Political institutions, banks, and economic growth: Evidence from the United States and Mexico, S Haber

Tags: Mexico, government, United States, banking system, federal government, institutions, Sylla, Economic Growth, Stanford University, Stephen Haber, Bodenhorn, state governments, economic development, Ross Levine, the United States, political institutions, Political Economy, James Robinson, democratic institutions, Pennsylvania, bankers, the state, the government, Maurer, Development, Mexico City, banking systems, Cambridge University Press, Bank of the United States, Simon Johnson, American History, National Bureau of Economic Research, Industrial Development, American Economic Review, Northern Illinois University Press, Journal of Financial Economics 58, Economic History, Journal of Economic History, Oxford University Press, Quarterly Journal of Economics, public finance, University of Chicago Press, Bank of North America, free bank, chartered banks, federal bonds, central government, barriers to entry, free banking, Kenneth Sokoloff, Richard Sylla, private banks, bank charter, Pennsylvania State Legislature, Banamex, Armando Razo, state governors, state legislatures, commercial bank, federal system of government, state government, bank director
Content: Political Institutions, Banks, and economic growth: Evidence from the United States and Mexico Stephen Haber Date of this Draft: December 16, 2004 Abstract: A broad social science literature has focused on the role played by political institutions in economic development. Social scientists remain uncertain, however, about a number of crucial questions. Do democratic political institutions cause growth, or does growth cause democratic institutions? Which institutions, in particular, are crucial for growth, and which are merely incidental to it? Where do political institutions come from and how do they evolve? This paper offers a contribution to the literature by focusing on how political institutions that foster (or hinder) competition in the political system affect the supply of a crucial input to the process of growth--the banking system. Drawing on the political and economic histories of the United States and Mexico, I argue that there were attempts by government in both cases to limit the supply of banks and the range of services they could provide. The political institutions of the United States, however, meant that those constraints tended to be quickly undermined. In Mexico, however, the lack of institutionalized political competition allowed the government to constrain the supply of banks over the long run. Support for this research was provided by National Science Foundation Grant SBR-9515222, the William and Flora Hewlett Foundation's Program in U.S.-Latin American Relations, and the Bechtel Initiative of the Institute of International Studies at Stanford University. Earlier versions of this paper were presented at the 2002 Meeting of the American Economics Association, at Stanford's Center for Research on Economic Development and Policy Reform, and at the 2004 Stanford Conference on the History of Congress. Richard Bensel, Jeff Frieden, Ross Levine, Noel Maurer, James Robinson, Paul Sniderman, Gabriel Sod Hoffs, Kenneth Sokoloff, Richard Sylla, Barry Weingast, and Scott Wilson made helpful comments on earlier drafts of this paper.
Over the past decade, social scientists have created a broad literature on the role of institutions in economic development. That literature demonstrates that there is a relationship between democratic governance, secure private property rights, and high levels of material welfare. (Barro 1991, 1997; Levine and Renelt 1992; Brunetti 1997; Przeworski et. al. 2000). Social scientists do not yet, however, fully understand the mechanisms that underlie this relationship. Which way does causality run? Do democratic institutions promote economic growth? Does economic growth promote democratic institutions? Or, are democracy and growth jointly caused by a third, unspecified, set of institutions? (Acemoglu, Johnson, Robinson, and Yared 2004). Social scientists are similarly unsure about how and why particular political institutions affect the security of property rights. (Keefer, forthcoming). Finally, social scientists are unsure about where institutions come from and how they evolve. Some broad theories have been proposed (Engerman and Sokoloff 1997; Bates 2001; Acemoglu, Johnson, and Robinson 2001, 2002, 2004), but there is, as yet, no consensus view. This paper offers a contribution to the literature on institutions and growth by exploring the origins and dynamics of a particular set of institutions--those that govern the development of banking systems. I focus on banking because there is a broad consensus among economists that banks are a crucial part of the financial system and that an efficient financial system is a necessary requisite for economic growth.1 World history provides no case of a society with a high per capita income 1 King and Levine 1993a, 1993b; Levine and Zervos 1998; Beck, Levine, and Loayza 2000; Rajan and Zingales 1998; Beck, Demirgьз-Kunt, Levine, and Maksimovic 2001; Levine 1997; Neusser and Kugler 1998; Rousseau and Wachtel 1998; La Porta, Lopez-de-Silanes, Shleifer, and Vishny 2000. 2
without a well developed financial system and no case of a well developed financial system without a banking system.2 The argument I advance runs in the following terms. The size and structure of banking systems are influenced both by the demand for, and the supply of, financial services. The demand for banks is an endogenous outcome of the size and structure of the real economy. When wealth is highly concentrated and when the overall level of development is low, the demand for banks is modest. As economies grow, and as wealth becomes more broadly distributed, the demand for banks increases. The supply of banks does not, however, respond automatically to increases in demand for their services. This can be the case for any of four reasons: bankers fear expropriation by the government; the government purposely limits the number of banks; bankers cannot enforce financial contracts with debtors; and investors (depositors and outside shareholders) fear that bankers will behave imprudently, and therefore do not deploy their wealth in the banking system. In order to overcome these problems, bankers, outside shareholders, depositors, debtors, and the government must create institutions that align their incentives. Societies have experimented with a number of different institutional solutions, but what they all tend to have in common is an active role for the government. To cite just a few examples, the government can compensate bankers for expropriation risk by raising bank rates of return through regulations that limit competition or by granting lucrative privileges to some banks. Similarly, the government can be charged with enforcing financial contracts through its system of laws, courts, and police. Or, the government can mitigate 2 In addition to a system of private banks, financial systems also include securities markets, a central 3
the risk to outside investors by granting limited liability to shareholders and insurance to depositors. It might also create rules of prudent behavior by bankers, and enforce these through various mechanisms, including supervisory agencies and deposit reserve requirements. The active role of government creates a thorny problem: at the same time that the government playes a role in the creation of the institutions that make the banking system possible, it looks to the banking system as a source of finance for its own operations. This means that the government has an inherent conflict of interest. It may structure the institutions that govern banking so as to meet the demand that comes from the private economy, or it can structure the institutions that govern banking so as to facilitate its own political survival. It might create bank monopolies that share rents with the government. It may allocate charters only to politically-favored constituents. Government officials may demand bribes in order to grant a bank charter. Or, the government may refuse to enforce financial contracts when the debtors are from politically crucial groups. Precisely because government has this conflict of interest, the political institutions that limit the authority and discretion of government play a crucial role in the development of the banking system. These political institutions vary across societies. There does not appear to be any single algorithm for their optimal organization. Nevertheless, these institutions operate according to a single unifying principle: they create institutionalized forms of competition within the political system that allow the ambitions of individuals or groups to be counteracted by the ambitions of other individuals or groups.3 bank, a system of money, and a system of public finance. 3 This paper therefore builds upon the literature on "limited government." The central idea of that literature--that political institutions must be structured so that ambition may be able to counteract 4
The presence of institutionalized competition--through such institutions as electoral suffrage, political parties, separation of powers, and federalism--increases the probability that there will be ex ante vetos over policies that allow the government's own short-term needs to take precedence over those of economic development. Similarly, the presence of institutionalized competition increases the probability that there will be ex post sanctions for public officials or government entities that exceed their authority or who engage in rent seeking at the expense of economic development. The lack of institutionalized political competition does not necessarily mean that there will be no banking system at all. It does imply, however, that the institutional solutions to expropriation risk, default risk, and imprudent behavior are likely to be inefficient. As we shall see below, the interaction of these inefficient institutions strongly works against the development of the banking system. Indeed, it gives rise to systems in which there may be few banks, and in which banks severely limit the range of services they provide to individuals and firms. I develop the logic and evidence of these claims in the pages that follow. Section II of this paper provides a theoretical framework. Section III discuss the methods and approaches to evidence of the paper. Section IV explores how the development of banking systems was conditioned by the existence of institutionalized political competition in two widely divergent cases: the United States and Mexico. Section V concludes. II. THEORETICAL FRAMEWORK ambition--goes back at least as far as James Madison's writings in the Federalist Papers. In recent years these ideas have generated a sizable social science literature, exemplified by North 1981: 154-57; Levi, 1988; North and Weingast 1989; North 1990; Weingast, 1995, 1997a, 1997b; North, Summerhill, and Weingast, 2000; Bates 2001. 5
The Problem of Expropriation: In order for individuals to invest their wealth in a bank they must believe that the returns to them outweigh the risk that the government will expropriate the bank. Expropriation, it should be pointed out, does not have to take the form of the government seizing bank assets--although history offers us numerous cases of this phenomenon. Expropriation can also take more subtle forms: the government can renege on loans made to it by banks; it can force banks to lend to it at belowmarket interest rates; it can raise taxes to the point that profits are below the opportunity cost of capital; it can require banks to hold reserves against deposits in the form of government bonds that pay negative real rates of interest; or it can expand the money supply, setting off an inflation that amounts to a tax on the holders of cash. The problem of expropriation means that in order for bankers to deploy their wealth there must be institutions that align the incentives of the government and the bankers. One solution is the creation of political institutions that limit the authority and discretion of government. In cases where the government's authority and discretion is not limited, however, the incentives of bankers and government are aligned by the creation of institutions that raise the rate of return to banking high enough to compensate bankers for the risk of expropriation. These institutions typically allocate a set of lucrative privileges to one or more banks and/or limit the number of banks allowed in any market. That is, the government takes a direct role in structuring the market because it is in the interest of both the government and the bankers that it do so. 6
The Problem of Contract Enforcement: Bankers will not write financial contracts if they cannot those contracts. They therefore need to align the incentives of debtors with their own by credibly threatening to take possession of the physical assets represented by debt contracts. There are a broad range of institutions necessary to make this threat credible. Bankers must be able to draw upon institutions that allow them to determine who owns a particular asset (a property register, for example) and that allows them to repossess that asset in the case of default (a system of laws regarding bankruptcy and foreclosure, an efficient system of courts, and a police force with the power of coercion). These institutions, it must be emphasized, emerge only in part because of demand by bankers. More fundamentally, they emerge because governments have a vested interest in creating mechanisms that allow them to monitor and enforce taxation, and because individuals and firms have an interest in protecting their property rights against encroachment by other individuals and firms. Indeed, firms and individuals have strong incentives to clearly demarcate (make transparent, in the parlance of property rights theory) and enforce property rights: transparent and enforceable property rights are easier to transfer--and hence are more valuable. The need of governments, individuals, and firms to create institutions that make property rights transparent and enforceable creates a fundamental dilemma. The same institutions that make property rights more transparent and enforceable make them more subject to expropriation by the government. In societies in which the authority and discretion of government is limited by political institutions, the threat of expropriation does not loom large. In societies in which the authority and discretion of the government are not limited, however, individuals and firms have weak incentives to lobby for the creation of institutions that make their property rights transparent and 7
enforceable. That is, if the government has unlimited discretion and authority, it is not in the interest of individuals and firms to have an efficient system of laws, courts, and police. Indeed, a government with unconstrained discretion that possesses efficient police and courts is likely to be a tyranny. It logically follows that when the government has unlimited authority and discretion the institutions that make property rights transparent and enforceable tend to be weak. It is not so much that the population tries to frustrate the development of efficient property laws, property registers, courts, and police--although it may do so. It is that the population does not lobby for reforms in these institutions that would make them more efficient. Over time, the effect is the same: disorganized property registers, arcane laws, poorly paid and inefficient police, and a court system that is cumbersome and subject to bribery. These inefficient institutions are reinforced by the government's inability to collect taxes. Precisely because property rights tend not to be transparent when government is not limited, the government cannot easily tax wealth or transactions involving wealth. Hence, even if the government wanted to make the institutions that specify and enforce property rights more efficient, it would not have the fiscal resources to do so. The result is a paradox: unconstrained governments tend to be poor and weak, while governments whose authority and discretion are limited tend to be wealthy and powerful (North and Weingast 1989). When property rights are not transparent and enforceable at low cost, bankers respond by limiting the types of contracts that they write. Alternatively, they may limit the range of individuals and firms with whom they write contracts. As we shall see below, one strategy they may pursue is to lend primarily to members of their own families, to other bank directors, and to themselves because 8
they can monitor and enforce those contracts without recourse to the society's inefficient legal institutions. The Problem of Imprudent Bankers In order for banks to grow beyond the wealth of their initial shareholders, they must attract the wealth of outside individuals and firms. These outsiders (most typically depositors and minority shareholders) will not, however, deploy their wealth if they perceive that the bankers will behave imprudently--that is, write credit contracts that have a high probability of default and/or a low probability of enforcement. At the extreme, bankers may loan the funds provided by outside investors to themselves, and then default on the loans--in effect looting their own banks. Thus, institutions must be created that align the incentives of the depositors, the outside shareholders, and bank directors. Alternatively, institutions can be created that minimize the extent to which the wealth of depositors and shareholders is at risk, even if the bankers are imprudent. Different societies have experimented with a range of institutions to align the incentives of bank directors, shareholders, and depositors. Some of these institutions, such as outside directors, are created out of the interaction of the bank directors and outside shareholders. Other institutions, however, are created by the government. These include laws that require banks to hold minimum levels of capital or that require banks to create reserve accounts that provision for risk. Governments may also create institutions that align the incentives of outsiders with those of directors by reducing the risk that those outsiders face. One common example is limited liability for shareholders. Another common example is deposit insurance. The Centrality of Political Institutions: 9
All of these institutions have a common thread: they directly involve the government. The government structures the banking system--allocating privileges, regulating entry by controlling bank charters, and limiting the types of contracts that may be written by banks. The government also enforces financial contracts through its system of laws, property registers, courts, and police. Finally, the government plays a key role in aligning the incentives of depositors and outside shareholders with bank directors and managers: it creates laws governing limited liability, reserves against risk, and the rights and responsibilities of bank directors; it creates and funds the agencies that supervise and regulate banks; and it creates (and often funds) the system of deposit insurance. The problem is that the government is not a disinterested party in this process. At the same time that the government specifies and enforces the rules that govern banking, it also looks to the banking system as a source of finance. These sources include revenues from taxes on bank capital or bank profits, dividend income from the ownership of bank stock, credit lines from banks, or the mandatory purchase of government bonds as a chartering or reserve requirement. Precisely because government has a conflict of interest, the political institutions that limit the authority and discretion of government play a crucial role in the development of the banking system. As we shall explore in detail below, the political institutions that limit the authority and discretion of government by promoting political competition interact, reinforcing one another. Indeed, the evidence that we present below suggests that political institutions that constrain government co-evolve. A somewhat parallel process occurs when governments are unconstrained by institutions that promote political competition. In these cases, bankers and governments create institutions that mitigate the problems of expropriation, contract default, and imprudent behavior. These institutions also interact, but they do so in ways that constrain the number of banks and the services that they will offer. 10
III. CASE SELECTION Operationalizing this argument requires that we proceed in three steps. First, we need to understand the origins and evolution of the political institutions of particular societies. Second, we need to understand how those political institutions structured the economic institutions that governed banking systems. Third, we need to understand how those economic institutions affected the supply of banks. Connecting the causal links in this chain requires that we depart from the kinds of crosscountry regression techniques that are standard in much of the financial economics literature, and instead draw on the historical records of particular countries. In an ideal world, we would focus on a large number of case studies. Understanding the evolution of political institutions, the economic institutions that govern banking, and the subsequent structure and performance of banking systems is not, however, an enterprise characterized by increasing returns. Inevitably, the more cases one includes, the less one is bound to know about each case. I therefore focus on two extreme cases--the United States and Mexico during the period 1781-1932.4 I focus on the United States because it was characterized by institutionalized political competition on four dimensions: electoral suffrage, organized political parties, separation of powers within the central government, and federalism. As the ensuing pages shall show, these different forms of competition worked to reinforce one another in ways that were not, in fact, fully foreseen at the time that the Constitution was written. 11
The net effect of these institutions was an equilibrium: political actors articulated their demands and adjudicated their differences by staying within the governance structure, rather than going outside it by making demands through the use of violence. There was, as Hofstadter (1969) put it, a "legitimate opposition"--one that used formal institutions to oppose the policies of the government, rather than oppose the government itself. This equilibrium broke down over only a single issue--slavery, requiring that differences in interests on this issue be adjudicated not through institutional channels, but through armed force. I focus on Mexico because it was characterized by an almost complete absence of institutionalized political competition of any type. This is not to say that there were not formal rules governing political action. In fact, Mexico had various constitutions--some federal and Republican, others centralist and decidedly anti-Republican. The problem was that the sets of political institutions that were created by these constitutions were not mutually reinforcing. Instead, Mexico had alternating periods of federal republics and centralized dictatorships, punctuated by coup d'tats, civil wars, and foreign interventions. Thus, during the first 55 years after its independence from Spain (1822-1876), most political competition in Mexico took place outside of the channels prescribed by the country's formal institutions--which is to say that it took place at gunpoint. Indeed, those first 55 years saw 75 presidents, with one military leader serving as president on 11 different occasions. 4 We break off our analysis in 1932, because the creation of the Federal Deposit Insurance Corporation in the following year gradually weakened the ability of states to charter and regulate banks independently of the federal government. 12
In the last decades of the nineteenth century Mexico finally obtained political stability, under the dictatorship of Porfirio Dнaz (1876-1911). Dнaz drew upon a pre-existing constitution to craft a governance structure that was nominally federal and Republican, but that was in fact rule by a network of political and economic elites who used the government's regulatory powers to create and share rents with one another. In short, neither during the period when the political system was unstable (1821-1876) nor when it was stable (1876-1911) was there institutionalized political competition in Mexico. These differences in political institutions had a powerful effect on long-run economic outcomes. In the United States, attempts by governments to limit the supply of banks in the face of increasing demand for banking services tended to be sustainable only for short periods of time. The nature of U.S. political institutions allowed the population to quickly undermine whatever restrictive arrangements had been made. In 1800, the U.S. banking system was characterized by segmented monopolies erected by state legistures and a private commercial bank with special privileges that served as the financial agent of the federal government. Within a few decades this organization of banking markets was completely transformed: there were thousands of banks, the overwhelming majority of banks had no branches, and the federal government had no bank at all. It was not until 1913 that the federal government would once again have access to an institution that could serve as a central bank (when the Federal Reserve was founded), but by that time there were more than 25,000 banks in operation in the United States. In Mexico, on the other hand, the absence of institutionalized political competition meant that the banking system always remained small--even in the face of increasing demand for banking services. During the decades before 1876, Mexico's political institutions meant that bankers faced 13
expropriation risk. They could also not count on the government to enforce their contract rights. As a consequence, there was virtually no chartered banking system at all. In the decades after 1876, the economy grew rapidly. Nevertheless, even as demand for financial services increased, the supply of banks was constrained. In exchange for limiting the number of charters granted, bankers shared some of their rents directly with the federal treasury and with government officials as individuals. The result was a market structure virtually the opposite of the United States: two banks with national branch networks controlled 60 percent of total assets; virtually no market had more than three banks in it. Moreover, unlike the United States, Mexican bankers faced an environment in which it was not possible to enforce their property rights. They therefore lent most of the funds in the banking system to themselves. The combination of a small number of banks and insider lending meant that there was differential access to capital in Mexico: a small group of entrepreneurs were awash in funds, while everyone else was starved for credit. IV. CASE STUDIES The United States During the early republican period, banking in the United States was characterized by segmented monopolies. In exchange for regulations that protected banks from competition, bankers shared their monopoly rents with state treasuries--and with state legislators as individuals. The central government had a commercial bank as well, the Bank of the United States--modeled on the 14
Bank of England, which had a series of special privileges, including the sole right to operate branches across state lines. These arrangements constraining the supply of banks, however, were inconsistent with the fact that U.S. political institutions fostered political competition. Moreover, that competition did not just come from the fact that the United States had democratic elections--though the existence of electoral suffrage figures as part of our story. In addition to electoral suffrage, four other institutions were crucial to the erosion of limits on the supply of banks: the development of political parties that could articulate coherent policy options; the existence of horizontal competition among states for capital and labor; the existence of vertical competition between states and the central government; and the existence of separation of powers within the central government, which gave states the ability to use their congressional and senate delegations to undermine federal initiatives.5 Political Institutions, Public Finance, and Bank Chartering The political institutions of the United States began to take shape even before independence was achieved in 1781. During the colonial period, each colony had essentially governed itself through a system of elected colonial assemblies and an appointed British governor. When independence came, the British governors and their supporters were thrown out, electoral suffrage was broadened somewhat, and the colonial assemblies became state legislatures. There was never 5 Relevant literatures on these institutions include McCubbins and Schwartz 1984; Weingast 1995; Tsebelis 1995; Haggard and McCubbins 2001. 15
any doubt that the United States would have democratic elections and a federal system of government. What was up for debate, however, was how strong the central government would be. The first experiment produced an extremely weak central government, under the Articles of Confederation. The new government did what almost all new governments do: it founded a commercial bank--the Bank of North America (BNA)--whose purpose was to serve as the government's fiscal agent. The problem was that the Articles of Confederation was ambiguous as to whether the central government had the authority to charter a bank. The BNA was therefore rechartered by the state of Pennsylvania ­and then almost immediately came into conflict with the Pennsylvania State Legislature. The BNA had been founded with the understanding that it would hold a monopoly on the issue of paper currency. It therefore took a strong position regarding the fact that the states were issuing bills of credit, which was paper money by another name. In retaliation, the Pennsylvania State Legislature revoked the BNA's charter in 1785. After a two year battle, the BNA got its charter restored, but only after accepting a series of restrictions on its activities. (Bodenhorn, 2003: 128). The extreme federalism of the Articles of Confederation soon proved unworkable because the central government lacked any independent taxation authority. In 1789, the Articles were replaced with the Constitution, which conferred on the central government a great deal more policy and taxation authority--in exchange for which it assumed the (quite considerable) debts that had been racked up by the states. Nevertheless, the American political system was still strongly federal: all policy-making authority not explicitly delegated to the federal government by the Constitution was vested in the states. 16
The delegation of power to the states meant that they had the right, and the incentive, to charter and regulate banks. Under the Constitution, the states lost both the right to tax imports and exports and the right to issue paper money--both of these powers were vested with the federal government. These constitutional provisions created two problems for state finances: tax revenues were dependent on cumbersome poll and property taxes; and states could no longer finance their expenditures in excess of taxes by issuing paper money. The response of the states was to sidestep the Constitution: It might have been the case that states could no longer issue paper money, but the Constitution said nothing about states chartering banks of issue, whose banknotes would circulate as currency. (Sylla, Legler, Wallis, 1987). The ability to charter banks allowed states to find a ready source of finance. A bank charter was not just a license to do business, it conferred two very valuable concessions on its holders--the right to issue banknotes (and thereby profit from seignorage), and the right to limited liability. This meant that potential bankers were willing to pay handsomely for a charter, especially if they believed that they were receiving the only one. States, for their part, had every incentive to sell charters so that they could fill their treasuries. Thus, bankers offered (and states demanded) the payment of charter "bonuses." These bonuses came in various forms--one-time cash payments, the promise to lend to the state government at preferential rates, or the ceding of blocks of bank shares to the state (which generated a stream of dividend income). (Sylla, Legler, and Wallis, 1987; Wallis, Sylla, and Legler, 1994; Bodenhorn, 2003: 15). state governments therefore owned sizable amounts of bank stock. In Virginia, for example, the state owned one-fifth of the stock in the Bank of Virginia (chartered in 1804)--a purchase that it financed by borrowing the funds from that very same bank. It later simplified matters by demanding 17
that new banks provide the state with shares gratis. In Massachusetts, the state government was a major investor in the Union Bank (chartered in 1793) and the Bank of Boston (chartered in 1803), giving it control of one-eighth of all banking capital until 1812. In Pennsylvania, the state owned more than $1 million in bank shares, most of it in the Bank of Pennsylvania, whose concession made it the state's fiscal agent. New states, as they entered the union, copied the chartering model of the original 13 states: all of them were major owners of bank stock. Kentucky, Tennessee, Illinois, Arkansas, and Alabama went even further: their first banks were 100 percent state-owned. Indeed, the basic pattern of state governments being major bank stockholders held just about everywhere--with the exception of New York, where (as we will discuss below) bankers did not just share rents with the state treasury, they shared them directly with legislators as individuals.6 The chartering of banks helped solve the problem of state government finance. Sylla, Legler, and Wallis have calculated that circa 1810, bank dividends and bank taxes accounted for six percent of total state revenues in New York, nine percent in Connecticut and South Carolina, 12 percent in Delaware and Virginia, 29 percent in Maryland, and 38 percent in Pennsylvania. These 1810 estimates do not include a number of states that we know (from subsequent observations) relied heavily on bank dividends or taxes. Thus, their first observation for North Carolina is in 1820 (when banks accounted for 31 percent of state revenues), and their first observation for Massachusetts is in 1830 (when bank taxes accounted for an amazing 61 percent of state revenues). Moreover, none of these figures include one-time cash payments by banks to state treasuries, the market value of bank 18
stock distributed gratis (or discounted) to states, nor the transfers created by bank-financing of public works projects or credit lines provided to states at favorable rates. (Wallis, Sylla, and Legler, 1994: 126). The financing of state expenditures via chartering bonuses created a problem of moral hazard: it created incentives for incumbent banks to offer bonuses to state legislatures to deny the charter applications of potential competitors; and it created incentives for state legislatures to accept those bonus payments--unless the newcomer was willing to offer a substantial share of its future stream of rent.7 It was, therefore in the interest of state governments to restrict entry into banking in order to maximize the amount of rent earned by incumbent banks, and then share in those rents via the dividend stream created by the state's ownership of bank stock. This meant that states had to make sure that there was not a competitive fringe of private banks that operated without charters. They therefore nearly all passed laws that required private banks to obtain a corporate charter from the state government--and then turned down many of those charter applications. (Bodenhorn, 2003: 17, 244). In some states, problems of moral hazard extended beyond the incentives of the state treasury and affected the behavior of legislators as individuals. The most notorious such case was New York. From the 1810s to the late 1830s, bank chartering in New York was controlled by the so-called Albany Regency--a political machine controlled by Martin Van Buren. Bank charters 6 Sylla, Legler, Wallis, 1987; Wallis, Sylla, and Legler, 1994; Bodenhorn, 2003: 15, 84, 234-35; Bodenhorn 2004. 7 The phenomenon of competing for bank charters by offering shares of the future rents appears to be an example of the phenomenon modeled by Krueger (1974), in which rent seekers expend rents up the 19
were only granted to friends of the Regency. The incentives of legislators were aligned with the bankers by allowing the former to subscribe to initial public offerings of bank stock at par, even though the stock traded for a substantial premium because the banks earned monopoly rents. In addition, banks also made direct bribes to legislators. Not surprisingly, in 1818 the Regency expanded the state laws (initially passed in 1804) that restricted unchartered institutions from providing banking services. It later amended the state constitution (in 1821) so that new bank charters required a two-thirds vote of the legislature, instead of the simple majority that had been required up to that time. (Gatell, 1966: 26; Bodenhorn, 2003: 134, 186-88; Bodenhorn 2004; Moss and Brennan 2004: 7). Banking in the Early Republican United States therefore tended to be characterized by segmented monopolies. In fact, the four largest cities in the United States in 1800--Boston, Philadelphia, New York, and Baltimore--had only two state-chartered banks apiece. Smaller markets typically had only one bank, if they had a bank at all. These banks, it should be pointed out, did not lend to all comers. Indeed, they discriminated on the basis of profession, social standing and political party affiliation. (Wallis, Sylla, and Legler, 1994: 135-39; Bodenhorn, 2003: 142; Majewski 2004a). The federal government pursued a similar strategy to that of the states, and chartered its own commercial bank, the Bank of the United States (BUS), in 1791. Unlike the states, the incentive of the federal government was not to produce a source of dividend income (dividends from the BUS were on the order of one percent of total federal revenues). Rather, it was to provide the federal point that they earn the normal rate of return. Most of the rent is captured by the government and by 20
government with a financial agent that could issue banknotes against customs duties and that could hold federal balances. Nonetheless, the BUS was founded and operated in much the same way as the segmented monopolies created by the states: it was a commercial bank fully capable of making loans to private individuals, and 20 percent of its stock was owned by the federal government. The federal government did not actually pay for its $2 million stake in the bank: rather, the bank lent the government the money for the shares, which the government then paid for out of the flow of dividends. In addition, 75 percent of the private capital in the bank could be paid for with federal government bonds at par, providing a mechanism to soak up federal debt. In other words, the BUS allocated a 20 percent ownership stake in the bank to the federal government in exchange for a series of valuable concessions: the right to hold federal government specie balances; the right to charge the federal government interest on loans from the bank (notes issued by the bank to cover federal expenses); and the sole right to open branches throughout the country. This gave it a tremendous competitive advantage over state chartered banks, which were not allowed to branch across state lines, and which did not have the advantage of having the federal government as their biggest depositor. Needless to say, the existence of the BUS generated considerable resentment from bankers who held state charters, and therefore from state legislatures. Some states even tried (unsuccessfully) to tax the bank notes of the BUS in order to constrain it from competing against their own banks. (Wettereau, 1942; Lane 1997; Sylla 2000). Political Competition and the Breakdown of Segmented Monopolies public officials. 21
The strategies of the state and federal governments made sense from their points of view. In addition, the Bank of the United States made sense from the point of view of a sound system of public finance. That sound system of public finance was, in turn, an important ingredient in the creation of a government that honored property rights: investors did not fear that the government would have to fund itself via inflation or the seizure of assets. (Sylla 2000; Majewski 2004a). Nevertheless, the system of segmented state monopolies and a single national bank was not an equilibrium that was stable given the political institutions of the United States. As the U.S. economy grew so too did the demand from the public for banking services. That demand was channeled via the country's political institutions--parties, elections, separation of powers, and federalism--and it quickly undermined the initial organization of the banking system at both the national and state levels. The first source of competition, and the one that has received the most attention from historians, was that between states and the federal government. Bankers with state charters, and hence state legislatures, had opposed the BUS from the time of its initial chartering in 1791. The reason for their opposition was straightforward: branches of the BUS undermined local banking monopolies. (Rockoff 2000). These state bankers formed alliances with the Jeffersonians, who were ideologically opposed to chartered corporations and "aristocratic" bankers. Indeed, though the Founding Fathers were all opposed to the concept of political parties, two parties (the Federalists and the democratic republicans) quickly formed around the broad set of debates initiated by Hamilton and Jefferson regarding banking and public finance. (Hofstadter 1969). Thus, when the BUS charter expired in 1811, it was not renewed by Congress. The War of 1812 demonstrated, however, the importance of a bank that could serve as the financial agent of the 22
federal government, and thus a new charter (for a Second Bank of the United States) was granted in 1816. The Second Bank of the United States was founded on the same principals of the first bank, and it met the same fate. The Second Bank was a private enterprise, but the federal government owned 20 percent of its $35 million in stock--which it paid for with a $7 million loan from the bank to the government. In addition, the bank paid a charter bonus to the government of $1.5 million. Much as happened with its predecessor, the holders of state banking charters, as well as state governments, resented the fact that the Second Bank held tremendous government balances-- balances which state bankers believed should be part of their reserve base. They also resented the fact that the Second Bank could compete in their markets by opening branches at will. The history of the closure of the Second Bank would take us beyond the space limitations of this paper. Suffice to say, however, that in the end the state bankers prevailed by forming an alliance with Jacksonian populists. When the bank's charter came up for renewal in Congress it passed by only a single vote, and was therefore successfully vetoed by Jackson. It was forced to close in 1836 when its charter expired. The United States would therefore be without a central bank until the Federal Reserve System was created in 1913. (Hammond 1947; Temin 1968; Engerman 1970; Rockoff 2000). A second, and less obvious, source of competition was that between states for business enterprise and population. The fact that the United States had a federal system meant that population and business enterprises could easily relocate to other states. The fact that it also had an expanding frontier meant that the pressures on state legislatures to hold their populations and businesses from emigrating were especially severe. 23
Competition between states undermined the incentives of legislators to maintain the monopoly banks they had earlier established. State legislatures were under considerable pressure to provide public works projects, particularly canals, because powerful constituents (urban merchants and large farmers) stood to gain from the increase in commerce they would generate. The pressure to carry out these projects was made particularly intense by the fact that the failure to do so would result in commerce (and hence labor and capital) moving to those states that had made these improvements. Indeed, state legislatures tended to be highly aware of canal and port improvement projects in neighboring states, and of the consequences of those projects for the flow of commercial traffic within their own borders. (Goodrich, 1961; Majewski 2004a). State legislatures tended not, however, to have the ability to fund public works projects out of their meager tax revenues. One response by states was to issue bonds, thereby funding public works out of future tax revenues. This strategy had a limit, because states could only issue debt to the point that investors believed that the state would have sufficient future revenues to repay it. (Grinath, Wallis, and Sylla, 1997; Sylla 2000; Wallis and Weingast, 2004). Many states therefore found supplementary funding for public works projects from "bonuses" levied on new bank charters. Such charter bonuses created, however, an incentive for state legislatures to renege on their arrangements with incumbent banks. As a consequence, quite a few banks were granted charters that undermined the monopolies of pre-existing banks, in exchange for which they provided financing for canals, ports, and (somewhat later) street-lights and railroads. These schemes were employed across the country, even including Southern states that tended to be highly resistant to undermining their initial monopoly banks. (Bodenhorn, 2003: 86, 148, 152, 22834). 24
One might wonder why incumbent banks did not simply outbid potential entrants by agreeing to finance state infrastructure projects themselves. Incumbent banks at times tried this strategy. In order to successfully outbid all potential entrants, however, they would have had to devote nearly all of their rents to public goods provision. They were better off forcing new entrants to shoulder the costs of infrastructure projects (costs which could sometimes be ruinous for those new banks). This meant that the rents earned by incumbent banks were dissipated by competition, but that process took place over time. Incumbent banks therefore avoided the "Krueger paradox," but they did so at the cost of their long run monopoly positions. Competition over capital and labor also drove states to expand the suffrage, and an expanded suffrage undermined the political coalitions that had supported restrictions on the number of bank charters. All of the original 13 states restricted the right to vote based on a person's economic standing: some had minimum property qualifications; others had a qualification based on a minimum tax payment. With the exception of Louisiana, all of the states that subsequently joined the union, however, either eliminated these qualifications entirely or reduced them to the point that they were not binding. The logic of these voting rules was transparent: new states were eager to attract population. As a consequence, the original 13 states were forced to respond by ratcheting their voting restrictions downwards. By the mid-1820s, property qualifications had been dropped or dramatically reduced in virtually all of the original states. (Engerman and Sokoloff, 2001). As we shall see below, the extension of the suffrage allowed citizens to bring pressure to bear on legislatures, voting in legislators who were willing to remove constraints on the chartering of banks. De Facto "Free Banking" 25
Political competition within and among states undermined the incentives of state legislatures to constrain the numbers of charters they granted. Massachusetts began to increase the number of charters it granted as early as 1812. This required that the state abandon its strategy of holding bank stock as a source of state finance (the value of those shares and their dividend streams would decline as the original banks faced increased competition), and instead levy taxes on bank capital. Pennsylvania followed Massachusetts's lead with the Omnibus Banking Act of 1814. The act, passed over the objections of the state's governor (he had vetoed an earlier bill), ended the cozy Philadelphia-based oligopoly that, until then, had dominated the state's banking industry. The act divided the state into 27 banking districts, and allocated at least one bank to each district. In all, it chartered 42 new banks. Rhode Island also followed Massachusetts' lead: in 1826 it sold its bank shares, increased the numbers of charters it granted and began to tax bank capital as a replacement for the income it had earned from dividends. It soon became, on a per capita basis, America's most heavily banked state. A necessary condition for this change in the institutions governing banking was that there be pent up demand within states for bank credit. That is, there had to be constituencies that demanded access to credit, but that been unable to obtain it under the earlier set of institutions. The existence of this demand can be clearly seen by four features of Pennsylvania's 1814 Omnibus Act. The first was that merchants, manufacturers, and farmers had long complained that the lack of capital and specie had constrained their enterprises. The second was a peculiar requirement of the new law: banks had to lend at least 20 percent of their capital to those same three groups. The third was that the law spread the new banks throughout the state, with particularly prosperous counties getting multiple banks. The fourth was that the stockholders of these new banks viewed them as being extremely 26
lucrative: demand for shares at the initial subscription prices far outstripped the supply. In fact, a Pennsylvania bank chartered prior to the 1814 Omnibus Act (the Mechanic's Bank in 1809) was viewed by potential shareholders as being so lucrative that they rioted in the streets of Philadelphia when they found that there were not enough shares to go around. (Bodenhorn, 2003: 142-43; Majewski 2004a; Majewski 2004b). Southern states tended to be a good deal slower than Northern states to make the shift toward more liberal chartering. This is not to say that they uniformly refused to grant any new charters beyond those issued to their original, partially state-owned, banks. It is to say, however, that they tended to grant additional charters quite sparingly by the standards of Northern states. Historians have not yet systematically pursued the reasons for this difference in the treatment of bank chartering. One explanation that is often advanced is that the demand for capital was lower in the South. While this is almost certainly true, it is difficult to reconcile with the fact that some Southern states later embraced free banking--and chartered large numbers of banks as a consequence. A second explanation is that, as Gavin Wright has pointed out, slavery reduced the demand for public goods. (Wright 1986). The lower political pressure for public works projects in the South would, in turn, have reduced the need for states to charter progressively larger numbers of banks to fund those projects via charter bonuses. As a consequence, the dividend streams generated by the state ownership of bank stock would not have been diminished by competition, and states would have had little incentive to shift away from the ownership of bank shares and toward more liberal bank chartering. Even with the relative slowness of southern states to undo their monopolies and oligopolies, the U.S. banking system grew remarkably quickly. In 1818 there were 338 banks in operation, with 27
a total capital of $160 million--roughly three times as many banks and bank capital as in 1810 (Fenstermaker 1965: 401-404). By 1835, there were 584 banks. Larger cities may have had a dozen or more banks, while small towns may have had two or three. (Bodenhorn, 2003: 12). To put the size of this banking system into perspective, consider the case of England, which is usually thought of as the world's financial leader in the nineteenth century. In 1825, the United States had roughly 2.4 times the banking capital of England, even though the United States had a smaller population (Rousseau and Sylla 2000).8 De Jure Free Banking By the late 1830s the de facto policies of many states in the Northeast to grant virtually all requests for bank charters became institutionalized as a series of laws known as free banking. Under free banking, bank charters no longer had to be approved by state legislatures. Rather, individuals could open banks provided that they registered with the state comptroller and deposited state or federal bonds with the comptroller as a guarantee of their note issues. Some states also had minimum capital requirements and specified a minimum number of directors. Many required that bank stockholders be doubly liable. These features of free banking laws were meant to encourage prudent behavior by the bankers: they prevented them from over-issuing notes, and gave shareholders strong incentives to monitor the bank directors.9 8 The relatively slower development of the British banking system can be tied to politically determined constraints on the supply of banks. The Bank of England was chartered as a monopoly. Until 1825, other banks were not allowed limited liability. Instead, they were restricted to unlimited liability partnerships of six of fewer people. Rousseau and Sylla 2000. 9 If a bank breached the requirement that its notes be redeemable on demand for specie, the comptroller closed the bank, sold the bonds held as collateral, and redeemed the bank's circulating 28
The first state to make the switch (New York in 1838) to de jure free banking was not one that had previously carried out a de facto reform.10 Indeed, free banking in New York was unambiguously a consequence of political competition undermining the coalition of upstate interests that had supported restrictions on bank charters. New York was among the last of the original 13 states to broaden its electoral laws: it was not until 1826 that it finally shifted to universal manhood suffrage. Once that happened, Whig candidates began to outpoll Democratic Republicans in elections for the state legislature. By 1837 the Whigs had gained a majority, ending the reign of the Albany Regency. In the wake of the panic of 1837, which highlighted the fragility of unsecured bank notes, the Whig-dominated legislature was able to push through the free banking law. The new law made the establishment of a bank an administrative, rathern than a legislative, procedure. Prudent behavior was enforced by requiring all bank notes to be 100 percent backed by high grade securities. By 1841, New Yorkers had established 43 free banks, with a total capital of $10.7 million. By 1849, the number of free banks mushroomed to 111, (with $16.8 million in paid capital). By 1859 there were 274 free banks with paid in capital of $100.6 million. (Bodenhorn 2003: 18692; Atack and Passell 1994: 105; Wallis, Sylla, and Legler, 1994; Moss and Brennan 2004). Other states soon followed New York's lead: Georgia passed a free banking law in 1839; Alabama passed one in 1849, and then New Jersey, Illinois, Massachusetts, Ohio, Vermont, notes from the proceeds. Some states also allowed banks to also use mortgages on unencumbered property to back their note issues. Many states, for obvious fiscal reasons, later amended their laws, requiring that all reserves be held in bonds issued by that state. Bodenhorn (2003): chap. 8. 10 Michigan was actually the first state to pass a free banking law, in 1837, but revoked the law in 1839. It reinstituted free banking in 1857. 29
Connecticut, Indiana, Tennessee, Wisconsin, Florida, Louisiana, Iowa, and Minnesota all followed during the 1850s. Some variant of the New York law was ultimately adopted in 21 states.11 Readers may wonder how such a system of free entry could have been compatible with the fiscal needs of state governments. The answer lies in the fact that under free banking all bank notes had to be 100 percent backed by high grade securities that were deposited with the state comptroller of the currency. The putative reason for this was that it protected noteholders (if the bank failed, the comptroller of the currency could sell the bonds and compensate the noteholders). States tended, however, to gradually restrict the range of securities that they would accept as backing for bank notes. Initially, they accepted federal government bonds, bonds issued by other states, and even bonds that secured mortgages on real property. Increasingly, however, they required that notes be backed only by federal bonds or by bonds of the state in which the bank operated (Bodenhorn 2003: chap. 8). As Moss and Brennan (2004) have shown, free banks were forced, in essence, to grant a loan to the state government in exchange for the right to operate. 11 There has been considerable debate in the literature regarding the impact of free banking laws. The traditional view, associated with Rockoff (1974, 1985), was that free banking reduced barriers to entry, increased competitiveness, and reduced profits. This view was challenged by Ng (1985), who argued that free banking acts, with the exception of that of New York State, did not reduce barriers to entry because the competitive pressures that existed among states prior to the free banking acts had already eroded barriers to entry. Ng's argument was subsequently challenged by Economopoulos and O'Neil (1995), who demonstrated that free banking states were able to respond more elastically to changes in demand for banking services than chartering (non-free banking) states. A series of articles by Bodenhorn (1990, 1993) has also cast considerable doubt on Ng's view. Bodenhorn has demonstrated that one of the effects of free banking was to put competitive pressure on non-free banking states to grant more charters than they would have otherwise. In short, the weight of the evidence supports Rockoff's original interpretation: free banking increased entry and contributed to the rapid growth of the number of banks operating in most markets. 30
Inasmuch as bankers could have earned higher returns than those available from investing in state bonds, this feature of the free bank laws constituted a form of financial repression. Bankers realized that this feature of free banking laws prevented them from being able to serve as credit intermediaries. They responded with an institutional innovation of their own: they began to pursue deposit banking, opening demand deposit (checking) accounts and then lending the funds in those accounts without restriction. This step meant that banks were now subject to runs on their deposits, even if their notes were fully protected. The growth in deposit banking explains, in fact, why state governments found it necessary to introduce double liability into subsequent free banking laws: they wanted to give shareholders strong incentives to monitor bank directors. (Moss and Brennan 2004). The problem of runs was exacerbated by the fact that banks were single unit enterprises. Individual small banks tended to be susceptible to failure because of changes in local business conditions. This problem would have been mitigated had there been branch banking, which would have allowed banks to transfer funds from one branch to another in the event of a run on any single branch. It also would have been mitigated had there been a central bank, which could have acted as a lender of last resort, by buying and holding the notes of banks that were in trouble. Unfortunately, the U.S. banking system after the closure of the Second Bank of the United States had neither. To a remarkable degree, this problem was largely mitigated by the specie standard and the Suffolk 31
System during the antebellum period. As research by Rockoff has shown, bank failure rates were surprisingly low.12 A reader prone to skepticism might be tempted to argue that the passage of these laws demonstrates that there no politically determined supply constraints in banking--as demand for banking services increased, states changed their laws. Such a view would be extremely difficult to reconcile, however, with one of the curious features of free banking laws: they almost uniformly precluded the chartering of branch banks. Virtually all banks in the nineteenth century United States, except those in some southern states were unit (single branch) banks. This unusual organization of the banking system was the outcome of an unlikely political coalition: populists who feared bank monopolies at the state level allied to bankers who wanted to create local monopolies. This system of free entry and unit banks did confer several advantages. One obvious advantage was that the number of banks and bank capital mushroomed. Circa 1860, the United States had 1,579 banks, with a total capital of $422.5 million (Lamoreaux 1991: 540). A second, less obvious advantage, was that large numbers of small banks (the average capitalization was only $268,000) made it easier for bankers to overcome information asymmetries. The only reliable information that bankers had about borrowers was what they could obtain via repeated business dealings with them, via information obtained from local (and informal) networks of businessmen and farmers, or via their kinship ties. This meant that bankers tended to lend primarily to people that they 12 Rockoff 1974, 1985. Also see Gorton 1996. Bank failures became more widespread in the 1920s, resulting in the creation of the FDIC in 1933. Calomiris and White 1994. 32
knew personally. Under these circumstances, a system made up of large numbers of small banks embedded in individual communities was a rational way to organize the banking industry.13 As Naomi Lamoreaux has shown, this fairly unusual organization of the banking industry allowed New England banks in particular (one third of all U.S. banks were in New England) to operate as investment pools--the nineteenth century equivalent of modern mutual funds. New England's banks were not the independent credit intermediaries of economic theory. Rather, they were the financial arms of kinship groups whose investments spread across a wide number of economic sectors and a wide number of enterprises. Basically, kinship groups tapped the local supply of investable funds by founding a bank and selling its equity to both individual and institutional investors. The founding groups then lent those funds to the various enterprises under their own control. Investors in the banks knew full well that the banks only lent to insiders. They bought bank stock precisely in order to invest in the broad range of business enterprises controlled by the founding group of entrepreneurs. (Lamoreaux, 1994). Had legal restrictions been placed on the founding of banks, these insider arrangements would have concentrated capital in the hands of a small number of kinship/business groups. The fact that entry into banking was essentially free, however, meant that large numbers of entrepreneurs could tap the capital of non-group investors by using banks as investment pools. The crucial point is that the rules of the game were not constructed so as to allow some entrepreneurs to play while excluding others. 13 Informal networks appear to have played an important role in commercial lending until the 33
Political Institutions, War Finance, and the National Banking System From the point of view of the federal government, unit banking had another major drawback: a system of small, state-chartered banks did not provide it with a source of finance. This had not been a major problem during the 1840s and 1850s (after the closure of the Second Bank of the United States in 1836), because the federal government had a low demand for debt. To the degree that it ran deficits, it was able to cover these by selling treasury bonds to the public and to European investors. The Civil War, however, dramatically increased the financial needs of the federal government. The response of the federal government was to do what most governments do when they need to finance a war: they turn to the banking system. It therefore passed laws in 1863, 1864, and 1865 designed to eliminate the state chartered banks and replace them with a system of federally chartered banks that would finance the government's war effort. The laws creating national banks were designed so as to centralize bank chartering in the hands of the federal government. The laws did not abrogate the rights of states to charter banks-- that would have violated the Constitution. They also did not abrogate the right of state-chartered banks to issue bank notes, as that too would have been unconstitutional. A follow up act to the National Banking Act (in 1865) did, however, impose a ten percent tax on bank notes, and then exempted federally-chartered banks from the tax. This created a strong incentive for state banks to obtain new, federal charters. In fact, the expectation of the federal government was that state banks would disappear. information revolution of the last decade. See Petersen and Rajan 2002. 34
The incentive of the federal government for doing this is not obvious until you consider a principal feature of the new law: federally-chartered banks had to invest one-third of their capital in federal government bonds (which were then held as reserves against note issues by the comptroller of the currency). Consistent with the goal of maximizing credit to the federal government, the National Banking Act made the granting of a charter an administrative procedure: as long as minimum capital and reserve requirements were met, the charter was granted. It was free banking on a national scale. (Sylla 1975). In the short run, the response of private banks was as the federal government expected: the number of state chartered banks declined from 1,579 in 1860 to 349 by 1865. (See Table 1). Federal banks grew dramatically: from zero in 1860 to 1,294 in 1865. They then continued growing, reaching 7,518 by 1914, controlling $11.5 billion in assets in that year. In the long run, however, the political institutions of the United States frustrated the federal government's goal of a single, federally-chartered banking system. They also undermined the barriers to entry in banking that had been created by the National Banking System. The federal government had effectively nationalized the right to issue bank notes by creating a 10 percent tax on the notes of state chartered banks in 1865. The 1865 law did not, however, say anything about checks drawn on accounts in state-chartered banks. State banks therefore aggressively pursued deposit banking, and checks drawn on those accounts became a common means of exchange in business transactions.14 Moreover, the states rewrote their banking laws, reducing even further the 14 Demand deposits were already becoming an important component of bank liabilities before the National Banking Act, as a consequence of the note security requirements of free banking acts. (Moss 35
requirements to obtain a charter. Most states did away with the requirement of "double liability" that had been part of most state free banking laws. State charters also tended to require lower minimum capital requirements than National Banks and had less onerous reserve restrictions. Finally, National Banks were not allowed to lend on real estate. State banks faced no such restrictions. (Sylla 1975: 62-73; Davis and Gallman, 2001: 272). The result was that state chartered banks actually outgrew federally chartered banks during the period 1865-1914. In 1865, state banks accounted for only 21 percent of all banks and 13 percent of total bank assets. By 1890 there were more state banks than national banks, and state banks controlled the majority of assets. Circa 1914, 73 percent of all banks were state banks, and state banks controlled 58 percent of assets. (See Table 1). The end result of this competition between states and the federal government was a banking system unlike that of any other country. In the first place, circa 1930 there were 21,309 banks in the United States. In the second place, virtually all of these banks were unit banks: many states had laws that prevented branch banking, even by nationally chartered banks; most other states did not explicitly forbid branching, but their laws provided no provision for branch banking. (Davis and Gallman 2001: 272). As Calomiris and White (1994) have shown, of the 21,309 banks in operation, only 723 had branches. The average number of branches operated by these banks was less than five. Large numbers of small unit banks created problems of volatility, which were only partially compensated by the fact that the national banks had correspondent relationships. It also made it difficult for banks to capture scale economies. (Bordo, Rockoff, and Redish 1994). and Brennan 2004). With the complete demise of state bank note issues after 1864, however, state 36
Unit banking did, however, confer two advantages. First, unit banking meant that all markets were contestable. Second, embedding banks into communities meant that bankers could overcome information asymmetries by tapping into local (and informal) networks for information about potential borrowers. Indeed, one feature that is particularly striking about this system was how many banks there were per person in the United States, and how geographically dispersed banks were. (See Table 2). In sum, constraints on the supply of banks in the United States tended to be short-lived. This was not because there were not attempts to constrain supply. Rather, it was because barriers to entry tended to be dissipated by the effects of competition--competition not among firms, but among (and within) the different political entities that could charter and regulate firms. MEXICO The colonial political institutions of Mexico were starkly different from those of the colonial United States. Rather than having elected colonial assemblies, Mexico had a viceroy, who ruled through provincial level colonial officials. Until the late eighteenth century, these colonial officials typically married into, and became part of, the local economic elite. Even after a series of reforms changed the policies governing the recruitment and compensation of royal officials, the basic point remained: colonial political institutions were hierarchical and non-democratic. Their purpose was to maintain a society that was composed of Native Americans and mestizos, but that was run by, and for the benefit of, a numerically small Spanish (and Mexican born, but culturally and ethnically Spanish) elite. banks had even stronger incentives to pursue deposit banking. 37
As a consequence, the process of Mexican independence was the polar opposite of the process in the United States. Mexican elites pushed for independence not because they resented royal authority, but because they were royalists. They had earlier sided with the viceroy against an independence movement whose goal was to challenge the colonial system of privilege and hierarchy (the Hidalgo revolt of 1810)--and had put down that insurgency with extreme brutality. They finally declared independence (in 1821) precisely because the King of Spain had been forced to accept a liberal constitution by his own army. Mexico's post-independence elite was not, however, all of one mind regarding the institutions that should govern the new country. Some sought to create a constitutional monarchy, and to maintain all of the other political and economic institutions of the colony, including the centralization of political power and exemptions from trial in civil courts for the army and clergy. Others wanted a federal republic--though one in which suffrage would be restricted on the basis of literacy, in a society where very few were literate. The one factor that unified both groups was their aristocratic inclinations: neither had any real intention of sharing power with the country's vast indigenous and mestizo populations. These two groups, one conservative and centralist, the other liberal and federalist, were unable to craft a set of political institutions through which political conflict could be channeled. Instead, they engaged in a series of coups, counter-coups, and civil wars--often mobilizing the indigenous and mestizo populations in the process. When one or the other emerged victorious they would scuttle earlier constitutions, and redraft a new one. The result was quite unlike what had occurred after independence in the United States. There, the Republicans and Federalists had each come to believe that they would be better off if they 38
competed with one another through the country's formal political institutions (e.g., elections, congress, state legislatures) rather than going outside them. In order to do that, they each built an additional institution--political parties with coherent ideologies. In contrast, the conservatives and liberals in Mexico each saw one another as illegitimate. The purpose of political competition was not just to take power, but to eliminate the other side. Political Institutions, Public Finance, and Banking, 1821-1876 The first government after independence was a monarchy, whose leader (Agustнn Iturbide) soon dissolved congress and declared himself emperor. The dissolution of congress left him with few supporters, and he was successfully overthrown by the liberals, who then instituted a federal republic. It lasted less than a decade, and was replaced by a conservative government headed by a military strongman (Antonio Lуpez de Santa Anna), who threw out the constitution and eliminated the federal system. (It was this move against the states' autonomy that prompted the Texans to secede from Mexico). A series of weak governments followed, most of them lasting only a few months. Their disorganization invited the United States to invade Mexico in 1846, and take half of Mexico's territory in the peace settlement. In the early 1850s the conservatives made another effort to build a centralized state. They re-installed Santa Anna (in all, he served as President on 11 separate occasions between 1832 and 1853), who soon decreed that his rule should last indefinitely. This provided a lightening rod for the liberals to build a coalition to depose him, which resulted in yet another liberal government and yet another federal constitution in 1857. This liberal government was then challenged by the conservatives in a three-year long civil war. The liberals emerged triumphant, but exiled 39
conservatives encouraged France to invade and occupy Mexico in 1862 (the nominal reason was to enforce payment of Mexico's external debt). When France withdrew its support in 1867 from the puppet regime of Archduke Maximilian the liberals retook power. Nevertheless, even this restored liberal regime was torn by internal conflict, and was overthrown in 1876 by Porfirio Dнaz, a popular army officer who led a revolt against the government because the president (Sebastian Lerdo de Tejada) was attempting to centralize power and arrange his own re-election, in contravention of the constitution of 1857. We will return to Dнaz at length, but for now, let us consider the impact that this long period of political instability had on the development of the banking system. All sides in Mexico's nineteenth century coups, rebellions, and civil wars preyed on the property rights of their vanquished opponents. Indeed, the logic of the situation virtually required that they do so: they had to reward their allies; and the most readily available resource at hand to do so was the wealth of their enemies. (Haber, Razo, and Maurer 2003, chap. 8). Every government that came to power also inherited a depleted treasury and no ready source of income that could be used to create a durable government. Desperate governments with short time horizons tend to raise taxes to the point that producers can recover only their variable costs. Mexico's nineteenth century governments were no exception to this general rule. The most productive part of the colonial economy had been silver mining. Indeed, in the eighteenth century Mexico had been the world's leading producer of silver, and the colonial government had steadily dropped the tax rate in order to keep the mines profitable. In the decades after independence, Mexico's various governments (state as well as central) pushed tax rates through the roof (they were on the order of 25 percent of revenues in the 1860s) in order to fill their empty coffers. In so doing, 40
they removed the incentives of mining companies to invest in new exploration and the resurrection of colonial mines. (Haber, Razo, and Maurer 2003, chap. 7). Mexico's governments also could have instituted a land tax. The reason why they did not do so, however, follows another general rule of political economy: when a small group of wealthy men control the state, they prefer not to tax themselves to fund its operation. A broadly based land tax in nineteenth century Mexico would have been expensive to administer: It would have required cadastral surveys and the creation of a public registry of land titles (and clear demarcation of parcel boundaries). It would also have required a collection mechanism. Under these circumstances, the easiest land to tax would have been the land held in large commercial farms. This land, however, tended to be owned by the same small elite that ran the government. Indeed, when governments changed hands in the nineteenth century they tended to allocate these estates to their military allies as compensation for their service. Taxing these estates was therefore politically difficult. The government therefore had to rely on taxes on imports and exports. The problem was that Mexico actually exported and imported very little. Moreover, the fact that Mexico's nineteenth century governments faced the very real threat of being overthrown meant that they needed large infusions of cash in the short run, not the modest stream of revenues that would trickle in from customs duties. Mexico's nineteenth century governments therefore borrowed from the country's private bankers (the agiotistas). The problem was that when governments changed, or when governments faced sufficient threat, they reneged on these debts. Private bankers therefore demanded interest rates that would compensate them for this risk. They also realized that they needed to collateralize the loans, and so the government mortgaged the customs revenues. The problem was, however, that 41
new governments could (and did) abrogate these agreements, which only lowered the mortgage value of the customs in the next round of loans. (Tennenbaum 1986; Walker 1986). In this environment--in which property rights were insecure and the government did not have a sound system of public finance--the incentives of private bankers to obtain charters were extremely low. Charters confer two advantages on bankers: they allow them to issue bank notes (and profit from seignorage), and they allow them limited liability (which allows them to sell equity to outsider investors). The disadvantage is that a charter makes the property rights of the banker transparent. In a situation in which bankers are subject to expropriation, they do not want their property rights to be more transparent--they want them to be less transparent, because transparency makes property rights easier to expropriate. As a consequence, Mexico had no chartered banks at all until 1863--and that bank chater was the product of very special circumstances. The specifics of this charter, in fact, give us an indication of why there had not been others previously: the charter was granted to a foreign bank (the British Bank of London, Mexico, and South America) by the puppet government of a foreign power (the Emperor Maximilian, who had been installed by the French). The British stockholders of this bank (known in Mexico as the Banco de Londres y Mйxico--BLM) believed that the French government would not expropriate them. One might be tempted to argue that Mexico had no chartered banks prior to 1863 (and then only one from 1863 to the late 1870s) because there was insufficient demand. Such an argument would not, however, square with two pieces of evidence. The first is that Mexico's fledgling industrialists were clamoring for more credit than could be provided by the private bankers. For that reason, they convinced the government to create a national industrial development bank, the Banco 42
de Avнo, in 1830. That bank was scuttled in 1842, when the central government, desperate for revenue, ransacked its vaults. That is, the government expropriated its own development bank. (Potash 1983). The second is that the Mexican government was itself clamoring for more credit at lower interest rates than could be provided by the private bankers. Clearly, it was in the government's interest to create a semi-official commercial bank (such as the Bank of England or the Bank of the United States) that would provide it with credit in exchange for a set of special privileges. The problem was that Mexican governments in the nineteenth century did not have a way to make a credible commitment to the bankers that they would not expropriate them as soon as they deployed their capital. The only threat that bankers could make was the threat implicit in all sovereign debt contracts: there would be no future loans if past loans were not repaid. When governments have long time horizons, this threat carries great force. But, when time horizons are short, threats of this nature have little impact. Regulated Entry and Insider Lending in Porfirian Mexico, 1876-1911 The unstable nature of Mexican politics, and the underdeveloped state of Mexico's banking system, changed dramatically during the 35-year dictatorship of Porfirio Dнaz (1876-1911). Dнaz's solutions to Mexico's weak institutions, however, only mitigated some of the problems facing the banking system. The banking institutions that developed mitigated the problem of expropriation and imprudent behavior, but they did not overcome the problem of contract enforcement. As we shall see, the institutional solution to expropriation risk interacted with weak contract enforcement to 43
produce a banking system that was stable, that served as a source of public finance, but that provided little credit to anyone who was not a bank director or a family member of a bank director. Dнaz confronted the same problem as all of the governments before him. He lacked sufficient tax revenues to finance a government capable of unifying the country and putting an end to internecine warfare. Borrowing his way out of this situation was difficult, because Mexico had a long history of defaulting on its debts to its international and domestic creditors. In fact, Dнaz himself had reneged on debts to some of the banks that had been founded in Mexico City during the early years of his rule. (Marichal 2002; Maurer and Gomberg, forthcoming). Dнaz did, however, have an advantage over earlier Mexican presidents. By the end of the nineteenth century the expansion of the U.S. railroad network and technological advances in transoceanic shipping had driven down international transport costs. Mexico could be integrated with the world economy in ways that were previously unimaginable. This meant that there was a tremendous source of rents that Dнaz could tap--from foreign direct investment in mining, petroleum, and export agriculture--that could be used to buy off opponents, or build a state strong enough to intimidate them. The problem for Dнaz was how to start the virtuous cycle of political stability, state capacity, foreign direct investment, and economic growth. The solution that Dнaz hit upon to jump start this process was one that had been used by European governments since the eighteenth century: create a semi-official super bank that received a set of lucrative privileges in exchange for providing a source of finance to the government. This solution mitigated the problem of expropriation risk (because the bank was compensated for risk with privileges that allowed it to earn supernormal returns). (Maurer and Gomberg, forthcoming). Later banks were rewarded with protected markets, via a system of taxes on new entrants. The fact 44
that this bank (and the ones that followed it) had to hold high levels of reserves against note issues mitigated the problem of imprudent bankers. (Maurer and Haber, 2004). Dнaz' solution did not, however, mitigate the problem of contract enforcement. Thus, Mexico's bankers employed an informal institution--they lent primarily to themselves and their family members. The problem with this strategy, however, was that the lucrative privileges that Dнaz bestowed on banks included restrictions on the number of banks in any market. This meant that credit was restricted to those few entrepreneurs in any state who happened to have family members who had obtained one of only a few bank charters. (Haber 1991, 1997; Maurer 2002; Maurer and Haber 2004). The Transformation of Mexico's Political Institutions Porfirio Dнaz, having just overthrown Sebastian Lerdo de Tejada in 1876, and having installed himself as president, inherited an economy that had scarcely grown over the previous six decades. He also inherited a constitution, a congress and senate, and a federal system of government--all of which had been created in 1857, but little of which had actually had much of an opportunity to function. Dнaz' goal--though he did not make this clear at the time--was to find a way to undermine whatever real bite these institutions had, and in so doing perpetuate himself in power. In the early years of his regime, Dнaz was conscious of the fact that his grip on power was weak and that the country's formal political institutions could serve as a lightning rod for opposition to any attempt to succeed himself in office. He therefore hand-picked a successor for the 1880-84 term (General Manuel Gуnzalez) and arranged for his election. Dнaz' allies in congress then amended the constitution, removing term limits and lengthening presidential terms from four to six years. This 45
step allowed Dнaz to legally return to office in 1884 and then "win" every presidential election until he was overthrown in 1911. By the late 1880s Dнaz had effectively undermined the independence of congress. As Armando Razo (2003) has shown, during the early years of his administration there was opposition to many of Dнaz public policy initiatives. After 1888, however, virtually all roll call votes in the Mexican Senate were unanimous. The decline in dissension in congress and the senate was coterminous with a decrease in the rate at which seats in those bodies turned over. During the period when Dнaz was still gaining control of the legislature (1876-1890), a senator or federal deputy had about a one-in-three chance of being reelected. From 1890 to 1910, however, the probability of reelection doubled--to two-in three. This change in legislative tenure was accomplished by the following simple mechanism: prior to elections, Dнaz would send the state officials who ran the election a list of his "preferred" candidates. Their job was to make sure that those candidates were then "elected." (Razo 2003). In order to pull this off, Dнaz had to undermine Mexico's powerful state governors, who commanded state militias that outnumbered the federal army. Dнaz' gradually appointed men loyal to him to state level posts--for example, chief of the federal police garrison--and then slowly promoted them into the governorship when the moment seemed propitious. These hand-picked appointees--who were often from outside the state and had few local ties--remained in power for decades, and owed that power to Dнaz. By the end of Dнaz's rule in 1911 over 70 percent of the state governors were presidential favorites "imported" from outside. (Knight 1986). In those states where Dнaz could not eliminate or undermine potentially recalcitrant governors, he coopted them. For example, Governor Enrique Creel of Chihuahua (of the Terrazas- 46
Creel clan, which had controlled the governorship of the state on and off since the 1870s) was named Dнaz's Foreign Minister. Another example is provided by Governor Olegario Molina, who ran the Yucatбn as a private business enterprise, who was named Minister of Development. (Knight 1986: 16). The undermining of the governors meant that rights that had formerly been conferred on the states were all centralized in the federal government. Given the fact that congress was a rubber stamp, this meant that policy-making authority was centralized in the hands of Dнaz and his long-term minister of the treasury, Josй Yves Limantour (1892-1911). As of 1884 the granting of mining titles and the establishment of mining taxes were made federal, not state, functions. (Dнaz got the governors to go along by allowing them to levy a 25% surtax on whatever federal taxes were charged. He later lowered the federal tax rates). The same act gave the federal government, and not the states, the authority to grant concessions and to tax the rights to ground water and petroleum deposits. By 1896, Dнaz was strong enough to force states to remove taxes on interstate commerce. Finally, in a series of laws enacted between 1884 and 1897 Dнaz centralized control of bank chartering in the hands of the federal government. The glue that held this governance system together was the distribution of rents from private economic activities directly to politically crucial governors, senators, and federal deputies. Recent research by Razo (2003) and Musacchio and Read (2003) on the networks of political and economic elites in Porfirian Mexico demonstrates that the boards of directors of the country's largest publicly traded companies--particularly those that needed federal largess and protection--were populated by a small group of powerful public officials. They went along with Dнaz, in short, because they got rich in the process. 47
Regulated Entry, Public Finance, and the Supply of Banks in Porfirian Mexico Once Mexico began to become stable and began to grow in the early 1880s, states-- particularly mining states on the U.S. border--began to charter banks. In addition, foreign-born merchants in Mexico City obtained charters from the national government to operate banks in Mexico City. By 1883, Mexico therefore had eight chartered banks. The fact that the Dнaz government was still, at this point, fragile gave it a strong incentive to monopolize bank chartering as a means to provide itself with a ready source of credit. Dнaz therefore made two moves. First, in 1884, he engineered the merger of Mexico City's two largest banks creating the Banco Nacional de Mйxico (Banamex). The intention of the government was to model Banamex on the Bank of England, granting it a monopoly over the issuance of paper money in return for providing a credit line to the federal government and acting as the treasury's financial agent. In addition, the federal government granted Banamex the right to tax farm customs receipts and the right to run the mint. (Maurer and Gomberg, forthcoming). Second, the government simultaneously federalized the chartering of banks. As of 1884, states could no longer grant charters (Haber 1991). What was crucial, from the point of view of Dнaz and Banamex, was that the commercial code of 1884 erected high barriers to entry. Not only had the federal government monopolized the granting of charters, it also required that new banks obtain the permission of Congress and the Secretary of the Treasury in order to obtain a charter. They also had to pay a five percent tax on the issuance of bank notes. Banamex was exempted from the tax. Finally, Banamex was permitted to issue banknotes up to three times the amount of its reserves. Other banks were not afforded this 48
privilege. In short, the federal government was attempting to exchange a set of special privileges for access to credit. (Haber 1991; Maurer 2002). Mexico's already extant banks, particularly the Banco de Londres y Mйxico, realized that the commercial code and Banamex' special privileges put them at a serious disadvantage. They therefore sued in federal court, and managed to obtain an injunction against the 1884 Commercial Code on the basis of the fact that the 1857 Constitution had an anti-monopoly clause. The ensuing legal and political battle ground on for 13 years, until a compromise was finally hammered out by Secretary of Finance Josй Yves Limantour in 1897. (Maurer 2002: chap 2). There were four groups that pressured the federal government in the crafting of the 1897 General Credit Institutions and Banking Act: the stockholders of Banamex; the stockholders in the Banco de Londres y Mйxico; the stockholders in other, smaller, state-level banks; and the state governors (who wished to award cronies with bank charters). The resulting law could easily be predicted from knowledge of the players in the negotiations: Banamex shared many (although not all) of its special privileges with the Banco de Londres y Mйxico; the state banks were given local monopolies; and the state governors were able to choose which business group in the state would receive a bank charter from the federal government. Holding the arrangement together was the fact that the federal government monopolized bank chartering. Legal barriers to entry into banking could not be eroded by competition between states, or between states and the federal government, because states did not have the right to charter banks.15 15 Had states had the right to charter banks, they would have been tempted to ratchet downwards the minimum requirements for a bank charter as they competed against one another for bank business. 49
The resulting competitive structure had the following features. Banamex and the Banco de Londres y Mйxico were granted a duopoly in the Mexico City market. In addition, only Banamex and the Banco de Londres y Mйxico had the right to branch across state lines. Banamex was also granted the exclusive privilege of providing financial services to the government: collecting tax receipts, making payments, holding federal deposits, and underwriting all foreign and domestic federal debt issues. In short, the compromise was that Banamex would retain the special privileges granted to it in 1884, and some of these privileges would also be extended to the Banco de Londres y Mйxico. (Maurer, 2002: chap. 3) State level banks, and their patrons--the state governors--were also protected from competition. The law was written in such a way that, as a practical matter, only one bank could be established in each state, although existing banks were grandfathered in. The law specified that bank charters (and additions to capital) had to be approved by the Secretary of the Treasury and the Federal Congress. In order to make this policy credible beyond the tenure of Josй Limantour as Treasury Secretary, the law also created three other barriers to entry. First, the law created very high minimum capital requirements, initially U.S. $125,000, which was later raised to U.S. $250,000--more than twice the amount for national bank in the United States. Second, the law established a two percent annual tax on paid-in capital. The first bank granted a charter in each state, however, was granted an exemption from the tax. Third, state banks were not allowed to branch outside of their concession territories. This prevented banks chartered in one state from challenging 50
the monopoly of a bank in an adjoining state. In short, the only threat to the monopoly of a state bank could come from a branch of Banamex or the Banco de Londres y Mйxico.16 The existence of these segmented monopolies was made incentive compatible with the interests of Mexico's most important public officials, who received seats on the boards of the major banks (and thus were entitled to director's fees and stock distributions). Indeed, the Dнaz government appears not to have chosen the groups that received a bank charter based on their entrepreneurial talents: it chose them based on their political connections. The board of directors of Banamex, for example, was populated by members of Dнaz' coterie, including the President of Congress, the Under-Secretary of the Treasury, the Senator for the Federal District, the President's Chief of Staff, and the brother of the Secretary of the Treasury. The Chairman of the board of the Banco de Londres y Mйxico was none other than the Secretary of War. Joining him on the board was a federal senator from the State of Sonora. The Banco International e Hipotecario (a mortgage bank) was similarly populated with political notables, including the President's son, the ambassador to Belgium and the Netherlands (who was also a federal senator), and the brother of the Secretary of the Treasury. The Banco Mexicano de Comercio e Industria was also a who's who of insiders. Its board chairman was the President of Congress. Joining him on the board was the Governor of the Federal District. 16 The law also allowed for the establishment of mortgage banks and "bancos refaccionarios," which were allowed to make long-term loans. These banks were not granted, however, the right to issue bank notes and were subject to a variety of restrictions on the types of investments they could make. Without the right to issue notes, and with few ways to actually foreclose on a mortgage, these banks could not compete. Few charters for mortgage banks were ever taken out. See Riguzzi 2002. 51
Banks with limited territorial concessions were also chosen based on their political connections. The only difference was that state governors, rather than cabinet ministers, sat on their boards and received directors' fees, stock distributions, dividends, and in some cases loans made with no expectation of repayment. In some cases, the governor himself received the bank concession. In fact, the system was deliberately conceived to distribute benefits to the state governors, and give them a stake in the maintenance of Porfirio Dнaz's rule. (Haber, Razo, and Maurer, 2003: 88-90; Razo, 2003: chaps. 8, 9). The resulting banking system had one major advantage, and one major disadvantage. The advantage was that the construction of Banamex created, for the first time in Mexican history, a stable system of public finance. Credit from Banamex meant that the Dнaz government did not have to prey upon property rights in order to maintain its fragile hold on power. Instead, it gave Dнaz the financial breathing room he needed to slowly recraft the tax codes governing mining, petroleum, and interstate commerce, gradually increasing government tax revenues to the point that he ran balanced budgets. (Carmagnani 1994; Haber, Razo, and Maurer 2003, chaps. 3, 6, 7). It also allowed Dнaz, with the help of Banamex's directors, to renegotiate Mexico's foreign debt--which had been in default for several decades. (Marichal 2002; Maurer and Gomberg forthcoming). Finally, the creation of Banamex allowed Dнaz to subsidize the creation of a national railroad system--which had a huge positive impact on the country's overall growth. (Coatsworth 1981; Kuntz Ficker 1995). The disadvantage was that was that Mexico had a very concentrated banking system. In 1911, there were only 42 formally incorporated banks in the entire country. The United States, for comparison purposes, had more than 25,000 banks and trust companies in that year. The capital available to the Mexican banking system was also small: total assets in 1911 totaled approximately 52
U.S. $385 million. (Mexico, Secretaria de Hacienda, 1912: 255). (See Table 3). For comparison purposes, total assets of the U.S. banking system were $22.9 billion. (United States, Department of Commerce, 1975: Series X 580-587). Finally, not only were Mexico's banks few in number and of small size, but the level of concentration was extremely high: Banamex and Banco de Londres y Mйxico accounted for more than 60 percent of all assets. (Mexico, Secretaria de Hacienda, 1912: 236, 255). The vast majority of markets had, at most, three banks: a branch of Banamex, a branch of the BLM, and a branch of the bank that held that state's territorial concession. It was not uncommon for there to be only one or two banks in some states.(See Table 3).17 The Economic Effects of Concentrated Banking A skeptical reader might argue that a concentrated banking system with branch networks might have been an efficient solution in a country with a low GDP. This hypothesis can be subjected to two kinds of tests against evidence, neither of which supports it. The first test is an analysis of excess liquidity in the leading banks. Noel Maurer has demonstrated that the two largest banks in the system (Banamex and the Banco de Londres y Mйxico, which jointly controlled 60 percent of assets) acted like inefficient monopolists: they held excess liquidity in order to ration credit and drive up their rates of return. As a result, their stockholders earned substantial rents while they incurred very little risk: the yield on common stock (the value of dividends divided by the market value of common shares) of these two banks was about equal to that of Mexican government bonds. (Maurer, 2002: chap. 5). 17 Three states on the U.S. border, Nuevo Leуn, Chihuahua, and Sonora, had more than three banks (four, five, and six, respectively) because of peculiarities of their banking histories. 53
The second test involves an analysis of the organization of Mexico's banking markets on downstream industries. Mexico's bankers and government had found a way to mitigate expropriation risk by limiting competition in banking markets--by constraining the number of banks. They had not, however, been able to build the broad range of institutions necessary to allow property rights to be enforced at low cost. The institutional solution that Mexican bankers hit upon was that they primarily lent to themselves. The evidence suggests that they hit upon this solution by trial and error, initially making arm's length loans that went into default. The collateral for these loans then proved to be either fictitious or unrecoverable. (Maurer and Haber 2004). In the context of a banking system in which bankers primarily lent to themselves and their family members, constraints on the number of banks in any market acted as a barrier to entry in downstream industries. It is not possible, of course, to observe the universe of potential entrepreneurs who did not found firms for lack of credit. It is possible, however, to determine whether industries that are usually characterized by near-perfect competition were structured that way in Mexico. If we observe that industries with modest scale economies had competitively structured markets, the implication would be that entrepreneurs did not face financial barriers to entry. If we observe, however, that industries with modest scale economies had market structures similar to industries characterized by increasing returns to scale, then the implication would be that entrepreneurs faced financial barriers to entry. 54
A series of papers by Haber (1991, 1997, 2003), and Maurer and Haber (2004), which focus on the industrial structure of the Mexican cotton textile industry, addresses this question.18 These papers specify three counterfactuals. The first compares Mexico to itself over time. Cotton textile manufacturing was an industry characterized by constant returns to scale technologies and the absence of entry barriers. We should expect that as the industry grew, concentration should have fallen. The second compares Mexico to three other countries (the United States, Brazil, and India) that had large textile industries, but which did not have Mexico's banking system. The third, following Sutton (1998), compares the Mexican textile industry's actual market structure to a hypothetical, fully competitive industry, in which the market structure was a function solely of industry size and a stochastic growth process. The results of these experiments are presented in Table 4. Concentration is measured in two ways: the four firm ratio (the percentage of the market controlled by the four largest firms); and the Herfindahl Index (the sum of the squares of the market shares of all firms in the industry). Regardless of the measure employed, the data indicate that Mexican cotton textile industry was "too concentrated." First, concentration in Mexico actually increased over time, even though the industry was growing rapidly. (In the United States, Brazil, and India, unlike Mexico, concentration fell or remained stable as the textile industry grew.) Second, the Mexican cotton textile industry was much more concentrated than the American, Brazilian, or Indian cotton textile industry. Third, the Mexican 18 This research focuses on cotton textiles because it was an industry characterized by the lack of barriers to entry created by advertising, branding, or technology. It was also characterized by a small minimum efficient scale of production and capital divisibilities. At the same time, the existence of financial barriers to entry in this industry would be expected to indicate similar barriers in other industries. 55
cotton textile industry showed much higher four-firm ratios compared to the ratio that would be expected in a perfectly competitive market. The implication is that some entrepreneurs were awash in funds, while others were starved for capital. In short, the evidence supports the view that the organization of Mexico's banking industry came at a cost to the real economy. A detailed discussion of Mexico's banking system after the fall of the Dнaz dictatorship would take us well beyond the space constraints of this paper. Suffice to say that the Mexican Revolution (1911-20) and the ensuing political instability of the 1920s did little to create institutionalized political competition. During the period 1911-29, political competition in Mexico took place entirely outside formal institutions, which is to say at gunpoint. The property rights of bankers were once again expropriated. The banking system, as a result, contracted until postrevolutionary governments forged a new set of institutions in 1924 that would govern the industry. These governments, much like Dнaz before them, were not constrained by institutionalized political competition. Indeed, until the late 1990s, none of their successors were either. Not surprisingly, they created economic institutions--and a resulting banking system--that were not dissimilar to that which had existed under Dнaz. (Maurer 2002; Del Angel Mobarak 2002; Haber, Razo, and Maurer 2003; Haber 2005). CONCLUSIONS AND IMPLICATIONS: In the past two decades social scientists have embraced the notion that a wide variety of economic outcomes can be explained by institutions. A number of questions, however, remain: where do institutions come from; which way does causality run; and which institutions are crucial (and which are merely incidental) to the process of growth? 56
This paper has offered a contribution to this literature by examining the causes and consequences of the economic institutions that govern banking. Obviously, sustaining the argument that institutions that encourage political competition translate into economic institutions that encourage competition in banking will require more empirical testing than I have provided here. Additional case studies, which focus on cases intermediate to the United States and Mexico, are required. Nevertheless, the analysis presented here makes a strong case for the argument that competitive markets do not happen all by themselves. All markets are embedded in political systems, and some of these markets are highly sensitive to regulation by governments. When governments are unconstrained by institutionalized competition, they (and the public officials within them) will structure economic institutions so as to maximize short-run government revenues and /or permit rent seeking by public officials. The unintended result is smaller and less efficient markets, which, in turn, constrains economic growth. One might be tempted to argue that one or another of the political institutions that encourage institutionalized competition are crucial, and the others are merely incidental. While I do not discount the possibility that research into cases beyond those analyzed here might support that view, the cases analyzed here suggest that political institutions mutually reinforce one another--they do not operate in isolation. One might be tempted to argue, for example, that electoral suffrage was the crucial institution, because it allowed voters in the United States to remove rent-seeking legislatures that maintained oligopolies. While it is true that voters did remove such legislatures, it is also true that they could not have done so had state legislatures themselves not voted to broaden the suffrage. The reason they did so had everything to do with federalism: states competed with one another for 57
business and population. In addition, federalism worked directly to undermine oligopolized banking systems, because states had strong incentives to increase the numbers of charters they granted: first, in order to fund public works projects necessary to keep business and population in the state; and second in order to keep business and population from migrating to other states where it would be easier to obtain credit. One might therefore be inclined to argue that federalism was therefore the crucial institution. This view, too, holds some water. Federalism explains why the United States had a banking system composed of tens of thousands of unit banks, instead of a banking system composed of a few large banks that could branch wherever they pleased. At the same time, however, federalism alone cannot explain why the United States had no central bank for most of the nineteenth century. Clearly, federalism played a role, because the congressmen who voted against the renewal of the Bank of the United States did so because they were representing the interests of bankers in their states. But, congressmen could only do this because of another institution: separation of powers in the national government. An analysis of the Mexican case produces similar conclusions. When Porfirio Dнaz came to power he inherited a constitution, a federal system, and a bicameral legislature. What he did not inherit, however, were two other institutions that could have given federalism and separation of powers real bite: effective electoral suffrage and political parties. Dнaz was therefore able to undermine the senate, the chamber of deputies, and the governors. The resulting rent-seeking network could not be challenged--at least not within the existing set of institutions. When it was eventually challenged, in 1911, it was by violent, not institutional, means. 58
The analysis presented here has an additional, practical, implication. Much of the research in financial economics about bank regulation in less developed countries suggests that a major problem for financial development is inadequate regulation and supervision. The prescription is to give regulators more power and authority. This view assumes view assumes that regulators and supervisers are immune to politics--that their goal, and that of the government, is to maximize social welfare. Levine (2004) has recently suggested that such a view might not only be naпve, but be counter-productive, because politically motivated regulators and supervisors with more authority might be tempted to use their power to structure institutions in ways that provide sources of short run revenues for the government (and for themselves), at the expense of economic development. The evidence presented in this paper is consistent with the Levine view: all of the major reforms in banking law, in both the United States and Mexico, were motivated in good part by governments seeking sources of public finance. The difference in long run outcomes was not caused by the motivation of government, but by the political institutions that limited the government. REFERENCE LIST Acemoglu, Daron, Simon Johnson, and James Robinson. 2001. "The Colonial Origins of Comparative Development: An Empirical Investigation." American Economic Review 91: 1369-1401. Acemoglu, Daron, Simon Johnson, and James Robinson. 2002. "Reversal of Fortune: Geography and Institutions in the Making of the Modern World Income Distribution," Quarterly Journal of Economics 118. Acemoglu, Daron, Simon Johnson, and James Robinson. 2004. "Institutions as a Fundamental Cause of Long Run Growth." In Phillipe Aghion and Steve Durlauf eds., Handbook of Economic Growth. Acemoglu, Daron, Simon Johnson, James Robinson, and Pierre Yared. 2004. "Income and Democracy." Mimeo. Atack, Jeremy and Peter Passell. 1994. A New Economic View of American History From Colonial Times to 1940 (Second Edition), (New York: W.W. Norton). 59
Barro, Robert J. 1991. "Economic Growth in a Cross Section of Countries." Quarterly Journal of Economics, 106(2): 407-443. Barro, Robert J. 1997. Determinants of Economic Growth: A Cross-Country Empirical Study. Cambridge, MA: MIT Press. Bates, Robert H. 2001. Prosperity and Violence: The Political Economy of Development. New York: W.W. Norton. Beck, Thorsten, Ross Levine, and Norman Loayza. 2000. "Finance and the Sources of Growth." Journal of Financial Economics 58(1): 261-300. Beck, Thorsten, Asli Demirgьз-Kunt, and Ross Levine. 2002. "Law and Finance: Why Does Legal Origin Matter?" World Bank Policy Research Working Paper 2904. Beck, Thorsten, Asli Demirgьз-Kunt, Ross Levine, and Vojislav Maksimovic. 2001. "Financial Structure and Economic Development: Firm, Industry, and Country Evidence." In Asli Demirgьз-Kunt and Ross Levine eds., Financial Structure and Economic Growth: A CrossCountry Comparison of Banks, Markets, and Development (Cambridge, MA: MIT Press), pp. 189-242. Beck, Thorsten, Asli Demirgьз-Kunt, and Vojislav Maksimovic. 2003. "Bank Competition and Access to Finance." Unpublished paper presented at the Federal Reserve Bank of Cleveland and Journal of Money, Credit, and Banking 2003 Annual Conference on Banking Consolidation and Competition. Bodenhorn, Howard. 1990. "Entry, Rivalry, and Free Banking in Antebellum America." Review of Economics and Statistics 72(4): 682-86. Bodenhorn, Howard. 1993. "The Business Cycle and Entry into Early American Banking." Review of Economics and Statistics 75(3): 531-35. Bodenhorn, Howard. 2003. State Banking in Early America: A New Economic History. New York: Oxford University Press. Bordo, Michael D. Hugh Rockoff, and Angela Redish. 1994. "The U.S. Banking System from a Northern Exposure: Stability Versus Efficiency," The Journal of Economic History 54(2): 325-341. Brading, David. 1971. Miners and Merchants of Colonial Mexico, 1763-1810 (Cambridge University Press). Brunetti, Aymo 1997. Politics and Economic Growth: A Cross Country Data Perspective. Paris: Organisation for Economic Cooperation and Development. Calomiris, Charles W. and Charles M. Kahn. 1996. "The Efficiency of Self-Regulated Payments Systems: Learning from the Suffolk System," Journal of Money, Credit, and Banking 28(4). Calomiris, Charles W. and Eugene N. White. 1994. "The Origins of Federal Deposit Insurance," in Claudia Goldin and Gary D. Libecap eds., The Regulated Economy: A Historical Approach to Political Economy (Chicago: University of Chicago Press), pp. 145-188. Carmagnani, Marcello. 1994. Estado y mercado: La economнa pъblica del liberalismo mexicano, 1850-1911 (Mexico City: El Colegio de Mйxico and Fondo de Cultura Econуmica). Coastworth, John H. 1981. Growth Against Development: The Economic Impact of Railroads in Porfirian Mexico. Dekalb, Illinois: Northern Illinois University Press. 60
Davis, Lance E. and Robert E. Gallman. 2001. Evolving financial markets and International Capital Flows: Britain, the Americas, and Australia, 1865-1914 (Cambridge Eng.,: Cambridge University Press). Del Angel-Mobarak, Gustavo A. 2002. "Paradoxes of Financial Development: The Construction of the Mexican Banking System, 1941-1982." Ph.D. Dissertation, Stanford University. Demirgьз-Kunt, Asli and Ross Levine. 2001. "Bank-Based and Market-Based Financial Systems: Cross-Country Comparisons." In Asli Demirgьз-Kunt and Ross Levine eds., Financial Structure and Economic Growth: A Cross-Country Comparison of Banks, Markets, and Development (Cambridge, MA: MIT Press), pp. 81-142. Demirgьз-Kunt, Asli and Ross Levine (eds.). 2001. Financial Structure and Economic Growth: A Cross-Country Comparison of Banks, Markets, and Development (Cambridge, MA: MIT Press). Economopoulos, Andrew and Heather O'Neill. 1995. "Bank entry during the Antebellum Period." Journal of Money, Credit, and Banking 27(4); 1071-85. Engerman, Stanley L. 1970. "A Note on the Economic Consequences of the Second Bank of the United States," The Journal of Political Economy 78(4). Engerman, Stanley L. and Kenneth L. Sokoloff. 1997. "Factor Endowments, Institutions, and Differential Paths of Growth Among New World Economies: A View from Economic Historians of the United States." In Stephen Haber ed., How Latin America Fell Behind: Essays on the Economic Histories of Brazil and Mexico, 1800-1914 (Stanford: Stanford University Press), 260-304. Engerman, Stanley L. and Kenneth L. Sokoloff. 2001. The Evolution of Suffrage Institutions in the New World. NBER Working Paper 8512. Fenstermaker, J. Van. 1965. "The Statistics of American Commercial Banking, 1782-1818," The Journal of Economic History 25:3. Gatell, Frank Otto. 1966. "Sober Second Thoughts on Van Buren, the Albany Regency, and the Wall Street Conspiracy." The Journal of American History 53(1): 19-40. Goodrich, Carter. 1961. Canals and American Economic Development. New York, Columbia University Press. Gorton, Gary. 1996. "Reputation Formation in Early Bank Note Markets." The Journal of Political Economy 104 (2): 346-97. Grinith, Arthur III, John Joseph Wallis, and Richard E. Sylla. 1997. "Debt, Default, and Revenue Structure: The American State Debt Crisis in the Early 1840's." National Bureau of Economic Research, Historical Paper 97. Haber, Stephen. 1991. "Industrial Concentration and the Capital Markets: A Comparative Study of Brazil, Mexico, and the United States, 1830-1930." The Journal of Economic History, 51(3): 559-80. Haber, Stephen. 1997. "Financial Markets and Industrial Development: A Comparative Study of Governmental Regulation, Financial Innovation, and Industrial Structure in Brazil and Mexico, 1840-1930," In Stephen Haber, ed., How Latin America Fell Behind: Essays on the Economic Histories of Brazil and Mexico, 1800-1930 (Stanford, CA: Stanford University Press), 146-178. 61
Haber, Stephen. 2003. "Banks, Financial Markets, and Industrial Development: Lessons from the Economic Histories of Brazil and Mexico," In Josй Antonio Gonzбlez, Vittorio Corbo, Anne O. Krueger, and Aaron Tornell, eds., Latin American Macroeconomic Reforms: The Second Stage (Chicago, IL: University of Chicago Press), 257-292. Haber, Stephen. Forthcoming. "Mexico's Experiments with Bank Privatization and Liberalization, 1991-2003." Journal of Banking and Finance. Haber, Stephen, Armando Razo, and Noel Maurer. 2003. The Politics of Property Rights: Political Instability, Credible Commitments, and Economic Growth in Mexico, 1876-1929 (Cambridge University Press). Haggard, Stephan and Mathew D. McCubbins eds. 2001. Presidents, Parliaments, and Policy (Cambridge University Press). Hammond, Bray. 1947. "Jackson, Biddle, and the Bank of the United States," The Journal of Economic History 7(1): 1-23. Keefer, Philip. Forthcoming. "What Does Political Economy Tell Us About Economic Development--and Vice Versa?" Annual Review of Political Science. King, Robert G. and Ross Levine. 1993a. "Finance and Growth: Schumpeter Might be Right." Quarterly Journal of Economics 108: 717-738. King, Robert G. and Ross Levine. 1993b. "Finance, Entrepreneurship, and Growth: Theory and Evidence." Journal of Monetary Economics 32: 513-542. Knight, Alan. 1986. The Mexican Revolution: Porfirians, Liberals, and Peasants. Cambridge, England: Cambridge University Press. Krueger, Anne. 1974. "The Political Economy of the Rent-Seeking Society." The American Economic Review 64: 291-303. Kuntz Ficker, Sandra. 1995. Empresa extranjera y mercado interno: el Ferrocarril Central Mexicano, 1880-1907. Mexico City: El Colegio de Mйxico. Lamoreaux, Naomi. 1986. "Banks, Kinship, and Economic Development: The New England Case," The Journal of Economic History 46:3. Lamoreaux, Naomi. 1991. "Bank Mergers in Late Nineteenth Century New England: The Contingent Nature of Structural Change." The Journal of Economic History 51(3). Lamoreaux, Naomi. 1994. Insider Lending: Banks, Personal Connections, and Economic Development in Industrial New England (Cambridge, England: Cambridge University Press). Lane, Carl. 1997. "For a `Positive Profit': The Federal Investment in the First Bank of the United States, 1792-1802," William and Mary Quarterly 54:3. 601-12. La Porta, Rafael, Florencio Lopez-de-Silanes, Andrei Shleifer, and Robert W. Vishny. 1998. "Law and Finance." Journal of Political Economy 106(6): 1113-1155. La Porta, Rafael, Florencio Lopez-de-Silanes, Andrei Shleifer, and Robert W. Vishny. 2000. "Investor Protection and Corporate Governance." Journal of Financial Economics 58(1): 327. Levine, Ross. 1997. "Financial Development and Economic Growth: Views and Agenda." Journal of Economic Literature 35(2): 688-726. Levine, Ross. 1998. "The Legal Environment, Banks, and Long Run Economic Growth." Journal of Money, Credit, and Banking 30(Pt. 2): 596-620. 62
Levine, Ross. 1999. "Law, Finance, and Economic Growth." Journal of Financial Intermediation 81(1/2): 36-67. Levine, Ross. 2004. "The Microeconomic Effects of Different Approaches to Bank Supervision." Mimeo. University of Minnesota. Levine, Ross and David Renelt. 1992. "A sensitivity analysis of Cross-Country Growth Regressions." American Economic Review, 82(4): 942-963. Levine, Ross, and Sara Zervos. 1998. "stock markets, Banks, and Economic Growth." American Economic Review 88(3): 537-558. McCubbins, Mathew D. and Thomas Schwartz. 1984. "Congressional Oversight Overlooked: Police Patrols versus Fire Alarms" American Journal of Political Science 28: 165-179 Majewski, John. 2004a "Jeffersonian Political Economy and Pennsylvania's Financial Revolution from Below, 1800-1820." Mimeo. UCSB Majewski, John 2004b. "Toward a Social History of the Corporation: Shareholding in Pennsylvania Banking and Transportation Corporations, 1800-1840." In Cathy Matson ed., The Past and Future of Early American Economic History: Needs and Opportunities. Marichal, Carlos 2002. "The Construction of Credibility: Financial Market Reform and the Renegotiation of Mexico's External Debt in the 1880's." In Jeffrey L. Bortz and Stephen H. Haber (eds.), The Mexican Economy, 1870-1930: Essays on the Economic History of Institutions, Revolution, and Growth. Stanford, CA: Stanford University Press. Maurer, Noel. 2002. The Power and the Money: The Mexican Financial System, 1876-1932 (Stanford, CA: Stanford University Press). Maurer, Noel and Andrei Gomberg. Forthcoming. "When the State is Untrustworthy: Public Finance and Private Banking in Porfirian Mexico." Journal of Economic History. Maurer, Noel and Stephen Haber. 2004. "Related Lending and Economic Performance: Evidence from Mexico." Mimeo, Stanford University. Mexico, Secretaria de Hacienda. 1912. Anuario de Estadнstica Fiscal, 1911-12. Moss, David, and Sarah Brennan. 2004. "Regulation and Reaction: The Other Side of Free Banking in Antebellum New York." Harvard Business School Working Paper 04-038. Musacchio, Aldo, and Ian Read. 2003. "Bankers, Industrialists, and their Cliques: Elite Networks in Turn of the Century Mexico and Brazil." Mimeo. Stanford University. Neusser, Klaus, and Maurice Kugler. 1998. "Manufacturing Growth and Financial Development: Evidence from OECD Countries." The Review of Economics and Statistics 80: 636-646. Ng, Kenneth. 1988. "Free Banking Laws and Barriers to Entry in Banking, 1838-1860," The Journal of Economic History 48:4. North, Douglass C. 1961. The Economic Growth of the United States, 1790-1860. Englewood Cliffs, N.J.: Prentice Hall. North, Douglass C. 1990. Institutions, Institutional Change, and Economic Performance. Cambridge, England: Cambridge University Press. North, Douglass C., William Summerhill, et al. 2000. "Order, Disorder, and economic change: Latin America versus North America." In Bruce Bueno de Mesquita and Hilton L. Root (eds.), Governing for Prosperity. New Haven: Yale University Press: 17-58. 63
North, Douglass C. and Barry R. Weingast. 1989. "Constitutions and Commitment: the Evolution of Institutions Governing Public Choice in Seventeenth-century England." The Journal of Economic History, 49(4): 803-32. Perotti, Enrico, and Paolo Volpin. 2003. "The Political Economy of Entry: Lobbying, Inequality, and Financial Development." Mimeo, University of Amsterdam. Petersen, Mitchell, and Raghuram Rajan. 2002. "Does Distance Still Matter: The Information Revolution in Small Business Lending." Journal of Finance 57: 2533-70. Potash, Robert. 1983. The Mexican Government and Industrial Development in the Early Republic: The Banco de Avнo (University of Massachusetts Press). Przeworski, Adam, and Fernando Limongi. 1993. "Political Regimes and Economic Growth." Journal of Economic Perspectives 7(3): 51-69. Przeworski, Adam, Michael Alvarez, Josй Antonio Cheibub, and Fernando Limongi. 2000. Democracy and Development: Political Institutions and Material Well-Being in the World, 1950-1990. New York: Cambridge University Press. Rajan, Raghuram G., and Luigi Zingales. 1998. "Financial Dependence and Growth." American Economic Review 88(3): 559-586. Rajan, Raghuram G., and Luigi Zingales. Forthcoming. "The Great Reversals: The Politics of Financial Development in the Twentieth Century." Journal of Financial Economics. Razo, Armando. 2003. "Social Networks and Credible Commitments in Dictatorships: political organization and Economic Growth in Porfirian Mexico, 1876-1991," Ph.D. dissertation, Stanford University. Riguzzi, Paolo. 2002. "The Legal System, Institutional Change, and Financial Regulation in Mexico, 1870-1910: Mortgage Contracts and Long Term Credit," in Jeffrey Bortz and Stephen Haber eds., The Mexican Economy, 1870-1930: Essays in the Economic History of Institutions, Revolution, and Growth (Stanford University Press): 120-160. Rockoff, Hugh. 1974. "The Free Banking Era: A Reexamination." Journal of Money, Credit, and Banking 6(2): 141-67. Rockoff, Hugh. 1985. "New Evidence on Free Banking in the United States," The American Economic Review 75(4). Rockoff, Hugh. 2000. "Banking and Finance, 1789-1914." In Stanley Engerman and Robert Gallman eds., The Cambridge Economic History of the United States Vol. II (Cambridge, Eng., Cambridge University Press). Rousseau, Peter, and Richard Sylla. 2000. "Emerging Financial Markets and Early U.S. Growth." Mimeo. Rousseau, Peter, and Paul Wachtel. 1998. "Financial Intermediation and Economic Performance: Historical Evidence from Five Industrial Countries." Journal of Money, Credit, and Banking 30(4): 657-678. Sinkin, Richard N. 1979. Mexican Reform, 1855-1876: A Study in Liberal Nation-Building. University of Texas Press. Stevens, Donald F. 1991. Origins of Instability in Early Republican Mexico. Chapel Hill: Duke University Press. Sutton, John. 1998. Technology and Market Structure (MIT Press). 64
Sylla, Richard. 2000. "Experimental Federalism: The Economics of American Government, 17891914." In Stanley Engerman and Robert Gallman eds., The Cambridge Economic History of the United States Vol. II (Cambridge, Eng., Cambridge University Press). Sylla, Richard. 1975. The American Capital Market, 1846-1914 : A Study of the Effects of Public Policy on Economic Development (New York: Arno Press). Sylla, Richard, John B. Legler, and John Wallis. 1987. "Banks and State Public Finance in the New Republic: The United States, 1790-1860," The Journal of Economic History 47:2. Temin, Peter. 1968. "The Economic Consequences of the Bank War," The Journal of Political Economy 76(2): 257-274. Tennenbaum, Barbara. 1986. The Politics of Penury: Debt and Taxes in Mexico, 1821-1856 (University of New Mexico Press). Tsebelis, George. 1995. "Decision Making in Political Systems: Veto Players in Presidentialism, Parliamentarism, Multicameralism, and Multipartyism." British Journal of Political Science 25: 289-326. United States, Department of Commerce. 1975. Historical Statistics of the United States: Colonial Times to 1970. United States, Board of Governors of the Federal Reserve System. 1943. Banking and Monetary Statistics. Washington, D.C. Walker, David W. 1987. Business, Kinship, and Politics: The Martinez del Rio Family in Mexico, 1824-1867 (Austin, TX: University of Texas Press). Wallis, John, Richard E. Sylla, and John B. Legler. 1994. "The Interaction of Taxation and Regulation in Nineteenth Century U.S. Banking," In Claudia Goldin and Gary D. Libecap eds., The Regulated Economy: A Historical Approach to Political Economy (Chicago: University of Chicago Press, pp. 122-144. Wallis John and Barry. R. Weingast. 2004. "Equilibrium Impotence: Why the States and Not the American National Government Financed Infrastructure Investment in the Antebellum Era." Mimeo. Stanford University. Weingast, Barry R. 1995. "The Economic Role of Political Institutions: Federalism, Markets, and Economic Development," Journal of Law, Economics, and Organization 11: 1-31. Weingast, Barry R. 1997a. "The Political Foundations of Limited Government: Parliament and Sovereign Debt in 17th- and 18th-Century England." In John N. Drobak and John V.C. Nye (eds.), The Frontiers of the New Institutional Economics. San Diego: Academic Press: 213-246. Weingast, Barry R. 1997b. "The Political Foundations of Democracy and the Rule of Law." American Political Science Review, 91(2): 245-263. Wettereau, James O. 1942. "The Branches of the First Bank of the United States," Journal of Economic History 2 : 66-100. Wright, Gavin. 1986. Old South, New South: Revolutions in the Southern Economy Since the Civil War. New York: Basic Books. 65
Table 1 Number of U.S. Commercial Banks, 1860-1932
Year 1860 1865 1870 1875 1880 1885 1890 1895 1900 1905 1910 1914 1919 1924 1929
State Chartered Banks
Number (Millions $)
National Banks Assets Number (Millions $)
1,294 1,612 2,076 2,076 2,689 3,484 3,715 3,731 5,664 7,138 7,518 7,780 8,080 7,530
1,127 1,566 1,913 2,036 2,422 3,062 3,471 4,944 7,325 9,892 11,477
Total Banks
Number (Millions $)
13,053 11,388
18,767 17,511
25,151 22,922
27,864 27,349
National Banks as % of Total Assets Number (Millions $)
Sources: Lamoreaux 1991: 540; Davis and Gallman 2001: 268; Calomiris and White 1994: 151. U.S. Federal Reserve 1943: 24.
Table 2 U.S. Commercial Bank Lending Resources Per Capita, By Region in 1909
Region New England Eastern1 Southern Mid-West Western Pacific
Number Population
of Banks (millions)
2477 4,961
20.1 25.4
7,059 4,276 1,326
26.0 6.7 3.7
Persons Per Bank 9,527 8,121 5,130 3,681 1,573 2,828
United States
1. Mid-Atlantic States from New York to Washington D.C.
Source: Davis and Gallman 2001: 270.
Table 3 The Mexican Banking Industry, 1896-1912
1896 1897 1899 1900 1901 1902 1903 1904 1905 1906 1907 1908 1909 1910 1911 1912
Number of Reporting Banks1 6 10 13 17 20 23 25 26 26 28 28 34 35 35 35 34
Total Bank Assets (M U.S. $) 50 54 78 113 107 107 130 184 205 264 301 339 283 302 385 342
Banamex Market Share 58% na 51% 39% 38% 35% 37% 41% 39% 40% 44% 40% 37% 39% 39% 36%
Source: Maurer and Haber 2004.
BLM Market Share 28% na 26% 25% 22% 19% 17% 15% 18% 16% 14% 12% 12% 12% 12% 11%
Herfindahl Index2 0.42 na 0.34 0.22 0.20 0.17 0.18 0.20 0.20 0.21 0.23 0.19 0.17 0.18 0.18 0.16
Table 4 Industrial Concentration in Cotton Textiles, Mexico, Brazil, India, and the United States
Circa 1888 1893 1895 1896 1900 1904 1909 1912 1913
Mexico 18% 29% 33% 30% 30% 33% 38% 30% 31%
Four Firm Ratio
Expected Brazil India
19% 37%
17% 35%
15% 21%
14% 14%
U.S.A. 8% 7% 8%
Source: Maurer and Haber 2004.
Herfindahl Index
Mexico 0.022 0.038 0.042 0.041 0.038 0.042 0.045 0.039 0.041
Brazil 0.058 0.059 0.028 0.014
India 0.018 0.018

S Haber

File: political-institutions-banks-and-economic-growth-evidence-from.pdf
Title: Haber, Political Institutions, Bankers, and Economic Growt...
Author: S Haber
Author: wb153545
Published: Mon Dec 20 16:47:43 2004
Pages: 69
File size: 0.15 Mb

Developmental education, 94 pages, 0.47 Mb

iPhones Have Consequences, 4 pages, 0.3 Mb

, pages, 0 Mb
Copyright © 2018 doc.uments.com