"Much of what investment bankers do is socially worthless"

The market will quickly converge to the consensus price, and then investors will buy if they deem it under-valued or sell if they deem is over-valued. I wouldn't so much call that mis-pricing (perhaps it is, but only if the consensus is wrong), but rather the difference between an individual's expectations and the market is what allows investment. I think that's what he was getting it, but I could be wrong.

Hang on, here; are you sure that what you're saying is actually a good thing? There was a housing bubble, after all, and there is a natural problem in using an overpriced market, like the instruments did, as backup. Yes, the instruments might have, as a side-note, allowed more homeowners to be funded, but if anything, that only represents a failure in allocation - the mortgages should never actually had been made, as the foreclosures caused by the post-bubble prices revealed. Perhaps I am misunderstanding you, but I don't think a misallocation of resources is considered a social good.

I'm not sure if this has exactly to do with his thesis, but I should point out that, with bubbles, there is largely a different mentality. To paraphrase from A random walk down wall street, the mentality is that "the market may be overpriced, but it's growing, so I can make profit if I buy and sell out before the other guy does". Investors, in this case, are not looking for fundamental value, but are speculating that the bubble would not collapse before they pull out. The skewed incentive here is to buy not what is underpriced, but what is overpriced.

For this particular case, however, I think the issue has more to do with information. Let's say you knew that housing would eventually go bust, but it's very pricey now. You could then stick a partial investment of housing inside a financial instrument, significantly increasing the value of that instrument, and convince consumers that your instruments are well-cushioned because of that housing thing. The consumers don't know your stuff is going bust, because you've made the instrument excessively complicated and they don't really get the damn thing; it's your firm's reputation that is getting them to buy. This is, at least from a social perspective, a form of fraud and what the author calls "rent seeking". What's the social value in making all these complicated instruments, aside from tricking consumers? I think something like this is Cassidy's point.

Why shouldn't the mortgages have been made?...The housing market only collapsed because people were allowed to walk away from their homes/mortgage payments when they realised their homes were in negative equity.

House prices are dictated by supply/demand so naturally there is no misallocation.

Likewise, you can blame the banks but they're only made because there is demand for them from investors. Investors should know what they're doing

They shouldn't have been made because of the high risk that people would walk away after the housing bubble burst.

Only if the supply/demand accurately reflect the utility achieved by society, which, judging by the extreme collapse afterwards, is probably not so (unless people suddenly started hated houses just that much ).

In an ideal world, perhaps. In the real world, it is evidently not the case, and this gives a very real chance for banks to seek rents (the socially worthless activity we were talking about).

Yes, there was risk... but investors know there is risk.

why didn't legislation stop homeowners walking away so easily? why isn't this question about politicians being socially useless?

I'm not sure what you mean. If I have a house worth $200,000 and try to sell it for $250,000 then certainly one person might buy it (happiness premium) but on a consistent basis across the country people wouldn't buy into that.

Certain things are complex by nature. We don't stop people selling gas ovens just because they could potentially blow up if you don't know how to install and use them properly. There's demand so there ought to be supply, the markets are driven by product knowledge and info release (or lack of it).