Many of the complex financial products which went into meltdown in the credit crunch were designed by physicists and engineers, the so-called "quants". E&T looks at the role of financial engineering in the crisis.
After the worst financial crash since 1929, and with today's economy in a recession that many fear could become as deep as the Great Depression of the 1930s, people are looking for someone to blame.
And engineers, surprisingly, are in the firing line. Not structural engineers or communications engineers, but financial engineers.
During the past two decades an increasing number of mathematicians, physicists and engineers have switched to work at investment banks and hedge funds, lured by the boom in financial engineering and quantitative finance. Talented engineers found they could make far more money on Wall Street or in the City than in academia or industrial research.
Monica Piercy, manager of Imperial College London's career and professional development service says that there "was a perception that the City was more alluring, dynamic, fast-paced". In three years, she says, you could earn "the salary that it would take ten years to achieve in science".
Some bankers referred to this new breed as 'rocket scientists', perhaps for no better reason than that aerospace was the most complicated form of engineering of which they had heard. But today the new breed is usually called simply 'quants'.
Perhaps the jewel in the quants' crown is the Black-Scholes theory. This options-pricing model, devised in 1973 by Myron Scholes, Fisher Black and Robert Merton, eventually won the Nobel Prize for Scholes and Merton, although not for Fisher Black, who had died. Black-Scholes gives a formula for valuing options in terms of the underlying security and cash. It quickly became used by all options traders and revolutionised finance. It gave banks a way of manufacturing options on a wide variety of securities and played a key role in the growth in derivatives over the last 30 years.
When Emanuel Derman, one of the best known quants, first went to work on Wall Street in 1985, his new boss gave him the task of improving a Black-Scholes-like model for bond options pricing. Derman proceeded slowly and carefully, as he had been trained to do in his previous career as a physicist. He studied the relevant papers, digested the theory, diagnosed the problem and began to rewrite the computer programme. But after several weeks his boss became impatient with the pace of his progress and told him sharply to be less detailed in his approach. Although Derman eventually did succeed in improving the model, he found that traders benefited more from the user-friendly interface that he provided than from small improvements in the maths.
Derman went on to become a quant whizz at Goldman Sachs - where he worked with Fisher Black - and at Salomon Brothers where he has been called the leading practitioner of his generation. Models he devised have been widely embraced. Looking back over his career, he says: "I feel very satisfied with the work I've done in finance, but I've learned that you can't expect the same success that is possible in physics. What I've done is find models that help you think qualitatively and quantitatively about what variables affect security values. But they're not gospel and they're not the principle of least action."
The Gaussian copula function
A quant whose work has been used widely, but whose influence has also been criticised, is David Li. He studied in China and Canada, and worked at Canadian Imperial Bank of Commerce and Barclays Capital. In the early 2000s, he pioneered the Gaussian copula function, which gave banks a way of modelling default correlation among baskets of mortgages. This was very useful to banks during the boom in securitisation of mortgages. But the model is useful at the expense of over-simplification. According to Viral Acharya, professor of finance at the Stern School of Business, New York University, the model has "some inherent weaknesses which were not taken seriously while managing risks".
Jean-Philippe Bouchaud, who still teaches physics at university, but spends most of his time at Capital Fund Management, a £3bn hedge fund which he co-founded, puts it more strongly. He says that the Li Gaussian copula "makes nonsense of the underlying reality - markets would not generate that sort of structure".
Bouchaud suggests that Black-Scholes itself leaves a lot to be desired. "Black-Scholes has zero risk - it's extremely beautiful but misses basic facts about why options exist," he says. "Yet it is still taught in textbooks and it creates in the minds of students and quants a wrong idea of what a model should be." For Bouchaud, both the Gaussian copula and Black-Scholes exhibit over-reliance on mathematical simplicity at the expense of humility before the empirical data.
Although over-simplification may be a fault of some financial models, that does not mean that non-quant investment bankers find them easy to understand. Many in senior positions in banks do not. "People created a complex language that management found difficult to understand," says Bouchaud. One of the elements in the run-up to the financial crisis was a sometimes almost superstitious faith by non-quants in the power of mathematics.
Derman says: "Financial engineering was certainly used in the development of over-leveraged complex products and in the attempt to make profits." In the aftermath of the crash, is there now a witch-hunt against quants? "I like the word 'witch'," says Derman, "Because it points out that deep inside people hope that one can magically predict human behaviour. 'When the hurly-burly's done.' There is a reaction against quants, and it's okay. If you want to get the upside, you have to take the downside too. No one should be exempt from that."
According to Bouchaud: "Everyone thought that someone else knew what they were doing. Management says: these guys [quants] know what they are doing. Quants say: management must know if these products work. Together they created a world in which they are happy and which allows them to get business."
Eton- and Oxford-educated Tets Ishikawa got a lot of business during the boom. He was a highly paid investment banker selling mortgage-backed securities for ABN AMRO, then Goldman Sachs and Morgan Stanley, before he lost his job in the credit crunch. He was not a quant but a salesman bringing in millions, and he relied on the work that quants do in creating financial products.
"I would not place a great deal of emphasis on quants as a cause of the financial crisis," he says. "There are quants sitting on both the investor and the banking side." He thinks that more to blame was the assumption that, as everybody was making money, "nothing could go wrong". He still feels that many of the ideas quants and bankers came up with were good. The problem was excess. "They took a good idea to the point it went barking mad," he says.
After he lost his job, Ishikawa wrote an entertaining book about his experiences called 'How I Caused the Credit Crunch', in which he details the life of excess, fast cars, fast women and the relentless pressure to do deals that characterised the lives of some investment bankers before the crash.
Although the book is fictionalised, the central character is "80-90 per cent me," he reveals, speaking on the phone from Thailand. "I was married to a Brazilian stripper and some of those things did happen to me," he says. Currently a freelance writer, he does not completely rule out going back into investment banking. "I don't intend to go back to it - just for now. Investment bankers are very opportunistic," he says.
The financial crisis of 2007-08 had many causes. Among them: years of low interest rates; the boom in mortgage lending; lowering of lending standards; growth in leverage and the property price boom. Before the crash, mortgage debt was generally seen as non-risky. Over-reliance on rating agencies, such as Moody's and Standards & Poors, was probably as least an important contributory factor as over-reliance on quants. Before the crash, mortgage debt was very highly rated.
Anne Richards is chief investment officer at Aberdeen Asset Management. Earlier in her career, she spent six years working in research and industry as an engineer. She says: "The financial crisis has its roots in the US housing market and people are ultimately to blame - from the guys who happily sold $100,000 mortgages to people who would obviously not be able to meet the repayments, to the bankers who bundled those mortgages together and added leverage in order to meet the inflated return expectations of their counterparties. Finally, regulators and rating agencies should also share some of the blame. The bundling or repackaging of mortgages has taken place since the mid-1980s, but people got gradually more greedy."
No jobs or different jobs?
Now, with many investment banks cutting staff, engineers whose career plan was to move into the City are stymied by a lack of jobs. The flow of engineers entering finance may be stemmed.
"Yes, I think this trend will begin to reverse," Piercy says. "As the competition for jobs intensifies, students with maths, physics and especially engineering backgrounds will look to alternative careers."
Didier Sornette, who after modelling earthquakes as a professor at UCLA moved to modelling crashes in financial markets and is now professor of finance at ETH Zurich (the Swiss Federal Institute of Technology), hopes the trend of engineers entering finance "reverses a bit".
He says: "We need engineers who solve the real problems of the world, like sustainability, new sources of energy and coping with the many new risks that society has to face. And not further developing a virtual financial world." But he does not think that the financial crash spells the end of financial engineering. "Is nuclear engineering over due to the existence of atomic bombs?" he asks rhetorically.
Despite Sornette's advice, many engineers are still looking for careers in finance. One area they are looking hard at is how to avoid such financial crises in future. Whether or not quants caused the financial crisis, they are now needed to help figure out what went wrong.
Dr Paolo Zaffaroni, programme director of the MSc in risk management and financial engineering at Imperial College Business School, says: "Finance needs quants and their role is secure. In fact, today they're tackling new problems which the crisis brought."
Acharya calls it a "Darwinian evolution". Those who create risks - traders and the quants supporting them - will become relatively less important; and those who manage risks will become more important - but "often quants are here too". He suggests that it would help "for quants not to enter banking careers directly." Instead, he recommends, they should take an MBA or Masters in finance "to get a sense of how the world functions, learn about capital structure and incentive problems at firms".
The availability of data and computer power are opening up a new era in quantitative finance. But let us also remind ourselves of the words of Sir Isaac Newton, who as well as being the world's most famous physicist, also lost twenty thousand pounds in the South Sea Bubble of 1720: "I can calculate the movement of the stars, but not the madness of men."