Most of academic research in QF is devoted to derivatives. My question is: why? What is the benefit to mankind? Clearly instruments for minimizing risk are very valuable, but at the same time complexity of products increases risk. I don't know how to interpret the appended graph, but nowadays it's clear that financial innovations are often not what they appear and that the ability of market participants to discount information is (very) limited.Source

it doesn't make much sense to me. An analogy would be forex market where speculative trading accounts for 90 or 95% of all currency exchange trades. If you graph utilitarian trades (i.e. the ones that exchange currency in order to purchase goods in foreing country or return profits from sold goods abroad) versus all trades you would get the same dramatic picture. If you take this ratio over the period of 20 years then the same graph as the one shown above. So what?If people found a way to count the same money twice (as most accountants have done for centuries with double record) it doesn't lead to a disaster, it doesn't mean anything.More meaningful picture would be if you show the total amount of underlying that derivatives hedge against versus total existing underlying then, probably, one could draw some conclusions.

Well, if you care about economic welfare it should be of interest. I don't think that speculation accounts for 90-95% in currency markets. The question is how one defines and measures speculation. I am talking about two different things: trading of derivatives and academic research on finance. The question is, why in the world, QF research spends 99% of it's time on pricing derivatives. That makes no sense at all, certainly from the standpoint of academia, perhaps not so much from the standpoint of business. Perhaps the reason is that people love their toy models and their fancy maths too much.The real question should be why and how markets are efficient and more importantly in what ways. It could be argued that financial engineering doesn't contribute to the functioning of markets. All the inventions made profits for Wall Street and now the whole thing imploded. On the one hand it is just the usual cycle of excess and contraction, on the other hand if academic research was independent such things wouldn't have happened, at least to some extent.

I read somewhere lately that at the end of 2006, the S&P 500 was valued at $12 trillion, futures and options contracts on the index was at $20 trillion, giving the expectations market a value of almost double the real market. Therefore, the $32 trillion stake in the S&P 500 was equal to 250 percent of GDP. The world's GDP is about $60 trillion, and aggregate financial derivatives are estimated at $600 trillion, with $45 trillion in CDS alone. As you said below, the question is how to define and measure speculation, but I don't think there is any doubt that speculation is dangerously rampant in current markets.

I got your point. Yes, it is difficult to define "speculation", but one way to estimate it to use inter-dealer trades versus all trades. That what is estimated as 90-95% of all trading in forex. So, it would be more informative to use numbers like total/average open interest to compare with world wealth rather than simply "derivatives vs...."But I understand that it is a bigger picture here. Indeed, I am often puzzled how derivatives pricing is disconnected from efficient resource allocation economical function.It happens throughout the history that in growth phase people think more about models and numbers and in contraction phase everybody suddenly remembers about real values.But I don't know how independent academia can possibly be. It is quite disconnected from reality already ... Perhaps, academic research should be made less independent? Regulators need to be independent.