1 A.P. Giannini (1870–1949) founded in 1904 the institution that would become the Bank of America. As an aside, it seems appropriate that a dissertation about his bank’s contributions should be written at Davis. In 1928, Bank of America set up an endowment of $1.5 million to establish the Giannini Foundation of Agricultural Economics at the University of California, whose activities largely involve the Davis, Berkeley, and Riverside campuses.

2 Bank of America’s “chief contribution was not in money, however, but in counsel and leadership in devising and putting through plans for refinancing,” according to James and James (1954, p. 406). Part of the success of branch banking was in the resemblance to universal banking, such as the ability of the central institution to facilitate upstream and downstream alliances with greater efficiency, owing to its access to information about all parties.

1 Bank of America’s loan-deposit ratio remained stable at 0.64 from 1929 to 1933, unlike the average for the rest of the state, which fell from 0.70 to 0.54 over the same period. This was driven by loans, not deposits; Bank of America’s lending decline from its 1929 peak to 1934 trough was half the California unit bank median over that period.

2 See Fishback (2017) for a recent review on the recovery from the Depression, particularly for the role of government policy.

3 An American Legion survey indicates that on average, veterans planned to spend 16 percent on buying a new home, 8 percent on housing improvements, and 31 percent on old bills and debts, including mortgages.

4 Bank of America advertised that it lent to college students and helped future attendees save for college, suggesting that using the Bank of America network as an instrument for economic activity changes would violate the exclusion restriction. Instead, we interpret Bank of America’s effect during the 1930s as a joint local credit and business cycle measure.

1 Such outcomes include a child’s educational attainment, economic mobility, and longevity (e.g., Chetty et al. 2014; Chetty et al. 2016; Chetty and Hendren 2018).

2 The HOLC gave letter grades A, B, C, and D to neighborhoods with map colors of green, blue, yellow, and red, respectively.

1 See, for example, Choudhri and Kochin (1980), Tortella and Palafox (1984), Bernanke and James (1991), Eichengreen (1992), Temin (1989, 1993, 2008), or Grossman (1994). Garcia Ruiz (1993), on the contrary, pointed to liquidity issues, while Martin-Aceña (1985) documented the monetary contraction of the 1930s.

2 See, among other more recent contributions, Rey (2015).

3 See, for a recent long-run take on the matter (Baron, Verner, and Xiong 2020).

4 See Accominotti and Eichengreen (2016) for the role of capital flow reversals and sudden stops during the Great Depression.

5 A revised version of this chapter is published in the European Journal of Economic History (Jorge-Sotelo 2020a).

6 A revised version of the appendix of this chapter is available in First View at the Revista de Historia Económica/Journal of Iberian and Latin American Economic History (Jorge-Sotelo 2020b).

7 The classic reference is Bagehot (1873), but see Ugolini (2017) for a discussion of more recent contributions on the rationale for raising interest rates under fiscal and/or external constraints.

8 The 1929 Wall Street Crash did not have a strong impact on the Spanish stock market, and asset prices remained relatively high during 1930.

9 Recent estimates put the contraction of economic activity and financial intermediation in Spain in a comparative perspective. Prados de la Escosura’s (2017) estimates of GDP per capita between 1929 and 1933 show that the contraction was similar to the one experienced by France, Italy, Belgium, and not too far from Germany. Higher frequency estimates of economic activity provided by Albers (2018) reinforce this idea, while Eichengreen and Mitchener (2003) show the drop in financial intermediation in Spain is comparable to the one experienced by the United States or Germany.

1 Democratic transition is another factor put forward in the literature (Glaeser, Ponzetto, and Shleifer 2007). The second phase of industrialization impacted positively education thanks to increased complementarity between formal skills and industrial tasks (Franck and Galor 2017; Diebolt, Le Chapelain, and Menard 2021). Inequality in land distribution, however, decreased both the presence of institutions promoting human capital (Banerjee and Iyer 2005; Galor, Moav, and Vollrath 2009) and private demand for education (Beltrán Tapia and Martinez-Galarraga 2018; Cinnirella and Hornung, 2016). Religious factors were also relevant, as Protestantism positively influenced education (Becker and Woessmann 2009).

2 It has been argued that the progressive introduction of large-scale factories was skill-saving, diminishing the demand for instruction and literacy rate (de Pleijt 2018). This was especially true if steam engines were favoring the use of machines on which children were performing secondary tasks in assisting older workers (Nardinelli 1980) and came under the name of the “deskilling hypothesis” (Sanderson 1972). However, in the case of France, there seems to have been a complementarity between technology and human capital (Franck and Galor 2017; Diebolt, Le Chapelain, and Menard 2017).

3 In the British case, literacy has been deemed useless to perform the vast majority of occupational tasks in the mid-nineteenth century (Mitch 1993) and irrelevant to explain growth at the country level (Allen 2003). More recent studies, however, found that primary education was the main driver of growth in Britain until the mid-eighteenth century (Madsen and Murtin 2017). Upper-tail knowledge has also been associated with pre-industrial economic growth (Baten and van Zanden 2008; Squicciarini and Voigtländer 2015). Average education was also related to the industrial development of Prussia during the nineteenth century (Becker, Hornung, and Woessmann 2011; Cinnirella and Streb 2017).

1 Notable exceptions include Carroll (2002), Lainer-Vos (2013), and Ridley (2018). O’Sullivan Greene (2020) was published after the submission of the thesis. None investigate the identities of the subscribers.

2 Hopkinson (2002, p. 26).

3 Mitchell (1995).

4 Carroll (2002, pp. 9, 15).

5 For example, Doorley (2005), Brundage (2016), Schmuhl (2016), Nyhan-Grey (2016), Keogh (2016), and Murray (2018).

6 UCD Archives, Bernard O’Rouke papers, P117/68, South Monaghan NL Register; Maura Ní Riada Private Collection (hereafter MNR), 1, Longford NL Register; East Tipperary NL Register, MNR/2/A.

7 Cork City & County Archives, Seamus Fitzgerald Papers, PR6/70/2/180; DÉ debates on DÉ Loans and Funds Bill, 20 Jun. 1928, 11 Oct. 1928, 6 Mar. 1929, 1 May 1929, 29 May 1929, 12 Jun. 1929, 20 Jun. 1929, 7 Nov. 1929, 5 Mar. 1930, 9 Apr. 1930, 14 May 1930, 4 Jun. 1930, 10 Dec. 1930, 18 Mar. 1931, 25 Mar. 1931, 27 May 1931, 21 Oct. 1931.

8 For example, TNA: CO 904/108-11; Thom’s Directory 1921; National Library of Ireland, Ms.3,110, Solemn covenant to resist Conscription, 21 Apr. 1918; O’Malley (2007); Monaghan County Museum, 1986/5N1, Interviews with IRA veterans; Bureau of Military History (hereafter BMH), WS, Witness Statements; BMH/MA, IRA military pension applications.

9 Controlling for relevant variables in each case.

10 National Archives of Ireland, DE/EXL/Vol.80, Manhattan subscriber list; 1920 Federal U.S. Census.