Political Origins of Banking Crises

(Adapted from Banks and Political Bargains, a Quant Perspectives column published by the Global Association of Risk Professionals)

We recently read, appreciated, and enjoyed the analysis and proposal of Charles Calomiris and Stephen Haber in Fragile by Design – The Political Origins of Banking Crises and Scarce Credit (Princeton University Press, 2014).  Calomiris and Haber analyze banking structure and operations over the past centuries and across varying countries and types of government.  These authors argue that “political bargains” in all cases determine banking structure and operations.  The bargains differ substantially depending on whether the form of government is authoritarian or democratic.  Within democratic societies, the degree of populism (majority rule) versus liberalism (protection of individual, corporate, and property rights from majority rule) also plays a large role in the bargain that underlies the banking system.

In the landscape of Calomiris and Haber, governments need banks as agents for state borrowing.  Thus, governments charter banks to enable and expand such borrowing and also to accomplish political tasks of providing loans to favored sectors and individuals.  In return, governments confer limited “charters” such that banks with charters have reduced competition and therefore enjoy higher returns than they would receive in an open market.  Direct and indirect government support activities, including deposit insurance, lending, and bailouts, are just additional features of the “bargain.”

Fragile by Design profiles the histories of banking in the United Kingdom, Canada, the United States, Mexico, and Brazil.  The authors’ theory explains, for example, why “the United States has had 12 major banking crises [since 1840] while Canada has had none.”  More specifically, the “political bargain” among government, banks, and regulators in the U.S. prohibited or strongly discouraged branch banking until 1980!  Unlike the Canadian banks with sizable branch networks, stand-alone (“unit”) banks have no risk diversification.  Calomiris and Haber describe the creation of deposit insurance by Federal legislation in 1933 as a government prop to unit banks.  Deposit insurance schemes at the state level had already failed by that point due to moral hazard and excessive taxpayer losses.

Calomiris and Haber argue that, since 1980, the political bargain detrimental to safety and soundness of U.S. banks is the government “encouragement” of residential mortgage lending.  Rather than wade through this contentious question of recent history here, we note that Morgenson and Rosner’s Reckless Endangerment (Times Books, 2011) shares the Calomiris-Haber view.  In the (majority) Financial Crisis Inquiry Report of 2011, however, the government commission disagrees that either government housing policy or the government-sponsored enterprises contributed significantly to the degradation of mortgage underwriting standards.

What does the financial risk manager gain from studying Fragile by Design?  The best answer may be simply that one interprets current events in a new light.  Consider, for example, the role of the (U.S. government’s) Consumer Protection Financial Bureau (CPFB).  The CPFB defines “qualified mortgage” (QM) and “ability to repay” (ATR) rules for mortgage origination.  When banks lend within the QM/ATR guidelines, they are far less likely to suffer future government penalties.  This is certainly a “bargain.”

Yet the CPFB’s mission is also to prod banks to lend rather than simply protect consumers from bank malfeasance and high fees.  One might call the CPFB-Bank relationship a “balancing act” or a “partnership” (in the words of Calomiris and Haber).  Further, the CPFB arguably encouraged the private firm FICO to change its credit score model in a manner that will boost apparent creditworthiness.  (See the credulous “FICO’s new scoring model to help lenders better assess risk,” Reuters, August 8, 2014.)

The quantitative analyst has a tremendous challenge!  How is it possible to incorporate bargains with government into default and valuation models?!