In this chapter, we look at regulation of financial institutions and how it differs from regulation in other industries.

Introduction: Why Regulate?

The debate surrounding regulation in general, not only of the financial variety, is polarized. There are those who are so suspicious of private enterprise that they believe almost everything should be regulated. Others are convinced that even well-intentioned regulation has such adverse unintended consequences that bureaucratic red tape should be cut to the barest of threads.

Worse than this polarization of opinion is the way prevailing opinion changes over time. During economic booms, when private-sector ambition knows no bounds, the laissez faire camp is triumphant in public debates—driving a ­liberalization agenda and skyrocketing sales of Atlas Shrugged.¹ At some point, this trend of liberalization overreaches itself, thereby contributing to an unsustainable boom that eventually collapses. Amid the wreckage of the too-lightly-regulated economy, the more dirigiste minded nag that they told us so. The indignant public debates that follow help drive a wave of public intervention. This begins to restrict some aspects of the economic recovery. The political, intellectual, and economic cycles create a system that do not check but amplify each other.

Before we get to financial regulation, it should be noted that financial firms are regulated like ordinary corporations. There are, however, some “carve outs” made for them—they are often exempt from both national and international competition laws on the grounds of financial stability. But they are generally subject to standard corporate laws and other laws of the land, such as ­employment law, health and safety standards, and product liability laws.

So, why do we need to regulate finance over and above the way other firms are regulated? Why would company, product-liability, health and safety, and other laws not be adequate to protect those buying a financial good as opposed to any other good? I try not to wear too much philosophy and theory on my sleeve in this book though. I am conscious that if it is not on your sleeve it is probably hiding inside it. John Maynard Keynes famously wrote that “practical men who believe themselves to be quite exempt from any intellectual influence, are usually the slaves of some defunct economist.”²

Let me say upfront, then, that my starting point for each issue that I grapple with in this book is that there should be regulation where there are significant “market failures.” Where there are not, the government should be reticent to regulate. Economists generally define a market failure as a situation where the outcome from unhindered private negotiations is biased in some way against another outcome with which society would be better off—even after any winners have compensated the losers.³ I recognize that this seemingly light idea is laden with value judgments.

Many will find this position too circumscribed. Even if a market failure does not appear to exist, surely government is required to consider the consequences of inequality and social capital. Governments routinely impose environmental levies on private activity to provide a better future for us all. These are vitally important issues. However, there are reasons for continuing with this narrow starting point in terms of financial regulation— as opposed to all other aspects of government policy. One of the most important but neglected reasons why financial regulation failed in the first decade of the 21st century was that it was given too many additional tasks that undermined the focus on systemic resilience. Tacked onto financial regulation were goals of tackling fairness, anti-money laundering, and combatting the financing of terrorism. In the way they are currently constituted, these tasks are so pervasive that they lend themselves to a box-ticking, legalistic approach, as opposed to one based around incentives to promote better behavior. This infected the culture of regulation. But the triumph of the box tickers came at the expense of systemic resilience.

This is not to say that we should ignore the serious concerns of money laundering, terrorist financing, and inequality. Where governments chose to pursue a wide variety of worthy financial policies such as combatting financial crime, broadening financial inclusiveness, and distributing taxes more fairly, there should instead be separate agencies working on dealing with these issues specifically. They should not be lumped into one parcel of regulation that is the responsibility of an omnibus agency. The separation of goals and policies would likely improve the overall effectiveness of the financial system’s resilience as well as act to prevent financial crime. The hope would be that these agencies talk to one another and share information—but let’s not be too ambitious. We will return to these issues in greater depth in Chapter 13 when considering the shape of financial regulation and its institutions.

While financial regulation should have narrow objectives, it still can, and indeed must, be carried out in a manner that recognizes some wider goals such as public trust. This might seem as if it’s nebulous and pandering to political correctness, but the consequences of a loss of public trust in regulation are tangible and substantial. During the global financial crisis that began in 2007, the public consensus was that a too-cosy relationship between government, regulators, and bankers had created a financial system that privatized the benefits of banking to a few and socialized the costs and risks to everybody else. Public trust in regulators to do the right thing evaporated. One result of this is that any governmental rescue operation became politically constrained. This contributed to the decision to let Lehman Brothers fail, with the attendant deep and wide-reaching consequences that followed.⁴ It also led to a substantial backlash and perhaps, in some areas, a swing in the pendulum of regulation too far the other way.

Market Failures in Finance That Require Specialized Regulation

There are two substantive market failures that are common in finance and less common elsewhere. It is these failures that require financial firms to be regulated over and above the regulation of firms in other sectors.

Lack of Consumer Protection

The first substantive market failure involves the financial market’s inherent weakness in protecting consumers. Markets protect consumers well when there are a large number of small, repeat transactions and when it is quick and easy to identify a “bad” purchase and avoid replicating it.

If a green grocer in Grantham, England, Alfred Roberts, sold bad apples, soon enough the good people of Grantham would stop buying his apples and purchase them from another green grocer. Alfred would have quickly spoken to his supplier to rectify the poor produce, and all the while his competitor would be reaping the benefits of the extra business. If Alfred’s supplier could not deliver better apples, Alfred might decide to turn his apples into a fine cider. Consumers would benefit from the availability of a new product. In this scenario, market discipline works, everyone is better off, and Ayn Rand⁵ followers are happy. Self-regulation works.

Apples are a healthy choice but they do not make the world go around. Many markets are more complex. Consider the market for taxicabs, for instance. In a big city, it is unlikely you will hail the same cab twice over a short period of time. If you are new to the city and the taxi driver takes longer to get to your destination than necessary, or overcharges, you may not be aware of the abuse. Furthermore, the taxi driver will not bear the consequences, as you are unlikely to be a repeat customer. Unchecked market forces could encourage more taxi drivers to shortchange their passengers. This would give all taxi drivers a bad reputation. It would drive customers away from using taxicabs, creating both an unsatisfied need for taxicabs and unemployment among taxi drivers⁶—a classic example of market failure. Nobel laureates Joseph Stiglitz and George Akerlof highlighted the many market failures that arise out of information asymmetries and also their possible regulatory remedies.⁷

The market failure that would arise without regulation is the rationale behind the widespread use of taxi-licensing schemes. Sometimes this regulation takes the form of a qualifying test or it may require tamperproof meters. Reporting infringements is usually made easy. Asymmetrical information is the reason why fares and taxis are more strictly regulated at train stations and airports, where passengers may be unfamiliar with the rules, and, pardon the pun, get taken for a ride. Few fear being ripped off when jumping into the back of a highly regulated “black cab” in London. Of course, while licensing systems protect consumers, they also protect industry incumbents—the cab drivers in this case—by creating barriers to market entry, reducing competition, and as a result raising profits. The strength of opposition to the mobile-taxi hailing “app” Uber by drivers of black cabs is a measure of how much they have to lose from greater competition.

When it comes to retail finance, like mortgages, life insurance, and pensions, consumers generally make a few large, seldom-repeated transactions—the exact opposite of the characteristics needed for a market to be good at protecting consumers. Moreover, financial transactions are often long term. Consumers only realize they have bought the wrong pension, life insurance policy, or endowment mortgage, often with severe consequences, long after it is possible to rectify the problem—because the original seller has moved on or closed down. Financial consumers need far more protection than consumers of apples.

The traditional response to market failures, especially those born of asymmetric information and conflicts of interest between “advisors” and “sellers,” has been to segment the market between retail and wholesale consumers. An ordinary retail consumer like “Aunt Agatha on the Clapham omnibus”⁸ is someone who cannot be expected to have specialist knowledge of finance. Wholesale consumers like banks, hedge funds, and insurance companies are expected to, and indeed are paid vast sums to, have sufficient specialist knowledge. In the name of consumer protection, how financial products are sold and represented to retail customers and any conflicts of interest arising from the transaction are rigorously regulated. This is one aspect of what is now referred to as microprudential regulation.

With wholesale markets between professional experts, the demand for consumer protection is less and the need to support innovation and liquidity greater. In these markets, regulators rely on caveat emptor.⁹ This is the kind of elegantly simple solution that economists love, where the points of regulation appear to connect with the points of market failure.

In practice, how best to protect consumers has been a live topic that is continuously evolving—invariably in the direction of greater protection. This trend has been partly powered by cash-strapped governments placing increased choice and reliance on consumers. The shift in the burden of choice and the provision of pension, health, and education to ordinary households has led to the creation of new markets and, in turn, additional consumer protection rules. It is with some irony yet inevitability that the banks that pioneered the retail finance revolution in the United Kingdom, during which private pensions were partly filled with the shares of newly privatized utility companies, were the same banks caught up in misselling scandals just a few years later. Lloyds TSB, for instance, was fined £1.9 million for “misselling” 22,500 pension-like policies through its branches between 2000 and 2001. No less than 44 percent of those policies were deemed unsuitable for the savers to whom they were sold.¹⁰

Each spectacular failure of the misselling rules ushered in new rules. Following the misselling scandals of the 1980s, rules on how and by whom products could be sold were tightened. Some would argue that the subprime mortgage scandal that happened ten years later suggests they were not tightened enough. In the 1990s, a wave of scandals, revealed by the dot-com boom and bust, centered on the conflict of interests by bank analysts (do you recall the cases of Henry Blodget and Jack Grubman?¹¹). This led to a strengthening of rules on conflicts and disclosures. By the early 2000s, those conflict-of-interest rules were again being scrutinized following the credit-derivatives scandals.¹² In practice, it is far from clear that regulating the sale of products to retail consumers has given them adequate protection. Instead, it has simply meant that they must sign ever-more-complicated waivers they do not understand.

Furthermore, during the previous and prior financial booms, wholesale customers revealed their ineptitude in handling complex financial products, making the distinction in expertise between retail and wholesale consumers appear to be less useful than had been hoped. In short, there are a set of market failures, fairly unique to finance, that requires all consumers—not just the vulnerable—to be better protected. Market forces alone are inadequate. This is a critical role of financial regulation that we will return to in Chapter 7.

Systemic Risks

The second type of market failure concerns the systemic nature of banking. Crucially, not all financial firms are involved in systemic activities. Unless they stray from the essence of insurance activity, insurance companies face risks that are not particularly systemic.¹³ Banking, however, is highly systemic. The essence of banking involves making loans that can be multiple times greater than the initial capital funded by the bank through short-term deposits or other borrowing that the bank promises to repay in full with interest. The less the initial capital, and the wider the gap between interest paid and interest received, the more profitable it is for the bank. But there’s also a greater risk that the bank will come undone if a wave of bad lending wipes out its capital or if enough depositors seek a withdrawal of the cash that the bank has lent over the long term in order to earn a margin.

While it is clear why banks are vulnerable to failure, this alone does not explain why the failure of one bank can so easily undermine all banks. The following illustration may help to do so. When a customer deposits $1,000 in a bank, the bank may keep 10 percent in reserve and lend $900 to a borrower. The borrower may then pay this money to a supplier who deposits this in her bank, which keeps 10 percent in reserve and lends $810 to another borrower, who pays this amount to someone else, who deposits this in his bank, which keeps 10 percent and lends $729. This goes on and on until on the back of a single $1,000 deposit and 10 percent reserve ratio, almost $10,000 is lent and spent.¹⁴ The credit multiplier is large. It is also partly dependent on the size of reserves and capital the bank regulators impose.

If one bank fails and the receivers try to recover all of their loans, it will start a domino effect of bringing down many banks. This is a key difference between banking and almost any other industry. If the high street insurance broker fails, or Alfred Roberts’s green grocers goes bust, neighboring competitors do not also fail. Often the opposite is true. The failure of a competitor can create more opportunities for remaining firms. Many believe that governments should let banks in difficulty fail rather than expensively bailing them and their highly remunerated employees out using taxpayer dollars. But these people neglect the interconnectivity of the credit economy. Governments are pushed into offering individual bailouts to banks because of the strong potential for failure of the entire banking system if they do not.

Connectivity runs deep in the financial sector. Systemically important firms in this sector do not even need to be lenders. Credit-rating agencies, custodians, and clearinghouses are not banks but they are highly connected parties. A failure of confidence in one of them will have a systemic effect. The systemic nature of banking is also related to but not wedded to size. Lehman Brothers was one of the smallest of the bulge-bracket investment banks. Yet, on September 15, 2008, the failure of this relatively small bank almost took down the entire global financial system because of its interconnectivity, especially in the credit-derivative markets.

Moreover, the path of contagion does not always follow actual connections. There are not many choices for collateralized lending besides mortgages and car loans. The failure of a small bank lending to home owners could make depositors fear that their own bank was involved in similarly poor lending. This mistrust can start a run on deposits, reversing the credit multiplier and leading to widespread bank failures. It is important to remember that the trigger points of the last financial crisis were not the kind of institutions to top anyone’s “too big to fail” list: Bear Sterns, Washington Mutual, Northern Rock, IKB, and Lehman.

Because of the multiple paths of contagion, just making banks smaller would not make the financial system much safer. In Chapter 4, we examine the issue of size in greater detail. But the wider point is one of externalities. Bank shareholders will invest in the safety of a bank to balance their private returns with the risk that they could lose their original investment. This private calculation does not take into account the wider social costs of the collapse of the banking system that could follow the contagious collapse of one bank. From the perspective of the financial system as a whole, left to their own devices, shareholders would underinvest in bank safety. History, replete as it is with a large number of banking crises, suggests that this is so.¹⁵

There is an anxiety that banks are becoming increasingly irrelevant and it is the shadow banks—such as the highly leveraged investment houses that are effectively in the lending and insurance businesses—where the risks are piling up. Consequently, over time many nonbanks and their activities have become subject to additional regulation. This is primarily done in the interest of consumer protection as well as ensuring these institutions are sound, operationally secure, and well governed.

There are genuine concerns, but they can be overstated from a systemic perspective. Like an ordinary shadow, these nonbanks are often connected to the banks. Most of them are only able to do what they do because they borrow from banks or because they appear to offer banking activities. While the institutions and instruments differ in every systemic crisis,¹⁶ what is common is that the growth of credit as well as the source of this credit through different avenues, invariably occurs at the banks. The regulation of bank lending from a systemic perspective and the enforcement of banking perimeters will influence what happens in the shadows to a large degree.

A Brief History of Financial Crises and Their Regulatory Response

Banking and financial crises prior to the 19th century, such as the South Sea Bubble¹⁷ or the Mississippi Bubble,¹⁸ were devastating but infrequent. These crises often arose in response to major wars that had bankrupted national treasuries, creating a supply of IOUs of various kinds. By periodically wiping out war debts, these crises had beneficiaries as well as victims. In the 19th and early 20th century, as the credit economy and international trade financed by credit grew, panic-occasioning bank runs were commonplace and the attendant victims more widespread. It is no surprise that this period of financial instability coincided with the rise of political and economic thought on the inherent instability of laissez faire capitalism.¹⁹

By the middle of the 20th century, banking crises had dried up. This was a direct response to four institutional and regulatory changes. In the first place, central banks had evolved into the lender of last resort. Bank deposits were also insured by the government. Then, there was a reduction of conflicts of interest through a fragmentation of the financial system, enshrined in the United States in the 1933 Banking Act (known as Glass-Steagall) that separated commercial and investment banking. And there was the Bretton Woods system, which fragmented international finance into financial flows that funded the trade of goods and services (relatively free of control) and flows for investment (more tightly controlled). We will return to each of these later.

Crises have many parents but the breakdown of the Bretton Woods system²⁰ of pegged exchange rates in August of 1971 may have contributed to the currency crises that spilled over into the banking crises that followed. In the decades of crisis after the breakdown of Bretton Woods, a critical aspect was the general worldwide liberalization of bank, capital, and exchange controls—including the formal repeal of the Glass-Steagall Act by the Gramm-Leach-Bliley Act of 1999.²¹ Banking crises were sporadic in the 1970s but became more frequent in the 1980s and even more so, with greater international spillovers, in the following decade.

The 1990s was a particularly wretched decade for international financial ­stability and this may have contributed to a policy response that brought greater exchange-rate stability in the first decade of the new millennium. In 1992–93, the European monetary system was in crisis, presaged by the abandonment of exchange-rate pegs in Scandinavia the year before. Britain and Italy were kicked out of the Exchange Rate Mechanism and +/–15 percent exchange rate bands replaced +/–2.25 percent bands. Then, there was the Mexican-led “Tequila crisis” of 1994/95. The years 1995–96 saw currency turmoil on the periphery of Europe.

The Asian Financial Crisis, which began in July 1997, unfurled into the Russia debt crisis of August 1998, via the Korean currency crisis. The Russian crisis helped to blow up Long Term Capital Management, a highly leveraged hedge fund, later in 1998, and the litany of crises ended with the Brazilian devaluation of January 1999. The first decade of the 21st century began with the dot-com debacle of 1999–2002. A prolonged period of growth and low volatility presaged “the great credit crunch,” which started in 2008.

Looking back at the history of the 20th century, it is tempting to argue that we do not need to overanalyze financial regulation. We must merely return to the Bretton Woods system and Glass-Steagall era of few banking crises and strong economic growth. This reasoning is persuasive but has a few caveats. For one, Bretton Woods would need to be substantially modernized, as it was designed for a different world with substantially less financial mobility. It is also vital to recognize that the golden years of 1949–70 may have been flattered by the postwar reconstruction of Western Europe, when the United States lent substantially to Europe for the purchase of its capital goods.²²

Another caveat is that the disruptive growth of emerging economies over the past two decades, which has led to a far-greater spread of economic power between countries and a more dramatic decline in poverty than occurred between 1949 and 1970, may not have occurred under the stability afforded by the Bretton Woods system. Stability is prized at times of crisis but it favors economic incumbents. Allowing all a fair opportunity may demand some disruption and turmoil at times.

Post–Bretton Woods

Let us focus further on the financial crises that have taken place since the collapse of Bretton Woods. Depending on the measure and definition of a banking crisis, there have been over 50 major crises post–Bretton Woods.²³ This number is enough for the empiricists to ask how costly and widespread financial crises are. Across countries and crises, the average cost of these banking disasters, in terms of lost output, is around 20 percent of GDP.²⁴ The cost of crisis resolution has also been substantial, amounting to an average of 30 percent of GDP.²⁵ Estimates of the cost of the 2007–09 crisis are a work in progress, but they run to several trillions of dollars and over 25 percent of GDP. Moreover, GDP measures do not capture the full range of effects, including rising inequality and poverty levels. Those living on the edge fall over. Armed only with child nutrition data, it is possible to correctly identify the Indonesian Financial Crisis of 1997–98.²⁶

Crisis avoidance also comes with costs. National regulatory bodies cost hundreds of millions of dollars and hundreds of staff members to run.²⁷ New regulations may lead to lost output and there may be many unintended distortions and costs. In any case, crises are not all bad and can have hard-to-quantify benefits. Allowing Barings Bank to fail in 1995 after one of its head traders, Nick Leeson, lost $1.3 billion in futures contracts sent a strong signal to banks that internal controls must be tightened. This included dealing with the previously neglected relationship between trading and settlement. Barings, the oldest merchant bank in London, had by 1995 become relatively small and not very connected. Its failure was not systemic and was an opportunity for banks to learn the positive lesson that internal changes in bank behavior could prevent another Leeson-like disaster. The logic behind the popular idea of not allowing banks to get too big to fail is something I discuss more critically in Chapter 4.

A “good” crisis is one where those doing the “right stuff” survive, sharpening incentives for banks to do what is right and for customers and investors to look more closely at the actions banks are taking. A bad crisis is one where a bank that has invested heavily and wisely in safety is knocked over by the panic triggered by another bank that had earlier taken a more reckless attitude to safety. The bad may bring down the good. And when all are brought down, it is hard to distinguish the bad from the good. What then is the point in being good? This question deters banks from investing in safety the next time around, making crises more likely. Based on this outcome, trying to reduce the likelihood and extent of a systemic banking crisis is worth the effort and cost. This is not just a financial equation. It is also worth the effort to avoid systemic crises if their wider economic and social costs are high and it would be hard to insulate companies and individuals from them.

Banking Regulation Today

Banking regulation has evolved, a little disjointedly, and today rests on a few main pillars and institutions. A crucial one is the role of the central bank in providing emergency liquidity to halt the failure of a bank due to a temporary loss of liquidity rather than the poor quality of its loan book.²⁸ This loss of liquidity might have stemmed from the failure of another, unrelated bank sparking panic and a withdrawal of deposits across the banking system.

The earliest incarnations of today’s central banks were established to help fund war debts, in part through the issuance of national currency required for the payment of taxes. The Bank of England, the Bank of Amsterdam, and the Sveriges Riksbank began life that way in the 17th century although they soon developed lender-of-last-resort facilities.

More recently, established central banks were developed specifically for the purpose of the lender of last resort. The US Federal Reserve owes its birth in 1913 to the Panic of 1907, when the financial system had to be bailed out by J. Pierpont Morgan, raising concerns that the sustainability of the financial system was improperly dependent on a few private actors. In carrying out lender-of-last-resort activities, modern central bankers cite Walter Bagehot’s dictum that central banks should lend freely but only to solid firms and only against good collateral and at interest rates that are high enough to dissuade borrowers without a genuine need.²⁹ This is eminently sensible. However, in the present day, where it is near impossible and considered deceitful to keep it secret that a bank has taken advantage of an emergency facility at the central bank, this knowledge is itself enough to create or deepen panic.

It is suggested that one of the contributing factors to the LIBOR scandal³⁰ during the last credit crunch is that banks were allegedly, with the tacit understanding of the Bank of England, unwilling to reveal the true costs of their borrowing lest signs of a relatively high cost would trigger a run on the bank. An ingenious way of cracking this “borrower’s curse” was used in 2008 when US Treasury Secretary Hank Paulson awarded the ten largest US banks $25 billion each—regardless of request or necessity. This eliminated the ostracism of those banks that actually were in difficulty. Those that were not simply repaid the loans when the time came to do so.

The second pillar of banking regulation is the use of deposit insurance. This insurance, given to individuals up to a certain limit, is backed by government guarantee—often through premiums paid by banks. This can contain panic from spreading among depositors concerned that their bank may be vulnerable to failure if another bank fails. “Bank runs” were frequent in the 19th and early 20th century, culminating in the many bank failures after the Great Crash of 1929. This experience, along with examples of mutual financial insurance in Hong Kong and elsewhere, led to a raft of deposit insurance schemes being established in the 1930s.³¹ During the Global Financial Crisis (GFC), there were signs of a return to bank runs. Perhaps because we had gone for a long period when insurance limits were not lifted, a significant number of savers felt insufficiently covered. The limits were promptly raised.³²

The next pillar of banking regulation surrounds issues of consumer protection as discussed earlier. These issues are different in character from those relating to monetary policy, emergency liquidity, and insurance to avoid runs. They have more to do with the legal minimum rights of consumers and investors concerning their treatment and disclosures made to them by financial firms. Many of these rights evolved out of old self-regulation exercises but when these proved inadequate, they were codified in laws and enforced by regulatory institutions. These institutions can sit within central banks, or, because of the more-legal-than-economic discipline of their employees or an area of activity, they are sometimes stand-alone agencies of government.

Modern banking supervision rests on the requirement that banks hold a ­minimum amount of capital. This, alongside scrutiny of operational and risk control, is designed to make individual banks safe and less of a risk both to their customers and investors and to the financial system as a whole. Minimum capital adequacy of banks is often seen as the quid pro quo for the government-backed guarantee and liquidity provision just mentioned. The design of the capital requirement seeks to be sensitive to the credit risks being taken—referred to as risk-sensitive capital adequacy. According to the approach, a bank that solely lends to the US government is taking less credit risks than one that only lends to small businesses in emerging markets.

The present starting point of capital adequacy requirements is the notion of risk-weighted assets. The required amount of capital is a percentage of a bank’s risk-weighted assets, where some assets have low-risk weights, like government bonds, while others carry high-risk weights—such as loans to weakly capitalized corporations. The purpose of the Basel 1 Capital Accord, concluded in 1988,³³ was to establish minimal capital requirements and international norms for risk weights to be used by internationally active banks to ensure that a rogue bank did not become a risk to the international financial system. Unintentionally, this accord and its later installments—Basel II and Revised Basel II—became the benchmark for the global regulation of all banks. There are many challenges with this risk-sensitive approach that we will touch on in the following section and return to in greater detail over the next few chapters.

An Overreliance on Bank Capital

The reliance on capital as the insurance against risk, and therefore the preference for institutions that appear well capitalized, is one of the historical reasons why regulators were uncomfortable with banking competition. Regulators did not want to face a phalanx of small, barely profitable, thinly capitalized banks so they were largely exempted from the competition and trade laws applicable to other firms. If they felt they would not be endangered, regulators often favored better-capitalized banks swallowing their less-well-capitalized neighbors. Competition concerns were waived aside in the name of financial stability. The 2008 takeover of HBOS by Lloyds Bank in the United Kingdom, and of Bear Sterns by J. P. Morgan in America, fit this long-tested bias. Over time, this preference for institutions with the most capital led to banking systems being dominated by a small number of profitable institutions that were deemed too big to fail.

The logic of the capital adequacy response to systemic risks is that if all institutions were well capitalized, there would be less chance of a bank failing. And if a bank failed, the capitalization of others, as well as the deposit insurance and the central bank’s emergency-liquidity provision, would stop bank runs. The core belief was that if microprudential regulation ensured that individual banks had strong internal credit controls and adequate capital, the system would be safe. There would be no need to agonize over a bank’s liquidity. A well-capitalized bank could always get access to liquidity. Were there ever to be a general panic that starved well-capitalized banks of liquidity, the only appropriate response would, in any event, be central bank action. Contrary to popular opinion, there is often a strong logic behind what regulators think they are doing.

The underlying model of what a banking crisis looks like, which underpinned this focus on bank capital, conceived a crisis as being caused by a rogue bank failing. This rogue bank would have perhaps held less capital than others or would have had worse credit controls. Given the systemic nature of banking, this failure would then spread to other well-managed banks. Yet this is not what we observe in practice. If it were so, then because this kind of rogue behavior is uncoordinated across banks, financial crises would be fairly random events that even in hindsight would be hard to predict. While such bank failures do exist, such as the Johnson Matthey Bank, which failed in 1984, Continental Illinois Bank (1984), British & Commonwealth Merchant Bank (1990), BCCI (1991), and Barings (1995), they are often well contained by the authorities and do not generally spread. Managing a rogue bank is not the critical problem that bank regulation needs to solve.

Widespread financial busts, on the other hand, with their enormous economic, fiscal, and social costs, is the problem we do need to worry about. But these are not random. The busts almost always follow financial booms. In the next chapter, we will examine the evidence for this assertion. Here we should merely observe that in previous financial regulation there was a fallacy of composition, namely that ensuring individual banks are safe does not make the system safe. In the middle of the boom, as asset valuations rise and risk measures fall, the banking system and all its major banks appear to be well capitalized and better capitalized than in the past. In the middle of the bust, as valuations collapse and measured risks soar, the banking system and all of its major banks appear to be too thinly capitalized, dependent on overstretched valuations and unduly concentrated in their lending. It is the system that goes rogue. Risk and capital are endogenous to the economic cycle. The problem is not risky loans not performing but so-called safe loans becoming risky at the same time. To combat crises and the financial cycle, microprudential regulation of individual banks is not enough. The next few chapters explore this problem and its solution.

___________________

¹Atlas Shrugged is the 1957 novel by Ayn Rand that describes a dystopia in which society’s most successful businesspeople abandon their fortunes and the nation in response to aggressive new regulations.

²John Maynard Keynes, The General Theory of Employment, Interest and Money (New York: Harcourt, Brace: 1936).

³Market failures are commonplace and economists have broadly identified five causes. The first is when there are differences in information between two sides of a bargain, where one side knows less than the other. There could also be the nonexistence of competitive markets, which could limit profit or expand the production of a monopoly. Then, there may be “public goods” where people can benefit from or access something without having to pay to do so, like clean air or a public park in the city. The next cause of market failure is private contracts failing to fully internalize their effect on others, such as the effects of noise or pollution on neighbors. The final broad cause of market failure occurs when there are conflicts of interest between principals and agents—for example, when proprietary traders bet shareholders’ funds to boost their own bonuses.

⁴I was in Washington at the time of the G-20 Summit that took place shortly after Lehman’s failure in September 2008. I recall, amid the debris of the Lehman fallout, Lorenzo Bini Smaghi, the Euro-area G-20 deputy, coming out of a meeting and saying to the assembled officials and journalists something like, “We have decided not to let any important financial institution fail” and someone posing the question, “Why didn’t you decide that before the collapse of Lehman Brothers?” Without pause, Bini responded, “Before Lehman Brothers, there was not the political mandate to save the banks.” Whatever the judgment on whether Lehman should or should not have been saved, I think he was correct in his assessment of the change in political climate.

⁵Rand is the author of Atlas Shrugged among other books espousing a free market perspective (see also Footnote 1).

⁶The English Parliament approved “An Ordinance for the Regulation of Hackney-Coachmen in London and the Places Adjacent” in June 1654 to remedy what it described as the “many Inconveniences [that] do daily arise by reason of the late increase and great irregularity of Hackney Coaches and Hackney Coachmen in London, Westminster and the places thereabouts.” In Acts and Ordinances of the Interregnum, 1642–1660, ed. C. H. Firth and R. S. Rait, reprint edition (Abington, UK: Professional Books, 1982).

⁷George Akerlof’s classic paper from 1970, “The Market for Lemons: Quality Uncertainty and the Market Mechanism” The Quarterly Journal of Economics, vol. 84, no. 3 (August 1970)—sometimes described as a take on Gresham’s Law, which is commonly summarized as, “The bad drives out the good”—is well worth a read if you have never done so.

⁸This is a modernization of the phrase “the man on the Clapham omnibus” used by courts in English law, where it is necessary to decide whether a party has acted as a reasonably educated and intelligent, but nonexpert, person.

⁹Caveat emptor, Latin for “let the buyer beware,” implies that the person who buys something is responsible for making sure that it is what he wants and that it will work to his satisfaction.

¹⁰Taken from Avinash Persaud and John Plender, All You Need to Know About Ethics and Finance: Finding a Moral Compass in Business Today (London: Longtail, 2007).

¹¹See Persaud and Plender, All You Need to Know About Ethics and Finance, for more on these scandals.

¹²Much has been written about the conflict of interest between credit-rating agencies, issuers of credit instruments, and investors. I think the case, while legitimate, is overstated given that the vast amount of credit ratings on single issues where the same issuer-pays business model did not fail. However, the point is that there is great pressure for rules to address the issue. In November 2012, Australia’s federal court ruled that the credit-ratings agency Standard & Poor’s (S&P) misled investors prior to the Global Financial Crisis by giving its safest credit rating, AAA, to complex securities, which later lost most of their value. At time of writing, S&P has stated its intention to appeal the decision.

¹³Indeed, the sector can play an important role in absorbing systemic risks, which I discuss in Chapter 6.

¹⁴This is a fair approximation of what is going on, but in reality banks create deposits when they give loans.

¹⁵See Carmen M. Reinhart and Kenneth Rogoff, This Time is Different: Eight Centuries of Financial Folly, reprint edition (Princeton, NJ: Princeton University Press, 2011).

¹⁶Some would say that Citibank turns up more regularly than most in post-1970s banking crises in the United States.

¹⁷The South Sea Company was founded in 1711 as a public–private partnership to consolidate and reduce the cost of the British Government’s war debts. The company was granted a monopoly to trade with South America; hence its name. Essentially, it was a bet on the future outcome of the War of the Spanish Succession, which Britain was involved in, over Spain’s control of trade with South America. Company stock rose greatly in value as it expanded its operations dealing in government debt, peaking in 1720 before collapsing.

¹⁸The Mississippi Company of 1684 was founded earlier than the South Sea Company, but it was only converted to a similar purpose in 1718 at the height of the euphoria around the South Sea Company, an early version of international contagion of financial euphoria and despair and of Parisian financiers following in the footsteps of their counterparts in London. Even without the help of 21st-century communication, both companies collapsed together. In 1718, John Law’s Banque Royale had subsumed the Mississippi Company and others with a monopoly on trade in French possessions as well as the right it had been given to issue notes guaranteed by the king.

¹⁹Karl Marx wrote The Communist Manifesto in 1848 and Das Kapital between 1867 and 1894. In the latter, he expounded the idea that human societies progress through class struggle and that capitalism created internal tensions, which would lead to its own destruction. Nineteenth-century financial panics were seen as symptoms of this impending doom.

²⁰The defining feature of the postwar Bretton Woods system was an obligation for each country to adopt a monetary policy that maintained its exchange rate parity to the US dollar. The International Monetary Fund was established to help bridge temporary balance of payments shortfalls, and there was provision for an adjustment of parities if an imbalance of international payments were to be persistent.

²¹The formal name is the Financial Services Modernisation Act. However, even before this act the demise of the Glass-Steagall Act had effectively occurred through increased liberal interpretations.

²²President Harry Truman signed the Marshall Plan on April 3, 1948, granting $5 billion in aid to 16 European nations. During the four years that the plan was operational, the United States donated $13 billion in economic and technical assistance (especially for rebuilding transport networks) to help the recovery of European countries that had joined in the Organisation for European Economic Co-Operation, the forerunner to today’s OECD. In current dollars, that would be approximately $150 billion, which represents around 5 percent of what was then the US GDP. And it worked. GDP in 1951, for Marshall Plan recipients, was at least 35 percent higher than in 1938.

²³See Reinhart and Rogoff, Eight Centuries of Financial Folly.

²⁴See Carmen M. Reinhart and Kenneth S. Rogoff, “Recovery from Financial Crises, Evidence from 100 Episodes,” American Economic Review 104, no. 5: 50–55.

²⁵Ibid.

²⁶See Steven Block, “Maternal Nutrition Knowledge and the Demand for Micronutrient-Rich Foods: Evidence from Indonesia,” Journal of Development Studies 40, no. 6 (2005).

²⁷The combined costs of the UK’s Prudential Regulation Authority (PRA) and Financial Conduct Authority reached £664 million (circa $1 billion) in 2013.

²⁸Think of this type of liquidity as cash flow. A bank may have assets that exceed its debts and so is solvent, but it may not be able to sell those assets overnight to meet an interest payment or a cash withdrawal and so is illiquid. Of course, in the dynamic of a crisis these neat conceptual distinctions become blurry.

²⁹Norman S. John-Stevas, ed., The Collected Works of Walter Bagehot, vols. 1–15 (New York, Oxford University Press, 1986).

³⁰A scandal in which banks appeared to manipulate the survey that set the benchmark interest rate: the London Interbank Offered Rate (LIBOR).

³¹The Banking Act of 1933 created the US Federal Deposit Insurance Corporation (FDIC). Today, it insures deposits up to $250,000 for approximately 7,100 institutions.

³²In the United States, where there is one of the longest histories of deposit insurance, the original limit in 1934 was $2,500. This was doubled in 1935 and held at that number until being doubled again in 1950 to $10,000. By 1980 it had been increased in four increments to $100,000, where it stood for an unprecedented gap of 28 years before being raised to $250,000 on October 3, 2008 in the aftermath of the Lehman collapse and signs of mini-bank runs developing.

³³In 1988 the Basel Committee on Banking Supervision (BCBS) of G-10 central banks and regulators published a set of minimum capital requirements (Basel I) for international banks in their member countries.