var gaJsHost = (("https:" == document.location.protocol) ? "https://ssl." : "http://www.");
document.write(unescape("%3Cscript src='" + gaJsHost + "google-analytics.com/ga.js' type='text/javascript'%3E%3C/script%3E"));
try {
var pageTracker = _gat._getTracker("UA-857684-6");
pageTracker._trackPageview();
} catch(err) {}
Actuarial sciences are primarily mathematical and depend on past data. Since there was no past data the risk of insuring the mortgage-backed securities was underestimated. AIG insured banks for the securities they bought and these debt swaps were an unknown collection of mortgages- some of whom were simply bad loans.
It is not yet clear as to what happened to the human factor in all of this. Maybe the insurance piece was handled from London and there was no real understanding of what was happening on the ground in the US as AIG went on taking on balooning risks. For example, several years ago someone mentioned about a real estate agent he knew who just collected her 6% commission on a home sale being totally uncertain as to how the buyer was going to pay the mortgage. This kind of information was available and discussed accross communities in America as the AIG crisis was building up.
The buyer,seller, real estate agents, real estate closing lawyer, the bank's lawyer in every now defaulting loan closing day would have had a sense of the dubious nature of the loan. Why did not the risk guys at AIG pick it up? Or did they -and nobody listened?
Banks know that there is risk and that is why they buy insurance and that is why AIG is paying up and US tax payers have to foot the bill of this now enormous risk.
Going forward, business will have to re-examine their methods of gathering not just past data but new data that is readily available but in this case had not been built into the actuarial models. Raw numbers and technology cannot replace the human factor where human, qualitative real time input must be sought from the ground to get a better sense of the real risks.