Nearly 10 years after Fannie Mae and Freddie Mac were placed into conservatorship, housing finance reform remains the single largest piece of unfinished business of the housing crisis. The failure of Fannie and Freddie, the taxpayer bailout and repayment that followed, and their unresolved conservatorship continue to demand final resolution, even if Congress does not.

Housing finance reform’s first priority should be to preserve the GSEs’ essential market support functions while maintaining the expectation that this support will be as broad and inclusive as possible without unnecessary barriers to responsible, sustainable credit for underserved borrowers and communities. It is an existential requirement that has not been reliably assured by any of the radical solutions of systemic reform that have been proposed to date.

Successful reform must ensure that the GSEs support housing finance liquidity by maintaining a healthy market for mortgages and mortgage securities. The GSEs should be flexible enough to determine their appropriate role given market conditions. At times, they’ll need to stay in the mortgage markets when others are retreating, and other times pull back when markets are becoming overheated, without withdrawing altogether and thereby eliminating their impact on the market.

During my tenure at the U.S. Treasury Department, I was part of the housing finance reform team under three secretaries and two presidencies. Under the leadership of Dr. Michael Stegman, Antonio Weiss and Craig Phillips, we met with hundreds of experts, read countless papers, and developed three unreleased white papers. We explored the utility model, the Ginnie Plus and half a dozen variations of Corker-Warner and Johnson-Crapo. It was like Policy Wonk Speed Dating: Sit down at the table, have a great conversation, go on a couple of promising dates, learn about some daunting flaws, go back to the table, repeat.

The fact is that anyone could redesign our housing finance system on the back of a napkin as long as you don’t care about the long-term, fully-amortizing, pre-payable, fixed-rate mortgage. But if you want a housing finance system that preserves the fundamental element that sets us apart from the rest of the world while providing the only meaningful wealth creation tool available to low- and moderate-income Americans, then it’s going to be an incredibly complex exercise, with enormous transition and counterparty risks and incalculable unintended consequences. Unfortunately, the law of unintended consequences is never repealed.

Last month, the Trump administration released its proposal to reform the federal role in housing finance, which I found to be an encouraging start to the difficult discussions we will be having this year. While there remain many details to work out, I was pleased to see the proposal would preserve much of the current system that is working while addressing structural flaws that must be dealt with, including the need to bring more private capital into the mortgage markets. Some elements are bipartisan non-starters, but in general, there’s a lot to work with in this proposal if we fill in its many blank spaces with thoughtful policy that is backed up by solid data and lessons learned during the housing crisis. This is especially true in the area of affordable housing requirements. Every federal benefit comes with a federal responsibility and housing is no exception.

An important first step is allowing Fannie Mae and Freddie Mac to exit conservatorship as private companies with access to a federal guarantee that is paid for rather than implied. We need to finish the work we began in the Housing and Economic Recovery Act of 2008, which made the government rescue of the mortgage market possible. We need to fix what is broken in the current system, not tear it down and hope that if we build a new one it will work for everyone, or at all. Field of Dreams was a great movie, but it is not a viable economic theory; if we build a new system, they won’t come. If they would, investors would be filling the halls of the Capitol insisting on passage of legislation like the Corker-Warner bill that envisions five or more GSEs. It’s not happening.

One common feature of many of the housing finance reform proposals made over the past 10 years has been the creation of multiple mortgage market entities to replace Fannie Mae and Freddie Mac’s “duopoly.”  Most envision five or more new private mortgage guarantors. I am a big believer that increased competition benefits everyone in a healthy market.  It not only brings better products at cheaper prices to consumers, it forces private companies to stay on their toes and demand excellence from their leadership and employees.

Fannie and Freddie had this kind of competition in the early 1990s, when dozens of lenders provided the vast majority of their business. Today, the mortgage origination market has consolidated to only five lenders that originate 80 percent of all mortgages. This consolidation has many positive attributes, bringing important economies of scale to a low-margin business, already struggling to make smaller loans due to the high cost of servicing and origination. But lender consolidation also means that those five large originators have enormous market control over the two GSEs, which distorts the competition in ways that undermine the market.

As an officer at Fannie Mae between 1994 and 2006, I watched this situation devolve. Countrywide Home Loans, one of the worst actors during the housing crisis, is a perfect example. In the 1990s, Countrywide was an important customer of Fannie Mae, but only one of many. When they made loans that were poorly underwritten, they were forced to repurchase them and they didn’t like it any more than the rest of the GSEs customers.  But repurchase requirements forced lenders like Countrywide to have “skin in the game,” the stake in a mortgage’s long term performance that Dodd-Frank failed to duplicate. As Countrywide grew, and as Fannie and Freddie fought over market share, Fannie Mae was pressured to stop making Countrywide repurchase bad loans.

By the early 2000s, as Countrywide joined the ranks of Fannie’s top loan originators, which together made up a growing share of Fannie’s total business, their demands grew to include ever-shrinking guarantee fees, always accompanied by the threat to abandon the company for Freddie Mac and make Fannie Mae the second largest GSE for the first time in its history. These low g-fees allowed them to offer better prices to small lenders who sold their loans to Countrywide instead of Fannie Mae, making Countrywide’s market share – and market power -- even greater.

By 2005, Countrywide was paying the lowest guarantee fee in the market, and the demands shifted to allow increasingly reckless underwriting, permitting multiple layers of product risk by combining interest-only adjustable rate mortgages, teaser rates with huge payment shocks after just two or three years, and fewer and fewer requirements for income verification. When I had the temerity to ask how people could possibly repay these loans, I was told that mortgage brokers would refinance them before they reset. Now the loans were backed, not by solid underwriting or appropriate capital reserves but by equity stripping and a permanent gamble on forever-increasing property values. 

Fannie Mae and Freddie Mac could have said no, as many of my colleagues argued, but they did not because they were afraid of losing so much market share that their stock price would plummet and they would become irrelevant, possibly even insolvent. It was a choice between drinking the poison or jumping off the ledge. Drinking the poison seemed like the better choice. 

Today, Countrywide is gone and the top five lenders in the mortgage market have learned their lesson, but for how long? In time, as the current generation of mortgage leaders retire, the same pressures for market share will emerge and lenders will press for lower fees and more permissive underwriting. Any new system has to recognize this basic dynamic of competition. 

In this market, more competition is not the answer, utility pricing is, and with utility pricing comes rates of return that do not attract capital in a competitive equity market. One alternative would be to have only one GSE, but then you lose the competition over execution that has worked well in the mortgage market. Five years of working at Treasury with some of the smartest people I know has taught me this: Much of what we need in a new system already exists in Fannie Mae and Freddie Mac.

Another essential element of any bipartisan approach will be how we quantify and enforce the enterprises’ responsibility to serve all Americans, not just the wealthy. Millions of Americans of all incomes are struggling to buy affordable homes or find quality, affordable rental units they need. Even with Fannie Mae and Freddie Mac, the system is not serving everyone it should. I will address this vexing roadblock to housing finance reform in another article.

If we can’t find common ground, we may have to wait to be moved to action by another housing crisis, as interest rates continue to rise and housing affordability continues to decline. That would be a truly tragic outcome and could cost us more than we can imagine. But it doesn’t have to happen. Not when Realtors, homebuilders, lenders, investors, and consumer and fair housing advocates are willing to work together to find common ground.