I am not so sure I see the advantage for a bank to want to buy a mortgage! Bank A has a mortgage to sell. To make a profit, they must sell it at a price over and beyond what the interest on the loan would have yielded. So, Bank B buys the mortgage for an amount GREATER THAN can ever be collected from the borrower. So, where’s the incentive for Bank B to buy the mortgage in the first place? It can’t be the hopes that oen will default on the loan for banks claim they are not in the real estate business.
And, if it doesn’t work this way, how does Bank A make a profit? Just by saving on the maintenance (i.e., paperwork) of each loan? Surely, that alone cannot make it worth their while!
I don’t know, but one factor has to be the uncertainty: there’s a certain chance that the mortgage will default, and the holder of the mortgage will have to foreclose. That entails additional expenses, and risks that the sale of the property will not cover the balance of the loan. Some banks might be able to manage that risk better than others; or evaluate the risk differently, which means that they’d find the loan more valuable.
If I owe 150,000 plus interest on my house, the bank is going to get that money, if ever, over the next 30 years. Bank B might give them 160,000 today. Bank A makes 10,000 on the loan today and Bank B takes on the risk and gets the reward.
Also, keep in mind, the loans typically get bundled. So the bank doesn’t good to Fanny Mae and offer them just your loan. They offer them a bundle of loans worth, say, 100 million dollars, comprised of 500 loans. This makes the risk much less if a few of them go bad.
After that, I think the loans end up getting traded like commodities.
Bank A gets the closing costs and doesn’t have to service the loan for 30 years. Bank B has the infrastructure to service the loan and gets the interest.
Note that banks have limits (internal and external) on the amount of credit risk they are willing to carry on their books. For instance, in order to mitigate their risk, they’re required to set aside capital to cover losses due to defaults and that’s a real restriction to the amount and type of deals they can have on the books. It is not uncommon for a bank’s executive committee to say “our risk appetite has changed, we need to free up capital for less risky deals”, which would involve selling off things like mortgages or CDOs or whatever.
But from a more philosophical standpoint, you’re asking “Why would person A ever sell something to person B? Shouldn’t that thing be worth just as much to A as it is to B?” Well, maybe A has too many of that one thing and wants to reduce his collection. Or maybe A knows something about the object in question that B doesn’t. Or maybe A needs the money to buy something he values more.
To make a profit the bank only has to sell the mortgage for more money than they loaned out + costs to make the loan + plus whatever holding costs they have in between when they make the loan and sell the loan. The holding cost portion (interest they have to pay to get the money) is essentially 0 right now, since they pay depositors virtually zero, and they pay the fed almost 0 interest if they borrow there too.
The biggest reason it is so profitable is that the government is involved; with Fannie Mae and Freddie Mac. They basically guarantee to buy all loans that fit certain guidelines (“conforming” loans). Failing that you can make a profit by pooling the loans and slicing up the risk so that you can sell low risk pieces to people who don’t want risk, and high risk pieces to people who do, thus increasing the amount of potential buyers and thus the price (CDOs). During the boom CDOs were even more profitable because the ratings agencies essentially said more of the package was low risk than actually was, which means you got more money for it.