I don't know if you know what you are talking about. Financial Engineering is a great tool to diversify risk and to provide flow of liquidity. Like all tools, they could be used (and abused) for good or bad. Learn more about it and you will appreciate their usefulness.
To estimate someone's job value for the society, we should consider the case when this someone fails at what he is doing, and see how much negative impact this has on the society.
...so, what if those bunch of guys I see across the trading floor (exotic rates traders) will fail miserably tomorrow, what will happen? ....well, the first order damage will be
to those same traders....which is under a dozen of people.
So what's the impact for them? Well, instead of driving maseratis some of them will have to drive BMWs, and instead of going with their families to a top notch ski resort in Switzerland they will go ski in Canada.
second order damage will be for people in infrastructure, but the damage for them will be minimal, cause they don't have direct impact on P&L, and their bonuses are not affected much by the bunch of gamblers upstairs.
...now, what about an average person on the street? Is those traders job productive enough for an average person on the street to feel the impact? How those traders' failure will affect Joe's standard of living?
Now, let's look at the complete opposite situation, that is: what if that Investment Bank does a terrific job and is enjoying big profits and bonuses. Now, about about our Joe? is he thriving as well? Well, as we all can see, today's unemployment reports are not as spectacular as bonuses at GS.
....all this "liquidity", "diversification", and "money allocation" bollocks has not much to do with what the most of the modern investment banking is about. ....you can apply "money allocation" argument to, say, venture capital business...but exotic options, high frequency trading??? Please....this is not a tiny bit more useful than the gambling business.