I'm linking you to the 1st in a series about the evolution banking and curious (here) --- the business of banks being only one of the new intermediaries in funding.  It ought to be enlightening considering the means by which the current crises developed without anybody being in charge or knowing the extent of the financial risks taken in the composite.... globally, I might add.

I'm curious about the whole intermediation chain and when or how it is curtailed or if there's even an end-point in the final analysis.  

That said though, my ultimate take at the moment is that the greater the chain of intermediation the more the risks being taken are absorbed by those who have the least ability to ascertain or dig out the real risk, much less the ability to back-out quickly if / when conditions indicate or pre-sage risk reduction is the prudent move.  In other words, it's in the interest of those taking the initial risk to lay that risk off on others who are less able to ascertain the risks taken... while the initial risk take is provided a fee for laying off their risk... hence they have no stake in having taken the risk, but are still paid as if they retained the risk. 

In other words, laying off the risk requires suckers to pick it up without knowing the real risk nor having the ability to ascertain it without going through services that in fact are benefiting by those risks being layed off in the 1st place.  It's fundamentally a less obvious form of buyer-beware, and because it isn't regulated as a bank or other financial institutions are regulated, it's a thinly veiled means of undertaking near (or absolute) pure laissez-faire financial dealing to avoid gov't regulation (which incidentally seeks only to insure and guarantee a level, fully visible, and informed playing field to protect the economy overall and the little guy (i.e. taxpayers and those providing the funds ... savings...in general).   

Here's another question ... if it's in a banks interest to provide a loan at what they deduce is the required risk premium (interest rate and terms), then what's the incentive they have for laying off that risks?  Presumably it would be to free up funds to make more loans... but this requires that they take an additional fee (or an effective mark-up in interest rate charged to the loan) when laying off the loan, otherwise there's no gain for the bank to lay off the loan.  

The other reason I can think of is that the bank decides the risk they deduced and hence risk premium they charged for the loan is insufficient... i.e. because they figured out their deduced risk was too low or because situations changed increasing their risks.... so they want to back out of it by reselling (laying off) that risk to others.  In the latter case they would have to disclose (fairly) the higher risk so that the bank would have to absorb the difference in risk premiums but without any assets retained to compensate .. thus losing money on the transaction to lay off the risk they didn't want to keep. The upshot is that in a fair market with full disclosure the bank would lose money on risks they were laying off because they decided the risk premium they'd initially charged was too low (for whatever reasons is immaterial).... so to minimize their losses on this type of laying off of risks, their incentive is to not disclose the full nature of the new risk (over and above that for which they initially set the interest rate & terms), leaving the purchaser of the risk to discover or figure it out for themselves.  .. another way to pull this off is to have a paid cohort tell the purchaser the risk level --- i.e. such paid cohorts being what we call Moody's or S&P ratings agencies.  Its not unknown though that ratings agencies have a huge conflict of interest... it's the banks and loan providers (those with the funds) that are paying them to tell the prospective purchaser of layed off risks what the risk level is... and if the risk level is set too high then the bank loses more to lay off the risk (so should retain it rather than sell it off), and if the risk level is set too low then the purchaser is absorbing more risk than they know they're purchasing.  So guess which way the rating agencies are going to go?   If one agency set's the level of risk too high for the bank's liking, then they simply go to another agency or threaten to... meaning that the rating agencies have more incentive to retain the bank's business than to see them take their business elsewhere, keeping a lid on the upper limit of risk they derive from the information they have available from what the bank provides them.  

The other problem with the paid cohorts is that they generally don't have all the information the bank has to make their own risk assessment of what the bank is laying off, or the complexity of the risk involved is beyond the rating agencies ability to properly assess at prices and in time-frames the banks are willing to pay for their cohorts assigning a risk level.  The same is true for corporate bond sales.  It's actually quite amazing to me that this has gone on so long already and without having been brought under any control.  What that tells me though is that the power and control that the capital owners association has over the industry and gov't has effectively kept this little charade alive and well despite it's glaringly obvious conflict of interest... it also tells me that the name of the game is really caveat-emptor to whatever level you can get away with.  It's a highly visible, but unregulated form of scratch-back theory --- you scratch my back and I'll scratch yours.   

The more transparent the system, the greater awareness of just how much our financial markets and business's rely on taking advantage of those less knowing or who are naive.... under the guise of "full disclosure". ... hence it makes sense that these things remain, as much as possible, less transparent... hence less regulation by gov't oversight and rules enforcement or investigative agencies.  

I don't know the ultimate answer to these issues, but I do know that a large segment of the population of human's serves to perpetuate it.  Therefore since the human incentive that perpetuates it is pecuniary advantage, the only dis-incentive would have to be a far greater pecuniary loss than that which is stood to be gained by being less than fully transparent.  Thus, rather than wrist-slap fines, perhaps simply having the gov't temporarily take over financial and business institutions that fail to meet that standard (when found out) is appropriate... putting the executives in jail and seizing all their (the executives') assets (both foreign and domestic)... i.e. losing it all and all control, plus imprisonment, might give one pause (and the pause of the players that go along) where today the simple slap on the wrist fines does nothing to curtail the practices at all. .. it's simply an acceptable cost of doing business.