Spotlight on Young Researchers: Eric Finn Schaanning

 

Eric Finn Schaanning was drawn to research by a thirst to understand what mechanisms drove the financial crisis. He has just defended his AFR PhD at Imperial College London, during which he developed an operational ‘stress test’ model that is already being used by two European Central Banks. The half Luxembourg, half Norwegian national is now a Senior Advisor at Norges Bank, where he continues to analyse and help improve understanding of how financial institutions react to economic shocks.

Throughout his undergraduate studies, Eric was convinced he would not be pursuing the PhD or researcher path. But all of this changed as Eric began to write his master thesis, as he explains:

“I began to realise how many open questions there are in understanding and modelling systemic risk, both from a financial as well as a mathematical perspective. This got me really excited! I wanted to understand in-depth what amplification mechanisms drove the financial crisis and how one could model these more realistically.”

Soon after, Paul Embrechts who supervised Eric’s master thesis at ETH Zürich, introduced him to his future PhD supervisor, Rama Cont at Imperial College London, and the collaboration with Norges Bank began.

Eric studies systemic risk in financial networks: ‘systemic risk’ is a field that analyses the mechanisms that can enable a localised ‘shock’ – such as the default of a medium-sized bank – to echo through vast parts of the financial system. It then has the potential to threaten the overall stability of the system, a recent example of this being the financial crisis of 2007-09 – the worst financial crisis since the Great Depression of the 1930s.

The challenge of developing a realistic model

The collective reaction of financial institutions is at the very core of the snowball effect that can amplify the impact of financial shocks. ‘Stress tests’, which serve to understand how well a financial institution can withstand an economic crisis, do not currently take the feedback effects from such reactions to the initial shock into account. This is where Eric’s work comes in, he explains:

“This photo shows the ‘indirect contagion network’, i.e. the network of overlapping portfolios between European banks through which fire sales contagion can take place. The network is generated based on data from the European Banking Authority (EBA).”

“My work focusses on developing models that can quantify how banks or other financial institutions react to a stress scenario by deleveraging[1] their portfolios in a realistic way, and infer what spillovers these reactions create.”

“On a conceptual level, this is not new at all. There are many models that describe these dynamics qualitatively. The emphasis of our work lies on ‘quantify’ and ‘realistic’: The challenge is to develop models that regulators can take to their datasets in order to estimate how bad a stress scenario could eventually become.

“Beyond prudential policy, this has also interesting implications for portfolio risk management at individual institutions: in presence of such effects the risk of a portfolio depends on what other institutions hold and do – our model can help to shed light on this.”

Evolving technology calls for new models

Why is it important to ‘update’ systemic risk models? Healthy financial markets are key to a prospering economy. Advances in technology not only impact how these markets work, but also how people and companies interact in them, as Eric explains:

“Think for instance of mobile payment apps, or ‘Robo advisors’ that can make automated portfolio allocation decisions for investors. Traditional economic models may not be pertinent anymore for comprehending and properly analysing the risks in these complex systems, so new tools are needed.

“Overall, markets have become much more intertwined and interconnected, so we need to understand the impact of automation, central clearing, algorithmic execution, high frequency trading and many more, on a deeper level in order to ensure that these markets in fact do what we expect them to, and that they serve society.”

Eric adds: “I would like to contribute to understanding these systems better, with the ultimate goal of helping to create policies that can contain the risks embedded in them. It is very difficult to design policies without having a very detailed knowledge of what it is that one is trying to regulate.” 

“This shows the number of banks that pass the stress test when one ignores feedback effects, but fail the stress test once price-mediated contagion (i.e. losses from feedbacks) are taken into account. The number of banks is shown as a function of the initial intensity of the stress scenario (in this case defaults in Spanish residential and commercial mortgage exposures) and the liquidation horizon, i.e. how fast banks liquidate their portfolios. The the intuition is simple, the faster they do it, the more intensely the prices will drop, and consequently the larger the losses.”

“I find it incredibly satisfying that my work finds such a direct application”

So how does one go about studying systemic risk? Eric explains it first involves reading many regulatory reports to get the institutional setup right, and then a lot of combining and analysing datasets. During his PhD, Eric’s work focussed mainly on describing the reaction of banks, and coding a simulation module that could be applied in practice.

Eric explains the interdisciplinary approach he took to developing the model: “Getting there was not entirely trivial: the modelling approach is somewhat unorthodox, both from a mathematical and economics point of view. I thought that a data-driven approach would be the most useful and practically relevant, but had no proof of course until recently. So some mathematicians asked me why I was not proving more theorems, while some economists asked me why the model was so mechanical and complex (making it impossible to write down closed-form solutions).”

Eric’s model is already in use by two European Central Banks – both the Bank of England and Norges Bank are in the process of implementing it for their-in house stress tests, to which Eric says:

I find it incredibly satisfying that my work finds such a direct application.”

Luxembourg has potential for highly innovative research

When asked what potential role Eric sees for Luxembourg in terms of research in his domain, Eric says: “Luxembourg is home to the world’s second largest fund market. I think it goes without saying that properly understanding (and mitigating) the risks around this position is of paramount importance. This is not just in the interest of Luxembourg as financial hub, but given its size, this is arguably a case where one should focus on interconnections, spillover effects and thus systemic risk in general.

“I think two central pieces towards achieving this goal are that (i) data is made available to researchers (under appropriate privacy protections, of course), and (ii) there is a close collaboration between academics, policy makers, and market participants. With these two premises given, I think Luxembourg has the potential to contribute to this field with highly innovative and important research.”

[1] i.e. reducing the size of their portfolios through rapid sales of assets.

Published 10 May 2017


More about Eric Finn Schaanning

Visit Eric Finn Schaanning’s homepage

You can also read about Eric on science.lu (German and French).

Eric Finn Schaanning

RELATED PROGRAMMES

About Spotlight on Young Researchers

Spotlight on Young Researchers is an FNR initiative to highlight early career researchers across the world who have a connection to Luxembourg. This article is the 10th in a series of around 25 articles, which will be published on a weekly basis. You can see more articles below as and when they are published.

This site uses cookies. By continuing to use this site, you agree to the use of cookies for analytics purposes. Find out more in our Privacy Statement