One of the most iconic scenes from the movie was the one wherein the Ryan Gosling character (named Jared Vennet in the movie but actually based on real-life Deutsche Banker Greg Lippmann) used a Jenga Tower to illustrate the vulnerability of Subprime Mortgage-Backed Securities.
The scene no doubt provides a creative visual analogy, but one that unfortunately is bound to have left those wanting to have a deeper understanding of mortgage-backed securities with a nagging feeling of inadequacy. For instance, while correctly stating that mortgage-backed securities consisted of "tranches" of mortgages, it stops short of explaining why there is need for such tranches to begin with. I'll try to fill that gap to the extent that I am able -
It was sometime in the late 1970s when bankers at Wall Street tried to make home mortgages into standardized bonds that they could sell to big institutional investors like pension funds, insurance companies, mutual funds, and the like... Precisely because these institutional investors are the big boys, they of course do not want to get into the more messy business of extending home loans to individuals. Note that it wasn't that they weren't willing to loan their investible funds to homeowners, but they did not want to have to deal with homeowners individually. To fulfill its role as a market maker, Wall Street needed to come up with a security that would enable the institutional investors to make loans to groups of homeowners. This seems simple enough, just group mortgages together into standard pools (use a Jenga tower to visualize). There remained a problem though. Unlike say, a corporate bond wherein the entire bond is issued by a single company, pooling individual mortgages to form bonds meant that behind each bond were groups of diverse individuals whose lives and decisions were separate and distinct from each other. This further means that the buyer of a mortgage bond would be subject to more uncertainty vis-a-vis other types of bonds. Imagine for instance, a pension fund buying a mortgage bond thinking it had locked in an investment at x% over y number of years, only to wake up next month to realize that Jack over at 100 Elm Street had pre-terminated his mortgage and that thus, some part of their investment had been returned in cash. It would wreak havoc on financial planning. To solve this problem, at least partially, the bankers separated each bond into tranches. Some tranches would pay more interest, but would be hit with prepayments and defaults first (the BBB / BB / B tranches), while some tranches would pay less interest but would be the last to be hit with prepayments and defaults (the AAA / AA / A tranches). An investor could choose which tranche he wanted to invest in depending on his appetite for yield vs. uncertainty. While this separation into tranches is not a perfect solution to the problem of diverse individuals behind a bond acting differently from each other, it at least provides a partial solution - an investor who values certainty for instance, could invest in the AAA tranche knowing that before the consequences of individual homeowner circumstances reach him, these would have to go through all the other tranches first.
When looking at the Gosling/Vennet's tower of mortgages, it is absolutely critical to understand that the AAA's are different from the B's not because the underlying mortgages are somewhat better, but primarily because the tranche is more senior (i.e.. gets paid first in the event of defaults) than the B's. Without this understanding, the more inquisitive viewer will be left to wonder why there is a need to mix different quality mortgages into a single bond to begin with. Why not just put them in separate towers to begin with? It is not that different quality mortgages were mixed together - the mortgages were all of roughly the same quality, but what made an AAA an AAA is that the tranche would be the least susceptible to the uncertainties of prepayments and defaults. What made a B a B was that it was the most vulnerable. Again, the need for tranches is a characteristic specific to mortgage securities because each bond is effectively issued by groups of individual homeowners, each of whom has a different financial story, with different consequences for the debt that he owes.
So, when Gosling/Vennet (or more accurately, the scriptwriter writing his lines) says that "somewhere along the line, these B's and BB's went from 'a little risky' to 'dogshit'!", it is not 100% accurate. I think it would be a lot more accurate to say "somewhere along the line, the whole tower of mortgages went from 'a little risky' to 'dogshit', and these B's and BB's, as the most junior tranches in the tower, are most vulnerable to the fact that the whole tower has turned into dogshit." There, no wonder I'm not employed as a scriptwriter; but if you're here to try to clarify the concepts, there you go..
While this is meant to be neither a review nor a critique of the film, I'd really like to say that it covers a lot of ground in that it then goes on to explain, creatively and accurately, why the crisis blew totally out of proportion. Synthetic CDOs, which allowed speculators to bet on the tower of mortgages many times over (as illustrated by Selena Gomez's blackjack hand), could probably be thought of as the gasoline that turned a fire into a firestorm.
Finally, yes, the most interesting question is why the original tower of mortgages turned into dogshit. To that question I think different people will have different answers, The film certainly presents one thesis, which is Michael Lewis' thesis. Again because of the difficulties of book-to-film adaptation, the film presents a somewhat diluted version the Lewis thesis. For a more forceful presentation of the Lewis thesis one has to, quite naturally, read his book. There are those, of course, who would as forcefully reject Lewis' thesis. The Wall Street Investment Banks and the Rating Agencies come to mind.
In the end, whatever the accurate explanation of the causes of the 2008 crisis, it further convinced me of two unassailable (for me at least) truths about human behavior - 1) we'll all respond to the most immediate incentives, and 2) we will build a house right on top of a fault-line if we haven't had an earthquake in 10 years.