24th Annual Financial Markets Conference - Mapping the Financial Frontier: What Does the Next Decade Hold? - May 19–21, 2019

Banner image for 2019 Financial Markets Conference: Mapping the Financial Frontier

Keynote Address: Donald Layton

Chief executive officer Donald Layton discusses his tenure at Freddie Mac during the conservatorship era. Federal Reserve Bank of Atlanta president Raphael Bostic moderates the discussion.

Transcript

Raphael Bostic: I hope you all enjoyed Chair Powell's remarks last evening. It's always good to hear from Jay, and I hope you enjoyed our little ad-libbing when the technology went out—I'm still going to remember that "Crazy Eddie's" thing, that was...[laughter]. I really enjoyed that, an unexpected treat. But the stuff on financial stability and how we manage innovation is really important, and we'll continue to have conversations on that. And I was really pleased that Jay was able to come and give his perspectives.

We have a pretty stimulating day today, so I figured we should just jump in and get to it. So I'm going to now introduce our keynote speaker for this morning, who is Don Layton. Don has a distinguished career in finance. He worked for nearly 30 years at JP Morgan Chase (and its predecessors), and he's covered a breadth of business lines and executive roles, and then he "retired" in 2004. And I say "retired"—I put that in quotes—because he didn't actually stop doing things, and in some ways he's been more busy since then. From 2007 to 2009, he was chairman and chief executive of E-Trade Financial, which is the online brokerage. He's also been a senior adviser to the Securities Industry and Financial Markets Association, or SIFMA, from '06 to '08.

Today, we're actually going to engage Don in his most recent gig, which has been as chief executive officer at Freddie Mac, which is one of the government-sponsored enterprises [GSEs] chartered by Congress to provide liquidity, stability, and affordability in the U.S. housing finance market by being one of the really important secondary market players. Don has led Freddie since 2012, and so he's been through some...we should call them fairly interesting times in housing finance. He oversaw Freddie's return to profitability, and has really changed the mind-set at Freddie to have it be much more competitive with Fannie Mae. And I would say there's a healthy competition that's going on now, and Freddie's trying to beat them up, which is a good thing for all of us.

His tenure has been in the conservatorship era. Freddie and Fannie both went into conservatorship in 2008, and that conservatorship included almost $200 billion of capital into the firms, and all of that has been paid back, so I think that investment has paid off, to some extent. This morning I've asked Don to talk a little bit about his views on the housing finance: where it's been and where it's going, how the conservatorship has helped lead to transformations in that sector, and then get his perspectives on what else needs to happen. And we're going to do this in a conversational format, so, Don, come on up, and let's start the conversation.

Donald Layton: So good morning, everyone. For full disclosure (for Raphael), we met when he became a board member at Freddie Mac, and then had to resign to take this little job in Atlanta [laughter]. So actually, one of my board members goes, "Does he really have to resign from Freddie? I mean, can't he do both jobs?" I had to explain to him about conflict of interest.

Bostic [laughs]: The challenges of a board, huh? And I should also say that Don is kind of on a farewell tour, so he is stepping down as the CEO next month. So you didn't bring, I didn't see tour T-shirts outside [laughter], you might want to get those put together, and we should be good. So anyway, welcome. Thank you for agreeing to be here. It's good to have you here, and can you talk a little bit just about what's happened during conservatorship for Freddie, and also for the mortgage finance system?

Layton: I want to give some predicate comments. So I'm a private sector banker, I was retired and I came to do this as public service. And by the way, I'm no defender of the old GSEs. The old GSE system, prior to conservatorship, had lots of problems that needed fixing, and so I came as a reformer to do so. And one of the things you have to understand right now about us is what I've found, here's a blinding insight: when government is behind about two-thirds to three-quarters (depending on how you define it) of mortgages, you get a lot of politics [laughter]. And so everything we do is politicized, everything you read, most things you read—the vast majority of the things you read—have political slants, for ideology, or rent-seeking special interests, or whatever.

So it's very hard for outsiders to tell what's going on. And in conservatorship, we're not allowed to really be loud—out there, saying what's going on—that's part of the rules, so our disclosures are in such scintillating documents as 10-Qs [laughter], status reports from the FHFA [Federal Housing Finance Agency], which I'm not sure how many people read them. And so again, there's this giant knowledge gap, and I'm going to address some of those today.

The first item of the knowledge gap is, everyone once in a while you see someone saying, "Nothing's changed in conservatorship." We 100 percent disagree. In fact, if you look at the big weaknesses of the old GSE system, almost every one of them has been addressed in conservatorship in a reasonable way. What were the big mistakes? The design flaws. The GSEs got to borrow implied guarantee, unsecured funding and build up these giant investment portfolios (discretionary). They produced most of the profits of the GSEs prior to the crisis. Part of the rescue, the government support agreement known as the PSPA [Preferred Stock Purchase Agreements], required us to run those down over 10 years. They've been run down, the FHFA made us run them down even more, they now just support the guarantee businesses. So, "check," that problem's been solved.

A sensitive topic, not really core to what all of you would think about in the mortgage industry, is what's called "level G fees." The GSEs gave volume discounts to the bigger mortgage lenders, and the small lenders—of which there are 1,000 of them or more—considered that to be inconsistent with, if you will, the public purpose of a secondary lender with government support. They mandated level G fees, we figured out how to do it. It's been solved for years, and the industry likes that.

My favorite, for Fed people who think about FSOC [Financial Stability Oversight Council]—and there's a lot of Fed people here—the concentration of $5 trillion of mortgage credit risk in two companies. When it was designing these companies, and you wanted to help housing and do housing policy, you go, "Wow, having a limited number of entities to execute housing policies is good, because you can set rules for them." They didn't think of the outcome, which was all this concentration of risk.

And that's where we—I, and a young man, and a few people at Freddie—invented GSE credit risk transfer for the single-family business, where it's been years of development, getting better and better, and what we have we call "fully effective CRT [credit risk transfer]." Starting about a year ago, we put (for new mortgages) almost all the catastrophic risk out to investors in a secure format (for those who worry about these things, it's structured more like a CAT bond, a catastrophe bond), so we don't have counterparty risk, they give us the cash up front.

And so we're actually completing the securitization business model. The first part was the MBS [mortgage-backed securities] for interest rate liquidity risk. Now we're securitizing the credit risk. So no longer is the taxpayer fully on the hook. The amount of exposure we have is running down rapidly—so when you see the credit statistics for our financial reporting, that assumes we own it, but actually, most of it's now going off to others. We do a GSE equivalent of the CCAR [Comprehensive Capital Analysis and Review] stress test, the next one is published by the FHFA in August. You'll see this continuation of risk during a stress environment, which people consider to be very excessive, continuing to have this big rundown, because the risk is for others now.

And for efficiency of the markets, the single security—so that we and Fannie can deal with investors on an equal basis (and possibly others' interests someday)—is being implemented in a few weeks (June 3). Just for the record, it took seven years and cost you—because our money goes to the taxpayer otherwise—$2 billion to do. So this is a big affair that's come in, and Raph mentioned, one of the others is that the GSEs were very focused on lobbying for profits to be pre-conservatorship. We went competitive, we decided to compete, and it's been great. This is the way it's supposed to be, instead of a complicit cartel.

Our technology budget is two and a half times the size what it used to be, so that we're starting to chip away at the mounds of paper the typical American had to bring to make a mortgage. There's just no reason for this stuff. This has been a backwards industry, and so we're starting to lead it in. So we think a lot of things have changed, and if you take the average industry leader, you read about it in the industry press and they'll tell you the mortgage system now is working better than it has ever been in terms of from their perspective.

Bostic: Well, it's really interesting, and with the single security, for example, I was kind of skeptical that we were ever going to get there. When I was on the board, just two very large organizations that didn't have a culture or a history of doing a project like that, really, to advance the sector. What kind of stuff did you have to do to change that culture, internally? And then, how did you work with Fannie and the FHFA to sort of help make that go through?

Layton: First of all, he's winging it, this was not a prepared question [laughs].

Bostic: I'm nothing if not unpredictable, right?

Layton: There you go. First of all, conservatorship has been a very odd status. Normally, conservatorship is this bank liquidation thing that lasts a few weeks or a few months, so no one's ever been in this status for a long period and large size. So first of all there's this whole track record of the FHFA saying, "Continue to do things separately, but this issue here I want you to be—the word is ‘aligned.'" So we've been having a culture of having alignment things, aligned to develop data standards so lenders can all send in the same automated data set to both of you, so that they can have one set of rules and be more efficient than two sets of rules (one for you, one for them).

So there's dozens of these things. So the muscles of cooperation have been getting going. If there are any lawyers in the room, this is accompanied by a piece of paper saying, "You can ignore antitrust, because the conservator can tell us to do this and override antitrust." Otherwise, we're not allowed to talk about these things.

And so, first of all there's been this whole system of kind of "frenemies"—right?—die-hard competitors, but we cooperate on this. Now, that's not that unusual among large FIs [financial institutions], because they often will lobby in Washington together or do sort of industry standards, like if you're a large bank and you compete like crazy, but when it comes to, say, managing DTCC [Depository Trust & Clearing Corporation], an industry utility, you're in it together and your people deal together.

So those muscles have been building, but even then it was too big a thing to be done by us in that manner. We formed literally a joint venture called CSS (Common Securitization Solutions) to do the implementation centrally. The requirements for the head of that were a lot of mortgage expertise and technology background, and couldn't have anything to do with either of the companies previously, so they were clearly neutral, and that's what happened. The man's name is Dave Applegate, and we hired him and he built our organization with some of our people, some of Fannie's, but mainly outside hires.

So, add all that up, it worked, with the FHFA playing a strong role. So if you have two joint venture partners and you disagree and 50-50 vote, they're the tie-breaker.

Bostic: Interesting. So I just want to remind you all, if you have questions please do put them in through Pigeonhole and we'll get to those a little later on. I wanted to say, though, you were at E-Trade, you were at JP Morgan Chase—then you go to Freddie in this funny kind of configuration. What did you expect going in, and then how did the actual experience compare with those expectations?

Layton: Yes, all right. The E-Trade was two years—kind of a special thing—but I regard myself as a banker. For those who remember the name, I started off as a trainee at Manufacturers Hanover (there's a name from the past); it was the fourth largest bank at the time. And in the mergers that created JP Morgan Chase, it was the first one to lose its name, so we went through all these things. I regard myself as an FI person—financial institutions manager—and I had broad experience, but a whole core of it (by the way) was I got assigned to work on all these things like Basel I.

Bostic: Basel I?

Layton: Basel I. And why we did Basel I and the original development of VaR [value at risk] and stress testing—these were in the late '80s, or late-ish '90s—so I have a big background in FI stuff. The first thing I found was everybody between Fannie, Freddie, and the FHFA were really mortgage and housing people. There was not much in the way of FI, or broad experience. Insight for everyone here: the government, in supporting housing, fundamentally created a parallel banking system. It was the thrifts, and the GSEs, and the home loan banks—and I'm probably forgetting a few others—and they were regulated separately by the Federal Home Loan Bank Board, and FSLIC [Federal Savings and Loan Insurance Corporation], remember these names?, and later OFHEO [Office of Federal Housing Enterprise Oversight], and it didn't go well. It had a lot of safety and soundness issues, but what's left over is the FHFA and the GSEs and home loan banks, as part of that separate banking system. The thrifts have moved into the regular banking system, with the OCC [Office of the Comptroller of the Currency].

So the skills level at FI, I ended up filling a lot working with the FHFA, so I was surprised that they had so low a skill base in those topics. Second—totally unexpected—when they made the FHFA conservator of both Fannie and Freddie, and the two GSEs are such a large share of the mortgage system, the FHFA became, if you will, a sort of central planner of the mortgage system. And normally when you say "government in charge," that means "noncompetitive." It actually helped the industry get away from being focused on politics and lobbying to running businesses. The first initiative they did, they came out with these things called "conservatorship scorecards," which were public expressions of how the FHFA, as the central planner, was going to upgrade the mortgage system.

Things to start with were very prosaic, uniform mortgage data sets, so that all these thousand lenders and hundreds of servicers could have a standard data set to automate, so eventually we can build automation on them. Later they came up with things like eligibility standards. There are a lot of articles today that nonbank seller-servicers are a large share of the market, and they're going to get in financial trouble, but you notice there hasn't been a lot of financial trouble, because about four years ago we started working on what were called "eligibility standards for nonbank seller-servicers." I'm going to be a little joking in the style, but this is basically the FHFA with the expertise of the two GSEs working with them, saying, "You want to deal with my kids—my son, Freddie, and my daughter, Fannie—you play by my rules. I don't regulate the nonbank seller-servicers, they're regulated at the state level. But you want to deal with my kids? You play by my rules." And they did that for MIs [mortgage insurance], and we've done it for a whole host of things.

Modification programs, post-HAMP [Home Affordable Modification Program], HARP [Home Affordable Refinance Program], the succession of HARP—all these things—the single security, standardizing CRT. So they've actually become this, and it has modernized what is the most backwards industry in American consumer finance. Everyone who is a banker in consumer finance knows the mortgage side is the most backwards part of the industry. So it's really been good having them do that, so that was one expectation.

Third—and I'm trying to do things that are interesting here and insightful—shortly after arriving at Freddie, I got called into Treasury (who, they support us so they got dibs to tell us things). And they went, "You're sitting on a lot of impaired mortgage assets." That is true. At that point, Freddie alone had $200 billion of impaired mortgage assets sitting on our books—in the form of either PLS [private label securitization] securities, end-payer PLS securities, or mortgages that had gone into default that we'd bought out of the securitization trust. That's a lot of money. Fannie had a larger amount, obviously, and they went, "You've never sold any of it, you've never done anything in your company except watch it run off. We want you to sell it. Wall Street will buy this stuff."

So the insight here was, these companies were sort of shell-shocked and didn't know how to handle this. The other insight is, I said to Treasury when they talked about "sell this stuff to Wall Street," I kind of went, "Did the people from Wall Street mention a price" [laughter]? The answer was no. And what I learned was...remember, I was a Wall Street guy, so I was on the other side now. Of course, Wall Street loves to buy assets when the markets are out of equilibrium and they're trading well under fundamental value. That happened in the thrift crisis with RTC [Resolution Trust Corporation], and they were looking for a redo of that, and I had been recruited at one point—I didn't do it—"Join our investment fund, to be on a bank board, because we're going to buy banks cheap from the FDIC [Federal Deposit Insurance Corporation], just like we did back in the RTC days."

That, in Wall Street terminology, is known as a wealth transfer to the private sector [laughter]. I don't do wealth transfers to the private sector. The Treasury was right, we hadn't sold anything, but I hired people from Wall Street to figure out how to dispose of the assets where you've got fundamental full value, as opposed to a distressed price. It took longer, but it got done, and we're very far down that road.

So it's those kind of things that were just interesting, how to establish these things. If you will, the GSE sector had been so much about politics and subsidies, pre-conservatorship, we had to establish a lot of basic economic skills and processes.

Bostic: Now I've got to say, I'm a little surprised that you've sung the praises of FHFA, but I've heard you talk about engaging with them. Was it all just milk and honey, or were there some challenges there?

Layton: No. On July 1, I can be fully honest [laughter]. No, I dealt with FHFA under two directors. I came in it was [Edward J.] DeMarco, and then it was [Mel] Watt. The nature of conservatorship and me being a reformer is, our fundamental desires were the same. So this is not like you had this normal tussle of a relationship between the regulator and the regulatee. We were in it together, we wanted to solve the problems together. But even your spouse and you don't agree, okay? So you have these disagreements. I found them—gee, what a surprise—too slow in some things, I think they were a little too worried about political reactions sometimes. They were brave a lot of times, but not all the times I would have liked them to have been.

And in 2012, the FHFA was smaller and the conservatorship was young enough that they had a light touch—every year the touch gets a little heavier and more intrusive, in that normal government way. And so, we're having more tussles over micromanagement issues, which we tell them, honestly, and so we have had an honest relationship.

But I do want to go off for a minute here on, I have tremendous respect for the leadership of the FHFA over those years. They were handed an odd job, no one knew how to do it, and because there's been so much good reform in the mortgage industry, when you write the history books they should go down as the good guys.

Bostic: Very good. So, lots happened, you've identified that. What's left? What needs to happen next?

Layton: So in college, I like history, right? History books. In history, they talk about eras or ages: the nuclear age, the age of exploration, the Victorian era, or whatever. It's messy, it's slightly inaccurate. The DeMarco years—which were like 2009 to early 2012—were really the era of dealing with the foreclosure crisis. This is before I got there, but the whole industry and the GSEs were just stunned, they didn't know what to do with all these foreclosures, they'd never had anything like this—deer in headlights. And eventually the government got its act together, and you had HAMP and HARP and HARP 2.0, because HARP, the first one, didn't really work well. And the giant issue of principal forgiveness on a broad scale, which was not done, but consumed a lot of time.

So you had that. Now again, it's sloppy. DeMarco also moved into the reform era. The Watt era, and the latter years of the DeMarco time, were really the era of GSE reform while in conservatorship. That was my era. And we have again, I think, addressed almost all the big problems, there are other problems, but a lot of the core ones have been addressed. One hopes in the future it's not lost. And this went well. You can read about all this. Every year they put out what's called the "conservatorship scorecard," it says, "Here's where we're going, the public knows."

And it's about building the single security, and it's about doing risk transfer, and it's about data standards and doing affordable right (i.e., safe and sound)—and it's out there in public domain. It's not noisy, and a lot of people will cover it up with other noise, but it's there. So you've seen that.

The [Mark A.] Calabria era, as he says—and it's sort of right there, again, we want to continue some of the reforms—is to get us out of conservatorship. It is hard to get us out of conservatorship. There's the administrative approach, there's the legislative approach, and there's the legalities associated with the rescue agreement and the shareholders of the underlying companies who were mostly—but not 100 percent—disenfranchised, and have some rights.

So it's both technically hard, and legally hard, and politically hard, but that's what they've said they're going to do. The administration put out their presidential memorandum, and it says, "Here's our general direction." What I can tell you—again, if you see more obscure sources of this information—is that the Treasury has the pen on the plan for the GSEs. They say they've kind of finished their plan and it's circulating inside the government, and it's going to come out by the end of June. I'm not sure I have much trust in deadlines, but it's pretty close.

And Calabria talks about then finalizing and working with them, and the plan should be finalized by the fall, and then ending the profit sweep—which is a big, symbolic deal—around year end. So they're embarked on it, it's the best shot they've had in years, but it's not assured, but one hopes it goes well. And that's kind of where things are now.

Bostic: All right. So I want to talk about the future, and in doing that there are a number of questions that have come up, so I'm going to go the questions, you'll see them. The first is really a reflection on the current model, and the question here is, would you say that the crisis experience has proved that the hybridized amount before is a bad idea? And when conservatorship ends, would you handicap that as the most likely outcome?

Layton: Okay, this is where there was a...some of the things I've learned in Washington: there is a testifier from one of the small lender groups who went to Congress a year or two ago, and said, "Listen, this is in response to all these interesting, galactic plans to redo the mortgage system. We've had everything from a government single utility to a cooperative model by the industry (it would be a big cooperative) to other versions of hybrids to pure private sector." And he went, "Listen, we"—"we" meaning his industry, which was small lenders—"are uninterested in someone's idea of what the mortgage system should look like, as if we were starting over again." Translation: big waves sink small boats, they don't want a lot of change.

And so he goes, "So why not just fix the five or six big things that didn't work in this system, and do that and let's move on." And so you sit back and think from this, and you realize, the "redesign it all" stuff means you start all over, with all the unintended consequences, all the loopholes, all the cross-subsidies that the next system would have. And so, yes, if we were starting it all over again, we might not have done it, because I agree with that testifier. But given the old phrase in restructuring: we are where we are.

Personally, the "bad idea" actually delivered something core, which is the average American homeowner gets the 30-year fixed rate with free prepayment, and you can lock in the rates ahead of time, and all that kind of stuff—and that is enabled by the GSEs. And if you fix the stuff that didn't work, maybe that's a better system than opening up the box and having the whole thing start over again with unintended consequences.

So as of this point, the relevant government officials, administration officials, have said publicly and privately, it all started off with all these plans all over the place, from far left to far right. People have kind of migrated into a range, and the range keeps the GSEs. It may increase the number of them, if they can ever get anyone to enter, to have more competition. But it kind of works, roughly, with the current system.

Again, reflecting this notion of a galactic new system would be so destabilizing, so opening up the Pandora's box, that they just don't have the appetite for it.

Bostic: So there's a question that is directly to this last point, which is around two GSEs versus six to eight of them, and you mentioned that some may enter. Do you think we'd be better off if you had six, or would that materially change the degree of competition in the marketplace and innovation?

Layton: The answer is, "Yes, maybe." I want to start with, the GSEs are not regular private sector companies. They never have been, they are hybrids—that other question had it. They are hybrids of our government objectives and restrictions. We can't go into credit cards, so we're very limited in what we can do, and government support, but private capital (for various reasons).

At 50,000 feet, I agree: it'd be better to have more competition, but let me ask two questions. Number one: who's going to enter? Who's going to raise capital with the following investment thesis: I'm going to start from scratch, and to make it work, to be able to deliver into the single security, I need a nationwide footprint because otherwise it won't be accepted as fungible with the others (so that's a market requirement). And so I have to go big, fast, and I'm going to compete with two massively entrenched competitors, who have scale advantages and 100 percent market share against me.

So what private equity firm is going to go, "I like that story"? So we sat there and go, "Who actually is going to enter?" I understand the theory that it'd be good if people would enter, but when you actually go and ask possible entrants, they go "It's just not a good deal." So I'm not talking theory, I'm talking about, will someone show up?

There was one of the MI leaders, you go, "Well, the MIs are kind of maybe in a position to do that." But then they have to choose. They're either "over 80 LTV" firms, or they're "under 80 LTV" firms. Under the rules, you can't be both now, right? So I don't know how that's going to work, maybe they'll make an exception, you could get some competition there.

The other practicality: anyone here ever worked for an actual mortgage lender? No? Okay. Was there a hand back there? Okay. Mortgage lenders have to deal with the GSEs, they deal with what's called "automated underwriting systems," the up-front systems to go: someone walks into one of their lending offices, they put the stuff in to know if it's going to be approved or not, and what data they need, and then to process it. And it's very complicated because if you don't have a W-2 there are alternative methods, and what documents it goes on. This stuff is like a thousand pages to describe this system.

Fannie has the more accepted tool, called DU—desktop underwriter. Lenders don't want to have two, they don't want to have three, they don't want to have five, they want one or two. We buy half our mortgages off of DU, because they have such a dominant market share and small lenders go, "I can't afford two systems." We're working to replace that, but it's chipping away.

So our struggles to compete: I do not understand how the front-end tools from a third or fourth or fifth are going to be accepted by the lenders. Then people go, "Well, then let's have a common one that everyone uses." Okay, that would be...well, my only data point is to do a comment on the back end of the security—took seven years and $2 billion. So, because our systems are our system, they're not railroad tracks that everyone can use, so you'd have to start from scratch, in the same sort of way. So who's going to pay for that? And this is where the debate goes around.

So I agree that, conceptually, competition will be better. I just don't know in this case, with this hybridized government thing where we're sort of a utility, whether it is practical. So the alternative model is the electric utility regulatory model. Maybe there are efficiencies for having just one or two here, what I'll call the "SIFMU model"—systemically important financial market utilities—where you have two or three clearing organizations competing, and that's all there's room for in a lot of financial markets.

So, people are kind of debating this at the working level, and we'll see what comes out.

Bostic: So that's interesting, the utility model's actually quite an interesting one. It requires a different type of regulatory approach that we'll have to work through, and figure out if there's appetite for that.

Layton: Ah, interesting, since you left the board, we are currently operating under a partial, electric utility-style regulation—single-family guarantee fees are now run under an electric utility rate-setting style by the FHFA bank in conservatorship. Their capital system, which they developed, is the denominator—this is for the entire single-family business—the G fees, the cash window fees (all defined over time), proper allowances for reserves, loan loss reserves and expenses. And we are given a monthly target to hit, every month, that our new flow must be a 9.7 percent return on conservatorship capital. That's the estimated capital we would need.

And so you can get a little bit lower price if I charge you a little bit more. And it's all risk-adjusted by the capital system, it's not a crude capital system like 50 percent risk-weighting; this really goes into detail. So, in fact, it's already half been assembled, the question is whether it'll be kept or not.

Bostic: Interesting. This has implications for another aspect of the mortgage market people talk about, which is private mortgages or the private mortgage market. What do you see in terms of the private market, relative to the GSEs?

Layton: The person who asked this question: did you mean banks or PLS, or both? Both, thank you. Okay, when people say "private mortgage market" in Washington, often they mean just the PLS market, okay. PLS means "private label securitization"—i.e., it was underwritten and securitized into a bond without government support through a GSE or some other mechanism, like FHA or something (or Ginnie [Mae]).

Okay, so first of all, start with (of course) we only do what I'll call "middle-class max mortgages" with a conforming loan limit, so above us it's a "no Fannie, no Freddie" market. What you see is that bank balance sheets dominate. The PLS market, which was flying as part of the bubble, has never really come back to be a major player. It's getting better over time. One of the heads of one of the major PLS players goes, "We're just chipping away at it slowly." But there's no Fannie or Freddie, it's all about them.

It reveals that the PLS market had a lot of fundamental flaws, it was a very bubble market. And, by the way, this is interesting, behind the PLS market was Fannie and Freddie and the GSEs (with their implied guarantee, cheap borrowings) buying PLS. At the height of the bubble, one out of $6 of PLS was owned by these implied guarantee institutions. In fact, that PLS market was not fully private, it was subsidized indirectly that way—we don't buy anymore, so they have to make it.

And there were real issues about who was watching the quality control that the documentation was right, and who was managing the servicer for high credit quality, and investors—the BlackRocks, the PIMCOs—they all feel they were burnt, that they got adverse selection against them, and they're not interested in this stuff. On top of it they go, "I like the mortgage asset class diversify, but the GSE CRT (credit risk transfer) series? That's how I'll get it, and I'll get it that way and it works."

So PLS is struggling, that leaves bank balance sheets. Bank balance sheets have a limited—not zero, but not too large—ability to take the interest rate risk. We all know this from the thrift crisis, but given their equity basis they can do some, they tend to do it for wealthier people. I remember when the PLS market started bragging, "We can come back." One of them, they had done a deal, people pointed out, "PLS is back." And I investigated, I found that the income behind the average mortgage in their portfolio was $50,000 per month—per month. In other words, this was not middle-class mortgages; this was wealthier people where the credit risk is low, and that's what investors would take.

So this thing struggles. PLS seems to have fundamental flaws, and is struggling to come back in real size. And the questions are, how big should it be—not how big it wishes to be, but how big in practical, real terms it should be—and how much can bank balance sheets take middle-class mortgages? You then get into the conforming size, and let me ask a fundamental question: if we, the GSEs, are there to channel a government subsidy to support middle-class, working-class mortgages, why should you expect others to compete well against us? Other than FHA [Federal Housing Administration], which is even more subsidized.

So I'm troubled by all this what I'll consider "thought process," which swirls around with what I don't think is the greatest rigor. If we're going to have a system in which the GSEs are there to support middle-class mortgages, then you expect a lot of the middle-class mortgages to end up with them (or again, FHA or VA [Veterans Affairs]). And above that, you would expect the private market to be very vibrant, since it has a completely open field. But even there, again, you run into this problem of the interest rate risk and the bank balance sheets, and PLS's inability to satisfy investor demands that they not be subject to adverse selection, which is still a struggle for them, that's the core.

And investors have moved on to CRT—GSE CRT—to satisfy their need for the diversification, so they have a tougher bar to jump...the PLS, therefore, has a tougher bar to jump to make investors happy.

Bostic: So we have about five minutes left, so I want to get a couple of questions. This one actually touches a little bit on what you were talking about in terms of possible models—your preferred view, we've talked a lot about utility, you've talked a lot about the private sector. And then the reason I like this is this question about sufficient capital, and what's the role that capital will play in a post-conservatorship world?

Layton: Okay, I'm going to dramatically look at my watch. On July 1, you get a different answer than today. Part of GSE conservatorship is, we're not supposed to take public policy positions, so I'm going to dance around this a little bit. Conservatorship, to get out of, is hard. You have to solve a bunch of political requirements, you have to solve some actual legal financial requirements that are like "How much capital?" You have to solve some regulatory requirements, "How much capital do we need?" Technically, the rule to set that, to get out of conservatorship—how much capital do you need?—was out for comment, has not been released. It's being worked on by FHFA, it's not out.

So we don't even know what the end of the goal line is yet. Presumably, in the next few months, they hopefully will do that. Now, the next step is, is it legislative, or is it administrative? "Legislative" meaning Congress changes the rules, it passes laws. The conventional wisdom in Washington—almost universally held—is that this is very likely to happen in the near term (maybe even the middle term). So they are focused on what's called "administrative reform," which can't change the rules—the legal rules—but it can get us out of conservatorship.

Administrative reform takes the form of amending the PSPA, the support agreement, so you go to the company, "You need our government support? Here's your rules. You have to adhere to a capital system that looks like this: you have to keep level G fees, a very important political requirement, for the industry support. You have to adhere to an investment portfolio limit, and five or six of these key things can be either kept or added in there, and yet you're now...and our support to you." We need that government support for our business model to work. No one's buying the MBS just because we're nice guys and a good company, they want government support.

So if the government support stays in the PSPA, investors again are very adamant, they need that government support, the PSPA is kind of it, it seems—no one's aware of any other alternative, absent legislation. It has to stay as part of this new deal, where we are supported by the government on contract terms, and the regulator is the regulator. So you have that.

There are also these legal requirements with the existing shareholders. Someone doesn't realize, when the government rescues companies—whether it's the GSEs or AIG [American International Group]—have you noticed how they do these weird things with senior-preferred stock, and then a warrant for 79 percent of the stock? Does anyone here, other than me, know why they do this? GAAP [generally accepted accounting principles], he says "GAAP"—close. The government is not subject to GAAP, but there actually is a part of the government that sets the accounting rules for what is on the books of the government and what is not. And the same way as GAAP, they basically say if the government owns 80 percent or more, the balance sheet gets added to—which means they would have blown the debt limit because these companies are so big, whether it's AIG.

So they do this very corporate finance, GAAP-style we use, even though GAAP is not there. I believe an office in OMB [Office of Management and Budget] sets the accounting rules. And so they do the senior preferred to get the equity that the world needs to support the liabilities, and then they take the 79, or 79.9, percent warrant.

So they never intended for the existing shareholders to have rights, but they had to leave rights—they had to leave them in place. So getting through that is not a minor affair. That's the lawsuits, there's talk about running the companies through receivership, which is not general receivership, receivership as specified in a particular law about the GSEs, to wipe these people out. We're not to do that, whether that's practical or too risky, there are all these things.

So I would just say that as a general result of my seven years in Washington, I would think...whatever choices you have, the lower-risk route I think should be highly considered by the government officials making decisions. Because there's lots of risk here, and if you get it wrong the possible damage is high.

The last item is raise sufficient capital. We don't have the final capital rule. If the one that was proposed—that's been commented on—is not changed, the number for us would be about $50 billion. Basically, the law kind of specifies what counts as capital, so the vast majority of that's going to be common equity. That's a lot of money. Fannie would obviously need proportionately more, so between the two of them you're talking maybe ballpark $125 billion—and in theory, that number comes down as more credit risk transfer is done, but in A basis, that is a large number. The largest IPO [initial public offering] in history, by the way, was Alibaba for $25 billion. The largest American one was, I believe, Visa credit card when they went public: $11 or $12 billion. In other words, you're not doing this in one week, in one gulp.

The low-risk way is to retain earnings, stop the sweep. That would put, in our case, about $8 billion a year on the books. If you take 50, declining with CRT at eight, you get about five or six years. You can talk about a baseline then being retained earnings for four or five years, and an IPO of some billion dollars to get you over the line. And the new investors come in and the company's out of conservatorship, so they get normal voting rights. Still have to clear the decks with the existing lawsuits, I don't know if they're going to invest with those existing lawsuits there, in fact, I question it.

Everyone goes, "That takes a long time, can't we accelerate this?" I'm not sure you can. I know there are a lot of people who are looking to do it, to try—we'll see how it goes. The fundamental issue in doing so will be: can you raise equity from investors who will not have classic voting control of a normal company with a board, and such? If the government exercises its warrant, it has 79 percent ownership. Are you going to be a minority investor to Treasury?

So it's not clear how this is going to work, or this sale. I just know equity investors are going to have this issue front and center, to be addressed. And if the government would like to accelerate it, versus that five-ish years, it's going to have to address those control issues to get over that hump. This is unprecedented, to my knowledge, and since at one point in my life I was running the investment bank at JP Morgan Chase, I feel pretty good about the knowledge base that this is highly unusual.

Bostic: Well, Don, thank you very much. We are out of time, but this has been very interesting. You've really merged the theoretical stuff with the practical, operational stuff—and, I think, really made clear how moving to that next level is going to be quite difficult. Thank you so much for being here.

Layton: Great, all right. Thank you, everyone.