There are risks. You have to think about them.

Economists have predicted 10 out of the last 5 recessions. Jokes aside, economic prediction is beset with false alarms and missed warnings these days, so why should we get excited about the author of the 2006 business best seller "The Crash is Coming", Max Otte?

Professor Otte, Worms Technical University, Germany (pictured, middle) and Kanwardeep Ahluwalia, Swiss Re (right) led the discussion at this evening's Swiss Re's Centre for Global Dialogue Risk Talk on forecasting the next crisis.

Otte seems to have been spot on with his 2005 prediction of a 'financial tsunami in 2008 and certainly before 2010'. But why should we have believed him at the time?

His reasoning was based on 'very simple measures' of total debt versus gross domestic product in the US, and the fact that indexes like the Dow were rising much faster than measures of real wealth.

But the simplicity of Otte's analysis masks what is perhaps a deeper level to his thinking. After all, relatively few people predicted this financial crisis. His ideas met with a significant headwind at the time. And his ideas where not the product of complex risk models. Quite the opposite, in fact.

"You can't simply put a number on risks,' Otte explained during the discussion at the end of the presentation this evening. Earlier he had expressed a certain cynicism towards the goal of precisely quantifying risk. In a jibe towards the widely used risk model Value at Risk, Otte questioned how any organization could rely so strongly on the assumption that past volatility alone could predict a company's future risk exposure.

Physicist Dr Ahluwalia shared some interesting insights from his final days at investment bank Bear Stearns. Luck or otherwise, Ahluwalia resigned exactly one day before the company's 'fatal liquidity problem' in March 2008.

"People were not confident to enforce the risk measures that were already in place," explained Ahluwalia. Here again we see that the problem is not simply a flaw in financial risk models. The problem concerns the limits to the very application of these models to risk management.

And that is what I think distinguishes Professor Otte from the crowd. "One needs to go back to first principles and ask when will the model itself break?"

"There are risks. You have to think about them," he said.