Factors driving systemic risk of banks in Latin America

Systemic risk is the risk “that cumulative losses will accrue from an event that sets in motion a series of successive losses along a chain of institutions or markets comprising a system… That is, systemic risk is the risk of a chain reaction of falling interconnected dominoes” (Kaufman 1995). We basically follow this idea by measuring the contagion from banks to the financial system and vice versa. European Systemic Risk Board and European Commission (2010) defines systemic risk as the risk of disruption in the financial system with the potential to have serious negative consequences for the internal market and the real economy. Similarly to this idea, Acharya et al. (2011) and Adrian and Brunnermeier (2011) quantify systemic risk by measuring a bank’s (risk) contribution to the overall financial system. For a list of more possible definitions of systemic risk in the literature, see Prokopczuk (2009).

Banking contagion, concentrating on the transmission of a bank shock to other banks or the financial system, lies at the heart of systemic risk. Early, Bagehot (1873) diagnoses as follows: “In wild periods of alarm, one failure makes many, and the best way to prevent the derivative failures is to arrest the primary failure which causes them”.

Fiordelisi and Marqués-Ibañez (2013), however, show that individual bank risk tends to be systematic (i.e. non-diversifiable) and has a direct impact on the European banking market. In this respect, it is useful to also consider individual bank risk for measuring systemic bank risk.

The problems of systemic risk / too big to fail have been characterized by using “one of the most famous phrases in the entire history of Supreme Court opinions”, i.e. the Judge-Potter-Stewart-Expression “I can’t define it, but I know it when I see it”. On the original (about a definition of pornography), see Gewirtz (1996). In the given banking context, see the citation of Dean Baker in Whitelaw (2009) and, more recently, Hansen (2014), p. 16..

The variety of systemic risk measures is growing fast: Bisias et al. (2012) and Billio et al. (2012) provide overviews of systemic risk measures in finance literature.

This criticism – the ignoring of the tails, backward orientation and sensitivity to model specifications – is similar to the general \( VaR \) criticism. As a response to the criticism and the bad experience with VaR modifications have been suggested. Real alternatives, however, remain being explored by academia or the financial industry.

To deal with the low number of tail observations in a heavy-tailed environment, Van Oord and Zhou (2011) propose an estimator of tail beta by an EVT approach.

For other studies that use the same 2006–2011 period as a reference for the financial crisis see e.g. Gilbert et al. (2013), Trunk and Stubelj (2013), Mukhlas (2012)and Kibritcioglu (2011).

Namely Argentina, Brazil, Bolivia, Chile, Colombia, Costa Rica, Dominican Rep., Ecuador, El Salvador, Honduras, Jamaica, Mexico, Nicaragua, Panama, Paraguay, Peru, Uruguay, Venezuela.

We manually checked missing accounting values, finding most of them. In five cases, however, we did not find the necessary data, which may bias our results since balance sheet composition may affect the bank opacity, see Flannery et al. (2013). In a recent paper on bank opaqueness, Mendonça et al. (2013) find that a decrease in bank opaqueness fosters an environment favourable to the development of a sound banking system and the avoidance of financial crises.

As such, Brazil’s role for our sample is considerable, but in harmony with its general role in Latin America: Here, Brazil is the dominating country with about one third of the Latin American population and GDP. As well, about 70 % of the Latin American market capitalisation is concentrated in this country and in the major Latin America Indices such as the MSCI Emerging Markets Latin America and the S&P Latin America 40, Brazilian stocks make up for more than 50 %. For means of comparison, we have conducted our analysis with an Latin-America-without-Brazil, too. The results can be obtained from the corresponding author upon request.

The Latin America Datastream Financials Index (Datastream code: FINANLA) offers the best available coverage for the Latin American financial sector. We also create our own indices by value weighting the stock returns of all banks in our samples (as proposed by e.g. Weiß et al. 2014), leading to the same core results for our regression. However, as we are more interested in analysing the determinants for systemic risk in the Latin American financial sector as a whole, those results are not presented in this paper.

Interestingly and in contrast to most of the literature, Dungey et al. (2012) find cases where firm characteristics make little difference to the systemic risks of banks.

Basel Committee on Banking Supervision (2014): The BCBS uses exposures (a method comparable to ours) as an indicator of systemic importance.