Today, I resume my analysis of the Financial Stability Improvement legislation draft (.pdf) released on Tuesday. I wrapped up the "Financial Stability Improvement Act of 2009" portion of this yesterday, after considering the new oversight council, how systemically risky firms will be chosen, what heightened regulation will look like, capital requirements and a future emergency provision. Today, I'll probably just tackle two more sections -- securitization and the resolution authority. Then I'll summarize.
The securitization provision is incredible. If enacted as is it may be the beginning of the end of securitization.
Congress calls this portion the "Credit Risk Retention Act of 2009." It essentially answers the call for lenders and investment banks to have some "skin in the game" when it comes to securitization.
First, let's say you're a lender who originates mortgages. If you want to hire an investment bank to turn a pool of mortgages into bonds to sell as mortgage-backed securities, then you must now retain some portion of the risk exposure to that pool. This will have a few outcomes.
The logistics to this are somewhat problematic, because they could inhibit innovation in bond structure. I know financial innovation has become a derogatory term these days, but if securitization is done properly, then such creativity can have huge benefits. One of the great things about securitization is that you can create bonds with different risk and return characteristics for different kinds of investors who have varying risk appetites. You may still be able to still do that, but it could get logistically messy to have the bank retain some precise percentage of the risk from each of the kinds of bonds sold.
This provision also necessarily limits a lender's origination volume and makes securitization more expensive. For example, let's say you're a mortgage company with $1 billion of funding. You can use that to create mortgages, so you do. Before, you could sell that billion in loans to investors through securitization. Now you must hold 10% of the risk, which presumably means you will need capital to cushion for that potential loss. Let's say that takes $3 million -- which implies a presumed loss rate of 3%.
First, this makes securitization more expensive, as you must hold additional capital. Your return will be lower. Second, since you have hold more capital, it limits the amount you can lend each time subsequent to a securitization. Next time you will have to subtract that capital from the amount of money you have to lend. Limiting credit might not be all bad, but it is something to note.
But this can also applies to "securitizers," which I take to mean investment banks that perform the securitization. It applies to them if there is no lender to take on the risk. I suspect that Congress had more esoteric asset-backed securities like collateralized debt obligations in mind when it decided to include securitizers. In this case, the investment bank originating the bonds themselves must be exposed to some portion of the risk. This will debilitate the markets for these kinds of bonds, because investment banks hardly want to have to hold additional capital for the deals they structure. That would wipe out the some or all of the fees they get for the transaction.