A One-Year Retrospective

In three days, Howard on Mortgage Finance will have been live for one year. I thought, therefore, that this would be a good time to look back on the first year’s inventory of posts with a piece that summarizes them, by category, and also provides links to each for reference purposes.

In my second post, Some Context, and the Coming Bailout Charade (February 15, 2016), I stated that the goal of the site was “to improve the quality of information available about the relative merits of various approaches to reforming the U.S. mortgage finance system,” noting “(t)here is a large amount of misinformation currently being circulated on the subject, which in many cases my background and experience put me in a position to correct.”

I’ve worked toward this goal in my first year of the blog. The majority of my posts fell into two broad categories—establishing the facts about Fannie, Freddie and the mortgage finance system, and (based on those facts) analyzing alternative proposals for mortgage reform. I also did six posts on a more specialized topic—the credit risk transfer securities issued by Fannie and Freddie, which, as I discovered during the course of the year, leave almost all of that risk with the companies.

I briefly discuss (and provide links to) each of these posts below, by category:

Establishing the facts (5)

My main reason for creating the blog was that so much of what was being written and said about Fannie and Freddie, and the causes of and potential remedies for the 2008 financial crisis, was demonstrably untrue. The true story, I knew, was “hidden in plain sight,” and could be backed up with verifiable facts. But someone had to showcase those facts, and also weave them together into a narrative that was both understandable and credible. Five of my posts were primarily aimed at doing that.

Thoughts on Delaware Amicus Curiae Brief was my first blog post, put up on the same day I submitted that brief to the U.S. District Court for the District of Delaware (February 2, 2016). The post had some commentary on aspects of the brief, but its greatest value as a reference was its link to the amicus itself (found here).

Although submitted for one specific case in Delaware, the Delaware amicus was written as a rebuttal to arguments made by counsel for Treasury and the Federal Housing Finance Agency (FHFA) in a host of lawsuits filed against the government for its post-conservatorship treatment of Fannie and Freddie. Treasury and FHFA’s counsel had brazenly put forth a provably fictional account of what the government had done to and with the companies and why, claiming that poor credit decisions by them required their rescue and a subsequent $187 billion bailout by Treasury, and later triggered the need for Treasury to sweep all of their future profits to prevent a “death spiral” of borrowing to pay the cost of that bailout. The Delaware amicus uses a multitude of footnoted source documents to tell a very different story: Treasury made a calculated and planned decision to take Fannie and Freddie over, against their will and without statutory authority, for its own policy purposes and to its own exclusive benefit.

The Takeover and the Terms (February 23, 2016) compares the mammoth non-cash losses recorded by Fannie and Freddie from 2008 to 2011—which forced them to take $187 billion in non-repayable senior preferred stock from Treasury—with the subsequent spike in their earnings between the fourth quarter of 2012 and the first quarter of 2014. During those 18 months Fannie and Freddie earned $158 billion, all of which went to Treasury because of the net worth sweep. That amount was one and a half times the companies’ cumulative earnings in their respective 70 and 38 years of existence, proving irrefutably that their 2008-2011 losses were not real.

Treasury and the Financial Establishment (April 14, 2016) is a hybrid piece that begins with a review of actions taken by the Federal Reserve and Treasury to try to substitute “private sources of capital” for Fannie and Freddie guarantees prior to the crisis, and ends with a discussion of the irony of “Treasury and the Financial Establishment” taking no responsibility whatsoever for the spectacular failure that resulted from their earlier efforts at reform, while continuing to pursue the same reform objectives today.

Getting From Here to There (May 2, 2016) goes back to the mortgage finance system of the 1970s to summarize the history of the battle between opponents and supporters of Fannie and Freddie for control of what at its peak (prior to the crisis) was an $11 trillion residential mortgage market. It then discusses more recent developments in both the post-conservatorship management of Fannie and Freddie and the court cases filed against the government challenging that management, linking them with the history of what I call the “mortgage wars” before speculating on how all of these interrelated events might ultimately play out.

Economics Trumping Politics (January 4, 2017) contrasts the two alternative approaches to mortgage reform—legislative and administrative. It reviews the history of the misinformation campaign conceived of and engineered by supporters of large banks and investment firms, which falsely blames Fannie and Freddie for the crisis and has as a goal legislation replacing them with a mechanism more favorable to themselves. The post asserts that with the change in administrations it now is more likely that reform will be accomplished administratively, with the consequence that “responsibility for accomplishing mortgage reform shifts from a large number of misinformed members of Congress with political incentives to get it wrong to a small number of informed finance professionals with economic incentives to get it right.”

Analyzing reform alternatives (7)

It is an axiom in Washington that “you can’t make good policy with bad facts.” Seven of my posts were attempts to apply “good facts” to the analysis of alternative ideas or proposals for mortgage reform.

Fixing What Works (March 31, 2016) was written in response to a request from the Urban Institute to contribute to its “Housing Finance Reform Incubator” project. The recommendation I gave in this piece was to build the secondary market finance system of the future around the entity-based model we know worked well in the past—with Fannie and Freddie at its center—but to make reforms to it in three key areas: putting “utility-like” limits on the returns Fannie and Freddie could earn; limiting their mortgage portfolios to a percentage of their outstanding credit guarantees and to purposes ancillary to the guaranty business, and having FHFA impose an updated, stress-based capital standard on the companies that requires them to hold sufficient capital, by product type and risk category, to survive a worst-case stress scenario defined by the government. I since have amplified on some of these recommendations, but not changed any of them.

Solving the Wrong Problem (June 8, 2016) is my analysis of the other nine essays on single-family mortgage reform submitted for the “Reform Incubator” series. The majority of the essayists said, either explicitly or implicitly, that the goal of mortgage reform was to solve some type of incentive problem they claimed existed at Fannie and Freddie. I thought that was the wrong problem (the companies’ incentive structure was good enough for them to have had the best loan performance of any pre-crisis source of mortgage credit). Instead, the real problem—to which reformers have paid scant or no attention—is how to set capital standards that simultaneously protect the taxpayer and allow credit guarantors to price their mortgage guarantees affordably for as wide a range of borrowers as possible.

Capital Considerations (June 23, 2016) followed “Solving the Wrong Problem” in sequence, and was my attempt to address the capital issue I claimed others had not treated adequately. This post was not as strong as it could have been, however; its numerical analysis of Fannie’s post-crisis credit losses was both too complex for the general reader and too simplified to have been technically accurate. Still, its main conclusion is valid: Fannie’s post-crisis loss experience makes clear that it as well as Freddie could withstand a future 25 percent nationwide decline in home prices with far less capital than their critics and opponents insist they hold.

The FHFA Letters Decoded (July 11, 2016) was written after FHFA Director Mel Watt received three policy-oriented letters from trade groups and members of Congress in less than a month. My interpretation of this flurry of correspondence was that, because of developments in the lawsuits contesting the net worth sweep, policymakers were becoming concerned that FHFA might soon escape from its subservience to Treasury, and were seeking to influence any decisions FHFA might make independently. I also noted some obvious inconsistencies in the trade groups’ letters to FHFA, and called on the leaders of these groups to be more responsible and disciplined when formulating public positions on reform issues.

FHFA Fails the Stress Test (August 19, 2016) was a response to misleading reports in the media of the results of the 2016 Dodd-Frank stress tests conducted on Fannie and Freddie by FHFA. The companies’ combined projected credit losses in this test were less than their projected revenues, meaning they “passed” it comfortably. But FHFA also added over $100 billion in non-cash expenses that were outside the scope of the Dodd-Frank test, to produce a headline loss figure of $126 billion (in support of Treasury’s contention that the companies were a “failed business model” that had to be replaced). The media universally reported on the inflated $126 billion loss, while this post addressed the results of the test more objectively and accurately.

A Solution in Search of a Problem (September 9, 2016) stemmed from an Urban Institute article claiming that loans acquired by Fannie and Freddie from 2011 through the first half of 2015 were “squeaky clean” and “hardly defaulting at all.” That prompted me to ask, “if that’s the case, what is the mortgage market problem for which getting rid of Fannie and Freddie is the solution?” The rest of the post discusses the bizarre state of today’s mortgage reform debate, where advocates for a financial mechanism that benefits them (but not borrowers) try desperately to find (or more often invent) a problem they can claim their proposal solves.

A Welcome Reset (December 12, 2016) circles back to my first post on this subject, Fixing What Works, which proposes an administrative solution to reform based on Fannie and Freddie. The statement by Treasury Secretary-designate Mnuchin that he wants to get the companies “out of government control… reasonably fast” significantly increased the odds that reform will be done through an administrative rather than a legislative process, involving negotiations with plaintiffs to settle the lawsuits against Treasury and FHFA. In this post I offer my advice to the participants in these negotiations: in determining what the post-settlement mortgage finance system should look like, “pick the best model, get the capital right, and be realistic about the role of government.” The post goes on to expand on each of these points.

Critiquing credit risk transfers (6)

Finally, six of my posts this past year addressed credit risk transfer transactions, principally Fannie’s Connecticut Avenue Securities (CAS) issuances, but also others.

Risk Sharing, or Not (March 9, 2016) was the first piece I wrote on the securitized risk-sharing programs Treasury and FHFA were requiring of Fannie (and Freddie). I made two criticisms of Fannie’s CAS in this post: Fannie was buying far more loss coverage than it needed, given the exceptionally high credit quality of the loans it was acquiring, and it was greatly overpaying for the insurance it bought. I noted that Fannie and Freddie had a natural advantage in taking mortgage credit risk because they could diversify it in ways that capital markets investors could not, therefore it rarely would be economic for the companies to buy insurance from those investors. The only reason Fannie and Freddie were regularly issuing risk-sharing securities, I said, was because Treasury and FHFA had made them mandatory.

A Risk-sharing Postscript (March 25, 2016) was just that, written in response to a proposal (“A More Promising Road to GSE Reform”) to replace Fannie and Freddie with a government corporation that would rely on programmatic securitized risk sharing, instead of equity capital, to support the credit risk it took. I noted the great irony of this proposal coming out so soon after our last experiment with credit risk securitization—the collateralized debt obligation, or CDO—had ended in disaster, and reiterated some of the criticisms I’d made in my previous piece on securitized risk sharing, the real “failed business model.”

Far Less Than Meets the Eye (August 8, 2016) resulted from my having read some prospectuses for Fannie CAS transactions to find out more about how risk-sharing securities worked. Much to my surprise, I learned from Fannie’s own disclosures that its CAS risk-sharing tranches are deliberately engineered to avoid taking credit losses. This post is more technical than most, but it’s worth plowing through to understand how these securities are structured and why they’re so inefficient. It astounds me that Treasury (and FHFA) would force Fannie to issue “risk-transfer” securities that require hundreds of millions of dollars in interest payments and leave virtually all of the credit risk with the company, but that’s what they’ve done.

Response to FHFA on Credit Risk Transfers (September 9, 2016) reproduces a submission I made in response to a “Single-Family Credit Risk Transfer Request for Input” issued by FHFA in June. In it, I advised FHFA that it needed to update Fannie and Freddie’s risk-based capital standards before it could accurately assess the effectiveness of any risk-sharing transactions the companies might contemplate, and that it should adopt a “borrower benefit” standard for these transactions—i.e., will they make the borrower better off, relative to the companies retaining the credit risk themselves? I also told FHFA it should not make risk sharing mandatory, that Fannie’s CAS (and similar securities issued by Freddie) were “clearly, and grossly, uneconomic,” and that “FHFA must ask itself how it failed to detect that fact.”

Getting Real About Reform (October 25, 2016) is another hybrid piece, mainly about risk sharing but also about mortgage reform. It was prompted by a second reform proposal—this one by the Milken Institute, following the “Promising Road” from the Urban Institute—that relies on securitized risk sharing as a substitute for equity capital. In this post I label both proposals “theoretical fantasies.” Advocates for these ideas pretend that a dollar in face value of a risk-sharing security is equal to a dollar of upfront equity capital, when in fact it is worth less than one-tenth as much. When you substitute capital reality for capital fantasy, both the Milken Institute and the Urban Institute proposals implode, as did CDOs about a decade ago.

Investors Unite Risk Sharing Call (November 16, 2016) is a written version of the remarks I delivered on this call (and includes a short postscript). In preparing for the call I drew on my previous three posts on risk sharing, and attempted to package the material in a way that would be more accessible to an audience less familiar with the subject. I covered risk sharing using forms of private mortgage insurance as well as securitized risk transfers, but spent more time on the latter. I concluded by calling Fannie and Freddie’s risk sharing securities “exercises in deception,” and by asking the audience to tell Treasury and FHFA to “stop trying to fool us, and instead turn their attention to devising reforms for the system that might actually work.”

*          *          *

Overall, I’m satisfied with the posts I’ve put up on the site during its first year. I’ve said much of what I wanted to say about Fannie, Freddie and the mortgage finance system, past and future. To avoid being repetitive, future posts may be more event-driven, possibly shorter, and perhaps even less frequent (if I don’t have something new or different to impart). But I’ll have the same goal for the site in the coming year as I had for its first one, and will seek to add as much value to the reform dialog as I can.