One reason why banks and other financial institutions committed so much to trading derivatives was because they had models that told them that their investments were sound. This all turned out to be false.
The attached article is a challenge, but illustrates the problem of assumptions and abstracting. The Black-Scholes equation was the first equation which valued investment options in a seemingly reliable way. It worked for a long time and encouraged the use of similar mathematical models that valued more complex assets in what were essentially gambles.
Regardless of what we may think of banks gambling with their clients money, the way these valuation models worked is fascinating. The bankers made assumptions on volatility and various other variables and the models became more ambitious. It turned out that the models did not survive reality with disastrous results. Yet it was quite clear that the models could never survive reality to anyone outside the financial sector.
The article is excellent reading for all those interested in financial economics and doing economics at university.
