Where the Models Are Wanting Part 2: Banks Networks, Risks and Modern Investment Theory

by Dr Constantin Gurdgiev, Adjunct Assistant Professor of Finance with Trinity College, Dublin

You may read part 1 of this series here.

In the previous post1, we covered some of the shortcomings of the core financial analysis and equity valuation models when applied to the banking sector stocks. This time around, lets take a look at the effects of networked banking systems on the models’ validity.

The timeline of the crisis in Europe, as well as the events in Ireland, Greece and Cyprus, highlight yet another, entirely distinct dimension of banks-sovereigns risk transmission channels. That dimension relates to the networks of banks – the complex web of links through interbank funding markets, but also cross-border assets exposures. Once again, these networks effects are not priced in the traditional equity valuations’ models.

In a sense, the banking sector is distinctly different from other sectors in the economy. For example, in manufacturing, two companies A and B competing against each other in a particular market segment face fully independent idiosyncratic risks and symmetric common shocks, allowing for simple separation of the two in factor analysis and the Arbitrage Pricing models.

In the banking sector, such competition is complicated by the fact that the two banks competing for the same customers can (and very often do) hold correlated assets and have linked exposures to each other through funding, derivatives, securitisation and repo markets. In this case, idiosyncratic risks that impact one bank balance sheet can significantly impact other banks valuations. Non-systemic risks become systemic, under specific conditions and idiosyncratic risks become correlated, including within a broader cluster of closely connected institutions.

In such a setting, the entire framework of factor models and Arbitrage Pricing models can fail.

To give you a flavour of the complexity involved in even attempting to realistically estimate these cross-links, Nier et al (2008)2, analysed the system fragility in hypothetical networks for varying parameter values. Their model was based on a one-period or static ‘game’ with no feed-back loops and no cascades of risks. It abstracted away from the possible pooling of risks across clusters of banks and ignored presence in the markets of non-banks counter parties – all inducing higher risks and more complex risk transmission channels in the system.

Source: Nier, et al's calculations using BIS locational banking statistics (quarterly).

Even in that case, a sub-system of just three banks interacting with each other in a network would generate non-linear risks to banks’ balance sheets. The authors also allow only a single idiosyncratic shock to one bank’s balance sheet at a time, which ignores the possibility of exogenous shocks affecting several banks simultaneously. Still, despite the vast oversimplification of the model, the results show that at low and high probabilities of a link in the system, system fragility changes with low predictability of both magnitude and direction of the shock impact. In addition, the reversal of links direction does not guarantee a symmetric reversal of the shocks’ impacts.

In a more empirical setting, Minoiu and Reyes (2011)3 study the topology of the international interbank networks between 1978 and 2009. They define two country categories, the core and the periphery, and show that both the number of links in the network and the size of exposures increase during economic or financial expansions and decrease following contractions. In simple terms, this means that timing of the shock is endogenous to its impact and the nature of network through which it is transmitted. But the network nature is also endogenous to the timing of the shock.

Reinhart and Reinhart (2010) provide empirical support for the above-mentioned cycles in network parameters4. Meanwhile, Eisert and Eufinger (2013) model establishes a network-based scenario where the banks incentives to become more interconnected in the markets is driven by the probability of a government bailout.5 This reinforces the first set of second order effects of banks links to sovereign bond risks discussed in the first part of this post and the networks channels for risk creation and transmission.

In summary, both second-order own effects of links between banks’ balance sheets and the sovereign bond risks (discussed in the first post), as well as the network effects, are inter-connected and reinforce each other. This creates highly complex, non-linear risk pathways from the sovereign bonds’ markets to banks’ balance sheets and, ultimately, banks’ equity valuations. These channels are not present in any other sector of the economy and do not apply to any other classes of equities.

All of which calls into question the validity of the core equity valuation models currently being used on analysis of banks shares. The traditional CFA toolbox appears rather outdated when it comes to the financial sector analysis.

References

1. See Where the Models Are Wanting Part 1: Banking Sector Stocks and Modern Investment Theory

2. Nier, Erlend W. and Yang, Jing and Yorulmazer, Tanju and Alentorn, Amadeo, Network Models and Financial Stability (April 1, 2008). Bank of England Working Paper No. 346:

3. Minoiu C, Reyes J, 2011. ft/wp/2011/wp1174.pdf">“A Network Analysis of Global Banking 1978-2009” Working Paper 11/74, International Monetary Fund,

4. Reinhart C, Reinhart V, 2010. “After the Fall” Working Paper 16334, National Bureau of Economic Research,

5. Eisert, Tim and Eufinger, Christian, Interbank Network and Bank Bailouts: Insurance Mechanism for Non-Insured Creditors? (July 9, 2014).

CFA Training - Free Trial with Learn Signal for WSO Members

Career Advancement Opportunities

March 2024 Investment Banking

  • Jefferies & Company 02 99.4%
  • Goldman Sachs 19 98.8%
  • Harris Williams & Co. (++) 98.3%
  • Lazard Freres 02 97.7%
  • JPMorgan Chase 03 97.1%

Overall Employee Satisfaction

March 2024 Investment Banking

  • Harris Williams & Co. 18 99.4%
  • JPMorgan Chase 10 98.8%
  • Lazard Freres 05 98.3%
  • Morgan Stanley 07 97.7%
  • William Blair 03 97.1%

Professional Growth Opportunities

March 2024 Investment Banking

  • Lazard Freres 01 99.4%
  • Jefferies & Company 02 98.8%
  • Goldman Sachs 17 98.3%
  • Moelis & Company 07 97.7%
  • JPMorgan Chase 05 97.1%

Total Avg Compensation

March 2024 Investment Banking

  • Director/MD (5) $648
  • Vice President (19) $385
  • Associates (86) $261
  • 3rd+ Year Analyst (13) $181
  • Intern/Summer Associate (33) $170
  • 2nd Year Analyst (66) $168
  • 1st Year Analyst (202) $159
  • Intern/Summer Analyst (144) $101
notes
16 IB Interviews Notes

“... there’s no excuse to not take advantage of the resources out there available to you. Best value for your $ are the...”

Leaderboard

1
redever's picture
redever
99.2
2
Secyh62's picture
Secyh62
99.0
3
Betsy Massar's picture
Betsy Massar
99.0
4
BankonBanking's picture
BankonBanking
99.0
5
kanon's picture
kanon
98.9
6
CompBanker's picture
CompBanker
98.9
7
dosk17's picture
dosk17
98.9
8
DrApeman's picture
DrApeman
98.9
9
GameTheory's picture
GameTheory
98.9
10
bolo up's picture
bolo up
98.8
success
From 10 rejections to 1 dream investment banking internship

“... I believe it was the single biggest reason why I ended up with an offer...”