3 ms·
No, it buys them from the banks and the banks know this going into the loans.
by Workaccount2 3y ago
No, it buys them from the banks and the banks know this going into the loans.
- JumpCrisscross 3y ago> it buys them from the banks and the banks know this going into the loans This isn't how the mortgage market works. Qualifying mortgages are guaranteed by Fannie and Freddie. The Fed's participation in the mortgage market has been for liquidity, not credit, purposes. Kill Fannie & Freddie and the 30-year mortgage goes away. Ban the Fed from buying mortgage securities and rates go up a bit.
- pandaman 3y ago>Kill Fannie & Freddie and the 30-year mortgage goes away. If this had been true, then 30 year jumbo loans would not exist (F&F cannot buy non-conforming mortgages).
- JumpCrisscross 3y ago> then 30 year jumbo loans would not exist (F&F cannot buy non-conforming mortgages) Fair enough, the 30-year mortgage wouldn't exist for the average American. I don't believe the Fed buys mortgage securities containing them either.
- pandaman 3y agoI don't follow. Jumbo loans exist now even without support from F&F, why can't smaller loans exist without F&F? F&F are not some charities, they make profit on their mortgage business. If they did not exist what would have prevented another firm taking over the same business (securitizing mortgages)? I mean, there are plenty non-conforming loans, which are not jumbo, written right now. It's just hard to compete with the F&F on the conforming loan market so nobody does that otherwise there is nothing magical about conforming loans.
- JumpCrisscross 3y ago> If they did not exist what would have prevented another firm taking over the same business (securitizing mortgages)? It doesn’t work without the implicit guarantee. If you run a neutral pricing model, you’d get a rate roughly double where they’re priced now. The only solution is to let the rate periodically reset.
- pandaman 3y agoWhatever model you run is most probably wrong because jumbo loans are not priced double and actually are priced pretty close to the conforming loans.
- JumpCrisscross 3y ago> Whatever model you run is most probably wrong because jumbo loans are not priced double and actually are priced pretty close to the conforming loans Modelling non-jumbo loans provided without support. I have a jumbo mortgage. I also had substantial assets when I took it out, substantial income and opted to put 25% down. Remove those factors and your credit component starts interacting with duration in complex ways. Remove that part of the market--the massive number of guaranteed, conforming mortgages--and the securitisation and hedging infrastructure that supports jumbo fixed 30-year loans falls apart [1]. [1] https://www.tandfonline.com/doi/full/10.1080/15214842.2020.1757357 https://www.tandfonline.com/doi/full/10.1080/15214842.2020.1...
- pandaman 3y agoIt's great that you have substantial assets but neither that, nor the article "model" anything. Also you are now changing your argument from "remove F&F" to "remove securitization and hedging infrastructure". As I already pointed out, if F&F did not exist, another firm could have done exactly the same securitization, as many do on non-government controlled markets right now.
- 3y ago
- dragonwriter 3y ago> Kill Fannie & Freddie and the 30-year mortgage goes away. Probably not; without them, the norm wouldn't have been created, but while rates for them may change those going away won't remove the expectation and, given the expectation, the market will find a rate af which it can fill it. Mortgages in general will be less attractive, though.
- JumpCrisscross 3y ago> while rates for them may change those going away won't remove the expectation and, given the expectation, the market will find a rate af which it can fill it Fixed-rate 30-year mortgages don’t exist, except for the very rich, in most of the world. The unsubsidised price of the instrument likely collides with popular conceptions of usury.