The US HOR are big girls
It's not a question of being smart. It's just that I write about this stuff for a living.
Kinda sorta. Technically, the US will own a beneficial interest in a trust which has security over the houses. To the extent that the US owns all the beneficial interests in a given trust, it will effectively own the houses. One of the problems with the plan is that there's no guarantee that the government will be able to buy up all of (or a controlling interest in) the beneficial interests, which will limit its ability to engage in foreclosure reduction and loan modifications and so on.
For the most part, the US government won't be buying defaulted mortgages (although as I understand there's nothing in the bill to stop them from doing so, and they could still make money if they bought them at the right price).
What the US will (mostly) be buying are mortgage backed securities, which as I said above represent beneficial interests in a trust with security over a pool of mortgages. What this basically means is that they will have the right to some of the cashflows from that pool of mortgages (ie payments of principal and interest), according to the terms of the notes and their position in the capital structure. If the position is a senior one (and by value the vast majority of the straight mortgage debt is senior, although the inclusion of CDOs complicates things enormously), then any losses on the underlying pool from defaults will be absorbed first by notes lower in the capital structure (this is called subordination or more generically credit enhancement). As a greatly simplified example, if you have a £100m pool of mortgages and a £90m senior bond, the pool can endure £10m of cumulative losses from defaulted loans (which will likely require mortgages with a nominal value of much more than £10m to default, given that you will recover something - 50% is the current estimate - from foreclosure) before the senior bond takes any loss at all. Now clearly there is some risk of loss for the senior bond (in practice because the senior tranches were often more than 90% and because cumulative losses are predicted to be much higher than 10%), even if it hasn't suffered any yet. Which is (among other reasons such as illiquidity and downgrades) why these bonds are marked at a discount. The question is whether the discount accurately reflects the likely ultimate losses on the bonds or incorporates other factors such as a liquidity premium which an investor like the government can legitimately ignore. That's a very, very tricky question (I just wrote a three page feature on the very subject and barely scratched the surface).
The (main) problem with the bailout plan is that its many objectives (as envisaged by different proponents) are at cross purposes. You can't recapitalise the banks and maximise taxpayer value at the same time through the pricing mechanism. The more you pay for the bonds, the more capital you provide to the banks, but the more risk there is for the taxpayer. Similarly, if you want to reduce foreclosures, you'd be better off buying the loans themselves out of the mortgage trusts rather than buying securities, but the only way you could do that without abrogating contracts would be to buy them at their face value, which would clearly cause massive losses for the taxpayer.
Personally, I think we'd be better off with a series of smaller, targeted programmes. An asset purchase scheme to clean up bank balance sheets and improve interbank liquidity. A recapitalisation scheme with the government taking equity stakes to shore up weaker banks. A loan modification or refinancing programme to reduce foreclosures in severely affected communities. But I'm not an economist, so I don't know how necessary they really are. It's clear that the bill as proposed is meant primarily to ward off a collapse of the stock market with all the political consequences that would entail.
GY is much smarter than me on this, so I'm glad he's here.
Does this mean that the US government is going to actually own all of those houses?
I too have wondered how Warren Buffet and others can say that the US government can make money on this deal if what they're buying are defaulted mortgages.
What the US will (mostly) be buying are mortgage backed securities, which as I said above represent beneficial interests in a trust with security over a pool of mortgages. What this basically means is that they will have the right to some of the cashflows from that pool of mortgages (ie payments of principal and interest), according to the terms of the notes and their position in the capital structure. If the position is a senior one (and by value the vast majority of the straight mortgage debt is senior, although the inclusion of CDOs complicates things enormously), then any losses on the underlying pool from defaults will be absorbed first by notes lower in the capital structure (this is called subordination or more generically credit enhancement). As a greatly simplified example, if you have a £100m pool of mortgages and a £90m senior bond, the pool can endure £10m of cumulative losses from defaulted loans (which will likely require mortgages with a nominal value of much more than £10m to default, given that you will recover something - 50% is the current estimate - from foreclosure) before the senior bond takes any loss at all. Now clearly there is some risk of loss for the senior bond (in practice because the senior tranches were often more than 90% and because cumulative losses are predicted to be much higher than 10%), even if it hasn't suffered any yet. Which is (among other reasons such as illiquidity and downgrades) why these bonds are marked at a discount. The question is whether the discount accurately reflects the likely ultimate losses on the bonds or incorporates other factors such as a liquidity premium which an investor like the government can legitimately ignore. That's a very, very tricky question (I just wrote a three page feature on the very subject and barely scratched the surface).
The (main) problem with the bailout plan is that its many objectives (as envisaged by different proponents) are at cross purposes. You can't recapitalise the banks and maximise taxpayer value at the same time through the pricing mechanism. The more you pay for the bonds, the more capital you provide to the banks, but the more risk there is for the taxpayer. Similarly, if you want to reduce foreclosures, you'd be better off buying the loans themselves out of the mortgage trusts rather than buying securities, but the only way you could do that without abrogating contracts would be to buy them at their face value, which would clearly cause massive losses for the taxpayer.
Personally, I think we'd be better off with a series of smaller, targeted programmes. An asset purchase scheme to clean up bank balance sheets and improve interbank liquidity. A recapitalisation scheme with the government taking equity stakes to shore up weaker banks. A loan modification or refinancing programme to reduce foreclosures in severely affected communities. But I'm not an economist, so I don't know how necessary they really are. It's clear that the bill as proposed is meant primarily to ward off a collapse of the stock market with all the political consequences that would entail.
Comment