Jim Parrott: Clarifying the Choices in Housing Finance Reform

The housing finance reform debate has often foundered under the weight of its complexity. Not only is it a complicated topic, both in its substance and its politics, but the way that we talk about it makes the issues involved indecipherable to all but a few. Each proponent brings a different nomenclature, a different frame of reference, often an entirely different language, making it enormously difficult to sort through where there is agreement and where there is not.

As a case in point, three prominent proposals for reform have been put on the table in recent months: one offered by Lew Ranieri, Gene Sperling, Mark Zandi, Barry Zigas, and me (Promising Road Proposal); one offered by Ed DeMarco and Michael Bright (Milken Proposal); and one offered by the Mortgage Bankers Association (MBA Proposal). These proposals have been discussed and debated in many forums, each assessed for its respective merits, risks, and likelihood of passage in Congress, but each largely in isolation from one another. That is, they are not compared in any intelligible way, forcing those hoping to come to an informed view to choose among what appear to be entirely different visions of reform, without any easy way to make sense of the choice.

In this brief essay, I thus bring these three proposals together into a single framework, making it clearer what they share and where they differ. Once the explanatory fog is lifted, one can see that they actually share a great deal and that deciding among them is not prohibitively complex, but a matter of assessing two or three key differences.

Figure 1 shows how the system supported by Fannie Mae and Freddie Mac (the government-sponsored enterprises, or GSEs) looked before conservatorship, which is a helpful place to start.

Figure 1GSE Reform Figure 1

The problem with this system was its dependence on a privately owned duopoly. Fannie and Freddie handled almost all of the securitization in this segment of the market and took almost all of its credit risk. Given our reliance on them, everyone in the market knew that we would bail them out if they ever stumbled. Their shareholders were thus incented to take excessive risk to chase greater profits, knowing that if their bets did not pay off, the taxpayer would step in to cover them.

And that, of course, is precisely what happened.

Coming out of the crisis, there was thus almost universal agreement that we should reduce our reliance on this duopoly and shift more risk into a competitive private market.

And that is what policymakers have done in the years since. In figure 2, you can see that the Federal Housing Finance Agency (FHFA) is taking steps to pull both the securitization infrastructure and much of the credit risk out of the GSEs, with the former going into a common securitization platform (CSP) and the latter being dispersed to credit risk investors through credit risk transfers (CRT). Once these processes are completed, the housing finance system will be much less reliant on the duopoly, reducing the risk that they pose to the system.

Figure 2GSE Reform Figure 2

Helpful as this transition will be, there is a relatively broad consensus that it is not enough. The system needs to be pulled out of the limbo of conservatorship, with the roles of the private market, the government, and the taxpayer better defined.

There are a great many ideas out there for how to do this, ranging from simply reprivatizing the enterprises with a modest increase in their oversight to winding them down without replacing them with any form of government support at all. Despite the range of views, however, there is a strong majority view that the best course is to further wean the system off of its dependence on Fannie and Freddie and more explicitly limit the role of the government to providing a catastrophic backstop and managing the secondary market infrastructure.

The three proposals mentioned are the dominant versions of that majority view and are thus worthy of some comparison.

Each begins from a common set of principles that drive their design:

  • To attract the interest rate investors critical to broad access to long-term, fixed-rate lending, the government should remove the catastrophic credit risk from securities backed by qualifying mortgages.
  • To minimize the chances that taxpayers are ever forced to bail this system out again, their risk should be insulated behind significant private capital from institutions whose viability the system does not depend on and a mortgage insurance fund paid for through mortgage premiums.
  • To ensure systemic stability and broad lender access to the secondary market, a government corporation should operate the securitization infrastructure. 
  • To minimize disruption to the market, as much of the current system’s infrastructure and processes should be used as possible.


To design a system consistent with those principles, the MBA builds off of the transition already under way. It mandates that Fannie and Freddie continue to share most of the non-catastrophic credit risk on the mortgages they guarantee. It checks their inclination to take on excessive risk by turning them into privately owned utilities with regulated rates of return and opening them up to the threat of competition from newly chartered guarantors. And it turns the CSP into a government corporation that issues the securities of Fannie, Freddie, and any other chartered guarantors and guarantees the interest rate investors the timely payment of principal and interest on their investments.

In figure 3, you can see how the system to which we are already migrating with the CSP and CRT is taken one step further in the MBA’s proposal.

Figure 3GSE Reform Figure 3

In the proposal that my coauthors and I have offered, the path of reform imagined by the MBA is taken another step. The CSP, Fannie, and Freddie are combined into a single government corporation that issues government-backed securities, sells off all the non-catastrophic credit risk on those securities, and guarantees interest rate investors the timely payment of principal and interest on their investments.

To see how our proposal is simply a step further down the path laid out by the MBA, take a look at figure 4 and then look back at figure 3.

Figure 4GSE Reform Figure 4

Bright and DeMarco choose a slightly different path to meet the common principles outlined above. Rather than build on the CSP and CRT processes being developed by the FHFA and the GSEs, they choose instead to build on the Ginnie Mae infrastructure and processes. So Ginnie Mae–approved issuers get credit enhancement from private institutions along with the Federal Housing Administration, US Department of Agriculture, and the Department of Veterans Affairs, and issue securities through Ginnie Mae, which guarantees interest rate investors the timely payment of principal and interest on their investments.

As in the MBA model, Fannie Mae and Freddie Mac compete over the management of non-catastrophic credit risk. Here, however, their inclination to take on excessive risk is checked by converting them to mutuals owned by the nation’s lenders, as lenders can be expected to be careful in the management of an infrastructure on which they will be so reliant for their own survival.

Despite their different route, as you can see in figure 5, Bright and DeMarco arrive in a very similar place as the first two proposals: the government corporation manages the securitization and guarantees the mortgage-backed securities, and the private market basically does the rest.

Figure 5GSE Reform Figure 5

There are thus two critical differences between the three proposals.

  • Ginnie versus CSP. For the securitization infrastructure in the new system, Milken uses the Ginnie Mae infrastructure, while the MBA and our proposal both use the CSP.
  • What to do with Fannie and Freddie. The MBA would turn them into privately owned utilities that compete with other market participants over the distribution of the system’s non-catastrophic credit risk, Milken would turn them into lender-owned mutuals that do the same, and we would combine them with the CSP to distribute that risk and manage the system’s securitization.

With these distinctions in mind, the proposals can be much more easily compared across the criteria that should ultimately drive our decisions on housing finance reform:

  • Access to sustainable credit. Which best maintains broad access to mortgage loans for those in a financial position to be a homeowner at the lowest rates?
  • Protecting the taxpayer. Which best insulates taxpayers behind private capital, aligns incentives systemwide and addresses the too-big-to-fail risk that undermined the prior system?
  • Promoting healthy competition. Which best maximizes the kinds of competition that will improve options and services for consumers, lenders, and investors?
  • Ease of transition. Which provides the least disruptive, least costly path of reform?

Stay up to date on Jim Parrott and his latest work here.