14-10 Club – 1st December 2011

Robert May, Baron May of Oxford

Stability and Complexity in Model Banking Systems

This was a talk at the 14-10 Club, the club at the Royal Institution for professionals who use maths in finance. http://www.rigb.org/contentControl?action=displayContent&id=00000004814

The speaker on 1^st December was Lord May (ex Chief scientific advisor to the UK government and ex President of the Royal Society).

Following are some incomplete notes from the talk and references for further reading.

Ecology is only 100 years old. During the first half of its development it was descriptive. Only since the 1960s has it developed a conceptual base. E.g. Hutchinson and Ken Watt.

In 2006-2007 the New York Fed and the National Academy of Sciences developed the report ‘New Directions for Understanding Systemic Risk’.

http://www.ny.frb.org/newsevents/news/research/2007/rp071031.html

Robert May built a simple model of banking. He checked with experts in banking whether this model was as simple as possible while still being a reasonably realistic representation of the banking system. They agreed that it was a reasonable model.

The model consisted of dividing a bank’s assets and liabilities each into 3 parts as follows:

Total Liabilities = Deposits + Interbank borrowing + Net worth

Total Assets = External Assets + Interbank loans + Capital

He found that to maximise instability you would want banks that hold both external assets and interbank loans. He showed some illustrations of model results that showed this result.

As an aside he quoted Joseph Stiglitz who said “Why is the invisible hand of the market invisible? Because it’s not there!”. May was also entertainingly rude about the banks behaviour pre-credit crunch, including the belief that B rated mortgage loans could be packaged together to make triple A rated bonds, because house prices in different parts of the US were uncorrelated. He quoted from Michael Lewis’s book “The Big Short”. Also he was dismissive about the notion that market price movements can be modelled by guassian distributions, suggesting that natural scientists would have tested this idea first rather than assuming it and using it in their models.

In the US 80% of assets are held by 1.4% of banks.

When banking crisis hit you can get fire sales of assets, driving down asset prices. However, worst of all is liquidity hoarding, it is contagious. John Vickers has worked on this issue.

Diversification of assets can be good for individual banks but bad for the system. Basle I and II promoted the spreading of assets.

May quoted some figures about how much the size of bank assets has ballooned in the UK compared to GDP. He asked the question, what purpose has this served?

He asked the fundamental question, what is the financial system for? The answer is it is for recycling capital into productive uses.

Benjamin Friedman carried out research to answer the question; what is the cost overhead of this capital recycling system as a proportion of the USA’s economy? This was published in the Proceedings of the American Academy of Arts and Sciences. He found that 30 years before the credit crunch the cost was 10% of all profits in the economy, 15 years before the credit crunch the cost was 20% of profits and just before the crunch the cost was 30% of profits.

May’s last slide was some references to papers relating to banking stability:

References:

Haldane AG, May RM, Systemic risk in banking ecosystems, Nature 2011

http://www.nature.com/nature/journal/v469/n7330/full/nature09659.html

Jones, Claire – Preventing system failure, Central Banking 21(1) 69;75 (2010)

http://www.centralbanking.com/central-banking-journal/feature/2043146/preventing-failure

Haldane AG, Rethinking the financial network (Bank of England speech 2009) 386

http://www.bankofengland.co.uk/publications/speeches/2009/speech386.pdf

May, Levin, Sugihara – Ecology for bankers, Nature 451, 893-895 (2008)

http://www.nature.com/nature/journal/v451/n7181/full/451893a.html

May, Arinaminpathy N – Systemic risk the dyanamics of model banking systems, Journal of the Royal Society Interface 7, 823-837 (2010)

http://rsif.royalsocietypublishing.org/content/7/46/823.full