The Financial Stability Board (FSB), established in the aftermath of the crisis by the G20, is an international entity that issues standards regarding global financial system. Since 2011, every November, it has been publishing lists of global systemically important banks (G-SIBs), based on the assessment methodology created by the Basel Committee of Banking Supervision (BCBS).
In order to designate a given institution as a G-SIB and to assess its level of systemic importance, the FSB relies on five indicators: size, interconnectedness, substitutability, cross-jurisdictional activity and complexity. Each of those factors should constitute 20% of the total score. Size is measured as total exposure. Interconnectedness constitutes a compilation of intra-financial system assets, intra-financial system liabilities, and securities outstanding. Cross-jurisdictional liability relies on cross-border claims and liabilities. The last two factors—substitutability and complexity—seem to involve the most complicated measurements. Four sub-indicators determine substitutability: assets under custody, payments activity, underwritten transactions in debt and equity markets, and trading volume. Those are meant to reflect areas in which banks are most likely to be irreplaceable and thus create systemic risk. In turn, complexity is understood mostly as engagement in market activities, as measured by notional amount of OTC derivatives, level 3 assets, and trading and available-for-sale securities. Depending on the total score each G-SIB is placed in a bucket from 1 to 5. Bucket placement corresponds with a G-SIB specific capital buffer each of these institutions has to hold on top of other capital adequacy requirements.
All this sounds pretty idyllic – every year the world receives a catalogue of the most systemically important banks, along with their ‘systemicness rank.’ Even though neither the methodology nor the list has any binding effect in and of itself, both reflect the result of global agreement among supervisors, and most national regulators follow similar criteria in domestic designation procedures and impose buffers reflecting bucket placement. Closer analysis suggests, however, some issues with the list.
1. An empty bucket will always be there.
On 22nd of November 2019, FSB published the newest version of its list of G-SIBs. It encompasses 30 banks allocated into four buckets. Here comes the first question—in four buckets? Why, if there are five available? The fifth bucket is supposed to serve as deterrent for banks from getting even bigger. BCBS methodology states clearly that, should a G-SIB land in that bucket, a new one will be added ‘to maintain incentives for banks to avoid becoming more systemically important’.
Currently, it is most likely JP Morgan Chase, occupying the next highest, fourth bucket, that could potentially reach the total score and move into the fifth one. However, the American biggest bank has been regularly placed within its present score range since the very introduction of the bucketing process in 2012. For 7 years, it has never reached the threshold of fifth bucket. One may wonder whether that incentive not to become more systemic actually works or whether banks exploit some other flaws of the methodology and still grow in their importance. This brings us to the next issue with the FSB’s list.
2. Systemic importance is not accurately assessed.
The FSB’s list can seem a little confusing to those who have knowledge of the business models and general riskiness of some of the G-SIBs. Bucket placement and differentiated capital buffers depending on the systemic importance constitute an attempt to tailor the framework according to each bank’s individual features. Thus, allocation in the same bucket implies equal systemic risk level created by given G-SIBs and results in the same additional capital requirement. However, one look at the composition of the buckets triggers some doubts regarding correctness of the methodology. For instance, investment-oriented Goldman Sachs and constantly troubled, internationally active Deutsche Bank are supposed to be as risky as mostly retail-focused, prevailingly domestic Wells Fargo. The same issue can be observed one bucket lower, which two of the world’s largest custodian banks, State Street and BNY Mellon, share with ING Bank, one of Europe’s largest consumer credit institutions. The designation process seems to overlook that G-SIBs do not constitute a monolithic group of financial institutions. On the contrary, they are very diverse, particularly in terms of funding, activities and created risks. By looking closer at the list, one realises that the systemic importance is unlikely to be being accurately assessed, if banks so different in terms of risk and activities end up in the same category. There are two main reasons for this inaccuracy.
First, it has been empirically proven that, as the list relies on end-year data, G-SIBs tend to off-load risky assets at year-end and thus decrease their systemic importance score. This practice not only undermines the credibility of the FSB’s list but is also detrimental to the market, as G-SIBs’ behaviour in this case is triggered by reporting obligations, not market forces.
Second, the two measures most closely related with the concept of systemic importance—interconnectedness and complexity—seem to be assessed in an overly simplistic and even random way. They take into account, as indicators, intra-financial system assets, liabilities, securities outstanding, OTC derivatives and level 3 assets, as well as trading and available for sale securities. However, these figures do not encompass all possible connections created between financial institutions, nor they indicate the actual place a given entity occupies in the network. Therefore, the potential magnitude and direction of contagion cannot be predicted using such indicators only.
3. Even though the list is not officially binding, national regulators treat it as such.
As mentioned before, the FSB’s list is a global standard and thus does not directly bind national regulators or supervisors. However, they tend to adjust the results of their designation procedures to the FSB’s selection, thereby turning the list into a legally binding tool. This tendency was recently visible in the example of Nordea. The Finnish Supervisory Authority overturned its previous decision to designate Nordea as systemically important, after the FSB did not include the bank in its G-SIB list. Consequently, given its actual direct impact on the national legal regimes and the regulation of G-SIBs, the list is much more important than it may seem.
What conclusions can be drawn from these issues related to the FSB’s list? To start with, it is of much higher relevance than many other non-binding global standards. Secondly, it fails to fulfil its goal of delivering an accurate systemic importance assessment and thus to indicate the level of systemic threat that each G-SIB poses. While the problem of G-SIBs’ window-dressing could be solved simply by requiring submission of average yearly data, the problems of systemic importance indicators and more individually tailored designation would require fundamental amendments to BCBS methodology.
Katarzyna Parchimowicz is a PhD candidate at the Centre for European Economic Law and Governance, University of Wroclaw. She was a junior academic visitor in the Commercial Law Centre at the University of Oxford during Michaelmas Term in 2019.