Collateralized Debt Obligations (CDOs)
aka financial weapons of mass destruction, according to Buffett
If you thought that stuff like mortgage-backed securities were complex enough, wait until you get a load of what we have in store for you today: collateralized debt obligations, or CDOs, are derivatives backed by a pool of loans which can then be sold to investors on the secondary market. In our article on mortgage-backed securities, we described how investment banks purchase a bunch of mortgages, turn them into a single corporation, and then sell shares of that corporation.
Here’s the problem with that strategy as described in our article: those shares are only going to be able to attract investors who’s risk tolerance matches up with the risk posed by purchasing the stock. In other words, investors who are looking for either safer places to park their money or investment vehicles that offer higher return (pun intended) are ruled out. From the investment bank’s perspective, this is a far less than ideal situation - why shouldn’t they be able to tailor to investors with all sorts of risk tolerances?
This is where CDOs come in handy. Instead of structuring the mortgages into a giant entity and selling shares, why not organize them into classes, or as it’s more commonly called in finance, tranches. There are typically 3 of them in your average CDO:
Level 1 (Senior): safest one, gets paid first, has the lowest yield
Level 2 (Mezzanine): a little riskier, offers moderate returns, gets paid only after the senior tranche owners have been
Level 3 (Equity): the riskiest of the bunch, highest returns but these ones are paid only after the first 2 have been and, because we’re dealing with mortgages in this case, this tranche takes all the hits from defaults
By dividing the mortgages in accordance with how risky they are, a more diverse range of investors can get in on the action. If you’re just looking for an investment that you don’t really want to have to worry about, well then look no further than some senior tranches. Interested in some more high-yield assets? Equity tranches are just for you.
Let’s turn to the example from our last article to explain how this works in further detail. Just to quickly refresh you on the numbers, we have 1000 people each borrowing $1000/month 30-year fixed rate loans at 4% interest, meaning a total of $360 million has been lent. Every year, the bank receives $14.4 million in interest fees. If we were to split this into tranches:
Level 1 (Senior): $140 million, 3% return, $4.4 million in returns (3%), 100k shares at $1400 each
Level 2 (Mezzanine): $110 million, $4.7 million in returns (4%), 100k shares at $1100 each
Level 3 (Equity): $110 million, $5.3 million in returns (5%), 100k shares at $1100 each
This is how the payments are structured for the various tranches, with, as we described, the equity tranches being paid the most and the senior tranches the least assuming no defaults. That last condition is particularly important, because say that 5% of borrowers default and the bank is able to recover half of that money by selling the houses, thus totalling to a net loss of $9 million (we go over these calculations in further detail in our mortgage-backed securities article). Guess who’s going to incur that loss? That’s right, the equity tranche holders - they now collectively own $101 million ($110 million - 9 million) and are making $5.05 million in returns (0.05 * $101 million). And if mortgage borrowers keep defaulting, more and more money is going to be pulled out of the equity tranche. For the mezzanine class to even be touched, a whopping 62% of the mortgage borrowers would have to default:
Again, a lot of this really depends on a given investor’s risk tolerance; because the senior tranches are paid first and thus most insulated from the defaults, they’re the safest but yield the least amount of money overtime if little or no borrowers default. The equity tranches, on the other hand, are meat shields once defaults start racking up, but its owners get the most money if all goes well.
Do keep in mind that CDOs are in no way limited to just mortgages - you can make one with basically any kind of security that’s backed by assets, such as corporate debt, bonds, or other CDOs. Yup, you read that right, you can have CDOs made up of CDOs. These are called CDO squared: they work the same way as mortgage-backed CDOs except with tranches of some other CDO instead of mortgages being sold. Banks take a pool of CDOs and structure them into tranches according to their risk and maturity, and then sell shares to other investors. The main reason banks in particular would want to do this is to make up for the credit risk, that is, the risk of their mortgage borrowers defaulting, by essentially selling it off.
The complexity, however, doesn’t stop there. In addition to squared CDOs, there’s synthetic CDOs. Put simply, these are bets on the performance of the underlying mortgage of an actual CDO. This doesn’t make them derivatives, since their value isn’t derived from mortgages or any other asset but rather the premiums offered on a specific synthetic CDO. Exactly how it works is a bit more intricate and requires an understanding of credit default swaps, which we’ll go over next time.
Econ IRL:
As the old saying goes, the 2 absolutes of life are death and taxes, neither of which people are big fans of. Focusing on the latter, one can really only wonder exactly how they’re taxes are being used by the government. The author’s of this week’s paper probably had similar thoughts in mind. Their research question was simple: do impressions of how governments spend taxes influence people’s willingness to pay those taxes? To answer this, the authors design an experiment that measures how 2,100 homeowners feel about paying taxes after being informed that they’re being funded towards school vs being redistributed to aid disadvantaged neighborhoods.
Aside from learning that a lot of people have an inaccurate interpretation of where taxes go despite most of that information being publicly available, the results are consistent with the principle of benefits-based taxation, in which an individual ultimately benefits from having to help fund public goods. When they learn that a higher share of property taxes are used to fund schools, households with kids who go to public schools are found to be more relaxed regarding how their taxes are spent. On the other hand, households without kids tend to be more stingy on that matter.
‘Till next time,
SoBasically