), followed by Navarreses (4) and Galicians (2): Socolow, Susan, The Bureaucrats of Buenos Aires, 1769–1810: Amor al Real Servicio (Durham, NC: Duke University Press, 1987)Google Scholar, p. 134.

³⁸ Cf. McCabe, Ina Baghdiantz et al. (eds.), Diaspora Entrepreneurial Networks: Four Centuries of History (Oxford: Berg, 2005)Google Scholar; Jarvis, Adrian and Robert Lee, W. (eds.), Trade, Migration and Urban Networks in Port Cities, c. 1640–1940 (St. John's, Canada: Research in Maritime History, Issue 38, 2008)Google Scholar.

³⁹ Although Lamikiz does not explicitly address Basque ethnicity, it certainly appears crucial: Lamikiz, Trade and Trust, esp. 10–14, 122–6. Additional regionally-focused studies on networks include Martínez del Cerro, Victoria E., Una comunidad de comerciantes: navarros y vascos en Cádiz (segunda mitad del siglo XVIII) (Sevilla: Consejo Económico y Social de Andalucía, 2007)Google Scholar; Imízcoz, José María (ed.), Redes familiares y patronazgo: aproximación al entramado social del País Vasco y Navarra en el antiguo régimen, siglos XV–XIX (Bilbao: Universidad del País Vasco, 2001)Google Scholar.

⁴⁰ Archivo Histórico Foral de Bizkaia, Judicial, file 62/34, (1713–1716).

⁴¹ Ibarra, Antonio, ‘Redes de circulación y redes de negociantes en Guadalajara colonial: mercado, élite comercial e instituciones’, Historia Mexicana, 56: 3 (2007), 1017–41Google Scholar. Ibarra goes further to argue that Guadalajara merchant networks posed a challenge to Mexico City well beyond its regional market. On northern predominance in various consulados, see the various articles in Ibarra, Antonio and Hausberger, Bernd (eds.), Comercio y poder en América colonial: los consulados de comerciantes, siglos XVII–XIX (México DF: Instituto Mora, 2003)Google Scholar.

⁴² Another corporate conflict of interests casting doubt on diasporic solidarity emerged between the consulados of Mexico and Cadiz: see Herrero, Pedro Pérez, ‘Actitudes del Consulado de México ante las reformas comerciales borbónicas (1718–1765)’, Revista de Indias, 171: 43, (1983), pp. 97–182Google Scholar.

⁴³ Interestingly, even after the introduction of additional Spanish ports to direct commerce with the Indies in 1778, Cadiz still commanded 40 per cent of ship departures to the Indies as a whole and virtually all departures to New Spain: Fisher, John, Commercial Relations between Spain and Spanish America in the Era of Free Trade, 1778–1796 (Liverpool: Liverpool University Press, 1985), pp. 121–4Google Scholar. Furthermore, the marriage applications dataset shows that almost all immigrants from the north arrived through Cadiz instead of the nearby ports of Santander and El Ferrol, attesting to the role of connections in the southern port city.

⁴⁴ The Pearson correlation coefficient is 0.98. Data on Cadiz includes 2,071 registered consulado merchants (1743–1823), excluding those born in Cadiz, many of whom were second-generation immigrants: the data used comes from Ruiz Rivera, Julián B., El Consulado de Cádiz: matrícula de comerciantes, 1730–1823 (Cádiz: Diputación Provincial de Cádiz, 1988)Google Scholar. Data on emigrants was taken from the passenger licences granted by the Casa de Contratación between 1751 and 1763, and Márquez's figures based on the same source for 1765–1824: Archivo General de Indias, Contratación, vols. 5492–5507; Márquez Macías, La emigración española a América, pp. 143–51.

⁴⁵ This problem is intrinsic to research based on notarial records, which are not only biased towards the wealthier tiers, but also only describe existing ties. Notarial records tell us about the operation of existing networks but they do not inform us about the significance of networks because what they observe – transactions between individuals – is by definition a tie within a network.

⁴⁶ AGN, Matrimonios, 14, 99.

⁴⁷ AGN, Matrimonios, 144, 64.

⁴⁸ AGN, Matrimonios, 157, 42.

⁴⁹ AGN, Matrimonios, 12, 62.

⁵⁰ One may assume that networks were not strictly based on paisanaje, allowing for mixed networks of, say, Riojans with Basques. However, that would downplay network embeddedness in ethnic identity, rendering the supposed role of informal institutions irrelevant.

⁵¹ In The Wealth of Nations Adam Smith regarded human skills as equivalent to physical capital in terms of increasing productivity. Key studies include Schultz, Theodore W., Investing in People: The Economics of Population Quality (Berkeley: University of California Press, 1981)Google Scholar; Becker, Gary, Human Capital: A Theoretical and Empirical Analysis, with Special Reference to Education (Chicago: University of Chicago Press, 1964)Google Scholar.

⁵² These methodological problems are discussed in Cipolla, Carlo M., Literacy and Development in the West (Baltimore, MD: Penguin Books, 1969)Google Scholar; Schofield, Roger S., ‘Dimensions of Illiteracy in England, 1750–1850’, in Graff, Harvey (ed.), Literacy and Social Development (Cambridge: Cambridge University Press, 1981), pp. 201–13Google Scholar.

⁵³ Nuñez has detected a strong correlation between regional literacy rates, economic growth and gender equality in modern Spain: Nuñez, Clara Eugenia, La fuente de la riqueza: educación y desarrollo económico en la España contemporánea (Madrid: Alianza, 1992)Google Scholar. Psychologists have argued that literacy also enhances cognitive processes, thus having a multiplier effect on human capital: Nuñez, Clara Eugenia and Tortella, Gabriel ‘Educación, capital humano y desarrollo: una perspectiva histórica’, in Nuñez, and Tortella, (eds.), La maldición divina: ignorancia y atraso económico en perspectiva histórica (Madrid: Alianza, 1993), pp. 26–7Google Scholar.

⁵⁴ Although business experience could have created an advantage, the previous acquisition of such skills at home or in Cadiz seems unlikely because wealthy immigrants usually came from rural families in the north and went directly to Mexico.

⁵⁵ The libranza was a type of a bill of exchange increasingly drawn by Mexico's merchants in the late eighteenth century. With widespread circulation in domestic transactions libranzas practically functioned as credit and currency. On the crucial importance of libranzas for commerce, and also for the entire late colonial economy, see Herrero, Pedro Pérez, Plata y libranzas: la articulación comercial del México borbónico (Mexico City: Colegio de México, 1988)Google Scholar, especially pp. 195–253.

⁵⁶ By comparison, the literacy rate of creole males in Mexico City based on a sample of marriage applications was 71 per cent, on a par with eighteenth-century European urban populations.

⁵⁷ Grubb, F. W., ‘Growth of Literacy in Colonial America: Longitudinal Patterns, Economic Models and the Directions of Future Research’, Social Science History, 14: 4 (1990), pp. 453–5Google Scholar.

⁵⁸ I attached annual rent data from a real estate census taken in 1813 to the population census of 1811. The census of 1813 was published as Datos sobre rentas de fincas urbanas en la ciudad de México (México DF: Oficina Impresora de Estampillas, 1903). Rent figures refer to entire buildings rather than households. Despite this limitation they are useful for analysing relative wealth across primary households. Total rents by cuartel confirmed the well-known centre-periphery urban disparity. Rent per inhabitant was 23 pesos in the centre and 5 pesos in the periphery.

⁵⁹ The rate of economic success is analytically defined as the share of wealthy immigrants out of the regional immigrant population. Overlap between this pool of the economic elite and the previous pool of wealthy merchants confirms that commerce was indeed the main route to wealth accumulation.

⁶⁰ The high success rate of smaller regional groups like the Riojans suggests that networks might not matter, but the figures are too small for any confidence.

⁶¹ Interestingly, five out of the 20 semi-literates employed in commerce worked in wine shops, suggesting that shaky handwriting was possibly related to lifestyle more than literacy.

⁶² On Spain's regional diversity, see Shaw, Carlos Martínez et al. , España en el siglo XVIII (Barcelona: Crítica, 1985)Google Scholar.

⁶³ Catalonia also demonstrated a similar pattern of development towards early industrialisation and economic growth: see the classic interpretation of industrialisation in Spain by Nadal, Jordi, El fracaso de la revolución industrial en España, 1814–1913 (Barcelona: Ariel, 1975)Google Scholar.

⁶⁴ Reher, David S., Perspectives on the Family in Spain: Past and Present (Oxford: Oxford University Press, 1997), pp. 48–73Google Scholar; Douglass, William A., ‘The Basque Stem Family Household: Myth or Reality?’, Journal of Family History, 13: 1 (1988), pp. 75–89CrossRefGoogle Scholar; Azpilicueta, Pilar Erdozáin and Peña, Fernando Mikelarena, ‘¿Existen las estrategias demográficas colectivas? Algunas reflexiones basadas en el modelo demográfico de baja presión de la Navarra Cantábrica en los siglos XVIII y XIX’, Revista de Demografía Histórica, 21: 2 (2003), pp. 13–58Google Scholar. Though focusing on French Basques, Arrizabalaga shows how impartible inheritance persisted despite adverse legal circumstances: Arrizabalaga, Marie-Pierre, ‘Succession Strategies in the Pyrenees in the 19th Century: The Basque Case’, History of the Family, 10: 3 (2005), pp. 271–92Google Scholar.

⁶⁵ Quoted in Carr, Raymond, Spain 1808–1975 (Oxford: Clarendon Press, 1982)Google Scholar, p. 19; Swinburne, Henry, Travels Through Spain in the Years 1775 and 1776 (London: Elmsly, 1787), pp. 327–34Google Scholar.

⁶⁶ Clara Eugenia Nuñez, La fuente de la riqueza, pp. 92–3.

⁶⁷ Despite Spain's linguistic-cultural regional differences, especially the unique case of the Basque language (Euskera) as a non-Romance language, schooling was in Castilian (Spanish), the ‘high language’ used in administration and commerce, and hence required for upward mobility: Atxurra, Rafael López, ‘Historia de las instituciones educativas en Euskal Herria. La enseñanza primaria en el antiguo régimen: pautas para la investigación’, in Gracia, Rafael Mieza y Juán (eds.), Haciendo historia (Bilbao: Universidad del País Vasco, 2000), pp. 419–45Google Scholar.

⁶⁸ AGN, Matrimonios, 230, 16. Located at a distance from home, school was also a site of socialisation long before emigration.

⁶⁹ I included Old Castile and Catalonia for additional comparisons, though they should be treated with caution because of the small numbers of migrants from these regions. Still, the Old Castilian population in Cadiz somewhat resembled its northern neighbours. The Catalan population had a low share of lower class migrants, similar to the prosperous Basques, Navarrese and Riojans.

⁷⁰ A similar northern predominance in commerce, excluding Galicia, was observed for Madrid in Sola-Corbacho, Juan Carlos, ‘Family, Paisanaje, and Migration among Madrid's Merchants (1750–1800)’, Journal of Family History, 27: 3 (2002), p. 7Google Scholar.

⁷¹ This is illustrated by the socio-economic indicators discussed. Andalusians were even more negatively selected as they could benefit from proximity to Cadiz. Out of the few servants (32) among peninsulares in Mexico City, 40 per cent were Andalusian: González Angulo, ‘Los inmigrantes’, p. 118.

⁷² Coatsworth argued that the crown prevented secure property rights, even for elites, in order to keep them unstable. It seems that under such circumstances economic elites were rather competitive, though not in a Smithian sense, a situation which may have facilitated the shift in origins. See Coatsworth, John H., ‘Inequality, Institutions and Economic Growth in Latin America’, Journal of Latin American Studies, 40: 3 (2008), especially pp. 556–62Google Scholar.

⁷³ Reasons to discount the importance of peninsulares' better connections in an administration dominated by compatriots are the same as those discussed above: the regional discrepancy between independent and official immigrants, and the mercantilistic thrust of a revenue-thirsty crown. Viceroy Iturrigaray of New Spain (1803–8), who even showed preference for creoles, was deposed by a peninsular-led revolt protesting against the Law of Consolidation: Caballero, Romeo Flores, Counterrevolution: The Role of the Spaniards in the Independence of Mexico, 1804–1838 (Lincoln: University of Nebraska Press, 1974), pp. 14–38Google Scholar; Burkholder, Mark A., Spaniards in the Colonial Empire: Creoles vs. Peninsulars? (Oxford: Wiley-Blackwell, 2012)Google Scholar, especially pp. 110–28. This is also suggested by the high socio-economic status of Spaniards after independence, especially during the Porfiriato: Lida, Clara E and Zamudio, Pilar Pacheco, ‘El perfil de una inmigración: 1821–1939’, in Lida, Clara E. (ed.), Una inmigración privilegiada: comerciantes, empresarios y profesionales españoles en México en los siglos XIX y XX (Madrid: Alianza Editorial, 1994), pp. 25–51Google Scholar.

⁷⁴ Engerman, Stanley L. and Sokoloff, Kenneth L., ‘Factor Endowments, Institutions, and Differential Paths of Growth Among New World Economies: A View from Economic Historians of the United States’, in Haber, Stephen (ed.), How Latin America Fell Behind: Essays on the Economic History of Brazil and Mexico, 1800–1914 (Stanford, CA: Stanford University Press, 1997), pp. 260–304Google Scholar; Acemoglu, Daron, Johnson, Simon and Robinson, James A., ‘The Colonial Origins of Comparative Development: An Empirical Investigation’, American Economic Review, 91: 5 (2001), pp. 1369–401Google Scholar; Coatsworth, ‘Inequality, Institutions and Economic Growth’. Despite diverging views on the relationship between institutions (property rights) and inequality, all agree that inequality institutionalised in the colonial era had a long-term impact.

⁷⁵ Games, Alison, ‘Atlantic History: Definitions, Challenges, and Opportunities’, American Historical Review, 111: 3 (2006), pp. 741–57CrossRefGoogle Scholar; Armitage, David, ‘Three Concepts of Atlantic History’, in Armitage, David and Braddick, Michael J. (eds.), The British Atlantic World, 1500–1800 (New York: Palgrave Macmillan, 2002), pp. 11–27Google Scholar; Greene, Jack P. and Morgan, Peter P. (eds.), Atlantic History: A Critical Appraisal (Oxford: Oxford University Press, 2008)Google Scholar.