In almost all financial businesses there will be unsuitable models. But how do imperfect models manage to exist for so long?
Claudio Albanese, Founder and Head of Development at Global Valuation – and our guest for this episode of Quantcast – thinks Darwin’s theory of evolution may offer some insight into this question.
For his latest study, Albanese, in collaboration with StÃ©phane CrÃ©pey from the University of Paris and Stefano Iabichino from JP Morgan, adapted Darwin’s principles to show how low-quality models that overestimate and over-cover structured products can survive – and even thrive – in the short term.
Since traders can sometimes benefit from over-hedging certain derivative structures, they may be tempted to ignore the weaknesses of particular models. The trade-off is that they will end up taking long-term risks that are not taken into account by the models, potentially leading to painful losses.
As an example, Albanese and his co-authors show the effect of using low-quality models for price bracket runs – an exotic structured product with a loaded history. But their conclusions are also relevant for other derivatives, especially those with built-in callability.
Albanese argues that a Darwinian lens can help banks spot bad models and ineffective hedging strategies – and avoid the dangerous exposures they create.
This foray into model risk is a departure for Albanese, who is known for his contributions to the study of derivative valuation adjustments. In this regard, he shares his thoughts on the concept of hedge valuation adjustment, which was recently introduced by Ben Burnett at Barclays and calls for its extension beyond transaction costs.
This could be a topic for another podcast.
02:50 Motivation for the model risk study
05:25 Range stacking case study
9:30 a.m. The risk of over-coverage
13:10 Visualize pattern risk as it materializes
17:15 The link between natural selection and derived models
21:15 What products are potentially affected by flawed pricing models?
25:00 Future research
To listen to the full interview, listen in the player above or download. Future podcasts in our Quantcast series will be uploaded to Risk.net. You can also visit the main page here to access all tracks, or go to itunes store or Google Podcasts to listen and subscribe.