Silicon valley types are the devil to the unwashed San Francisco hippies. As an investment banker you also provide a service. When Google buys a startup for $50 million, that transaction doesn't make itself happen.
No -- but the banks are overcompensated for their work on M&A. You could hire a strategic consulting firm -- McKinsey, Bain, BCG -- to do M&A due diligence for half of what the investments banks charge. They hire from the same pool of 25-year-old Ivy League graduates. I doubt they'd do a worse job than the banks.
IPOs? You shouldn't need an underwriter to do an IPO. That's the whole idea behind OpenIPO -- which is how Google itself went public. It's true that the investment banks have access to a larger pool of investors. But that's circular reasoning. If the investment banks didn't exist, then these capitalists would be investing in IPOs through a Dutch auction like OpenIPO.
Structured products? The investment banks earn a 50% profit margin on structured products. The people who buy structured products could double their returns by constructing their own structured products on the options and future markets. And if they're not sophisticated enough to do that without an investment bank, then they probably don't understand what they're buying!
Theoretically, the main reason to prefer traditional IPO's to dutch auctions, is the auctions in general may not give sufficient rewards for "price discovery". That is, everyone is relying on everyone else to research the firm and discover its true value. That's not to say there is NO incentive for price discovery in auctions, just not as much as through the traditional process.
In my corporate finance class they described how some countries (I forget which) had experimented with auctions instead of using i-bank underwriting, and in the end they had all gone back to using i-banks.
I don't know anything about M&A, but on structured products I agree they are just for customers who don't know any better.
