Too Big to Fail: Measures, Remedies, and Consequences for Efficiency and Stability

Bank failure was almost unthinkable in Europe long before “too big to fail” became a byword for U.S. regulatory policy on big banks. But the 2007−2009 global financial crisis, which for some countries grew to a full-blown crisis, made the unthinkable a real possibility.

U.S. and EU regulators were responsible for much of the increase in bank size and concentration over the previous three decades, and the crisis forced them not only to contemplate allowing banks to fail, but to plan for banks’ orderly demise. To that end, they have proposed and implemented remedies for the too-big-to-fail, or TBTF, problem. One major goal has been to reduce the risk of contagion: the spread of financial trouble through links among many banks.

The regulators are trying four approaches to TBTF: (1) restrict bank size; (2) ring-fence bank activities into distinct legal and functional entities (in the U.S., through the Volker rule); (3) require higher capital levels; and (4) provide a framework for orderly resolution. These approaches are not entirely independent, with one potentially having repercussions for the others.

But size is not the only determinant of risk. The complexity of a bank’s business model can also increase the risk of one bank’s troubles spreading to other institutions. Thus, in a new paper, we evaluate whether current TBTF reforms enhance the financial system’s stability, and we explore trade-offs between maintaining system stability and efficiency.

The very designation of TBTF can affect stability and efficiency. It conveys an implicit guarantee to a bank’s creditors,[1] creating shareholder incentives to shift more risk to the government.[2] What’s more, a TBTF bank’s structural complexity often reduces the transparency of its risk-taking and can weaken market discipline. The implicit guarantee also distorts competition to the benefit of TBTF banks—distortions that can be amplified by “regulatory capture,” whereby a few very large banks can wield influence over regulators. Finally, TBTF links bank risk to sovereign risk.

But the different treatment of large banks has been formalized, with the designation of 30 banks as globally systemically important banks (G-SIBs),[3] and an even larger number designated as systemically important financial institutions (SIFIs). The designations have implications for regulation and supervision, as well as for the degree to which they may enjoy implicit subsidies to costs of funding. No designation standard exists for SIFIs; each country’s regulator determines its criteria and the rules and supervision that apply.

The fiscal and political costs associated with bank bailouts of the 2007−2009 crisis, and the euro debt crisis thereafter, suggest that the large banks in many countries aren’t only TBTF, but also too big to save. Relatively small countries whose large banks operate internationally are particularly vulnerable. In 2009, for example, Iceland lacked the resources to rescue three major banks engaged in large international operations and nearing collapse.

The dilemma for policymakers? If reforms don’t credibly eliminate the TBTF problem, a “too big to save” crisis may make bank resolution even more doubtful. The problem may become particularly acute for banks with large operations in multiple countries whose interests may conflict over the burden of crisis management.

To solve problems of excessive size and complexity, we must understand why banks have grown to their current size and structural complexity. Thus, we look at various measures of bank size and growth, since such measures are essential for implementing appropriate remedies. We present data on the world’s 100 biggest publicly traded banks, in various dimensions—e.g., total assets; total assets reported on a gross (IFRS) versus net (GAAP) basis; total assets as share of the home-country GDP; and equity capital relative to home-country’s GDP. As we show, no single measure tells the full story on bank size.

Different measures of concentration, for example, produce starkly different rankings among the 100 biggest banks. The largest 25 banks account for about half of the assets of all banks worldwide, and their assets total almost two-thirds of world GDP. But with the growth of bank assets, we are seeing not just increased concentration, but also what has been called the “financialization” of many advanced economies, i.e., market-oriented financial activities not directly linked to intermediation.

Most of the world’s biggest banks have a mix of businesses holding different assets and engaged in different cross-border activities. This mix is a key factor affecting complexity, especially when there is no financial independence of a bank’s functional and legal entities, and no consistency among legal, functional, and financial organization.

The Financial Stability Board counts complexity as just one factor among others that capture systemic risk in banks designated as G-SIBs. We argue that the FSB criteria for complexity are both too limiting and too broad, and that the weights of different factors contributing to systemic risk are important for the relative systemic risk ranking of banks with different business models. To show this, we present empirical evidence on the links between bank characteristics associated with TBTF and risk. One important factor for both a bank’s default risk and contribution to systemic risk is the number of subsidiaries, which is an indicator of complexity.

We examine the four TBTF reform approaches in terms of financial stability and efficiency. And while we agree that banks have grown too large and complex, we doubt that explicit limits on size and scope can be defined or enforced without risking great losses of efficiency.

Restrictions in size, for example, don’t differentiate between big banks that pose a systemic risk and those that don’t—and adjusting restrictions according to varying degrees of “bigness” offers no guarantees of solving TBTF. Another example: while it is possible to count domestic and foreign subsidiaries of the largest banks, those numbers say nothing about the size of the foreign operations.

Ring fencing requires banks to separate risky activities legally, functionally, and financially, or to cease these activities altogether. The Volker rule prohibits an insured depository institution or its affiliates from engaging in “proprietary trading,” and bars insured institutions from sponsoring or acquiring ownership interests in hedge funds or private equity funds.[4] We ask, in the end, whether the rule targets the wrong firms. The more regulators limit banking activities, the more likely they are to create incentives for those same activities at nonbanking or offshore firms.

A conflict of interest may arise, with banks preferring a subsidiary structure with mutual default insurance (rescue) arrangements, though this contributes the most to systemic risk. Regulators prefer standalone banks, but these cannot exploit financial synergies. This conflict between the objectives of banks and regulators can be mitigated by capital requirements. If these are made effective and the probability of a government bailout is relatively low, both bank and regulator prefer the subsidiary structure. If the probability of a government bailout is relatively high, both bank and regulator prefer the standalone structures.

We examine two issues arising from capital requirement proposals: whether the capital requirement is in fact effective, and what the consequences may be from special capital requirements imposed on systemically important banks. Setting a very high capital requirement creates a large discrepancy between required and desired equity financing and is likely to create strong incentives to manipulate or avoid the requirement. The stronger these incentives, the greater the costs to supervise and examine compliance unless rules for bank governance can also be reformed.

The fourth approach requires banks to devise their own “living wills.” But no one knows how or whether such a plan will work. Moreover, loopholes in the rules will always exist, meaning that bailouts could still occur. And conflicting interests of countries in crisis suggest that mutual recognition of resolution authorities is very unlikely.

We support higher capital requirements to support both efficiency and stability—but only in terms of a relatively simple leverage ratio. If market discipline and capital requirements increase together, these two reforms will support each other. We also suggest reducing resolution costs by requiring banks and other financial institutions to form legal entities that coincide with their functional entities. This would complement both resolution procedures and capital requirements.


[1] There are estimates of the implicit subsidy associated with TBTF. For example, an International Monetary Fund study estimates that the implicit subsidies given just to the G-SIBs in 2011–2012 represent around $15–$70 billion in the United States, $25–$110 billion in Japan, $20–$110 billion in the United Kingdom, and $90–$300 billion in the euro area.

[2] Some observers argue that the size of a bank’s failing is less important for a government’s decision about whether to bail out the bank than is the size (share) of the banking system under stress. For example, some emphasize “too many to fail.”

[3] This is the number of G-SIBs identified by the Financial Stability Board as of November 2014.

[4] The so-called Lincoln Amendment in Dodd-Frank withholds FDIC insurance and Federal Reserve borrowing from derivatives dealers, which could force banks to establish separate affiliates in which to engage in many derivatives activities.

This post comes to us from James R. Barth, Lowder Eminent Scholar in Finance at Auburn University, and Clas Wihlborg, the Fletcher Jones Chair in International Business at Chapman University. It is based on their recent article, “Too big to fail: Measures, remedies, and consequences for efficiency and stability,” available here.

Leave a Reply

Your email address will not be published. Required fields are marked *