Death Of A Middleman: The Banking Crisis And The Role Of The Securities Markets In Comparative Historical Perspective

Death of a Middleman: The Banking Crisis and the Role of the Securities Markets in Comparative Historical Perspective Oren Litwin1 15 July 2009

Introduction As the financial crisis unfolds, interested parties from all sides of the political spectrum have been quick to assign blame. To some (for example Krugman [2009]), the banking crisis was caused by deregulation and the fruits of unrestrained capitalism. To others, the crisis was set off by government incentives for risky behavior by mortgage lenders, particularly through asset securitization by Fannie Mae and Freddie Mac. While each of these explanations is partly true, both miss the larger structural reasons that led to the conditions the two sides decry. This paper hopes to remedy this lack by employing a comparative approach. The history of the American banking system, which at present is tightly intertwined with the securities markets, will be compared to that of Germany, which has a reputation for excessive control by banks and weak securities markets. Surprisingly, neither of these two characterizations used to be true—and the history of how each system changed over time illuminates the roots of the modern crisis, and where to go from here. Briefly, the American banking sector has been slowly dying for the last few decades, unable to compete with the strengthening capital markets—and this paper argues that we should stop propping up the corpse and let it finally molder away. This paper argues that the financial systems of the United States and Germany diverged in their logic because of the effect of bank regulations on the availability of capital to industry. In Germany, a banking environment with relatively few legal restrictions allowed for rapid industrial growth by granting easy access to capital—not least through the Berlin securities exchange, among the best in the world—despite Germany’s status as a late developer. The United States, despite having built an advanced banking system fifty years earlier than Germany’s, ran up to limits on growth in the second half of the 1800s due to the unit banking system, which made the banking system more fragile and limited banks’ ability to provide capital, stifling the development of large-scale industry. United States securities markets also developed more slowly as a result. Perversely, this resulted in powerful banks gaining more control over the American economy than would have happened otherwise. The trajectories of both systems were radically altered in the 1930s. In Germany, the Nazis deliberately throttled the securities markets as part of their campaign against finance capitalism; after the war, the German economy was dominated by direct lending relationships with powerful banks as a result, while the securities markets languished. In the United States, a federal government hostile to the great power of the banks passed a series of laws to curb their power, starting with the 1933 Glass-Steagall Act which divorced investment banking from commercial 1

This article is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 3.0 United States License, described here: http://creativecommons.org/licenses/by-nc-nd/3.0/us/ 1

banking. After an initial period of capital scarcity and accompanying economic stagnation, this had the effect of supercharging the development of the securities markets, as specialized investment banks had strong incentives to foster financial innovation in securities. Interest-rate controls enacted in 1966 accelerated this process and caused the banks to lose business to money-market mutual funds and other direct investments. The competitive pressure caused large commercial banks to take greater and greater risks to maintain profits and to circumvent GlassSteagall restrictions, culminating in the passage of the Gramm-Leach-Bliley Act of 1999 (GLBA), which repealed the provisions of Glass-Steagall preventing commercial banks from participating in investment banking. Ultimately, however, banks are becoming increasingly obsolete, and the damage caused to the financial system by their behavior is only made worse by protective subsidies and preferential regulation by the government and the Federal Reserve. Background This paper takes inspiration from Hall and Soskice’s (2001) theory of institutional complementarities. The authors argue that specific types of economic institutions will become mutually reinforcing and lead to very different patterns of behavior. Liberal market economies will tend to encourage interaction between firms based on supply and demand, and arm’s-length transactions. Coordinated market economies, on the other hand, encourage interaction based on strategic bargaining and large-scale coordination between actors. While these are ideal types and both styles of behavior can be found in any economy, the tone will be predominantly set by one of these logics. “In any national economy, firms will gravitate toward the mode of coordination for which there is institutional support” (Hall/Soskice 2001:9; cf. Baliga/Polak 2004, Monnet/ Quintin 2007). In particular, access to capital in a liberal market economy is granted most often through the capital markets; in coordinated market economies, capital is accessed via long-term relationships with powerful banks (Hall/Soskice 2001:28). The United States is considered the ne plus ultra of liberal economies, while Germany is a good example of a coordinated economy. A given pattern of institutions constrains future development of the system, by making some changes more attractive than others in the short-term—even when other changes might have better effects in the long-term (North 1990:7). Institutions will also guide the behavior of actors operating within them, by encouraging particular kinds of behavior, skills, and learning over other kinds (North 1990:74). Consequently, the constellation of actors and institutions will constrain each other’s development over time, causing the system as a whole to move according to an inherent logic guided by past history (North 1990:99). This is often called path dependency. Path-dependency effects are immensely powerful in the financial industry, in particular when considering the interaction between two components: the banking sector and the securities markets. Firms trying to raise capital from investors 2 need two related things: information and control. Investors must be able to tell whether a given firm is a good investment; additionally, they must 2

Firms have two means of raising capital: debt and equity. Borrowing money, either in a straight loan from a bank or by issuing bonds, imposes a fixed payment obligation on the firm. Issuing equity, i.e. stock, carries no fixed obligations, but rather a proportional claim on the profits of the firm. Equity is thus riskier but potentially more lucrative for investors. 2

have some assurances that the firm will stay a good investment, and that managers will not suddenly adopt risky practices once they can use other people’s money. Banks solve both problems by extracting information from firms and exerting control over behavior through the threat of withholding future credit. The public capital markets, on the other hand, are less able to monitor firms and exert control than are banks. Thus, the markets will tend to only fund companies that they trust (Calomiris 1995). In general, high-quality borrowers will raise capital in the financial markets, while lowerquality borrowers will turn to bank lending (Macey/Miller 1995). Whether a given firm will be considered high- or low-quality depends heavily on the relative sophistication of the banking sector and the financial markets. The better the markets, the easier for firms to get funding through them (Boot/Thakor 2004). This means that the banking sector tends to lose monopoly power as the markets reduce transaction costs, becomes more transparent, and develops institutional responses that allow it to exert control over firms. The foregoing stylized discussion assumed a dichotomy between the stock market and banks. In fact, there are different kinds of banks and different banking systems. A bank that engages in lending is called a commercial bank. An investment bank, on the other hand, provides services to firms seeking capital on the securities markets; investment bank fortunes are tied to the markets, and not to lending. A bank that does both is called a universal bank. A bank might want to help a firm issue securities, instead of lending money directly, in order to expand its client base and not to tie up its own capital in a limited number of firms. Additionally, the fees for underwriting (facilitating and marketing) a securities issue can be quite high, partly compensating for lost lending business. In environments of low information and high transaction costs, universal banks can fill an invaluable role. A universal bank can establish long-term lending relationships with new firms, giving them access to capital in exchange for a large degree of control over their activities. Ideally, as a firm become more seasoned, the bank can help shepherd it up the “financial pecking-order” of funding sources, moving from bank loans, to venture capital, to bonds, to stock (each requiring more trust from investors), taking advantage of its lengthening track record, visibility to investors, and the credibility of the sponsoring bank. Ultimately, the firm will graduate to relying heavily on the capital markets, using banks for short-term credit and as expert advisors in navigating the financial world (Calomiris 1995; cf. Fama 1985). This sequence depends on the quality of the financial markets—if the markets are sluggish, firms will be forced to rely more on banks for capital. Universal banks will be less likely to foster the development of the capital markets and the innovation of new financial products, all else equal, because the additional profit they receive is offset by cannibalized lending business. Furthermore, such innovations are soon adopted by competitors, reducing the potential gains if the banking system features heavy competition. Specialized investment banks, on the other hand, are far more likely to innovate because they have no lending business to sacrifice. And as the markets grow more sophisticated, they allow the creation of new products that are even more sophisticated. The reciprocal development of the markets and financial products conditions future innovation, in a classic example of a form of path dependency called increasing returns (Boot/Thakor 2004). 3

The German banking system allowed universal banks throughout its history. In such systems, universal banks tend to overpower their specialized competitors because of sheer size and their consequent influence over regulators. The United States, in contrast, enforced a segregated banking system after 1933 where commercial and investment banks were rigidly divided. In such systems, investment banks are given free rein. Due to the accelerated financial innovation in segregated banking systems, commercial banks will end up losing much more market share than they would in universal-banking systems (Boot/Thakor 2004). As we shall see, the regulatory structure can have a large effect on the functioning of a financial system. Regulations can improve the functioning of public markets by improving transparency and lowering transaction costs (cf. Zingales 2009), and can improve the banking system by reducing incentives for excessive risk-taking (cf. Lorenzoni 2008). Regulation can also be harmful, especially when it impedes activity that is natural to the financial industry and the prevailing state of technology. Transaction costs especially are a crucial driver of behavior and institutional change. Activities that are prohibitively expensive within a particular institutional framework can be extremely profitable in a different one (North 1990:95). In particular, “The costliness of information is the key to the costs of transacting” (North 1990:27). Therefore, the effectiveness of a particular regulation is directly related to the costs of sidestepping it through some other means (Kane 1977). Ultimately, as long as market participants are able to work around regulations, such regulations will fail in their intended purpose: “Prototypically, bureaucratic controls and market adaptation chase each other round and round, generating additional problems, confrontations, and costs for society at large” (Kane 1977:56). Indeed, while the typical large firm is seen to be hidebound and set in its ways, large firms facing burdensome regulation will be driven to seek new and innovative methods of corporate organization or product design that allow them to escape regulatory burdens, while performing much the same functions as before. This leads to what Kane (1981) calls the Law of One Tax-Adjusted Price. For example, early Australian banking law forbade banks from making advances on real estate.3 But by the 1850s, these restrictions had become a dead letter due to legal and financial “ingenuity,” and Victoria repealed them in 1888 (Hickson/Turner 2002). In the discussion to follow, the following arguments will be made. Germany’s system of large, branching universal banks featured very low transaction costs, allowing firms to easily raise credit both from banks and then later from the securities markets. In contrast, the early American banking system had very high transaction costs due to the unit-banking system, which impeded the pooling of capital. Industrial development was stifled, and those banks that did accumulate large amounts of capital developed tremendous power. Ironically, it was this power that ultimately led to Glass-Steagall. German Banking, 1830-1918

3

While a full discussion is outside the scope of this paper, banking-industry behavior is a primary contributor to speculative bubbles in real estate. See Herring and Wachter (1999, 2002) for a theoretical and empirical analysis, and policy proposals. 4

Nineteenth-century Germany is a classic case for understanding the effects of a strong banking system. Germany is considered to be a late industrializer, having developed heavy industry considerably after Britain and other international competitors. Nevertheless, in short order the German economy developed into the powerhouse of Europe. Gerschenkron (1962) attributed Germany’s rapid industrialization to its powerful banks, who used their power over industrial firms to organize them for large-scale, efficient production. While later research has added some caveats to this story, the period remains a good demonstration of the interaction between banks and the financial markets, and how banks benefit from large branching networks. In Germany, private banking evolved in two flavors (Guinnane 2002). The first group of private banks, many centered around the port of Hamburg and the trading hub of Cologne, expanded into the lending and bill-discounting business from their core activities as merchants. The second group, centered around Berlin, emerged only in the 19th century and was chiefly involved in lending to the Prussian state and marketing its debt. German banks were generally forbidden to issue private notes (not least because such notes competed with state debt), and therefore initially relied on investors for their capital (Calomiris 1995, Guinnane 2002). This is in contrast to the British experience, in which goldsmiths pioneered deposit banking and note issue very early on, shaping the development of British banking practices (and later those of the United States). Private banks tended to be small, made up of individuals or family groups or partnerships between investors (Guinnane 2002). (The Rothschild banking network is the archetypical example, unusual only in its eventual size.) By about 1830, private bankers were providing limited lending and investment banking services to new firms. Over time and as restrictions on incorporation loosened, private bankers organized larger Kreditbanken (credit-banks), which made it easier to raise money and finance large industrial projects like railroads. Between 1848 and 1870, forty credit-banks were founded by private bankers; the creditbanks operated in much the same way as private banks, but they tended to have much higher levels of capitalization (Da Rin 1996). The Grossbanken tended to favor dealing with large firms, creating what Gerschenkron called a “sectoral bias” (1962:10). “As early as 1853, the stated policy of the Bank of Darmstadt was to concentrate on firms with a turnover of 50,000 Guilders” (Baliga/Polak 2004). Another form of banking institution, which pioneered the use of mass deposits, was the Sparkassen, or savings banks. These were meant to provide a place for the poor and middle class to keep their savings; they were government-supported, sometimes government-owned, and initially were required to invest their reserves in government debt, providing the provincial governments with a cheap source of funds. (Competing banks often complained that the savings banks were shown favoritism for that reason.) The savings banks developed the “banking habit” among a wide strata of society, making it easier for the universal banks to begin accepting deposits themselves in the 1890s (Guinnane 2002, Da Rin 1996). Additionally, credit cooperatives formed in many places beyond the reach of the larger banks, allowing small, riskier borrowers some access to finance (Guinnane 2002). Private banks and credit banks were soon playing a key role in helping client firms issue securities. Until the 1870s, almost all German bonds were issued by governments and railroads; stock (much less common than bonds) was issued by railroads, banks, and industrial firms. To a 5

degree not found elsewhere, German private banks were heavily involved in helping the railroads secure funding during the early part of the century. The banks typically purchased a large part of the securities issues themselves for later resale (aside from what they took as their fee), as the trading markets of the day were underdeveloped and could not handle high-volume issues (Guinnane 2002). As a result, private banks were able to place their own representatives on the boards of the companies they financed, giving them considerable control. Da Rin (2002) argues that the banks’ equity holdings and control over industry gave them the incentives necessary to actively organize the German economy for better growth. In 1870, Germany liberalized the rules for forming joint-stock companies. This stimulated a rush of corporate reorganizations and subsequent equity offerings; at the same time, many private bankers cooperated to form new joint-stock banks (Burhop 2006). The 1870 law required that new issues of securities be fully subscribed. Therefore, underwriters tended to be large banks who could purchase the whole issue, and then resell it into the market over time (Fohlin 2002). Banks marketing shares often chose to retain small equity stakes in their firms, or were forced to retain equity if they could not market the whole issue. This gave them voting rights and membership on the corporate board, much as the private banks had done previously (Fohlin 1999). Banks also held a great deal of stock in trust for clients, who typically signed over their proxy votes to the bank. In this way, even once a bank had completely marketed a stock issue, it could retain a great deal of power over the firm’s board (Guinanne 2002). Germany saw a rapid boom in equity issues as banks helped the new firms raise capital; this soon produced a glut of offerings, leading to a market crash in what was called the Panic of 1873 (Da Rin 1996).4 The Panic tended to increase firms’ reliance on the large banks. Investors were much more cautious with their money, and banks became more important as a facilitator of capital investment. Firms that lacked close ties with a major bank appear to have had much less access to capital than those with such ties (Becht/Ramirez 2003). Additionally, Germany instituted a series of shareholder protection laws, most notably the Corporate Law of 1884. This included a mandatory one-year waiting period between incorporation and equity offerings, and strengthened the power of the Aufsichtsrat (supervisory board) over management. As bank directors often sat on these boards, the Corporate Law and others like it (such as the closing of the futures markets) may have had the effect of strengthening the position of the banks over management (Becht/Ramirez 2003).5 One law that had a significant impact was a new tax on market trades introduced in 1881. This gave an advantage to banks, which could hold investors’ shares in trust and execute transactions internally, only incurring taxes on net purchases and sales across the whole bank. As a result, banks quickly gained custody over investors’ securities and trading shifted away from the open exchanges—at least until the tax was extended to internal trades in 1900 (Fohlin 2002). The banks took advantage of their new power to enforce exclusive relations with their firms, 4

The similarities between this sequence and the one leading up to the 1929 crash, described below, are instructive. While a full discussion is outside the scope of this paper, it seems plausible that any sudden increase in liquidity or investment capital leads to speculative bubbles and subsequent oversupply, as the market adjusts to the new conditions. Cf. Fisher (2007:248-251). 5

However, Fohlin (2002) finds little quantitative evidence to support this point. 6

allowing the banks to gain more information about their borrowers and monitor them more effectively (Da Rin 1996). However, by the 1890s, the great banks had lost their power over industry; given the easy entry of new banks into the business, the banking sector had become fully competitive (Calomiris 1995). Additionally, firms were growing large enough to regain some control from the banks and play them off against each other. While the early process of cartelization was often bank-led, by the 1890s it was being run by the firms themselves, driven by technological concerns (Da Rin 1996). In response, the banks went through a period of consolidation, trying to regain their competitive edge by buying competing banks, incorporating more activities, more size, and more geographic scale—becoming true “universal” banks, capable of accumulating the savings of many depositors and providing financing and investment banking services to even the largest cartels (Burhop 2006). Even then, however, the old era of exclusive relationships could not last; large industrial groups often had relationships with several banks at once, and relied more on self-financing through retained earnings and their increasing access to the financial markets (Da Rin 1996; cf. Becht/Ramirez 2003). As part of this, industrial firms started placing their own directors on the boards of their associated banks, especially banks that facilitated securities offerings, in a pattern of interlocking directorates (Fohlin 1999). Ultimately, the prewar German banking system consolidated into six Konzerns, affiliated networks of local and regional banks tied closely to universal banks. The smaller banks allowed universal banks to extend their reach across Germany and continue serving smaller borrowers and local communities, which made the banking system as a whole much more stable due to diversification; the networks also gave the large banks a ready market for securities issues (Guinanne 2002, Da Rin 1996). Finally, the local banks could more easily monitor the sprawling operations of German cartel groups, centralizing information and allowing the universal banks to focus on relationships with the top management. The universal banks gave the smaller local banks access to the financial markets and much greater capitalization. While both sides benefited, there was no doubt that the universal banks ran the show (Da Rin 1996). At the turn of the century, German firms relied considerably less that American firms on long-dated debt issuance; with more advanced financial markets, they were able to raise money with more advantageous terms, such as through equity (Calomiris 1995). The underwriting spread (fee) for new issues was between 3 and 5 percent of an issue, much less than in the United States in the same period. “German bankers’ spreads on equity were less than one-fourth those in the United States. Small German firms were able to issue equity for less than the cost large American corporations paid for issuing bonds” (Calomiris 1995:297). The German universal banks were able to efficiently market new issues to their base of clients. This had the effect, however, of cutting into their traditional lending business (Da Rin 1996). The largest banks continued to grow, aided by a strong stock market that richly benefited their investment banking business (Fohlin 2002, 2007). At the eve of World War I, 17 of the largest 25 German firms were banks; the Konzernen also controlled access to the Berlin financial markets, which were second only to the London markets on overall size and liquidity (Baliga/ Polak 2004, Fohlin 2007). WWI greatly accelerated the concentration of the banking sector, as the state employed the largest banks to help raise funds for the war effort (Fohlin 2002). The savings banks in particular played a key role; by the end of the war, German war loans made up 7

more than 35% of the Sparkassen’s assets (Guinanne 2002). This gave them crucial experience that they later used in marketing general securities to the public. American Banking, 1780-1932 In the 1780s, there were only three, brand-new banks in the United States—one each in Boston, Philadelphia, and New York. Additionally, the country was awash in devalued paper currency issued by the separate states. In 1791, Alexander Hamilton proposed that the federal government charter the Bank of the United States (a public-private bank) as a means to sell federal debt, retire the debt of the states, introduce a more uniform currency backed by gold and silver, and develop the banking system. In the next year, the new bank issued 6%-yield securities that were fully subscribed, trading at or above par in newly established securities markets in Philadelphia, New York, Boston, and Baltimore. These new securities markets soon traded in other, more local securities as well, growing more formalized and sophisticated in the process (Rousseau/Sylla 2005; cf. Baliga/Polak 2004). Had the BUS remained a monopoly bank comparable to the Bank of England, American history would have been very different. Instead, entrepreneurs across the country applied for bank charters of their own from the states; the states, seeking to defend their financial systems against the encroaching BUS, were cooperative (Rousseau/Sylla 2005). In the 1790s, 28 new banks were chartered, and 73 more in the next decade. By 1860, there were some 1600 state banks across the United States. “From the 1790s to the middle of the 19th century nowhere else in the world was the banking corporation developed as a competitive business enterprise to the extent that it was in the United States” (Rousseau/Sylla 2005:5). This includes Britain; as early as 1825, America had more than twice as much bank capital as Britain, despite having a smaller population. Nonbanking firms, in turn, were able to secure charters of incorporation from the states which facilitated investment; many of these new corporations were in and around the cities with active securities markets, which let them market shares (Rousseau/Sylla 2005).6 The apogee of early American banking was reached during the so-called free banking period, which lasted from the demise of the Second Bank of the United States in 1832 to the beginning of the Civil War. In this period, there was no central bank, states regulated their own banking industries, banks could issue private notes backed by state bonds, and in most states, anyone with sufficient collateral could gain a bank charter (Briones/Rockoff 2005; cf. Sechrest 1993, Shambaugh 2006). While entry into the banking field was free, states often imposed restrictions on bank practices. The most significant of these for our purposes was called unit banking. Banks in most states were allowed to operate only a single location. Branching was forbidden, including branches from out-of-state banks (Briones/Rockoff 2005). This limited banks’ size, available capital, and ability to diversify their lending portfolio. Later American regulation limited the size

6

Early stock exchanges had very low liquidity. Stock par values were quite high, typically $1000 per share at a time when per-capita annual incomes were typically between $100 and $300. Rousseau (2009) finds that equities trading in the Boston stock exchange were predominantly bank stocks until par values came down into the low hundreds in the mid-1800s, stimulating liquidity and allowing large numbers of industrial firms to issue shares. 8

of a bank loan to 10% of its surplus plus capital; therefore, a firm’s access to credit was directly proportional to the size of the bank or banks from which it borrowed (Giedeman 2005). Until the second half of the 19th century, unit banking did not cause shortages of credit in practice. In antebellum America, as in Britain, most manufacturing took place in small-scale artisan shops or relatively modest factories, which the banks were able to finance adequately (Bodenhorn 1999). Early American banking, particularly in New England, was quite similar to the German model of close integration with industry (Calomiris 1995). The Civil War period saw several changes to the operation of the banking system. First, the governments of both sides turned to the banks to help them market war bonds. This gave the banking system valuable experience in forming underwriting syndicates and finding buyers. The banking networks formed during the war persisted and deepened afterward, forming the basis for the later emergence of the great underwriting empires of the 1880s and 1890s. Nevertheless, the process was made costly by prohibitions against branching and interstate consolidation, which prevented the economies of scale and of information-processing that were emerging in Germany at the same time (Calomiris 1995). Second, the federal government instituted the National Banking Act of 1864 and began to charter its own banks. National banks had the privilege of issuing national bank notes, which were fully backed by U.S. Treasury bonds. At the same time, the federal government imposed a new tax on note issues by state banks, with the result that state bank notes were soon replaced by national bank notes (Calomiris/Mason 2008). After the Civil War, large differences in the cost of capital developed across regions. These differences were starkest in the industrial sector, which had returns on capital far exceeding those of other sectors of the economy. This indicated the difficulty of raising money for large industrial projects (Calomiris 1995). 7 Giedeman (2005) finds that large American firms had insufficient access to capital even during the comparatively late period of 1911-1922. By this time, industrial technology was beginning to favor modern, large-scale manufacturing instead of the old pattern of small artisan shops. Large firms, which needed more capital than most banks could handle, instead relied on the short-term commercial-paper and long-term bond market when they could. Debt issues allowed firms to access larger pools of capital than local banks could provide. Commercial paper, in particular, was for a long time a unique feature of American finance (Calomiris 1995). By syndicating debt offerings on behalf of their client firms, banks could provide the capital that they themselves did not have. Moreover, banks also issued their own debt on the markets, increasing their capital reserves and allowing them to tackle larger-scale deals (Calomiris 1995). However, debt issues were not a costless alternative to bank loans. Early American securities exchanges still tended to be thinly-traded by modern standards, and limited by region. Additionally, only high-quality firms could market their debt or equity on the exchanges, leaving many firms cut off from access to capital (cf. Rousseau 2009). Unit banking also restricted the ability of banks to market securities offerings. They lacked a ready market for shares and the economies of scale that a large branch network would provide.

7

One consequence is the fragmented American electrical grid. While in Germany the utilities were financed by large banks and could afford to install a bigger and better-integrated electrical system, American utilities were financed by electrical-equipment manufacturers for lack of bank capital, limiting the size of their installations (Calomiris 1995). 9

While they could and did organize networks of banks into underwriting syndicates, the process involved considerable transaction costs and friction (Calomiris 1995). As a consequence, bank underwriting fees for bringing new issues to market were extremely large, and varied with the risk of the issue and the consequent difficulty of finding buyers. Even when the great investment banking firms such as J.P. Morgan emerged in the late 1800s, the typical underwriting spread for equities was 20% of the entire issue.8 It is worth emphasizing how large these spreads are. A 20 percent spread indicates that a firm only receives 80 cents for every dollar of claims it issues. This places a substantial cost on investments, especially by young, unseasoned firms. An investment opportunity must be able to generate enough income to pay interest or dividends to claimants and compensate existing shareholders by an amount (in present value) in excess of 20 percent of the project’s cost (Calomiris 1995:296).

The shortage of capital had costs. American industry depended more heavily on labor and was more natural-resource-intensive than did German industry, as a substitute for more efficient methods of production which needed more capital. Additionally, much more of the economy was devoted to consumer goods than capital equipment; as investment capital was scarce, it made more sense to make what you could sell to consumers (Calomiris 1995). 9 As noted, those banks with expertise in securities issues reaped huge fees from issuing firms. Additional fees were generated through restructuring and mergers, particularly in the troubled railroad industry; for some firms, fees could go as high as 10% of the value of the company (De Long 1991). As in Germany, American investment banks often took an equity stake for their fee in lieu of cash. The difference was in the consequent degree of bank control: because the underwriting fees were so high, banks controlled much larger fractions of their firms. This would soon have political consequences. Even in the heyday of American universal banking, only a handful of investment banks had the capital, the connections with investors, and the expertise to market security issues of more than $10 million; the largest of these were J.P. Morgan and Co. and Kuhn, Loeb, and Co. (De Long 1991).10 J.P. Morgan in particular benefited from strong connections to the London capital markets, allowing the firm to raise tremendous amounts of capital from European investors.11 Other investment banks simply did not have access to enough capital. As a result, not banks in general but particular bankers such as J.P. Morgan himself gained concentrated control over large swaths of American industry, seating his allies on the boards of dozens of industrial firms and railroads. “Morgan’s Men” did not (only) exploit the firms they oversaw, however; analysis of the firms indicates that “Morganized” firms were more profitable, and their stock traded at a 8

That this fee accurately reflected the cost of bringing issues to market, and was not simply an expression of bank power over industrial firms, is indicated by the decline in spreads in the 1920s even as banks grew more powerful. The most likely explanation for the decline is the emergence of large institutional buyers such as life-insurance firms that made bulk placement feasible (Calomiris 1995). 9

Could the much-maligned American consumer culture have gotten its start due to government restrictions on banking? 10

Later, with the development of bank-retailed investments, they would be joined by the First National Bank and the National City Bank (eventually renamed Citibank). 11

Nor was J.P. Morgan the last bank to do so, as we shall see. 10

premium to the market estimated between 7% (Simon 1998) and 30% (De Long 1991). Morgan was able to gain his power by providing good value for investors, in part by constructing industry cartels. “Morgan made companies, effectively warranted them against failure, and stood behind the warranty; and the securities industry evidently knew it” (Sabel in De Long 1991:246). Starting at about 1890, the New York Stock Exchange (which had listed few industrial securities previously) saw a sudden growth in equity issues and a consequent deepening of the financial markets, partly from the ongoing restructuring of industry. The growth in trading volume increased during the depression of 1893-1897, as equities were viewed by banks as less risky to hold than loan portfolios (Smiley 1981). When the economy recovered in 1897, the stage was set for a sustained growth in the equities markets—the stock exchanges had become more advanced and major institutional buyers such as banks were more willing to hold equities than ever before. Meanwhile, the growing power of the investment banks started to alarm the political establishment: Banks became victims of their own success; by 1912, 18 financial institutions sat on the boards of 134 corporations with $25.325 billion in combined assets. Of these 18 institutions, five banks held the lion's share: J. P. Morgan & Co., First National Bank, National City Bank, Guaranty Trust Co., and Bankers' Trust sat on the boards of 68 nonfinancial corporations with $17.273 billion in assets. To put this number in perspective, U.S. GNP in 1912 was $39.4 billion; hence, these five banks controlled industrial assets (on behalf of others) representing 56 percent of the country's GNP (Simon 1998:1081).

As a result of their growing power, the “money trust” drew the enmity of Progressives. Presidents Theodore Roosevelt, Taft, and Wilson all fought vigorously to curtail the banks; their cause was given intellectual heft by the writings of Louis Brandeis, who believed that the social effects of monopoly and cartels were so harmful that interlocking directorates should be outlawed, even if they were in some sense the “best” way to organize industrial production (Simon 1998). Congress ultimately passed the Clayton Act of 1914, which outlawed interlocking directorates and placed limits on the amount of stock banks could own in nonfinancial firms. J.P. Morgan, hoping to derail the legislation, had announced just beforehand that his representatives would resign from the boards of thirty firms (but not all of them). The move was partly successful: the Clayton Act did not forbid bankers from sitting on industrial boards entirely, as had been initially proposed (Simon 1998). Nevertheless, the restrictions were enough to preclude the emergence of German-style universal-bank control over industry (Calomiris 1995).12 The anti-banking campaign was abruptly halted by World War I. Bankers were heavily involved in financing the Allied war effort, and in return the Wilson administration gave tacit approval to continued bank influence over the industrial sector (Ramírez 1999).

12

It is noteworthy that while bank control over management was weakened, there was nothing with which to replace it. Shareholders remained dispersed and unable to monitor or discipline companies on their own, and it would be decades before the rise of powerful institutional investors. In the meanwhile, the manager class had unchecked power over firms they did not own (De Long 1991; cf. Mises 1996:306). Even now, institutional shareholders in the United States are still not powerful enough to truly control management (Macey/Miller 1995, Zingales 2009). 11

Yet again, the needs of wartime finance stimulated new developments in the financial system. This time, the key change came when the government enlisted banks to sell Liberty Bonds directly to the public (De Long 1991). Banks developed sales teams to sell the new issues, gaining expertise in selling securities to unsophisticated homeowners. The experience made a deep impression on one man in particular—Charles Mitchell, eventual president of National City Bank. “[He] came to recognize that a financial empire does not have to be built by slowly creating a reputation as a shrewd judge of investments but can be built through direct salesmanship by uninformed representatives” (De Long 1991:231). Over time, national banks began engaging in securities dealing and underwriting, which took them outside of the bounds of the National Banking Act of 1864. Regardless, it became the common practice, and when the McFadden Act of 1927 authorized banks to operate internal securities departments, it was only ratifying and formalizing an existing practice (Kroszner/Rajan 1997). Banks benefited from new forms of corporate organization such as the bank holding company, which allowed formally distinct banks to be consolidated within the same corporate structure. This allowed banks to partly overcome the restrictions of unit banking and develop larger distribution networks for securities (cf. Kane 1981). Securities affiliates of large, universal-style commercial banks had a distinct advantage over investment banks in that their large branch networks let them more easily market securities issues (Calomiris 1995). A prime example was National City Bank. Charles Mitchell became president of NCB in 1921, and with great fervor set about building a massive sales organization to mobilize the “large, new army of investors” that had purchased their first bonds during the war (Wilmarth 2005). National City Bank, as well as Chase Bank, were among the most aggressive of the new breed of securities marketers, and securities they sold ended up having especially poor performance. Bonds underwritten by both Chase and National City sold at discounts to their peers, indicating that the market recognized the banks’ risky practices—even if the true magnitude of the problem was still larger than thought (Kroszner/Rajan 1997). Commercial banks underwrote 22% of new securities in 1927, and 45% in 1929 (Kroszner/ Rajan 1997). Over a hundred banks opened up new securities affiliates or developed internal securities departments in the decade ending in 1931. The increasing access to the capital markets spurred many firms to issue securities instead of relying on loans. “Beginning in the latter half of 1928 and continuing through October of 1929, equity issuance grew rapidly in popularity, and firms switched from issuing bonds to issuing stocks” (Kroszner/Rajan 1997:509; cf. Wilmarth 2005). The quality of bond issues was also deteriorating, as underwriting standards loosened. Easy access to investment capital led to a glut of manufacturing capacity. Large industrial projects were begun based on overly optimistic projections of future demand, which also discounted that competing projects were doing the same thing. Soon, matters came to a head and deflation resulted (Schweikart 1991, Wilmarth 2005). Large firms suddenly faced huge deficits, making their heavy debt loads unsustainable. As a result, many firms defaulted. A similar process was playing out on the stock exchanges. The rush of new stock offerings had saturated the market; stock returns were being kept artificially high by the aftereffects of wartime inflation and easy access to margin, where banks let traders finance as much as 90% of

12

their stock purchases on credit (Wilmarth 2005). This could not last; the market ultimately reverted to the mean in the Crash of 1929. Banks, who were especially exposed to macroeconomic risk due to their changing business practices, soon faced a crisis of bank runs and failures. A crucial element in the mass failures of the banks was the declining value of their asset portfolios (Schweikart 1991, Wilmarth 2005). First, many borrowers defaulted. Second, banks often held large amounts of securities, particularly international bonds, which plummeted in value as the market crisis unfolded. Often, big-bank securities affiliates had retailed the riskiest securities to unsophisticated local banks, which consequently went bankrupt at a high rate (Wilmarth 2005).13 The carnage was exacerbated by the unit-banking system. In Canada, which had strong branching banks, not a single bank failed during the Depression (Bordo/Rockoff/Redish 1996). Similarly in California (which allowed branching), even singleton banks tended to last longer during the crisis if they faced competition from a branch bank like Bank of America, in part because they had already become more efficient to survive (Carlson/Michener 2009).14

German Banking, 1920-Postwar The Weimar hyperinflation of the 1920s was devastating to the banking sector. Worst hit were the Kreditbanken, whose clients in the industrial sector were taking huge losses. Several banks failed in the banking crisis of 1931, and the survivors became even more concentrated than before—the “Big Six” became the “Big Three” (Guinanne 2002, Kopper 1998). The large universal banks had depleted capital reserves and an uncertain future when the Nazis rose to power. Being dependent on state funds to survive, they were forced to sell large stakes to the government, the Reichsbank, or its subsidiaries: 91% in the Dresdner Bank, 70% in the Commerzbank, and a minority stake in Deutsche Bank (Kopper 1998). Ultimately, banking policy became totally dependent on state goals. The Nazis believed that the power of the banking sector had grown too great, and that the industrial sector had been weakened by overborrowing and dispersion of control. They put in place policies designed to encourage self-financing of new development by industrial firms (Kopper 1998). The maximum allowable level of dividend payments was sharply restricted, discouraging new stock issues; at the same time, banks were discouraged from making long-term loans to industry by capital controls that mandated a low interest rate.15 Industrial firms instead ran up large debts to suppliers of raw inputs, and as the German economy recovered they also 13

This is in contrast to the behavior of the German Konzernen; because universal banks marketed their securities to local banks affiliated with the same Konzern, they had a strong interest in ensuring the securities’ quality. 14

In part, however, their performance was dumb luck. As competition increased, Californian banks tended to shift their portfolios away from securities (which were seen as safe investments) into loans (which were more lucrative, but seen to be riskier). As a result, they were insulated from the capital-market crash (Carlson/Michener 2009). 15

The capital controls were also meant to facilitate borrowing by the state. As before, the chief role here was played by the savings banks, who held nearly half of Reich debt by 1939 (Kopper 1998). Nazi capital policy also favored particular sectors key to economic autarky, particularly synthetic fuel, mining, and metals production. Banks were pressured to invest large sums in these industries, despite low projected profits. 13

reinvested more of their profits instead of acquiring equities, as the Nazis had hoped. “Despite economic recovery, the issue of new shares and industrial debt fell to levels below those that had prevailed at the nadir of the economic crisis in 1932” (Kopper 1998:54). In short, the Nazi regime deliberately starved off what had been one of the best capital markets in the world, and tied the banks closely to government policy: National Socialist credit policy prevented the provision of credit from resuming its traditional routine by directing the banks into a rather riskless, uninnovative and boring business with limited, but guaranteed profit margins. Allocating credits and loans became less and less a matter of skilful banking through being more and more determined by government interests (Kopper 1998:62).

In the turmoil after World War II, the new West German political order ultimately settled on a form of cooperative corporatism in which government and universal-bank ownership played a key role in the ownership and management of the large corporations. The new economic system in West Germany was carefully designed to produce stability first and foremost, in a political climate in which labor needed to be incorporated into economic governance and instability would have been fatal. Growth, on the other hand, was sacrificed. This tended to make equity investing relatively less attractive than lending, reducing the incentives to turn to the capital markets for financing and strengthening the position of the banks (Macey/Miller 1995).16 In particular, banks have continued to benefit from their control of clients’ equity proxies, and bank representation on the advisory boards of firms is among the highest in the world (Santos/Rumble 2006). The securities markets never recovered their former glory (Fohlin 2007). “German firms borrow $4.20 from banks for every dollar they obtain from capital markets, whereas American firms borrow $0.85 for every dollar they raise in the capital markets” (Macey/Miller 1995). As of 2002, there were only about 700 firms on the German stock exchange, while in the United States there were over 7000 (Guinnane 2002). In modern Germany, most large-scale investing is done via personal connections with executives in the banks and other firms, instead of relying on the public capital markets (Hall/ Soskice 2001:23). Of the European countries, Germany has the highest percentage of firms controlled by other firms; around 80% of direct equity stakes in public firms are owned by firms instead of individuals (Goergen/Manjon/Renneboog 2008). This system makes it difficult for new firms to start up and attract financing; at least as late as 1990, Germany had no dedicated venture-capital sector (Hart 1992:188). More recently, there has been some reform of the capital markets to encourage liberal-style investing. In large part, this was due to competition from aggressive American investment banks that were able to penetrate the German market. German universal banks, relying on their dominance of the economy, had become sluggish and uncompetitive (Boot/Thakor 2004). Advances in the European securities markets since the 1970s have transfered some power to borrowing firms and away from the incumbent universal banks, who now face a more challenging environment. In particular, competing banks now tend to assume much more of the risk when bringing new issues to market (Battilossi 2000). 16

“As a generalization, the more easily others can affect the income flow from someone’s assets without bearing the full costs of their action, the lower is the value of that asset” (North 1990:31). 14

American Banking, 1933-Present In response to the Crash of 1929, the Senate began an investigation of Wall Street practices. The investigation became known as the Pecora Commission, after Ferdinand Pecora—who was not a senator, but the chief counsel of the investigation. Pecora and his assistants exposed widespread malfeasance by investment and commercial banks, especially National City Bank and Chase. For example, banks had often knowingly marketed securities on behalf of failing companies to enable those firms to repay bank loans (Wilmarth 2005).17 The Roosevelt Administration used the Pecora hearings as a springboard for its vast program of banking and securities regulation. Chief among the new laws was Glass-Steagall, which forced commercial banks to give up their investment banking business. Krugman (2009:157) argues that the division between commercial and investment banking was enforced in order to allow tight regulation of the commercial banks’ deposit-taking practices, to prevent bank runs in the future. However, the main focus of Senator Carter Glass and his allies was not on bank runs, but on the banks’ role in creating the speculative boom in stocks (Wilmarth 2005). He and his allies argued that aggressive marketing of new issues, excessive credit given to firms and to stock traders, and conflicts of interest between underwriters and brokers, led to the glut of shares and the eventual crash. The Glass-Steagall Act outlawed the underwriting of securities by commercial banks or any of their subsidiaries. “From a governance perspective, the importance of Glass-Steagall was that it no longer allowed private bankers to exist” (Simon 1998:1091). The Act also forbade national banks from owning any corporate stock (Santos/Rumble 2006). The industrial sector was suddenly cut off from its former access to capital. Firms responded in different ways. Many firms replaced their former relationships with banks by allying with insurance firms, which controlled large amounts of assets; over the next several decades, representatives of insurance companies began showing up on corporate boards and playing much the same role as bank representatives once had (Ramírez 1999). American firms also responded by entering into subsidiary relationships with large, well-capitalized holding companies, which could then operate internal capital markets and allocate resources among subsidiaries, acting as a universal bank would (Baliga/Polak 2004). In general, however, firms’ access to capital was sharply constrained—particularly those firms without ties to insurance companies (Ramírez 1999). Indeed, Senator Glass himself was soon dismayed by the shortage of capital. He had expected that investment banks would be able to adequately finance the industrial economy on their own; when it became clear that they were not up to the job, Glass proposed an amendment to the Banking Act of 1935 that would have given commercial banks limited authority to issue bonds (but not stock). But the Roosevelt Administration opposed the amendment and it was soon dropped (Wilmarth 2005). The new banking regulations had the additional effect of disrupting both bank and market control of firm managers, contributing to managers’ growing power in the American economic 17

For a more critical view of the Pecora Commission, particularly for its ignorant characterization of stock pools (which has sadly become conventional wisdom), see Mahoney (1999) and Jiang/Mahoney/Mai (2005). 15

system (Calomiris 1995). Furthermore, the position of unit banks (reeling from crisis-related losses at the time) was buttressed by the new legislation, delaying reform of the unit-banking system for decades. “Once the Great Depression legislation was passed, it resuscitated unit banks as a powerful special interest resisting reform or repeal of Great Depression protections” (Calomiris 1995:307). Of all the effects of the new legislation, two are crucial for our argument. First, commercial banks were barred from investment banking, and investment banks were barred from commercial banking. Thus, investment banks had strong incentives to build up the financial markets, and develop new types of securities products to poach business from the commercial banks (cf. Baliga/Polak 2004). This was sharply different from the normal pattern followed by universal banks, in which securities innovation was limited by the need to retain lending business. While it took some decades for the capital market to recover, in time financial innovation by investment banks became a direct threat to the commercial banks. Second, the new legislation outlawed paying interest on demand accounts, and directed the Federal Reserve to impose capital controls on the interest rates banks could pay their depositors. This the Fed did on November 1, 1933 with Regulation Q, which limited deposit rates to 3%. This rate was actually above the prevailing market rate at first; the Fed did not see it as its job to restrict the industry as a whole, as Congress had wished, but instead to control the most aggressive banks who might attract deposits with excessively high interest rates (Gilbert 1986). Regulation Q policy therefore did not distort the credit market. As late as the 1960s, banking was a safe and lucrative profession. Banks had a simple, effective business model: attract interest-free demand deposits and low-interest savings deposits, invest in safe government debt, and engage in judicious lending to high-quality businesses (Wilmarth 2002).18 When prevailing interest rates started creeping up, the Fed raised the Regulation Q limits to accommodate them (Gilbert 1986). Then, in 1966, Congress passed legislation radically changing Regulation Q policy (Gilbert 1986). Concerned that mortgage rates were increasing, Congress felt that rising interest rates were the fault of competition between banks and thrifts to attract deposits. The new interest-rate caps were set below the yield on Treasury bonds, in an attempt to halt the rise in rates. This attempt was doomed from the start, as rates were rising due to growing borrower demand, rather than increasing bank costs (Gilbert 1986). A key goal of the new deposit-rate ceilings was to favor particular interests such as the thrifts, savings-and-loan banks, mutual savings banks, the housing industry, and construction unions at the expense of commercial banks, credit unions, and small savers (Kane 1981, Gilbert 1986). Favored institutions linked to the mortgage industry were given higher interest-rate ceilings, allowing them to attract more deposits and hence offer more mortgages. To sustain the implicit subsidy, Federal regulators were forced to extend regulation over more and more activities as households tried to find new places for their money; not least, minimum denominations for Treasury bills were raised to take them out of reach of small households. On the other hand, regulations were relaxed for large depositors. For example, repurchase agreements transformed large-dollar deposits into contractually guaranteed securities 18

This business model is summarized by the 3-6-3 Rule: borrow at 3%, lend at 6%, play golf at 3:00. 16

transactions, allowing banks to compensate depositors without falling afoul of interest-rate restrictions (Kane 1977). In June 1970, deposits of more than $100,000 were exempted from Regulation Q entirely (Gilbert 1986). All in all, banks and thrifts benefited at the expense of the poor and middle class, while the rich were largely spared.19 20 The inability for small households to protect their savings from inflation fundamentally changed their spending patterns. Consumers stocked up on tangible goods like food, commodities, and real estate. Purchases on credit increased, as lending rates were kept artificially low as well. The result was increased consumption, increased inflation, reduced savings, a shortage of business investment, and a government-subsidized boom in real estate (Kane 1981, Herring/Wachter 1999). Capital controls in all of the industrialized Western countries also had the entirely predictable effect of stimulating the development of international credit markets. The Euromarkets, as they were called, allowed banks in one country to evade domestic regulations by borrowing money in another country, especially through the London exchange (Battilossi 2000). As capital controls tightened, the Euromarkets became larger and more sophisticated—not least because of innovations from aggressive American investment banks. For example, Citibank created the first negotiable CDs (which banks could trade in the money market) in 1961, increasing the liquidity of American bank assets. Soon, world markets adopted Euro-CDs, which traded on a London secondary market run by specialized dealers (Battilossi 2000). Tight regulation once again ended up favoring those banks with branches overseas. They were able to bypass capital controls by financing the foreign operations of American multinationals, and could raise money in the European markets to lend back in the United States, weakening the impact of American monetary policy. Finally, American commercial banks could expand into investment banking in foreign markets, circumventing Glass-Steagall and preparing the ground for later regulatory erosion (Battilossi 2000; cf. Schweikart 1991). 21 In this manner, the regulatory system favored securitization of credit, and the increasing power of large banks over small ones.22 When interest rates spiked in the 1970s, interest-rate caps provided a golden opportunity for new competitors to emerge. While American banks were subject to interest-rate controls, foreign 19

“Controls on prices and quantities challenge the market’s solution [to conflict between social groups] without themselves resolving conflict. Controls perpetuate the conflict of economic interests and divert this into noneconomic channels” (Kane 1977). 20

In partial response to this blatant discrimination, “[P]ublic-interest lobbyists… called down upon banks a veritable plague of what we may call financial fair-play regulation” (Kane 1981:364). Rather than ending the distortion caused by interest-rate controls, Congress first imposed new mandates on banks based around social objectives. These included several Fair Credit laws, the Community Reinvestment Act, and laws regulating the provision of mortgages. Kane (1981) accurately predicted that these laws would remain in effect “long after deposit-rate ceilings have gone to regulation heaven,” introducing new distortions into the economy. 21

These investment banking activities were in fact carried out under the authority of Federal Reserve Regulation K (Wilmarth 2002; cf. Kane 1983). 22

“The dynamic of banking multinationalisation during the 1960s and 1970s can be conceptualised as a selfreinforcing interaction between derived demand and regulation arbitrage. Relationship banking provided a particularly strong incentive to expand multinational branch networks in order to supply services to corporate customers with large international and multinational activities” (Battilossi 2000:169). 17

banks operating on American soil were not—nor did they have to contend with “relationship banking” patterns of customer loyalty. As a consequence, domestic banks lost a tremendous amount of deposits to the invaders (Battilossi 2000, Kane 1981, Hackethal 2000). Additionally, the interest-rate caps fueled the rise of money-market mutual funds (MMFs). Money-market funds were able to pay the true prevailing interest rate to savers, causing a dramatic erosion of bank deposit accounts as large savers switched over in search of better returns (Kane 1977, Gilbert 1986, Edwards/Mishkin 1995). Demand deposits fell from 60% of bank liabilities in 1960 to less than 20% in the 1990s (Edwards/Mishkin 1995). Nonbank finance companies were also coming into their own (Calomiris 1995; cf. Glaeser/ Scheinkman 1998). In 1975, business lending by finance companies exceeded their consumer lending for the first time (Marquis 2001). Nonbank finance companies, such as credit-card companies or GE Credit, engage in several types of commercial lending not by taking deposits, but by raising money on the capital markets by issuing debt and equity (aside from retained earnings). This allows them considerable freedom to structure their obligations optimally. Without demand deposits, finance companies do not face the risk of a run on the bank. Debt issues can be made for any length of time desired. And equity imposes no hard financial constraints at all, unlike debt (Cowen/Kroszner 1990, Marquis 2001). Additionally, specialized finance companies can often better understand their borrowers than can banks. Many finance firms are subsidiaries of businesses such as auto manufacturers, existing to finance big-ticket purchases and allowing the firms to internalize the profits once earned by bank financing. Similarly, a great deal of finance is concentrated in the equipmentleasing industry; firms such as Caterpillar have expertise in evaluating large capital-equipment deals and in repossessing and reselling equipment if borrowers default. They thus have significant competitive advantages over banks, eroding the traditional role of banks as informational monitors (Marquis 2001). The fortunes of finance companies have been closely linked to the commercial-paper market,23 which was being quickly expanded by investment banks eager to seize lending business from the commercial banks. When transaction costs are low enough, commercial paper makes an excellent substitute for bank loans, since the short maturities allow the markets to discipline firm behavior much as banks would (Macey/Miller 1995). From 1960 to 1985, the commercial paper market grew four times as quickly as bank loans; large corporations financed over half of short-term borrowing with commercial paper, and only a quarter with bank loans (Cowen/Korszner 1990).24 The growth in commercial paper was driven in large part by the increasing popularity of money-market funds, which invest in paper. In turn, the deepening money markets have allowed nonbank financial firms to raise more money and expand their own lending, cutting further into bank business (Edwards/Mishkin 1995). 23

And more recently, to the growth in asset-backed securitization; finance companies have been able to securitize many of their loans and transfer the risk onto the open market (Marquis 2001, Wilmarth 2002). 24

To make things worse, banks began losing their lower-quality borrowers to the new market in junk bonds (Schweikart 1991, Edwards/Mishkin 1995). By 2000, there was almost $600 billion in junk bonds outstanding, over twice the outstanding business loans by banks (Wilmarth 2002). 18

In response, banks scrambled to exploit regulatory loopholes in order to evade capital controls, and to encroach on the securities, underwriting, and insurance industries.25 In this, they were helped by several new technologies, particularly the EFT (electronic funds transfer) system. By dramatically reducing the cost of processing customer transactions, EFT made several new products feasible and increased the profits for providing multiple services to existing customers (Kane 1981). EFT also made long-distance banking possible, obliterating the regulatory logic behind unit banking and several other regulations. Banks responded with creative new corporate structures to allow them to expand operations; regulators struggled to keep up. What was often discussed as a period of deregulation was in truth one of “re-regulation,” as authorities tried desperately to reassert control over a rapidly changing industry (Kane 1983). One factor conditioning regulatory change was the competition between several regulatory agencies, such as the Federal Reserve, the FDIC, the SEC, and state-level authorities. If one agency’s regulations grew too onerous, banks could adjust their corporate structure so as to exit that agency’s control and move to another, more hospitable, regulator. This encouraged empire-building regulators like the Federal Reserve to loosen their guidelines over time, while tightening them in particular areas to extend their power (cf. Kane 1999, Calomiris 2006). As banks continued hemorrhaging their deposit base to money market funds, Congress finally abolished interest-rate controls between 1980 and 1986 (Gilbert 1986). Banks were able to slow the loss of funds, at the cost of a permanent erosion in their profit margins—they were now obliged to pay market rates for most of their capital. Banks were thus exposed to a new source of risk: the spread between long-term and short-term interest rates. Recall that banks borrow short and lend long; if the spread narrows, bank profits evaporate. If short-term rates rise above long-term rates, banks are in great peril (Wilmarth 2002). The increasing depth and liquidity of capital markets brought about a new concept of what banking liquidity meant. Banks increasingly relied less on equity investors and depositors, or assets kept in cash or readily marketable bonds, than on their ability to arbitrage between different markets for capital, borrowing cheaply on the open market and lending out at higher rates. The largest banks become spinning-plate artists: their solvency depends on borrowing enough money to meet ongoing obligations, while their long-term loans and investments yield rich returns in the long run. “[I]t appears that the principal business of large banks is no longer the intermediation of credit but has instead become the management, transfer, and trading of various types of risk in the financial markets” (Wilmarth 2002). Unsurprisingly, the pioneers of this strategy were American banks facing competitive pressure in the 1960s (Battilossi 2000). When all goes well, this creates a flexible, profitable banking system able to meet nearly any client’s needs. But should this great river of capital suddenly turn against the banks—either if short-term rates rise too much relative to long-term rates, or if the markets lock up so that banks cannot borrow at all—then banks will be in trouble. Either they must quickly sell their illiquid assets (at bad prices, most likely), or they will implode (Iori/Jafarey/Padilla 2006). 25

“In the 1970s, loophole mining and fabrication became the main business of modern depository institutions…. It was the main avenue left along which bankers could exercise old-time yankee ingenuity. Their innovative behavior stands as a convincing contemporary manifestation of the can-do spirit that made our country great” (Kane 1981:359). 19

This transformation of the banking industry into one of risk arbitrage was exacerbated by government-subsidized sources of credit, which banks used to replace their eroding deposit base. Over time, the federal government has provided banks with several subsidized (or “non-risk priced”) sources of capital. Among these are brokered deposits, which are insured by the FDIC, and long-term advances from the Federal Home Loan Bank, which must be fully collateralized by real estate (King/Nuxoll/Yeager 2006).26 FHLB advances had historically been restricted to thrifts, but in 1989 they were extended to commercial banks as well. The attractiveness of FDIC deposit insurance and subsidized sources of cheap capital led to “structural arbitrage” in which banks increased their dependence on government failsafes over time, dramatically increasing the government’s risk exposure and the general fragility of the system.27 28 Moreover, these subsidies encouraged nonbank financial firms such as insurance companies and brokerage firms to open up their own subsidiary deposit institutions, so that they could extend the benefits of government protections across all of their operations (Kane 1983, Billet/Garfinkel/O’Neal 1998, King/Nuxoll/Yeager 2006, Wilmarth 2002).29 With Congress dithering over how to respond to the new crop of nonbank banks, the Fed took action. In 1983, it unilaterally broadened its interpretation of what activities fell under the Bank Holding Company Act, and therefore under its regulatory control. At the same time, the Fed created new privileges for BHC subsidiaries, making it more attractive for financial firms to recharter themselves as BHCs. One result was to open the door for deposit-taking banks to ease their way into the securities and insurance businesses, while at the same time raising new barriers to securities and insurance firms who wanted to enter banking (Kane 1983). In the 1980s, banks faced a difficult environment of interest-rate instability, coupled with increasing competition from finance companies and foreign banks. As a result, some banks (especially those with “entrenched” bank managers benefiting from the disproportionate power that the managerial class has gained over investors and the board) began taking more risks to maintain their profit margins (Gorton/Rosen 1995, Edwards/Mishkin 1995). In particular, banks sought to replace their shrinking corporate lending with commercial real-estate loans, fueling a speculative boom during the 1980s. Troubled thrifts turned to subsidized FHLB advances to plug 26

Note first that FHLB advances therefore privilege real-estate lending, adding permanent inflationary pressure to the market; and second, that if real-estate prices decline sharply, FHLB loans will suddenly lack collateral. 27

The FDIC Improvement Act of 1991 replaced the flat rate for deposit insurance, which was unresponsive to credit risk, with a series of graduated rates ranging first from 0.23% to 0.31% (Hovakimian/Kane 2000), then from zero to 0.27% after 1994 (Billet/Garfinkel/O’Neal 1998), depending on banks’ capital levels and supervisory ratings. The higher-rate tiers were intended to penalize banks for excessive risk. However, in 1996, a full 95% of banks were paying zero percent. Regulators apparently had become more permissive, not less (Billet/Garfinkel/O’Neal 1998). And in any event, the rate tiers were still too cheap relative to actual credit risks (Hovakimian/Kane 2000). 28

This is an example of how subsidized programs tend to create their own constituencies, who adjust their behavior to rely on the programs and become more dependent on them over time, even if such dependencies create perverse behavior (Pierson 1994:42). 29

Such subsidized access to capital interferes with the disciplining effects of the market. This has been demonstrated in the interbank market, where banks provide short-term loans to each other. Lending banks charge interest rates that are sensitive to the riskiness of the borrowing bank; however, because of the presence of subsidized capital, banks that have become too risky simply stop borrowing on the interbank market instead of changing their behavior. Thus the market “may exhibit monitoring but not influence” (King 2008:313; cf. Furfine 2001). 20

the holes in their balance sheet, allowing them to pursue even riskier lending in hopes of making back their losses (Stojanovic/Vaughan/Yeager 2008). This ended in disaster; while banks had failed at a rate of 6 per year between 1946 and 1980, in the next decade failures skyrocketed to 104 per year (Gorton/Rosen 1995). Federal regulators, responding to the rising instability of the banking system, set the first general balance-sheet rules in 1981-83. In response, banks shifted away from traditional lending and expanded their use of off-balance-sheet activities such as asset-backed securitization and standby lines of credit, which were often riskier than traditional practices (Wilmarth 2002). Starting in 1975, regulators had also put in place rules limiting financial firms’ ability to buy junk bonds; in short order, the credit-rating agencies became “captured” by their clients, taking large fees in exchange for the investment-grade bond ratings so crucial to expanding the market for an issue (Wilmarth 2002). In 1987, Federal Reserve regulators took advantage of imprecise language in Section 20 of the Glass-Steagall Act to allow BHCs to operate “Section 20” subsidiaries making up to 5% of their revenue from investment banking. This limit was soon raised to 10% in 1989, and then to 25% in 1996 (Wilmarth 2002). These Section 20 subsidiaries proved quite profitable for their parent organizations; at the same time, entry by the commercial banks into the debt-underwriting field ended up lowering underwriting spreads, making the commercial-debt market more efficient and adding to the competitive pressure (Mamun/Hassan/Maroney 2005). Thus, well before Congress passed GLBA, commercial banks had already reentered the investment banking field in the United States. By 1994, off-balance-sheet income (mostly from advisory and underwriting fees and active market trading) had grown to 35% of total bank income (Edwards/Mishkin 1995). Such business is riskier and more heavily exposed to macroeconomic factors than traditional lending (Wilmarth 2002). Banks that failed between 1995 and 2003 depended much more heavily on fee income than other banks. Moreover, general industry trends showed increased competitive pressure and a worrying accumulation of risk. Bank deposits are steadily shrinking, banks have engaged in more commercial real-estate lending, banks depend more on Federal Reserve funds, and cash reserves are at historic lows (King/Nuxoll/Yeager 2006). The push toward securitization was accelerated by the 1998 Basel Accord, which set riskbased standards for banking soundness. To evade these standards, banks increased their use of securitization to transfer assets off of their balance sheets, reaping immediate underwriting fees instead of sustained growth over time (Wilmarth 2002). Increasingly, banks only retain those assets they cannot securitize—meaning those that are opaque, highly uncertain, and difficult to understand (Macey/Miller 1995). After the banking crises of the 1980s, policymakers began to realize that the many federal subsidies for banking, and the general policy to protect banks that were “Too Big To Fail” (TBTF), had created malign incentives that were contributing to the weakening of the banking system. In response, several new laws were passed, including the FDIC Improvement Act of 1991, and National Depositor Preference legislation that went into effect in 1994 (King 2008). These new laws were meant to restrict the ability for risky banks to abuse the federal safety nets, and included several provisions meant to discourage regulators from favoring troubled banks. The 1994 implosion of Long-Term Capital Management, a large hedge-fund, 21

demonstrated that regulators had no intention of listening. The Federal Reserve orchestrated a massive capital infusion to LTCM’s largest creditor-banks, though as it turned out the banks were never in danger of collapse, and the market perceived the Fed’s action as a reinforcement of the TBTF doctrine (Furfine 2006). In part responding to the legislative attack on their TBTF subsidies, banking firms initiated a wave of massive mergers in 1998, starting with Citicorp-Travelers (Kane 1999). The idea was to make the resulting firm such an integral component of the system that the government would be forced to protect it from failure, whatever the law said (cf. Hovakimian/Kane 2000, Wilmarth 2002). Additionally, many of these mergers were between traditional competitors such as banks and insurance companies, straining the limits of Glass-Steagall (Wilmarth 2002). These megafinancials increased the systemic risk of the financial system, by concentrating asset and counterparty risks onto a very few actors who turned out to be unable to handle the late crisis (cf. Wilmarth 2002). 30 The Citigroup merger is remarkable because it explicitly violated the core prohibitions of Glass-Steagall: Citibank was a commercial bank, while Travelers was a financial services and insurance conglomerate. The merger, and the regulatory-financial alliance that fostered it, was the crowning moment in the long campaign to defeat Glass-Steagall once and for all. An extraordinary feature of the Citigroup transaction was that Citicorp's and Travelers' chairmen consulted with, and received positive signals from FRB chairman Alan Greenspan, Treasury Secretary Robert Rubin, and President Clinton before the merger was publicly announced. Eighteen months later, during intense negotiations between the Clinton administration and congressional leaders over the final terms of the GLB Act, Citigroup appointed Mr. Rubin—a close confidant of President Clinton and a prominent supporter of the legislation—as its new cochairman (Wilmarth 2002).

The Federal Reserve’s approval of the Citicorp-Travelers merger was a direct challenge to Congress. While technically the merger was granted only a 5-year temporary approval, this in fact added to the pressure—no one seriously imagined that in five years the two companies would split apart again. Glass-Steagall’s restrictions had just been set aside by regulatory fiat. Unsurprisingly, Congress passed the GLBA within the year, having been successfully “dragooned” by the banking industry (Kane 1999, Wilmarth 2002). That said, Congress was also responding to the undeniable changes in the financial industry. “Congress adopted the GLB Act because it recognized that advances in information technology and the development of innovative financial instruments had subverted the prior legal barriers between banks and other financial firms” (Wilmarth 2002). Electronic funds transfers and the sharp competitive pressure from the capital markets had made the banks’ position untenable under the old restrictions. Additionally, there was wide consensus among economists and finance academics that Glass-Steagall had harmed the American banking system and needed to be repealed (see for example Wilmarth 2002 and 2005, Calomiris 1995, Macey/Miller 1995). 30

Large banks also contribute to the instability of the system through the interbank market, where banks lend each other short-term funds. While such lending tends to make the system more stable in general, when crises occur they can quickly cause “avalanche” effects by transmitting losses from bank to bank. This is particularly so when some banks are heavy borrowers and others are heavy lenders (Iori/Jafarey/Padilla 2006). In fact, the largest banks tend to be net borrowers, and small banks are net lenders (Furfine 2001). Thus, if a large bank fails, it could bring down several smaller banks with it. 22

As the largest banks were consolidating, they were also engaging in riskier activities to keep profit margins up. A small group of giant banks accounted for almost all foreign lending, particularly in emerging markets, and in 2000 controlled 8 of the top 10 subprime mortgagelending companies (Wilmarth 2002). Banks reduced their capital reserves to “dangerously low levels” in the 1990s to meet earnings targets, in part because of pressure from the SEC for banks to be more aggressive (Wilmarth 2002, footnote 113). Large banks were also using equity investments to generate a large portion of their income. Ten major banks had over $30 billion in venture-capital stakes in technology start-ups right at the top of the bubble (Wilmarth 2002). Notably, many studies have concluded that banks stop becoming more efficient when they grow larger than $10-25 billion, and efficiency in fact declines. The megamerger trend had more to do with preempting hostile takeovers, becoming an irreplaceable part of the banking system so as to extort regulatory lenience and TBTF status, and managerial hubris (Wilmarth 2002). At the beginning of 1998, the spread between short and long interest rates shrank to nearly zero. This hurt banks tremendously, as their profit margins from lending practically vanished. In early 2000, the spread actually turned negative and remained so for over a year. In response, the Fed cut interest rates from 6.5% all the way down to 1.75% over the course of 2001, eventually reversing the negative spread.31 This came at a substantial cost—cheap credit provided the impetus to the real-estate speculative boom (Wilmarth 2002). Additionally, falling interest rates made mortgage refinancing attractive, placing mortgage-servicing companies under more competitive pressure. Profit margins in home mortgages fell steadily, leading to reckless lending standards (Wilmarth 2002). We know how that turned out, of course. Policy Recommendations Hall and Soskice argue (2001:49) that a liberal economy, because of its pattern of institutional development, can usually be reformed most effectively through deregulation. Broadly speaking, this article concurs. Our economy has been moving toward a financial system built around public capital markets, and we should see that as a good thing. The great power of financial intermediaries like banks is corrosive to systemic stability, free enterprise, and democratic society itself, while direct unmediated access to the capital markets benefits lenders, borrowers, investors, and innovators alike. Intermediaries can still exist in such a financial world, but they must prove their value over direct access, instead of relying on political power to enforce their position. Krugman (2009:163) argues for more stringent regulation of “shadow banking,” i.e. illiquid structured debt and the over-the-counter derivatives market: “Influential figures should have proclaimed a simple rule: anything that does what a bank does, anything that has to be rescued in crises the way banks are, should be regulated like a bank.” This pronouncement neglects the history driving the development of these products—it was precisely the heavy regulation of 31

Another reason for the cuts was to arrest the stock-market crash of 2000. Chairman Greenspan made explicit reference to the danger of “a sharp deflation in asset prices.” While the stock market was now tied to the pension wealth of most working Americans, generating heavy political pressure to intervene (Zingales 2009), the fall in asset prices also harmed the banks specifically (Wilmarth 2002). 23

banks and the resulting advantages for securities that made them popular. Merrily extending the regulatory net to each new thing that comes along will only make the next new thing come along that much faster (cf. Kane 1977, 1981, 1983). The more that governments try to constrain banks through regulations alone, the more strain is put on the regulators until keeping up is beyond their ability. Additionally, regulatory oversight is notorious for focusing on the old threats and missing new ones, or being most lenient precisely when greater scrutiny is needed (Herring/Wachter 1999, Zingales 2009). Increasing the banks’ exposure to market discipline, paradoxically, makes it easier for regulators to do their jobs. Therefore, the general policy must be to cut back on government-subsidized benefits for the banks. In particular, the price for deposit insurance or free-flowing credit must be responsive to the actual risks, or else banks will have fewer incentives to moderate themselves (Billet/ Garfinkel/O’Neal 1998, Herring/Wachter 1999). Furthermore, while the concern over bank-run-style collapse is real, it is the regulatory morass in the financial sector that has impeded the growth of alternatives such as mutual-fund banking (Cowen/Kroszner 1990), which would be relatively safe from bank runs. Banks are currently the beneficiaries of privileged access to the Federal Reserve System’s wire-transfer and check-clearing networks, the Fed discount window, and myriads of special subsidies and regulatory “moats” with which to deter competitors (Cowen/Kroszner 1990, Kahn/Roberds 2009).32 Yet even so, these moats have been insufficient to totally prevent the rise of credit-card companies, money-market mutual funds, Paypal, Prosper.com, and a host of nonbank alternatives. Banking system regulations need to be dramatically cut down to enable the financial sector to rationalize itself and move away from the antiquated yoking of long-term assets to short-term liabilities. Traditional banking is a vestige of coordination problems that no longer exist. To keep defending banking from nonbank competitors would be like outlawing email to save the Post Office. This analysis suggests that the health of the public capital markets is of great importance to the economy. It is therefore alarming to see that increasingly heavy regulation of publicly traded securities, such as the Sarbanes-Oxley Act of 2004, has driven many firms to go private, or else issue “unregistered” securities that trade on new private-equity exchanges, controlled by the megabanks and unavailable to small investors (Zingales 2009; cf. Calomiris 1995, Boot/ Gopalan/Thakor 2006).33 Since SarbOx was passed, over half of all new equities issues have been private placements—a sharp change from 2002, in which only 22% were. If left unchecked,

32

One reason for government favoritism toward banks that this paper has not addressed is that the Federal Reserve employs the banking sector to carry out monetary policy. This would be much harder to do with nonbank financial firms, and the government will likely fight hard to prolong the death-throes of the banks for that reason. While it is outside the scope of this paper to discuss the ills of macroeconomic policy, Sechrest (1993) has a good treatment. 33

Two examples of such exchanges are the Goldman Sachs Tradable Unregistered Equity (GSTrUE) exchange, and the Open Platform for Unregistered Securites (OPUS), initially run by Citigroup, Lehman Brothers, Merril-Lynch, Morgan Stanley, and Bank of New York Mellon (Zingales 2009). The much-reviled collateralized debt obligations (CDOs) that facilitated securitization of risky mortgages were private placements, and were not traded on the open markets. 24

this trend will radically shift the balance of power away from small investors34 and toward the banks and other “Qualified Institutional Buyers.” This should be anathema to a free society, and government regulations should certainly not be fostering such a plutocracy. We should strive to make the public markets as attractive to issuers as possible. It should also be a general policy goal to improve the capital markets’ ability to control firm management. At present, the managerial class has tremendous influence with state and federal governments that prevents effective shareholder discipline. Additionally, shareholders have no control over which directors stand for election to the board—the slate is chosen by the incumbent board (Zingales 2009). If the capital markets are to effectively control American companies, these privileges must be sharply reduced. Zingales (2009) proposes that investors be able to propose their own slate of board candidates, which would be a good start. Additionally, I propose that board members no longer be paid stipends by the companies they ostensibly oversee, but should be paid instead by shareholders, on whose behalf they are supposed to act. Ultimately, the overriding policy goal must be to remove the protections and subsidies enjoyed by the banking system, to allow free competition with new alternatives so that the system as a whole can adjust to modern technologies. There is no inherent reason—none—why bank instability should be a danger to the modern financial system. Existing regulations that maintain the banks as favored gatekeepers to the capital markets have kept us under threat of economic collapse, merely to benefit those financial powers that have bought government cooperation. America needs to open the gates and let us all through.

34

Or, more realistically, the large investment intermediaries like mutual funds that work on small investors’ behalf (cf. Macey/Miller 1995, Zingales 2009). 25

Works Cited Baliga, Sandeep, and Ben Polak. 2004. "The Emergence and Persistence of the Anglo-Saxon and German Financial Systems." Review of Financial Studies, Vol. 17: 129-163. Battilossi, Stefano. 2000. "Financial Innovation and the Golden Ages of International Banking: 1890-1931 and 1958-81." Financial History Review, Vol. 7: 141-175. Becht, Marco, and Carlos Ramírez. 2003. "Does Bank Affiliation Mitigate Liquidity Constraints? Evidence from Germany's Universal Banks in the Pre-World War I Period." Southern Economic Journal, Vol. 70, No. 2: 254-272. Billet, Matthew, Jon Garfinkel, and Edward O'Neal. 1998. "The Cost of Market Versus Regulatory Discipline in Banking." Journal of Financial Economics, Vol. 48: 333-358. Bodenhorn, Howard. 1999. "An Engine of Growth: Real Bills and Schumpeterian Banking in Antebellum New York." Explorations in Economic History, Vol. 36: 278-302. Boot, Arnoud, Radhakrishnan Gopalan, and Anjan Thakor. 2006. "The Entrepreneur's Choice between Private and Public Ownership." Journal of Finance, Vol. 61, No. 2: 803-836. Boot, Arnoud, and Anjan Thakor. 1997. "Banking Scope and Financial Innovation." Review of Financial Studies, Vol. 10, No. 4: 1099-1131. Bordo, Michael, Hugh Rockoff, and Angela Redish. 1996. "A Comparison of the Stability and Efficiency of the Canadian and American Banking Systems, 1870-1925." Financial History Review, Vol. 3: 49-68. Briones, Ignacio, and Hugh Rockoff. 2005. "Do Economists Reach a Conclusion on FreeBanking Episodes?" Econ Journal Watch, Vol. 2, No. 2: 279-324. Burhop, Carsten. 2006. "Did Banks Cause the German Industrialization?" Explorations in Economic History, Vol. 43: 39-63. Calomiris, Charles. 1995. "The Costs of Rejecting Universal Banking: American Finance in the German Mirror, 1870-1914." In Lamoreau, Naomi, and Daniel M.G. Raff, eds. Coordination and Information: Historical Perspectives on the Organization of Enterprise. National Bureau of Economic Research/University of Chicago Press. Calomiris, Charles. 2006. "The Regulatory Record of the Greenspan Fed." American Economic Review, Vol. 96, No. 2: 170-173. Calomiris, Charles, and Joseph Mason. 2008. "Resolving the Puzzle of the Underissuance of National Bank Notes." Explorations in Economic History, Vol. 45: 327-355. Carlson, Mark, and Kris James Michener. 2009. "Branch Banking as a Device for Discipline: Competition and Bank Survivorship During the Great Depression." Journal of Political Economy, Vol. 117, No. 2: 165-210. Cowen, Tyler, and Randall Kroszner. 1990. "Mutual Fund Banking: A Market Approach." Cato Journal, Vol. 10, No. 1: 223-237. Da Rin, Marco. 1996. "Understanding the Development of the German Kreditbanken, 1850-1914: An Approach from the Economics of Information." Financial History Review, Vol. 3: 29-47.

26

Da Rin, Marco. 2002. "Banks as Catalysts for Industrialization." Journal of Financial Intermediation, Vol. 11: 366-397. De Long, J. Bradford. 1991. "Did J.P. Morgan's Men Add Value? An Economist's Perspective on Financial Capitalism." In Peter Temin, ed. Inside the Business Enterprise: Historical Perspectives on the Use of Information. University of Chicago Press: 205-249. Edwards, Franklin, and Frederic Mishkin. 1995. "The Decline of Traditional Banking: Implications for Financial Stability and Regulatory Policy." Federal Reserve Bank of New York Economic Policy Review, July: 27-45. Fama, Eugene. 1985. "What's Different About Banks?" Journal of Monetary Economics, Vol. 15: 29-39. Fisher, Ken. 2007. The Only Three Questions That Count: Investing By Knowing What Others Don't. Hoboken, NJ: John Wiley & Sons. Fohlin, Caroline. 1999. "The Rise of Interlocking Directorates in Imperial Germany." Economic History Review, Vol. 52, No. 2: 307-333. Fohlin, Caroline. 2002. "Regulation, Taxation and the Development of the German Universal Banking System, 1884-1913." European Review of Economic History, Vol. 6: 221-254. Fohlin, Caroline. 2007. "Does Civil Law Tradition (or Universal Banking) Crowd Out Securities Markets? Pre-World War I Germany as Counter-Example." Enterprise & Society, Vol. 8: 602-641. Furfine, Craig. 2001. "Banks as Monitors of Other Banks: Evidence From the Overnight Federal Funds Market." Journal of Business, Vol. 74, No. 1: 33-57. Furfine, Craig. 2006. "The Costs and Benefits of Moral Suasion: Evidence from the Rescue of Long-Term Capital Management." Journal of Business, Vol. 79, No. 2: 593-622. Gerschenkron, Alexander. 1962. Economic Backwardness in Historical Perspective. Cambridge, MA: Harvard University Press. Giedeman, Daniel. 2005. "Branch Banking Restrictions and Finance Constraints in EarlyTwentieth-Century America." Journal of Economic History, Vol. 65, No. 1: 129-151. Gilbert, R. Alton. 1986. "Requiem for Regulation Q: What It Did and Why It Passed Away." Federal Reserve Bank of St. Louis, February: 22-37. Glaeser, Edward, and José Scheinkman. 1998. "Neither a Borrower nor a Lender Be: An Economic Analysis of Interest Restrictions and Usury Laws." Journal of Law and Economics, Vol. 41: 1-36. Goergen, Marc, Miguel Manjon, and Luc Renneboog. 2008. "Recent Developments in German Corporate Governance." International Review of Law and Economics, Vol. 28: 175-193. Gorton, Gary, and Richard Rosen. 1995. "Corporate Control, Portfolio Choice, and the Decline of Banking." Journal of Finance, Vol. 50, No. 5: 1377-1420. Guinnane, Timothy. 2002. "Delegated Monitors, Large and Small: Germany's Banking System, 1800-1914." Journal of Economic Literature, Vol. 40, No. 1: 73-124. Hackethal, Andreas. 2000. "How Unique Are US Banks? The Role of Banks in Five Major Financial Systems." Finance and Accounting series, working paper No. 60, Johann Wolfgang Goethe-Universität, Frankfurt-am-Main. 27

Hall, Peter, and David Soskice. 2001. Varieties of Capitalism. Oxford University Press. Hart, Jeffrey. 1992. Rival Capitalists. Cornell University Press. Herring, Richard, and Susan Wachter. 1999. "Real Estate Booms and Banking Busts: An International Perspective." Working paper, Wharton Financial Institutions Center. Herring, Richard, and Susan Wachter. 2002. "Bubbles in Real Estate Markets." Zell/Lurie Real Estate Center, Working Paper #402. Hickson, Charles, and John Turner. 2002. "Free Banking Gone Awry: The Australian Banking Crisis of 1893." Financial History Review, Vol. 9: 147-167. Hovakimian, Armen, and Edward Kane. 2000. "Effectiveness of Capital Regulation at U.S. Commercial Banks, 1985 to 1994." Journal of Finance, Vol. 55, No. 1: 451-468. Jiang, Guolin, Paul Mahoney, and Jianping Mei. 2005. "Market Manipulation: A Comprehensive Study of Stock Pools." Journal of Financial Economics, Vol. 77: 147-170. Kahn, Charles, and William Roberds. 2009. "Why Pay? An Introduction to Payments Economics." Journal of Financial Intermediation, Vol. 18: 1-23. Kane, Edward. 1977. "Good Intentions and Unintended Evil: The Case Against Selective Credit Allocation." Journal of Money, Credit, and Banking, Vol. 9, No. 1: 55-69. Kane, Edward. 1981. "Accelerating Inflation, Technological Innovation, and the Decreasing Effectiveness of Banking Regulation." Journal of Finance, Vol. 36, No. 2: 355-367. Kane, Edward. 1983. "Technological and Regulatory Forces in the Developing Fusion of Financial-Services Competition." Journal of Finance, Vol. 39, No. 3: 759-772. Kane, Edward. 1999. "Implications of Superhero Metaphors for the Issue of Banking Powers." Journal of Banking and Finance, Vol. 23: 663-673. King, Thomas. 2008. "Discipline and Liquidity in the Interbank Market." Journal of Money, Credit and Banking, Vol. 40, No. 2-3: 295-317. King, Thomas, Daniel Nuxoll, and Timothy Yeager. 2006. "Are the Causes of Bank Distress Changing? Can Researchers Keep Up?" Federal Reserve Bank of St. Louis Review, Vol. 88, No. 1: 57-80. Kopper, Christopher. 1998. "Banking in National Socialist Germany, 1933-39." Financial History Review, Vol. 5: 49-62. Kroszner, Randall, and Raghuram Rajan. 1997. "Organization Structure and Credibility: Evidence from Commercial Bank Securities Activities Before the Glass-Steagall Act." Journal of Monetary Economics, Vol. 39: 475-516. Krugman, Paul. 2009. The Return of Depression Economics and the Crisis of 2008. New York: W. W. Norton and Co. Lorenzoni, Guido. 2008. "Inefficient Credit Booms." Review of Economic Studies, Vol. 75: 809-833. Macey, Jonathan, and Geoffrey Miller. 1995. "Corporate Governance and Commercial Banking: A Comparative Examination of Germany, Japan, and the United States." Stanford Law Review, Vol. 48, No. 1: 73-112. Mahoney, Paul. 1999. "The Stock Pools and the Securities Exchange Act." Journal of Financial Economics, Vol. 51: 343-369. 28

Mamun, Abdullah, M. Kabir Hassan, and Neal Maroney. 2005. "The Wealth and Risk Effects of the Gramm-Leach-Bliley Act (GLBA) on the US Banking Industry." Journal of Business Finance & Accounting, Vol. 32, No. 1-2: 351-388. Marquis, Milton. 2001. "What's Different About Banks—Still?" FRBSF Economic Letter, April 6. von Mises, Ludwig. 1996. Human Action. Fourth Ed. San Francisco: Fox & Wilkes. Monnet, Cyril, and Erwan Quintin. 2007. "Why Do Financial Systems Differ? History Matters." Journal of Monetary Economics, Vol. 54: 1002-1017. North, Douglass. 1990. Institutions, Institutional Change, and Economic Performance. Cambridge University Press. Pierson, Paul. 1994. Dismantling the Welfare State? Reagan, Thatcher, and the Politics of Retrenchment. Cambridge University Press. Ramírez, Carlos. 1999. "Did Glass-Steagall Increase the Cost of External Finance for Corporate Investment?: Evidence From Bank and Insurance Company Affiliations." Journal of Economic History, Vol. 59, No. 2: 372-396. Rousseau, Peter. 2009. "Share Liquidity, Participation, and Growth of the Boston Market for Industrial Equities, 1854-1897." Explorations in Economic History, Vol. 46: 203-219. Rousseau, Peter, and Richard Sylla. 2005. "Emerging Financial Markets and Early US Growth." Explorations in Economic History, Vol. 42, No. 1: 1-26. Sechrest, Larry. 1993 [2008]. Free Banking. Auburn, Alabama: Ludwig von Mises Institute. Santos, João, and Adrienne Rumble. 2006. "The American Keiretsu and Universal Banks: Investing, Voting and Sitting on Nonfinancials' Corporate Boards." Journal of Financial Economics, Vol. 80: 419-454. Schweikart, Larry. 1991. "U.S. Commercial Banking: A Historiographical Survey." Business History Review, Vol. 65, No. 3: 606-661. Shambaugh, Jay. 2006. "An Experiment with Multiple Currencies: The American Monetary System From 1838-60." Explorations in Economic History, Vol. 43: 609-645. Simon, Miguel Cantillo. 1998. "The Rise and Fall of Bank Control in the United States: 1890-1939." American Economic Review, Vol. 88, No. 5: 1077-1093. Smiley, Gene. 1981. "The Expansion of the New York Securities Market at the Turn of the Century." Business History Review, Vol. 55, No. 1: 75-85. Stojanovic, Dusan, Mark Vaughan, and Timothy Yeager. 2008. "Do Federal Home Loan Bank Membership and Advances Increase Bank Risk-Taking?" Journal of Banking and Finance, Vol. 32: 680-698. Wilmarth, Arthur. 2002. "The Transformation of the U.S. Financial Services Industry, 1975-2000: Competition, Consolidation, and Increased Risks." University of Illinois Law Review, Vol. 2002, No. 2: 215-476. Wilmarth, Arthur. 2005. "Did Universal Banks Play a Significant Role in the U.S. Economy's Boom-and-Bust Cycle of 1921-33? A Preliminary Assessment." Current Developments in Monetary and Financial Law, Vol. 4: 559-645. Zingales, Luigi. 2009. "The Future of Securities Regulation." Journal of Accounting Research, Vol. 47, No. 2: 391-425. 29