The crisis-prone United States
Since 1792, the United States has experienced at least 17 major banking crises (more by some counts), where a banking crisis is defined as either (a) an episode of substantial banking insolvency (negative net worth of failed banks relative to GDP), such as in many agricultural states during the 1920s, (b) a dramatic episode of sudden withdrawals (commonly called a banking panic), such as in 1907, or (c) both phenomena occurring together, such as in 1933.
US politicians often responded to banking crises with new regulatory policies. For example, in response to the frequent banking panics of the post-Civil War era (1873, 1884, 1890, 1893, 1896 and 1907), a National Monetary Commission (NMC) was created to propose reforms that would end banking panics. The NMC produced many voluminous and careful studies of the banking systems of other countries, which documented the uniquely fragmented structure of the US banking system. Shocking as it may seem, state laws not only prohibited banks from opening branches across state lines, they prohibited banks from opening branches even within the state! This ‘unit banking’ structure made the US banking system uniquely prone to banking panics; banks could not diversify risks across regions or head off runs by moving funds across branches.
The NMC studied the crisis-free history of Canadian banking, and the relatively stable histories of many other countries. Canada was especially noteworthy because it was America’s neighbour to the north and shared many attributes with the United States. Then as now, it relied on commodity exports, which implied cyclically volatile income. And yet, as American commentators noted, Canada had never experienced a severe banking crisis. To scholars at the turn of the 20th century the cure for US banking fragility was simple: end the fragmented unit banking system and allow US banks to branch, which would permit banks to compete with each other, lower their overhead costs, and diversify their portfolios.
But such a change was not on the political menu. Unit banking was an extremely durable rent sharing arrangement, agreed at the level of individual states - which controlled bank chartering - that brought together landowning farmers and unit bankers. It was a triumph of the joint lobbying efforts of local bankers (who benefited from limited competition in their local markets) and agrarian populists (who saw advantages to limiting local banks’ lending options to the local economy). This alliance was especially powerful because it spanned different political parties and ideological predispositions – apparent in the antipathy farmers often demonstrated toward bankers, illustrated, for example, in books such as The Grapes of Wrath.
A history of ineffective policy responses
Instead of fixing the bank instability problem, the NMC and Congress came up with a second-best solution – the Federal Reserve System (the ‘Fed’) – to reduce the extent of seasonal and cyclical liquidity risk in the banking system. Unit banks that joined the Fed were given access to Federal Reserve Bank lending. Although there is evidence that this did reduce liquidity risk, it did not end banking crises. The 1920s saw a massive wave of bank insolvencies in rural areas in the wake of commodity price declines that bankrupted undiversified unit banks. And the 1930s continued the same pattern on an even more massive scale, culminating in the 1932-1933 waves of bank failures that eventually caused President Roosevelt to declare a nationwide banking holiday to allow examiners to sort out which banks were solvent and which were insolvent.
Post-Depression banking legislation – including the 1933 Glass-Steagall Act – doubled down on unit banking by limiting the consolidation of the banking industry (a process that had begun in many states during the 1920s and early 1930s) by creating a new deposit insurance system specifically designed to protect small, risky rural banks. People like Representative Henry Steagall and Senator Huey Long made support for risky unit bankers the price of their support for other reforms that people like Carter Glass favoured. Amazingly, despite the costs of the Great Depression, the United States continued its uniquely unstable unit banking system for more than five more decades.
After the unusually low macroeconomic volatility of the 1950s and early 1960s, it returned with a vengeance. At the same time, deposit insurance protection expanded dramatically. The resulting increases in risk, and in protected bankers’ taste for risk, set the stage for the banking catastrophes of the 1980s.
This time the unit banking-agrarian populist coalition was not strong enough to resist the reform of consolidation policy. The prior decades had seen a steady shift of population from rural areas to cities. ATMs and other technological changes had eroded the local monopolies enjoyed by unit bankers. Widespread bank distress in the 1980s due to interest rate rises, farm price collapses, an energy boom and bust, and a commercial real estate crisis, created high resolution costs for the Federal Deposit Insurance Corporation (FDIC) and credit scarcity in many of the areas with high rates of bank failure. When banks from some of the states that did permit branching (California, Ohio and North Carolina) expressed their willingness to take over failing banks, and thereby cut the fiscal costs of FDIC resolution and limit the credit contraction of the affected local economies, they were welcomed.
The banking crises of the 1980s had significant regulatory consequences. Prudential capital requirements were imposed in the early 1980s, and were raised in the Financial Institutions Reform, Recovery, ad Enforcement Act (FIRREA) of 1989 and the Federal Deposit Insurance Corporation Improvement Act (FDICIA) of 1991. FDICIA also imposed new limits on Fed lending and restrictions on bailouts of uninsured debts of banks, which were advertised as ensuring that risky banks would never again be a burden on taxpayers. In 1994, the Riegle-Neal Act allowed unfettered nationwide branching throughout the United States by 1997.
As we now know, that apparently promising start to banking reform in the 1990s did not end well. Enhanced risk-based capital ratio requirements under FIRREA and FDICIA were easily manipulated. Capital was overstated by hidden loan losses, and risk-based capital requirements were further eroded by the all-too-forgiving measures of risk used by banks and regulators (which particularly understated mortgage-related risks).
Why did the numerous and highly publicised prudential reform efforts of the 1980s and 1990s fail? Why weren’t regulations enacted and enforced more seriously? The answer is that the restructuring of the banking system to permit nationwide branching did not mark the end of risk-creating political influences on banking, but rather the beginning of a new era, and the forging of a new political coalition that would sponsor banking fragility of a new kind, based on the partnership between a new set of strange bedfellows: too-big-to-fail megabanks and urban activist groups.
Once nationwide branch banking and the construction of megabanks would be permitted, the new question was: according to what criteria? Congress and the Administration gave the answer: the Federal Reserve Board was charged with approving mergers, and the most important requirement for a would-be acquirer was demonstrating ‘good citizenship’. Good citizenship would be demonstrated by the testimony of urban activist groups at Fed merger hearings, and the activist groups would point to clear evidence of good citizenship in the form of contractual commitments by merging banks to route lending through activist groups – which, from 1992 to 2007, totalled $867bn. The Fed was caught in the middle of a political game for creating and distributing economic rents within the new megabank-activist partnership, and the Fed had no interest in resisting the role it was given for the simple reason that Congress and the Administration had enough power over the Fed to pressure it into playing along.
To raise the stakes further, and enable the banks to increase their commitments to risky lending, the Federal Housing Enterprises Financial Safety and Soundness Act of 1992 (known as the GSE Act) was passed by President George HW Bush, and strengthened by his successors. That Act effectively required Fannie Mae and Freddie Mac to purchase large amounts of the risky loans originated by the new too-big-to-fail merging banks. The banks and Government Sponsored Enterprises (GSEs) could be further prodded to be generous in their willingness to originate and hold risky mortgage exposures if they were rewarded with low prudential capital requirements and other regulations that effectively allowed banks and GSEs to finance themselves at very low cost (thanks to the debt guarantees provided by taxpayers, who were left to clean up the inevitable mess).
The only way for the GSEs to meet their increasingly demanding mandates was to debase their underwriting standards by reducing required mortgage down payments and removing limits on documentation requirements for mortgages. The GSEs, who set the standards for the whole mortgage market, could not relax standards selectively just for the mortgages of targeted groups. That meant that everyone had equal access to imprudent mortgage credit, and people in all income classes and locations took advantage of the lax standards, resulting in a mortgage crisis of unprecedented proportion.
Will this time be different?
The Dodd-Frank Act of 2010, like the Federal Reserve Act of 1913, the Banking Act of 1933, the FIRREA of 1989 and the FDICIA of 1991, was advertised as the solution to American banking instability. Is that likely?
Consider the recently issued rules by the Consumer Financial Protection Bureau governing ‘Qualifying Mortgages’ (QMs), loans that are supposed to ensure low risk to lenders and affordability for borrowers. The idea was that mortgages could only obtain QM status if they met strict standards, but the final version of the rules watered those standards down dramatically, and also provided that mortgages repurchased or securitised by Fannie or Freddie were exempt from even the watered-down standards.
The Volcker Rule was intended to prevent risky trading by banks, but as enacted the rule specifically exempts real estate–related securities from its limits.
What about GSE reform? Proposals in the House would entail serious reform but they have no chance of passing the Senate, where the favorite current approach is to create a new substitute government mortgage guarantee system. The absence of agreement means the continuation of the government conservatorship of Fannie and Freddie by the Federal Housing Finance Agency. Its recently appointed director is former Congressman Mel Watt, a longtime political supporter of taxpayer-subsidised mortgage risk. Watt began by announcing a delay of increases in Fannie and Freddie guarantee fees, which is bound to ensure that banks will continue to have strong incentives to buy and hold Fannie and Freddie securities as a way to lower the amount of capital backing their portfolios. Then in May he introduced a new GSE initiative to spur increased risk taking in the mortgage market.
And Dodd-Frank’s abolition of too-big-to-fail bailouts? By designating certain banks as Systemically Important Financial Institutions (SIFIs) and creating a committee headed by the Secretary of the Treasury that can respond to the threats that these institutions present to the stability of the financial system by bailing out SIFIs, Dodd-Frank institutionalises bailouts and even establishes the authority for taxing surviving institutions to fund bailouts.
Another regulatory history
In sharp contrast to the US experience, crisis-free Canada has seen a very different history of highly effective bank regulation. Bank charters in Canada expire every five years, and a new banking law is drafted to coincide with the renewal of bank charters. Prudential regulation has been proactive and effective. Not only has Canada avoided crises, its banking system has also been more successful than the United States, historically and presently, in supplying abundant credit, fostering competition, and giving remote areas access to credit on terms not very different from urban centers. And the continuous adaptation of regulation has also avoided the disintermediation of deposits that occurred in the United States in recent decades, which helped reduce the illiquidity risk that plagued US banks that were heavily reliant on wholesale funding in 2007-2009.
Remarkably, the absence of crises in Canada has not created regulatory complacency. While the United States seems incapable of learning from its crises, Canada has been able to adapt its regulatory system to avoid banking crises despite never having had one. What explains that difference?
Canada’s political institutions made it very hard for bankers and populists to form coalitions to use banking regulation as a means of extracting benefits from the rest of the population. In the United States, the agrarian populist-unit banker partnership, and later the megabank-urban activist one, succeeded at the expense of the rest of the population, but in Canada, banking laws and regulations were made and enforced in the broad public interest. This was an outcome of the Constitutional design of Canada which combines the centralisation of power in the national government with a powerful appointed Senate. For example, the same agricultural populist-unit banking factions that formed to control banking policy in the United States attempted the same in Canada, but they consistently were defeated in the Canadian Senate.
Canada’s anti-populist Constitutional design, and its effective bank regulatory policy, were not accidents. They reflected the conscious intent of the British government to facilitate Canadian development and population growth in the 19th century. After winning Canada from the French in 1759, Britain found itself ruling a colony filled with French settlers who longed for independence. French Canadians also occupied a crucial geographical blocking position for transportation to the Atlantic economy. After a history of French Canadian attempts to prevent western development, and an uprising in 1837, the British hit upon a solution to deal with their enemy within – namely, diluting the voting power of the French by unifying Canada and giving disproportional power to the British immigrant population.
The United States, in contrast, was a country born in revolution, with irrepressible populist tendencies apparent from the start. Its Constitution included a number of steps to limit the power of populist ‘factions’, to use James Madison’s term. Ironically, the separation of power in three separate branches of the national government, and federalism’s granting to the states of significant power (including banking regulation, for most of US history), did not thwart populist influences on banking policy. On the contrary, the fragmentation of power under federalism made it easier for the agrarian populist-unit banker coalition to prevent nationwide branching.
The checks and balances of the US system blocked on-budget legislative remedies to urban poverty and inequality of opportunity in the late 20th century, which encouraged the formation of the megabank-urban activist coalition as an alternative. Although it was wasteful and cruel to fight poverty by giving the urban poor a lottery ticket on a future (that is, a low probability of being able to keep their dream house, financed with a zero down payment mortgage and underwritten without any evidence that the borrower had a job), that path appealed to urban activists, their constituents, and their political allies in Washington as better idea than nothing at all.
It is as unlikely that the United States will end its history of banking crises as it is that US citizens would swap our Constitution for Canada’s. After all, very few Americans would favour the idea of having the UK’s Queen appoint our senators. The price of American populism, and its institutional foundations, however, has been a history of banking instability and empty regulatory theatre, and those patterns are likely to continue for many years to come.
Charles W Calomiris is Henry Kaufman Professor of Financial Institutions at Columbia University and a visiting fellow in the research department of the International Monetary Fund
Stephen H Haber is the AA and Jeanne Welch Milligan Professor in the School of Humanities and Sciences at Stanford University and the Peter and Helen Bing Senior Fellow at the Hoover Institution
The authors have written a new book, Fragile by Design: The Political Origins of Banking Crises and Scarce Credit (Princeton 2014)