System-wide risk and systemic importance: an incomplete review of metrics and data

Duration: 49 mins 36 secs

Share this media item:

Embed this media item:

<iframe width="_width_" height="_height_" src="https://sms.cam.ac.uk/media/1810966/embed" frameborder="0" scrolling="no" allowfullscreen></iframe>

Choose size:

About this item

Available Formats

About this item

Description:	Tarashev, N (Bank for International Settlements) Thursday 25 September 2014, 14:00-14:45


Created:	2014-09-29 16:42
Collection:	Systemic Risk: Mathematical Modelling and Interdisciplinary Approaches
Publisher:	Isaac Newton Institute
Copyright:	Tarashev, N
Language:	eng (English)
Distribution:	World (downloadable)
Explicit content:	No
Aspect Ratio:	16:9
Screencast:	No
Bumper:	UCS Default
Trailer:	UCS Default


Abstract:	I. Two major competing types of systemic-risk metrics are: (i) quantiles, e.g. VaR; and (ii) tail expectations, e.g. expected shortfall (ES). I.a. The choice of a risk metric is often motivated with: (i) properties of the metric’s estimator; (ii) computational convenience. But different metrics correspond to different economic concepts and thus to different policy objectives (e.g. to reduce system-wide risk vs. to build a war chest for a systemic crisis). Similar tension between alternative metrics exists in an investment portfolio context. I.b. Given a metric for system-wide risk, the Shapley value offers a convenient prism for comparing alternative measures of the systemic importance of individual institutions. It also allows for deriving important properties of these measures and for identifying the policy contexts in which they should be used. I.c. There is a subtle difference between the impact of an institution on system-wide risk and the presence of an institution in systemic events. Ignoring this difference could lead to grossly misleading conclusions as regards systemic importance. II. Data availability shapes existing approaches to measuring system-wide risk and systemic importance. One approach builds on a structural model of system-wide risk and relies on balance sheet data and commercial vendors’ estimates of individual riskiness. Another approach builds on reduced-form models and relies on market-price data. II.a. The first approach has revealed a material impact of the system's network structure on system-wide risk and the systemic importance of individual institutions. The scarcity of real-world data on such structures is thus a major problem in practical applications. II.b. The second approach makes it possible to study tail interdependence across institutions. Tools based on extreme-value theory deliver estimates of such interdependence, which contributes materially to measures of systemic importance. Related Links: http://www.sciencedirect.com/science/article/pii/S1042957313000326 - "Measuring the systemic importance of interconnected banks" http://www.bis.org/publ/work308.htm - "Attributing systemic risk to individual institutions" http://www.bis.org/publ/qtrpdf/r_qt1306g.htm - "Looking at the tail: price-based measures of systemic importance"

Abstract:

I. Two major competing types of systemic-risk metrics are: (i) quantiles, e.g. VaR; and (ii) tail expectations, e.g. expected shortfall (ES). I.a. The choice of a risk metric is often motivated with: (i) properties of the metric’s estimator; (ii) computational convenience. But different metrics correspond to different economic concepts and thus to different policy objectives (e.g. to reduce system-wide risk vs. to build a war chest for a systemic crisis). Similar tension between alternative metrics exists in an investment portfolio context.
I.b. Given a metric for system-wide risk, the Shapley value offers a convenient prism for comparing alternative measures of the systemic importance of individual institutions. It also allows for deriving important properties of these measures and for identifying the policy contexts in which they should be used.

I.c. There is a subtle difference between the impact of an institution on system-wide risk and the presence of an institution in systemic events. Ignoring this difference could lead to grossly misleading conclusions as regards systemic importance.

II. Data availability shapes existing approaches to measuring system-wide risk and systemic importance. One approach builds on a structural model of system-wide risk and relies on balance sheet data and commercial vendors’ estimates of individual riskiness. Another approach builds on reduced-form models and relies on market-price data.

II.a. The first approach has revealed a material impact of the system's network structure on system-wide risk and the systemic importance of individual institutions. The scarcity of real-world data on such structures is thus a major problem in practical applications. II.b. The second approach makes it possible to study tail interdependence across institutions. Tools based on extreme-value theory deliver estimates of such interdependence, which contributes materially to measures of systemic importance.

Related Links: http://www.sciencedirect.com/science/article/pii/S1042957313000326 - "Measuring the systemic importance of interconnected banks" http://www.bis.org/publ/work308.htm - "Attributing systemic risk to individual institutions" http://www.bis.org/publ/qtrpdf/r_qt1306g.htm - "Looking at the tail: price-based measures of systemic importance"

Available Formats

Format	Quality	Bitrate	Size
MPEG-4 Video	640x360	1.93 Mbits/sec	721.39 MB	View	Download
WebM	640x360	437.64 kbits/sec	158.99 MB	View	Download
iPod Video	480x270	522.02 kbits/sec	189.64 MB	View	Download
MP3	44100 Hz	249.8 kbits/sec	90.81 MB	Listen	Download
Auto *	(Allows browser to choose a format it supports)

Streaming Media Service Upload

System-wide risk and systemic importance: an incomplete review of metrics and data