Felix Salmon has an excellent writeup on the Treasury report on the future of housing finance released today. The Treasury report walks the reader through what went wrong in the housing markets, particularly with the GSEs. There’s a clear distinction between short-term and long-term goals in reforming the housing market.
The first goal is to wind down Fannie Mae and Freddie Mac by increasing the guarantee fee to bring in more private capital (Increasing the g-fee as a mechanism to wind-down the GSEs has been suggested by many, including John Hempton). They’ll then take steps to wind down the investment portfolios, reduce conforming loan limits and increasing the down payment. That’s a serious improvement over the status quo.
As Ezra Klein noted a lot will depend on Dodd-Frank. The CFPB will have to both create and implement stronger underwriting standards. Regulations for the ratings agencies will have to go forward and new rules for securitization, servicing, etc. will have to be written. Very important for the purposes of this blog, they propose reforming servicing compensation and establishing national standards for mortgage servicing, as well as dealing with the treatment of second/junior liens. These are much needed reforms that should have been in place years ago; it’s good to see Treasury writing that these are essential to getting the housing market started again.
Treasury puts out markers on three potential options for the housing market that would be implemented over the next 5-7 years. The first is full privatization. Felix worries, as do I, about whether or not private capital can meet the needs of the housing market with any kind of sufficient liquidity:
More generally, it’s far from clear that there’s enough private money at all which is willing to fund such a system. The main problem I have with Treasury’s report is that it simply assumes that if government support for the housing market is slowly removed, then private money will come in to take its place — at a higher price, to be sure, but at some price.
The big risk is that private money won’t come in, at any price, if there isn’t a guarantee — that the amount of private funding for the US mortgage market will be substantially lower than the demand for mortgage loans. The result would be a broken, non-clearing market, with people stuck in their homes because they can’t sell them, and the idea of a “market price” being somewhere between a purely theoretical entity and an outright joke.
That’s why my preference would be for Treasury’s third option, where the government guarantee remains extant, just with a lot more safeguards than it has right now, in a system where it’s priced rationally rather than well below market. Fannie and Freddie would go away, and be replaced by private mortgage insurers; the government would reinsure the mortgage insurers, rather than insuring the mortgages directly. And the government would only lose money after the private insurers had lost all of their money.
I think that’s all correct, but I’d go a step further. If the implication is that the private-label securitization system is going to take over the full housing economy, that worries me. Looking back on the past 10 years, I would characterize the private-label securitization system as one rife with fraud. Fraudulent originations by lenders and sometimes by borrowers. The origination of loans that had deceptively high costs, that thrived on pushing any definition of fraud up to the line and then cherry-picked the pools afterwards. We ended up with short-term mortgages designed to generate liquidity crises with perpetual resetting and that risk-shifted interest rate risk onto the homeowner’s budget. Why would we want to turn over the entire housing sector to this failed system? Treasury has references to covered bonds growing to meet this need, but that would need to be made more clear.
Perhaps the CFPB can fix the fraud issue, but that’s asking a lot. It’s worth noting that some of the biggest supporters of privatization also have eliminating the CFPB as their number one goal and are pushing stripping securitization and ratings agency reforms from Dodd-Frank. So much for worrying about fraud in the private label securitization markets.
The Treasury report makes a small reference to it (“Government support can also increase access to secondary markets for smaller lenders and community banks, promoting a more competitive market and minimizing consolidation”), but a likely result of privatization would be a more consolidated financial sector, with fewer and larger banks disconnected from communities. Maybe you like that or maybe you don’t, but smaller lenders would likely have a harder time competing as they did under the previous growth of the subprime market.
And I’d also be worried that the money that comes into a privatized market will be from hedge funds, people with 5-year expectations who will want volatility in the housing markets. Taking insurance out on your home and then setting it on fire is the hedge fund housing market play, and if it turns out they’ll be the ones capitalizing our housing markets I can see them wanting something very different than the normal, boring housing market most want to see.
Break In Case of Emergency, Hybrid, Can We Ever Credibly Not Bailout?
The second option is what we’ll call the Last Resort, Break in Case of Emergency option. Here there’s a government-option priced far away from market rates that can start to lend mortgages in the time of a crisis, one that would be competitive in times of crisis where private capital disappears. I’m not sure if it is a good idea to have a backstop we aren’t charging someone for. Having that backstop, making sure liquidity is provided in the middle of a crisis, is still a bailout in the sense that it protects private market agents and we aren’t asking for anything in return.
And this gets to my my problem post-TARP: the backstop is there. I’m not sure we can ever really say with certainty that the government won’t jump in and save the financial and housing markets. So why not just go ahead and charge them for it?
The third option which would go that route, similar to ones Center for American Progress, several academics and financial interests have put out, is the hybrid model. Conceptually it can be thought of as similar to the FDIC model. Through one lens, FDIC insurance is a corporatist subsidy, where the government takes on tail risk in exchange for a fee that isn’t always correctly priced. Through another, it’s a way of providing a backstop to prevent panics, that standardizes a market that can help block out fraudulent elements (a vanilla option?) and that puts losses onto the institutions first and winds them down when they fail. Banks that are taken over from FDIC get can fraud convictions easier if needed; private-market labels, not so much.
FDIC has had its troubles over the years, and I’m not sold on the hybrid model. But is the alternative simply hoping that the private financial market cleans up its act and the government double-seriously promises to not step in at the last minute again?