0:43:04 … If we’re going to offer this new advantage to the homeowner, and allow him to borrow at today’s current market rates, but only because the government’s going to guarantee it, we should have full recourse to every borrower. – Alan Boyce
Doom Transcripts: Index & Guide
Housing Doom is pleased present an almost12 complete unauthorized annotated transcript V.A for the American Enterprise Institute’s disturbing7 March 26, 2009 event "Can Elements of the Danish Mortgage System Fix Mortgage Securitization in the United States?"1 The event site has a variety of resources including a summary and both an audio and a video of the proceedings. There is as yet no official transcript.
UPDATE 6/24 2009: I’ve added a short summary and some comment here.
Table of Contents
[link navigation works best when full article displayed]
- 0:00:00 Peter Wallison intro
- 0:07:20 Alan Boyce presentation
- 0:46:08 Wallison & Boyce dialog
- 0:48:18 Bert Ely discussion
- 0:59:23 Boyce response
- 1:04:31 Olivier Hassler discussion
- 1:16:30 Boyce response
- 1:19:50 Alex Pollock discussion
- 1:31:30 Boyce response
- 1:33:52 Panel discussion
- 1:45:22 Q&A
- 1:56:14 (end)
Peter Wallison: [0:00:00] We have a really interesting program today, and one [crosstalk] … That’s right. I’m sure you’re going to "tax" us up here.
We really have an interesting program today, and let me tell you how we’re going to work this. We’ll have a presentation by Alan Boyce about his plan, and then we’ll have each of the people on the panel here comment, and then some crosstalk in the panel, and then some questions from the audience.
So if things come up in the course of Alan’s presentation or elsewhere, please make a note so you’re able to ask some questions when the time comes.
I’ll start with a small introduction and then … I’ll introduce Alan, and then the members of the panel before they get started with their commentary.
Although members of Congress and the Obama Administration have sweeping ideas for how to regulate the US economy in the future, few of them, apparently have thought very deeply about how we finance mortgages. Yet at the root of the country’s current financial crisis is a dysfunctional mortgage system.
The central actors in that system are two bankrupt companies, Fannie and Freddie. And they are likely cost taxpayers $400 billion by the time all their losses are toted up, and their conservatorship brought to an end. Their lack of adequate regulation and their domination of housing finance were, more than any other factor, responsible for the financial crisis we now confront.
In light of this, one would think that, rather than planning to extend regulation to the farthest reaches of the financial system — regulation, incidentally, that has not worked, as we can see particularly well with the banking industry — policymakers would spend a little time thinking about how to reform a mortgage finance system that has obviously got major problems.
But in listening to Barney Frank and others, the most serious thinking they have done is to require all participants in the securitization process that we have today to have some skin in the game by taking a piece of the risk associated with the securitized mortgage.
Now this is a perfectly reasonable idea, but it raises all kinds of thorny technical issues, and is not remotely as substantial a change as they are promoting in the regulation of the financial system as a whole.
To its credit, the Bush Administration, as one of its last policy efforts in the financial area, published an interesting and favorable report on how a covered bond system would work. We had a conference on this issue in September,2 and Alex (to my right) and Bert Ely, two of our panelists today, were discussants there.
Suffice it to say that there were plenty of questions about the Treasury and the FDIC plan, about the covered bond idea that had to be answered before that plan could be considered workable. We’ll probably have some serious questions today about Alan’s approach. But at least someone was thinking at the time.
In any event, we’ve heard no more about it in the 6 months since the conference, and given the fact that Treasury is now woefully understaffed, and the FDIC is fully engaged in the current problems of the banking system and the financial industry, it will be a while, if ever, before anyone at either agency returns his or her attention to considering the implementation of a covered bond program.
One of the useful things that we do here at AEI, I hope, is to look a little bit ahead of the day-to-day controversies that absorb Washington and try to consider the policy issues that need attention. The restructuring of a housing finance system certainly qualifies on that score.
That’s why we’re having this conference today. And despite the fact that everyone’s attention is on the sexier question of who will regulate whom, and how much, [laughter] it’s worth talking about mortgage finance.
I have long been intrigued, myself, by the Danish mortgage system. This is the second or third conference in which it has been discussed in some way. But it’s the first one at which the Danish system is actually the focus of attention.
It is attractive to me because it does not involve any explicit or implicit taxpayer liability. It permits homeowners to take advantage of declines in housing values, which is a particular advantage in the US market today, as we can see, and appropriately separates interest rate and credit risk.
If these advantages can be retained, if it is integrated in some way into the US system, [0:05:00] we would indeed have an improved system over what we have today.
When Alan Boyce [PDF], an expert in the Danish system, volunteered to describe how the Danish system might be grafted onto the US securitization system, it was an irresistible opportunity to me to see what advantages — how the advantages of the Danish system could be integrated or grafted onto our own. As you will see, Alan’s plan does not quite do away entirely with government involvement in liability. Probably nothing that emerges from a Soros shop ever will. [laughter] But he outlines how other advantages of the Danish system might be available in a restructured US market.
So I want to thank all of you for coming, and displaying your interest in mortgage finance. Alan has been talking to people throughout the Administration. I’m sure he’s getting positive responses from them, so we might actually see something like this coming along. There are a lot of issues associated with it, as you will understand. But it is a very interesting approach.
Let me just introduce Alan for you, and then we’ll proceed with the rest of the conference.
Alan is the Chief Executive Officer of Absalon, which is a joint venture between George Soros3 and the Danish financial system that is assisting in the organization of a standardized mortgage backed securities market in Mexico.
Alan was the Senior Managing Director for investment strategy at CountryWide Financial Corporation, where he was responsible for secondary markets, the hedging of mortgage servicing rights, and the balance sheet for CountryWide Bank and Balboa Insurance.
Alan was the Director of Special Situations at the Soros Funds Fund Management from 1999 to 2007, where he managed a portfolio of assets of the Quantum Funds and had principle operational responsibilities for the bulk of the fund’s investments in Latin America.
Before joining Soros, he served for 14 years as a Managing Director in charge of fixed income arbitrage at Banker’s Trust.
Alan, the floor is yours.
Alan Boyce: I’ve got a big slide deck here. [slide4 1] I’m not going to be able to touch on every point, because I really want to get to the discussion phase, so I’m going to try to fly through this quick. But I urge you guys to go to AEI’s excellent web site and download it. And I’m sure you can figure out how to get in touch with me afterwards if you have any other questions.
Peter Wallison: I should mention here that we are no longer passing out all the information that comes at a conference. We are posting it on our web site, including the bios for everyone who is here at the speaker table.
Alan Boyce: So to start off with, I’m not going to spend too much time on the Introduction. [slide 2] I think all of you guys in this room understand the serious nature of the financial difficulties facing the planet. And I’d say the root cause of them could be traced back to the problems in the US residential mortgage market. [slide 3]
As Peter described, my job function now is actually to try to encourage the policy makers in this city to think and implement the positive attributes of the Danish mortgage system. We are actually, in particular, got some action steps here. And the 3 action steps that we really want people to focus on is [slide 4] –
- We need to be reducing interest rates for all mortage borrowers who are current on their mortgages today.
- We need to focus directly on trying to reduce the number of homeowners who have negative equity, and this is a situation I’m not going to discuss a lot, but I’m a firm believer that we are now in a single-trigger default model in the United States, and at the expectation of housing price depreciation based on the Case-Shiller index for the next 12 months, that would put about half the mortgage … homeowners with mortgages in the United States in an underwater situation. That would be 30 million mortgages in the United States. This is a systemic crisis that needs to be addressed, and in particular, and in all seriousness.
- The last one is while we’re in the process of trying to fix this problem, this would be the perfect time to introduce a new mortgage system based on the positive attributes that are embedded in the Danish mortgage system, in particular there are 4 points:
- Standardization;
- Transparency;
- Interest Alignment; and,
- Counter-cyclical Properties;
sadly, very few exist in our current system.
I’m going to try to skip through a lot of this background, but I think there’s a few important points here. The old US mortgage system is gone. It has been entirely replaced by one that is completely dependent upon US government guarantees. There is no way, [0:10:00] no how, for the foreseeable future, that any mortgage backed security, or covered bond with mortgages as its asset, can be sold without the Federal Eagle.
Now I have been one of the leading fighters in the GSE wars over the last decade, and it is going to amaze people in this room that know me forever, but we’re actually going to have to rebuild the US mortgage system on the back of a US government guarantee. That is being realistic. I’m not going to try to exclude the possibility that we can migrate to a private sector based system in the medium term, but I’ve been on Wall Street since the early ’80s, and I think we’ve broken things so badly that it won’t happen while I’m working.
The second point, which I think we’ll all agree on now, is that Fannie and Freddie have been de facto nationalized. I spent a good part of my life fighting against the most egregious behaviors that came out of those two GSEs, but I believe now that they’re actually a key to help solving the problem. Another thing that I’m amazed I’m saying these days.
The comprehensive reform faces a lot of big obstacles, obviously. It …
Alex Pollock: … shows you’re intellectually flexible Alan, and that’s good.
Alan Boyce: [laughs] The financial problems have now leaked into the real economy, and we’re in a serious global recession right now, which makes it much harder to deal with problems. As unemployment skyrockets, as aggregate demand plummets, as every homeowner curls up into a ball and tries to protect himself from all real and perceived evils.
The past behavior, in particular in the non-Agency mortgage market, has created a complex web of almost impossible to untangle securitizations, one in which the owners of the various tranches — AAAs down to BBBs and even within AAAs — have distinct and often conflicting goals. And being in a litiguous society, and I think there may be a few lawyers in this room, it’s something that isn’t going to be solved easily.
I think, just to talk about before, that actually to reduce quickly the number of homeowners that have negative equity is a very hard problem, and in all fairness I believe that most of the current solutions being worked on and proposed are based on historical models from 20 years ago, which everyone believed required a two-trigger event to happen before there were defaults: that is, negative equity and a life event. And I believe that basing today’s solutions on our parents’ mortgages is incorrect, we’re in a single-trigger universe.
And then — let’s just go a little bit faster here — so just to talk quickly about successful reform: lowering mortgage rates, limiting foreclosures, and moving the system forward on a sound basis. It’s this last one that I’m really going to talk about.
And the trick here is, that we need to cleanly separate credit risk from interest rate risk at birth. We need to get the mortgage intermediaries — mortgage banks, mortgage credit institutions — to become advisors to homeowners.
We’ve got a situation in the United States where the biggest financial transaction of everybody’s life, which is taking on a mortgage, your counterparty is actually somebody who’s not aligned with you at all, and is trying to make as much money out of the transaction as possible.
And we employ financial advisors for all our other financial decisions. In those situations, those financial advisors are aligned with us. They sell their services based on their past performance. We employ advisors for our brokerage accounts, for our pension funds, and every other major financial decision. And we need to get a situation where our mortgage bankers, the mortgage credit institutions we deal with, become our liability advisors. Because this is the biggest and most important financial transaction of any homeowner’s life. The second thing we need to do is we need to make sure that their interests are perpetually aligned with us, which I’ll talk about more.
By separating credit and interest rate risk, you really allow the system to leave the risks in the places where they are best able to handle them. The fact of the matter is, nobody will know as much about the credit risk of an individual homeowner as the person that they immediately deal with. His branch manager will know a little less. The regional manager will know a little less. The underwriters at the regional office will certainly know less. The main office will know less, and when they are securitized, and the whole origination process sheds all that credit risk, we’ve proved to ourselves that tranching out credit risk actually does not transmit information, and the people who bought that credit risk didn’t know what they were doing.
So I argue strongly that credit risk should be retained, as much as possible, at the lower level. I would love to copy [0:15:00] Denmark exactly, where the credit risk is entirely retained by the mortgage credit institutions, who are private sector players with no government guarantees. However, I don’t believe that’s possible.
I do however, believe that some percentage of that credit risk can be mandated to be perpetually retained, and we’ll talk more about that in a second.
The second thing on splitting interest risk from credit risk is that the bond market is actually particularly good, and in this crisis retained the ability to manage interest rate risk. One of the things that didn’t break in this terrible financial crisis was the bond market. The treasury market, the interest rate swap markets, swaptions, caps floors, the Eurodollar futures market, those all functioned perfectly. And in fact in almost all the G7 Countries those were the only markets, along with foreign exchange, that continued to function.
So if we can cleanly separate credit risk from interest rate risk, the bond market has proved its ability to handle this risk, hedge it out, without creating systemic difficulties.
Now just to talk about quickly, in the United States, when we talk about the old system, this is what the old system is in the United States. [slide 5] Homeowners, often through brokers, go through mortgage banks. The mortgage banks decide, in a rather opaque fashion, whether the loan is going to go through a non-Agency securitization or a GSE securitization. This [presumably the "non-Agency"] doesn’t exist any more. But in both cases, the credit risk is passed right through. In the GSE case, the credit risk stops at the GSE, but in general both pass through together.
Now it’s also a misnomer for people to believe that all the good loans were guaranteed by the GSEs, and all the bad loans were in the non-Agency securitizations. At the January conference,5 Ed Pinto [scroll down] did a very good job of elaborating on how much inappropriate risk was guaranteed by the GSEs in their search for yield. But equally so, a lot of conventional mortgages, ones with especially low LTVs and high FICOs, ended up in non-Agency deals as well.
This misconception is actually leading to some, in my opinion, bad policy mistakes, because we’re focusing on trying to refinance people if their loan happened to end up in a GSE guaranteed pool in the past. We’re excluding these [non-Agency] borrowers. There’s actually a lot of good borrowers who need help up here, and a lot of bad borrowers that probably don’t deserve the help down here. [slide 5 -- upper and lower paths]
So that’s our system that we’ve had up to date. The system can be fixed, in my opinion, by trying — sorry, can you guys hear me? in the back OK? –
Karen: [inaudible]
Alan Boyce: oh … all right, sorry … The system can be fixed by trying to emulate the Danish system, [slide 6] and the Danish system, the mortgage credit institution, all the risk that they’re allowed to take is credit risk. In fact the Mortgage Credit Act, which is the only involvement of the government in the Danish mortgage system, stipulates that all other risks, like interest rate risks, and roll risk, are not allowed to be taken. The idea being that a mortgage credit institution is only allowed to take the risk that it is in particular best placed to take and deal with, and that all other risks it’s not allowed to because any other risk is a potential to blow up that mortgage credit institution.
So what happens is the MCIs [that's the Danish Mortgage Credit Institutions] wash the credit risk off of the bonds. And then what the bond market gets ultimately is pure interest rate risk. And you don’t have to deal with rate risk, slope of curve risk, curvature risk, volatility risk, everything that the bond market is trained and has the tools and the ability to hedge.
The MCIs all compete with each other on a level playing field. There is no government subsidy to one side or the other. The MCIs, as I discussed before, in Denmark are perfectly aligned with the borrower. So they do everything possible to improve the borrower’s situation. And in that respect they are really an MI [mortgage insurance?] company that does the paperwork, but not only that, they’re like a property casualty insurer. They go through and say, "Wow, if you add fire exit doors here, and put smoke alarms there, and maybe some sprinklers here, your building will be safer, and your rates will be cheaper.
So in the case of mortgages in Denmark, the liability advisor is constantly searching for the most expensive bond that the homeowner could potentially issue a mortgage into. And they make fees by advising the homeower to refinance their existing mortgage and do a more expensive bond, which is a cheaper liability. And every time they can find a positive net present value transaction for the homeowner, they are both reducing the risk to the homeowner, and therefore the risk to themselves. It’s a perfectly aligned system.
In Denmark, the mortgage banks are so separated from interest rate risk — and this is very important — that they don’t actually fund the loans. [0:20:00] And this is something that is key.
The mortgage banks set up what called "bond series" in Denmark, It’s basically a trust run by the Central Securities Depository, and the homeowner gets the price of where the bond market trades the day he wants his loan funded. So for example if you want to take out a 5 percent 30 year callable mortgage, and the 5 percent 30 year callable bonds are trading at 99 1/2 today, you issue your loan into that bond, and you get the proceeds from the sale of that bond.
In Denmark, they have what’s called the Principle of Balance, and I’m going to refer to this time and time again. And if you download my paper [6 - PDF] from the web site, you’ll find a … "PoB" all over the place.
Principle of Balance means that all loans are exactly funded by the issuance of a bond with the exact same characteristics. And this is a very important going in, because it leads to all sorts of positive externalities for the life of the loan, which actually leads to a better situation for both the bond holder, the mortgage credit institution and the borrower.
As we’ve talked about before, the bond market knows how to deal only with their interest rate risks. The asymmetries that come from the US system are eliminated. The key thing, if you stick with the Principle of Balance in Denmark is what they call Optional Redemption. And that means, if interest rates go up, the homeowner’s liability is the face amount of his debt times the price of the bond which, by the way, is how every other bond market in the world works. That’s how sovereign debt works, that’s how corporate bonds work.
Let me just give you a quick example. When Larry Summers was Treasury Secretary, we were running a budget surplus. He decided to buy back some Treasury Debt. He bought back some long bonds that were trading at discounts. He didn’t pay par for those long bonds, if he did he would have been run out of office. He paid the market price.
When corporate treasurers want to buy back debt, if some of their debt trades at a discount, they don’t pay par, they pay the market price. But we have created a system in the United States where legally the homeowner is required to redeem his debt at par, regardless of where it trades.
So now, in this room, I’m sure a few of these people have mortgages, so let’s take Alex for example. Alex has got a nice house somewhere outside of Washington, DC. He’s got a non-Agency mortgage, that non-Agency mortgage is in a mortgage backed security that’s trading at a price of about 60. All of us in this room can go buy the cash flows from his mortgage at the market price. 60. Except Alex.
Economically, the market doesn’t matter for him, he has to follow the legal contract, which doesn’t have the Principle of Balance written in, so he has to pay par.
Now Alex is a respectable citizen, and he’s an older generation, so he’s of a twin-default model, so he’s going to follow his contract.
But maybe his kids say, "Wow! I’m the fool at this party, I’m the patsy at the poker game. Why should I pay par when everyone else can buy it for 60?" And that’s actually the situation that we’re in.
So what I’m arguing is, if we had had this system existing today, we would not have the problem in the US mortgage market, and we probably would not have blown up the global financial system.
So, let’s just go over this one more time. I’m going to pound this into your skulls.
The current system. [slide 7] If interest rates decline, housing prices go up, home owners — we have callable mortgages — they can prepay their existing mortgage at par. This allows for equity withdrawal. It’s a great way to put real action into Federal Reserve easing, it drops mortgage rates, everyone refinances, right? … puts cash in peoples’ pockets, pumps up aggregate demand, it’s great. It’s one of the greatest transmission mechanisms for monetary policy known to man.
If interest rates go up, sadly, house prices go down. The value of the mortgage drops in the market as interest rates went up, but the homeowner can’t take advantage of that. He can’t prepay his existing mortgage as we just discussed. And what happens is you get into a situation where the homeowners maybe in aggregate owe trillions of dollars more than their houses are worth. It’s a systemic based on a set of individual contracts.
How it could be [slide 8]:
Interest rates decline, it’s the same. The fact is, in Denmark they’ve got all the identical loan products that we do in the United States, with the exception of pay-option mortgages. They’ve got 30-year fully amortizing callable mortgages, they’ve got 30-year interest-only, they’ve got 5-year hybrids, they’ve got 1-year floaters, they’ve got ‘em all. But it’s predominantly a 30-year fixed market.
But the flip side is, when interest rates go up, [0:25:00] the homeowner has the ability to buy his debt back. They can prepay by purchasing back their loan by buying a pari passu amount of the bonds.
And this is a very important thing, because when interest rates rise, this is a market-based mechanism that maintains equity in the home.
It’s not actually a system whereby you can take advantage of the fact that your house went down, it’s the fact that the major systemic driver for making housing prices go down is mortgage rates went up. And you can take advantage of the fact that your bond is trading at a discount. And the key to this is to insert this, [points at slide] which requires the Principle of Balance at loan origination, into the United States.
It’s actually quite simple, and I’ll talk more about that later, but given the situation that the US mortgage market is in today, where 99 percent of the mortgages go through a Ginnie, Fannie or Freddie securitization process, which is highly standardized, we’re like 2 or 3 very small changes in, literally software code, to go to set ourselves up for this situation.
Now lets do a bit of T-accounts here. [slide 9 -- speaker gets up and goes to screen] All of us at one point in time had to take a little bit of accounting, but this is very simple.
At origination, both Denmark and the United States have sort of a standard mortgage, where if you buy a house for $100[,000] your max LTV is 80 [percent] and you have to put up $20[,000] of that [the 100]. So I’ll go through two different examples, and I did these examples on September 9, [2008] so this was when the crisis was getting going, but before Lehman blew up and the world almost ended.
So we’ll take the first one of an Agency loan. Under our existing system, housing prices on average in the United States were, say, down an average of 10 percent from their highs. So your house had gone from par to 90. On that day, Fannie Fives [FN 5% mortgage bond prices], which is sort of the biggest coupon of Agency guaranteed mortgages, were trading at a dollar price of 94. Generally all issued around par.
But in the US system it doesn’t matter where the bond trades, you own par. So your loan’s still worth 80, so what that means is very simply your equity has dropped by 50 percent.
If we had the Principle of Balance, something that would be no skin off the backs of the bondholders, your loan would be 75 [==] 80 [times] Market Price. Or, another way of looking at it is, your equity would have dropped by half as much.
hmmm … Wow, ex post maybe it looks like this would have been a pretty good idea.
But lets take a more extreme example. Let’s take the non-Agency loan. And the sad fact of the matter is in the United States, that we’ve got entire metropolitan areas that can only be financed by a non-Agency loan. The GSE loan limit has not kept up with housing price depreciation [sic] in certain metropolitan areas.
Some of these metropolitan areas are geographically constrained by oceans and rivers and mountains and deserts, so … and they’re desirable places to live, so housing prices actually are always going to be higher there.
So under the existing system, those houses have dropped by 30 percent. You get a situation where the homeowner is actually upside down on his house. And what we see in those areas is delinquencies rising at insane rates, you’ve got huge percentages of homeowners underwater, it feeds on itself with no hope of stopping it.
If we had the Principle of Balance, at that point in time, non-Agency mortgage backed securities of high quality were trading around 75. If we had the Principle of Balance system, your liability would have been worth the face amount times the market price, for an economic value of 60. Your house would still be down 30 percent, but you would still have positive equity in your house.
This is very simple T-accounting. But this is, right here, is the whole trick about the Danish mortgage market. This and only this.
Now sadly, today, we’re March 26th, this number here [non-Agency MBS prices] is nowhere close to 75. It’s more like 50. And we’ve got an alphabet soup of Federal government programs that I’m really hoping can try to get these prices up. Because the lower those prices go, the more negative equity also flows through to all the financial institutions in the world that own these assets. Herein is the simple math.
Don’t for a second [video is showing "Exhibit 6" from Boyce's paper [5] ] here think that there’s something particular about Denmark and the Danish people that makes this work in Denmark and not in the United States. As we’ve already discussed, they’ve got the same kind of mortgage products. Denmark had a housing bubble just like we did. Some places in Denmark, housing prices went up 5 or 6 times, like appartments in Copenhagen. Now they came from a very low base. But it was an amazing move.
Well, there were also places in Denmark like, [0:30:00] you know, little towns out in the farm country of Jutland where housing prices didn’t move. Just like a lot of places in the United States. And if you go to Des Moines, they never had the boom of the 90s and this decade.
But the flip side is, in Denmark this housing bubble did not lead to a giant pile of foreclosures. In fact, the number of foreclosures in Denmark, and the number of underwater homeowners is a very low rate. And this is a country — you think like, "oh, great tidy Denmark" — lets talk about some of the things — the attributes of Denmark.
It’s a little European country, outside the Euro, highest percentage of mortgages to GNP [sic] on the planet, and any one of those things was enough to blow up another small European country. Yet what did Denmark do? Their 10-year sovereign debt tighened to Bunds[ph], it was the only country in Europe where their 10-year sovereign debt spreads actually tightened. They held their peg to the Euro, their central bank extended less credit to the mortgage sector than any central bank in the world, their mortgage market was open every single day, and, because their mortgage market was open, it enabled foreign investors to have some asset that they could possibly sell.
So foreign participation in the Danish mortgage market shrank by about 40 percent. It was essentially a bunch of German money managers and insurance companies that sold their Danish mortgage bonds, because they could. They couldn’t sell the Pfandbriefe[ph] because that market ceased to function. And you can find all these statistics on the Danish central bank’s web site, but I view that to be a remarkable progress.
Here we go, second slide … and avoided foreclosure. [slide looks similar to the last chart in the Boyce paper] Now you guys might want to talk, just go aside. Denmark had high levels of foreclosure in the early 90s. Just as an aside here, we’re constantly talking in the United States about whether we should follow the Japanese or the Swedish model for dealing with bad banks. It’s a complete — we’re asking the wrong question.
Both Sweden and Denmark ran really bad fiscal and monetary policies in the 70s and early 80s. And they both decided in the late 80s to go on a potato diet. In Sweden they had a poorly structured mortgage market which blew up their banks. So then they had to decide how to fix their banks. In Denmark they had a well structured mortgage market, and through the Principle of Balance, and separating clearly the credit risk from the interest rate risk, their mortgage market didn’t blow up their banks. Therefore, we should be trying to copy the Danish model so that we don’t get in a situation where we have to fix the banks later on.
But it did blow up quite a few homeowners, so in the early 90s the Danish system did have to flush out quite a few homeowners that took inappropriate financial decisions.
Now, to sort of reduce people’s fear of the uncertainty, there’s also a lot of talk in this town about the need for transparency. Turns out that Denmark, which is by the way about the second biggest mortgage market in the world outside after the United States — only 5 million people in Denmark. It’s the size of New Jersey. But it’s bigger than Germany, it’s bigger than Spain, it’s bigger than the UK. €360 billion in size. OK, and it’s almost $500 billion, its almost as big as Ginnie Mae’s. That their mortgage market, all the details of their bond system, are reported every day.
They’re reported on the Copenhagen Stock Exchange. And the fact is the Copenhagen Stock Exchange merged with the other Nordic stock exchanges into a company called OMX, and OMX merged with NASDAQ. So NASDAQ — a functional market in the United States, because it is a transparent market, publishes every day, on their web site, [slide 10] everything that happens in the Danish Mortgage Market.
These are freely available systems. The fact is, nobody in the United States knows about it. We’re clamoring around trying to find, "how to get transparency? how can we publish this information?" and NASDAQ already has the systems.
If you go into NASDAQOMX Nordic web site, where you can go into NASDAQOMX and have to click one more time, you get to this point-and-click screen. [slide 10] And it’s very easy. Click on the mortgage and credit and special institutions button, you can choose an issuer if you want, you can choose an interest rate, you can choose a year of expiry, they do their mortgages by the year of final maturities, so the 2038s was the 30-year mortgages that mature 30 years from 2008, so this is where the recently closed issues were issued into. If you click on any one of the securities, [slide 11] you get a bunch more tabs so, for example, on this one it gives you all the statistics about the bond that was traded today — there’s some price history, [0:35:00] however you want to go. It gives you news, [slide 12] so there’s a series of news releases — each one of these news releases is basically comes with an attachment.
This the publicly available web site, but NASDAQ, most of these are, they turn into feeds to the broker/dealers who trade this. And these are basically a bunch of flat files that come out at 4PM Copenhagen Time every day, and at 4:01 the broker/dealers are able to transmit those flat files into their daily research reports.
Daily, you find out information on what the … how many bonds were created, where they traded, if you click on the trade here, every single trade in the system is logged. There’ll be a note on whether it was an exchange-based trade, or OTC, [over-the-counter] … skip ahead [slide 13] … so for example today’s key figures. [slide 14] This comes from New Credit, which is one of the biggest mortgage banks and also has a big broker/dealer, so this is detailed information about a bunch of different bond series. So it’s information about whether the … how many people … the pre-pays that have accumulated to date, whether it’s a private homeowner or a commercial building, the break-down by loan size with smaller loans being more valuable to the bondholders, as these are callable loans, the smaller borrowers are less likely to efficiently exercise their call options. This comes out at 4:02 or 4:03 every day.
Some of the other broker/dealers have faster computers and will take that flat file and turn it out to investors by 4:01.
Everybody in Denmark gets the information at the same time, there’s no informational advantages, something that plagues the US mortgage market … let me just go back a second here [slide 13] … On a monthly basis, every mortgage bank publishes a 10-page report, which is not yet up on NASDAQ, but they’re freely available. This is a screen shot of one of the reports which shows buy-vintage, so that’s, say, loans made in ‘05 or ‘06 or ‘07; what the original status was, and this would basically be the LTV bucket at the time of origination, so the number of loans, and then, as of here, I think December 31st, what the current LTV is.
So what you’re seeing is, there has been an increase in the number of 80+ loans, but the current LTV here is the current house price, done by their equivalent of an automated valuation model, divided by the fair value of the mortgage. The fair value of the mortgage being the face amount times the price.
Now I asked the Danes whether they could give me this report with another line in it — how we would calculate it, right? Which would be the par value of the mortgage, and they all responded, like, "Why would we give that? Who cares about the par value of the mortgage? The homeowner owes the lower of par or the market price." But they’re going to get me the information anyway.
Just to point out, if … and I will tell you one thing. If we do it the way the US shows, this would show a whole bunch of loans in the 80+, and maybe even in the 100+ [i.e. underwater] bucket. But it’s irrelevant, because the market price of the debt is what matters to the homeowner. … sorry here [moving slides around] …
All right. [slide 15] We’re going to do a little bit of line graphs, because I want to … I want to pound this into everybody’s head.
Simple price/yield graph. And just to make sure where credit is due, I’ve copied this form of analysis from a good friend of mine, Hakeem Dubal(ph), who’s a very outspoken and absolutely correct financial economist in Europe, who’s probably the leading mortgage expert in the world.
And the very simple idea, when interest rates go up, housing prices go down, and vice versa.
When it comes to mortgage bonds, mortgage bonds will also follow that path. From a homeowner’s perspective, if you take out an adjustable rate mortgage, it’s always worth par, and if you take out a fixed-rate mortgage, when rates go up, you still owe par.
Now you may be in a situation where your house is going down. Economically, this creates negative equity. Another way of looking at it is you might say, "well, I can still afford the payment," but now you’re locked into your house, right? At a minimum, what we’ve done is guarantee labor mobility goes to zero for anybody that has a low coupon mortgage in the United States and their house went down in price. They’ve just been nailed to the ground.
And last I checked, labor mobility was a positive attribute of our country.
If you switch to the Principle of Balance, the homeowner’s liability is where the bond trades. [slide 16] So let’s see what happens. Interest rates go up, his asset’s going down, his liability’s going down. Wow! That seems much more symmetric. It’s actually the right way to do it.
All right. Let me go talk a little bit, because I promised Peter I’d try to delve in the real world. [0:40:00]
This is a unique opportunity to get it right. [slide 17] And I believe that the GSEs should be transformed, it can happen slowly, into Principle of Balance guarantee vehicles only. This requires first that they be merged, they’re going to have to build some new systems, there’s no reason to build 2 new systems, they don’t compete with each other in any way shape or form any more. I think their portfolios should be run off. It was never something that helped the US mortgage market. They used their portfolios to try to earn money and make their share prices go up.
Credit risk should be shared between the originators and the federal government. I would love to have the mortgage originators in the United States have most or all of the credit risk, but practically, you’d never be able to sell a mortgage bond.
I argue that 10 percent is about the right number. I originally picked it as a place holder, and it turned out that that’s sort of where all the arguments center on. The mortgage bankers in the United States think that 5 percent is as much as they can hold, and a lot of people in Washington DC and elsewhere think that they ought to hold 20 percent.
If it’s 10 percent, that, I think, leads to the mortgage banks having to hold about 2 percent capital against that 10 percent guarantee.
Borrower, obviously, it’s a market-rate based on transparent bond pricing; that’s actually what’s happening today. Fannie 4 1/2s trade at 102, Fannie 4s trade at par. Where that [unintelligible] you issue into the bond. The loan is cancelable at the lower of par, just like today’s callable mortgage rates, or where it trades. So if it trades at a discount, the homeowner has the ability to buy it back.
The trick here though is we need to issue very standardized, transparent, and very large bonds. So this requires mortgage banks to make big bonds. In Denmark, they do this via tap issuance [PDF] over long periods of time. I’d say that in the United States, since we’re 80 times bigger than Denmark, we don’t have to have multi-year tapping periods, we could have tapping periods that last for 30 to 75 days at the most.
The fact is that before Freddie Mac even mutualized, it ran most of its mortgage program through the cash window, and Freddie Mac created bonds on a monthly basis and every day had an auction, and they tap-issued into that bond. And the loans that came into the cash window were priced where the auction priced out every day.
So we’ve actually done this in the United States in my career. Now when Freddie Mac demutualized, they realized that that was too transparent for them to make money out of. They ceased that, and started some opaque practices instead that benefited themselves, but not the homeowners or the bondholders.
That’s something that, literally, the software code exists at Freddie Mac today, and at all the big mortgage banks. We just have to find some old retired guys, pull them out, because they wrote the code in COBOL, and get them back and figure out where they hid that code.
I also argue that we’ve made a few other mistakes in the United States. If we’re going to offer this new advantage to the homeowner, and allow him to borrow at today’s current market rates, but only because the government’s going to guarantee it, we should have full recourse to every borrower.7
And not enforced at the state level because, number 1, the States are pretty good at ex post changing how laws are interpreted. It should be enforced at the federal level.7a
The … I think it’s really important that we create a sense of responsibility among homeowners. Because today, recourse is not being employed.
I think the Federal Reserve needs to assist. There’s a very interest-aligning system here. If you go into the later slides, which I won’t get a chance to go over, this system reduces long-term interest rate volatility, it creates a bunch of shock absorbers for the system, it turns pro-cyclical behavior into counter-cyclical behavior. These are things that the Federal Reserve is increasingly concerned about.
This system should benefit from lower risk-capital ratings, lower margins at the various Federal Reserve borrowing lines, and so on.
This system is actually very supportive to Credit Unions and community banks, the people who actually know their customers.
It’s actually the financial system that didn’t blow up this time. I believe that there should be a unitary financial regulator, if possible, that’s going to be difficult.
And I also believe that — this is a regulatory discussion that we don’t have in the United States. But we talk about who should regulate what, and give them, what, death-sentence powers? But the fact is, in Denmark the regulator is a prudential regulator. He gets all the data real-time, just like everyone else, and decides, "I don’t like that loan. I think your LTV is wrong. I disagree with your appraisal. Take 40 percent of that loan out."
So they actually stop problems from happening at origination, which means that they never [0:45:00] have to wield the guillotine. And I think that’s very important. So all those last points, is what I think a financial regulator should do.
And then, just some other things here. [slide 18] I’m a firm believer that we should be offering a refinancing to every performing borrower in the United States. The fact is, you didn’t know where your loan went, … you could have been a bad borrower and your loan went to Fannie Mae and now you get a refinancing opportunity, you could have been a good borrower and it’s in a non-Agency. I think the GSEs should be reducing pooling and guarantee fees. They ought to be offering refinancing up to … if the go up to 900,000 it will get 99 percent of the performing borrowers in the United States.
And I think that the GSEs ought to be supplying this process, requiring appraisals, that industry completely fell down and didn’t do anything.
There’s a few other things. This modification program, the federal government’s decided on their own initiative, so I don’t want to discuss that too much, and from there on, I’ll leave to the rest of the panel, because I’m using up too much time.
Peter Wallison: Well you’ve used a lot of time, but you’ve used it well. Thank-you very much, Alan. Excellent, excellent presentation.
Let me just ask you one question I don’t yet understand, I’m afraid: why it is that in Denmark the mortgage banks function without any government support, but you believe it cannot work in the United States? Why is that?
Alan Boyce: I think the mortgage banks could function, but you couldn’t sell a bond.
Peter Wallison: Why not?
Alan Boyce: I don’t think any investor in the world would trust a private sector player right now.
Peter Wallison: Oh, right now …
Alan Boyce: … right now …
Peter Wallison: … but …
Alan Boyce: … I’m leaving open the possibility that we could transform to that in the medium term, but I’ve … I mean I’ve talked to investors around the world, like, I think we’ve so damaged people’s confidence in our processes and procedures that it’s very unlikely we’ll be able to sell pure private label mortgage backed securities, even done this way, for a very long period of time.
[crosstalk]
Peter Wallison: … there isn’t any theoretical problem …
Alan Boyce: No threoretical problem …
Peter Wallison: … you simply … the opinion of people around the world …
Alan Boyce: … yes … it’s purely practical.
Peter Wallison: OK. … Bert. Let me just say a few words about Bert, who is who is no stranger to our dias here. He is a financial institutions and monetary policy consultant. He’s got his own firm, Ely & Co. He’s quoted widely in the newspapers and elsewhere about his opinions on federal financing of mortgages and many many other things, the condition of banks. He’s well known for having predicted the decline and fall of the S&L industry well before anyone really thought it was a problem, and Bert has been one of the great critics of GSEs over his entire career, and certainly in the last 10 years or so when we have worked together a lot.
He’s worked on and developed an excellent plan for the privatization of the GSEs. And I’m delighted to have an opportunity to introduce him here and ask him to comment on Alan’s approach. … Bert?
Bert Ely: Peter, thank-you very much, and Alan, thank-you. [slide8 1] I will just say up front that I have some reservations about Alan’s proposal, but I do commend him for the work he’s put into this.
I think it raises lots of interesting questions. More than anything else, I very much appreciate his comments about the GSEs, because of course they’re very much in line with my own, and Peter’s. [laughter] And I’m always supportive of folks who agree with me.
Now let me just put a little caveat. I have studied, closely, the materials that Alan distributed. I also did some work a couple of years ago at Peter’s suggestion, on the Danish mortgage system. But I will say right up front, there are possibly some aspects of what Alan is proposing here, and aspects of the Danish mortgage system, which I misunderstand. And to the extent that that’s the case here, I look forward to correction.
So let me begin by summarizing just very quickly my points I’m going to make. First of all, I agree with Alan that the US residential mortgage market needs fixing, but I’ve always thought that. [slide 2] But we first need to understand the causes of the crisis, and I might add I’ve got copies of my paper here called "Bad Rules Produce Bad Outcomes",9 which will be published shortly in the Cato Journal, and I go into great depth on that. I encourage you to read it.
The Danish mortgage system has some interesting features; however, in my opinion, based on what I understand about it, the system is neither a model of efficiency, nor a model of efficacy and I’ll talk about that more. [slide 3]
PoB seems to have a contradictory feature that I’ll talk about at the end of my remarks, [slide 4] and then I want to outline two reforms [0:50:00] that would improve upon the present US system for financing home mortgages that would not involve anything like the Danish system. [slide 5]
First of all, some interesting features of the … of Alan’s proposal. [slide 6] First of all, the relatively high down payment required, as I understand it 20 percent, that you’ve got to have skin in the game. And at least that’s my understanding of it from also my earlier research on the Danish system, and I think that’s very positive. I think that one of our fundamental problems is people have not had enough equity in their homes to cushion them during downturns.
It protects homeowners if house prices decline as interest rates rise, [slide 7] but the question I’ll pose is what happens when both house prices and interest rates are declining. And we actually have that phenomenon today, and I would also suggest that negative equity is less if a problem in a mortgage system which doesn’t produce bubbles and busts, and that in many ways goes to issues I address in this article here, the Cato article.
Mortgages will have full recourse to borrowers. [slide 8] A great idea. I think that’s a real weakness in the US system but, and Alan touched on this, how popular will that be politically? You know, there ain’t no free lunch, [slide 9] and as I look at it lending is not [sic] a zero-sum game in nominal terms. And loans are in nominal terms. If the borrower pays back less than the face value of a loan, then the lender has received less than face value. And as I understand the Principle of Balance, the borrower can buy back his mortgage at a discount from face value if market rates rise above the rate on the mortgage borrower’s bond, but the lender does not get paid back in an amount above face value if interest rates drop below the mortgage rate. [slide 10]
This is a chart I’ve developed [slide 11] after looking at some of Alan’s charts, and I would suggest that you have an asymmetry here, in that the borrower — the homeowner — wins here, [lower right of chart] the lender loses over here [upper left of chart] because the lender, when rates go down, the lender isn’t able to, if you will, sell the bond back to the homeowner for higher than par.
So this is where I think there’s an asymmetry that I think causes a problem. To me it’s, for the borrower, "Heads I win, tails you lose," [slide 12] to the lender, "Heads you win, tails I lose," [slide 13] and therefore I consider Principle of Balance to be asymmetrical in the borrower’s favor, [slide 14] and I think that serial refinancing, which this system essentially fosters, are great for the borrower.
Now here is a chart. [slide 15] This is my chart, I take full responsibility for this, although there was one of Alan’s charts that suggested this idea to me, that essentially what this will do is actually stimulate refinance activity, which I have a lot of problem with, because of all the transaction costs associated with it.
That when rates are high, you’ll refi to reduce principal balance, then at the bottom of the cycle, you’ll refi to reduce the interest rate. And I haven’t had time to run the numbers on this yet, but my sense [slide 16] is that this means that upon entering into refinancing, the lender either gets back less than the face value, which he reinvests at a higher rate, but for a shorter expected duration, because remember when rates are high, if they come back down people will refi. And he gets back only face value, which he can only reinvest at a lower rate, but for a longer expected duration — that’s when rates come down.
So this is what I see as a problem, and that is … and my theory on this, or my sense of this, and again Alan may strongly dispute this, is that for a given mortgage rate and a given set of interest rate fluctuations, over time, a Principle of Balance lender will earn a lower return on his capital than a non-PoB lender. [slide 17] And that’s due to the fact that the lender has less to reinvest when high rates trigger refinance, and the PoB lenders must compensate with higher mortgage rates. Again, there’s no free lunch.
And so, I think that mortgage rates would be higher, and … but I’m going to touch on the … come back to this in just a minute, –
Second of all, you have higher transaction costs. [slide 18] Costs of refinancings occur both when rates are both peaking and bottoming. And as I’ve talked about in other sessions at AEI, transaction costs are a very significant element of home … of mortgage finance in this country, and something that I think needs to be attacked. It’s one of the reasons that I’ve never liked the GSEs.
But the other thing is borrowers must be good market timers, and credit-worthy in order to fully capitalize on this system. [slide 19] It’s hard to predict when rates have peaked or bottomed, [0:55:00] and credit impaired borrowers cannot take advantage of the PoB. And effectively they subsidize those who can refi. And it’s just the same problem we have in the US.
If everybody could refi automatically at the same time without credit constraints, then the cross-subsidy exists. But I would suggest to you that the PoB has cross-subsidy in it from the weaker credits to the stronger credits that we have in the US.
Now, I’m also troubled by the Mortgage Credit Intermediaries [MCIs] [slide 20] It’s a high concentration — again a lot like the mortgage insurers — then in effect the MCIs are monoline credit insurers who could go bust during a housing bust. And that increases taxpayer risk, and I also, along that line, question using GSEs as transitional MCIs.
The requirement for a highly standardized mortgage loan potentially introduces rigidity into mortgage lending, impeding innovation. [slide 21] Are highly standardized loans practical for the US, and is transparency needed if credit risk is not transfered?
Now, what’s this contradiction that I see. [slide 22] And I’m quoting now from Alan’s paper I got, "MCIs act as liability advisors to the homeowners. Mortgage advisors have an incentive to get homeowners only into those loans that make sense to that family." But then, a little bit later I see [slide 23] this statement, "MCIs must also make all efforts to make the prepayment characteristics of its bond series no worse than that of its competitors for investors. This creates an incentive for MCIs to make loans to those homeowners who have a lower probability of refinancing, including first time homeowners and others who have been underserved in the past."
I would suggest to you that this is a conflict of interest that the MCIs have. Now how they resolve it I don’t know. [note: comment above that small loans worth more implies this ...] That it is acually not in their interest to promote optimal refinancing. And I was involved as an expert witness in some litigation a couple of years ago that kind of got at this very issue.
So here are my own — Peter, I’m getting to the end — my recommended innovations. [slide 24]
First of all, covered bonds. I’m a great fan of covered bonds.6 It’s on-balance-sheet borrowing secured by mortgages, this is what a bank balance sheet would look like with covered bonds and other assets. [slide 25] It eliminates the need of securitized mortgages. [slide 26] I think we all agree that that’s not efficient, but by enabling the mortgage originator to hold the mortgage in portfolio without incurring severe maturity mismatch. It permits mortgage credit risk to be held within diversified lenders, something that’s very important. It greatly reduces the need for mortgage standardization, and of course this is something that was touched at at a [Sept 19th '08 AEI] conference,2 and I have at the bottom to the screen there the link to that conference.
Now, the other recommendation is Ratchet Mortgages™. [slide 27] Many of you probably haven’t heard about that, in the interests of full disclosure, I’m the co-inventor of the Ratchet Mortgage. [slide 28] My co-inventor and I have a patent pending on it. And just to discuss that briefly, I will but … [58:00]
Ratchet mortgages work with covered bonds. [slide 29] So those are the two innovations I see working together. Ratchet mortgages is a functional equivalent of a series of refinanced fixed-rate mortgages. [slide 30] The Ratchet Mortgage’s rate indexes down in sync with the published index, but the RM rate does not rise if the index later rises. The Ratchet Mortgage gives the homeowner the benefit of well-timed, automatic, no-cost reductions in mortgage interest rate and monthly payment. Therefore you’re taking transaction costs out of the system. The initial all-in interest rate on a Ratchet mortgage can equal the rate on a fixed-rate mortgage made to the same homeowner. And Ratchet Mortgages would be funded with Ratchet pass-throughs or Ratchet bonds, [slide 31] whose rates decline in sync … as the Ratchet Mortgage rate drops.
This is the key to the invention. [slide 32] This illustrates it, how you can get a step-down effect. It’s kind of a one-way ARM [Adjustable-Rate Mortgage]. And our patent application is based on the linkage of the mortgage to the Ratchet pass-through, or Ratchet bond.
The Ratchet Mortgage would bring many benefits to originators also, [slide 33] including better customer retention, a more profitable mortgage business, less volatile mortgage origination activity — because you basically eliminate refis — and greater cross-selling opportunities.
And with that, I welcome the discussion. [slide 34]
Peter Wallison: [laughs] Thank-you, Bert. It was a breathless tour through Alan’s proposal, but … and yours …
Alan, let me give you just a few minutes…
Peter Wallison: … two or three minutes … to respond to what Bert said … the most important points that he made … and we’ll do that in each case here, but I think while it’s fresh in everyone’s minds it would make some sense for you to be able to respond right now.
Alan Boyce: OK, so let me try to respond to a few of these things.
Denmark has an 80 LTV max for their mortgage systems. There’s a lot of homeowners that have combined LTV ratios well in excess of 80. Their 2nd mortgage market is an individual bank credit. 2nd mortgages in Denmark are always subsidiary to the 1st. If you refinance the 1st, the 2nd mortgage stays in a second position. [1:00:00] So … their 2nd mortgages assist first-time homeownership, but don’t block up credit problems, which I think is very, very important. It’s a subtle legal distinction, but in situations like today …
Bert Ely: … Could I say a question about that? So, does the 2nd have to be paid off when a new 1st is put on?
Alan Boyce: … no …
Bert Ely: OK, that’s different — that’s a key difference from the US.
Alan Boyce: … right … It [the 2nd] is basically an unsecured loan. That’s the way to look at it. OK? It has no rights.
Second thing, when you’re talking about idiosyncratic risk, you are absolutely correct. There are places in the United States where housing prices went down when interest rates went down. Generally those … that isn’t this current housing crisis, those are things associated with, like, single-industry towns.
OK, the fact is that there is no mortgage instrument you can devise for Flint Michigan, where all the jobs went away, OK?
This is actually one of the considerations where it gets into arguing how much the US government should guarantee. Because when you have high idiosyncratic risk, you’re in situations where it’s only the US government that can basically guarantee that town.
To just discuss the … I’m sure everyone in this room has been following The Economist magazine’s Housing Bubble-O-Meter over the last 7 or 8 years. The fact is, every country in the world’s got their own unique mortgage system, and every country in the world had a housing bubble.
There is actually no mortgage system there has been out there that could have prevented housing bubbles with lower rates, OK? The countries where housing didn’t go up in the last 20 years, was like Japan, where they had a gigantic housing bubble in the ’80s, that took the better part of the last 20 years to deflate.
To go over your … the asymmetry is actually a very important point here that I need to address.
There is no compulsion in Denmark. It is a misnomer to believe that the homeower can make the lender sell him his loan where the market price trades.
The whole point is, there is a market price. The market price is determined by the marginal buyer and the marginal seller.
That means there is somebody who is happy to sell that bond at that price. If there is nobody that’s happy to sell the bond at that price, the price is higher, OK? So there — this idea that somehow the lenders aren’t … are asymmetrically treated here is wrong.
Now if you go to this graph, [Boyce deck's slide 15] it’s very important to distinguish between a callable mortgage that’s principle-balanced and a non-callable mortgage.
In particular, if you want to eliminate asymmetries entirely, you have to have a situation where you have non-callable bonds. And that would be this blue line here.
The fact is that Denmark has about 40 percent of its mortgages are non-callable. They were made at times where interest rate volatility was high and the borrower said, "well, I don’t want to pay to get the option to call," right? "when rates fall I can call it at par, I’ll take the non-callable one."
In that case, it’s perfectly symmetrical, and the homeowner has only the right to redeem his loan where the bond trades.
Now I’ve been trashing the United States here, but the fact is the United States has the 2nd best mortgage market in the world. It just turned out to be not good enough.
Germany has a non-callable mortgage market that’s not Principle-of-Balance’d. They have the worst situation. So a German homeowner faces this line. When rates are low, they have to buy their mortage back where the bond market trades, but when rates go up, they owe par.
So, I mean, we could have done things way worse in the United States than we did.
So I actually do argue that there is a free lunch here, as long as there is a secondary market. The fact is, in Denmark the secondary market is actively traded, there’s always somebody who’s willing to sell their bonds at a discount. They’re crossed … yes … sorry, go ahead.
Peter Wallison: We’re going to go on to Olivier at the moment …
Alan Boyce: … OK …
Peter Wallison: … but there’s plenty of time for you at the end or after each of our other commentators.
So, Olivier Hassler has been a senior housing finance specialist at the World Bank since 2001. He coordinates there a team of housing experts within the financial and private sector development vice-presidency.
He’s worked for a French financial institution and specialized in mortgage finance, especially for lower-income households, and in municipal finance.
He was in charge of funding and capital markets operations and had responsibilities in regulatory issues, and the real estate portfolios management. … Olivier?
Olivier Hassler: Thank-you Peter. Thank-you very much for having me here. I’m a little outsider because I know the government in Europe, I have experience in emerging economies, but what I don’t know as well as you do is American market, obviously. So my comments might be a little bit off the point, and I apologize for that first. [1:05:00]
I have a few comments, and I will start with … to commend Alan Boyce’s approach. I think that in what you normally read about mortgage markets, you … people talk a lot about allocating risk. And it’s kind of a slicing the risk between the lender as a capital market, investors, the Bourse, but what Alan is doing with this approach I find is that his is a more comprehensive approach. It’s not completely systematic … systemic, but he relates strongly the housing market with the financial markets, which of course is a core issue in the crisis that we are experiencing now.
I understand that his core tenent is that through this Danish style mortgage bonds the bondholder [== homeower] is hedged, in a way, because on one side, he may lose on one asset, his house value, but on the other side he will gain on his debt. So I think that’s a core central proposition of the argument is … and that’s of course extremely interesting.
That being said, the orientation is extremely valuable and really it’s worth exploring. But also, at once, it raises two questions that we started discussing by the way.
First of all, is this hedge mechanism really good? Do housing prices really fluctuate symmetrically to interest rates?
I think it’s, unfortunately, much more complex than that, and it’s not just because of that Flint [Michigan] experience or unemployment locally in certain areas, but there are a lot of factors which drive housing prices, and interest rates are one of them but there are many other ones. So the hedge is far from being perfect.
As a second question that this approach raises is, "does it have such an impact?" Does this situation where you can have negative equity, does it … does everything turn around? … is that the central issue? I’m not so sure. For instance what comes to mind is when you want to refinance you have to borrow in general, because it’s a new loan. And you’re going to borrow at a higher cost if you want to take advantage of the fair value which has been appreciated of your debt.
So it’s just a trade-off in time. The net present value of your operation is neutral, theoretically.
So I wonder how much importance this ability to buy back a debt, which is either lower market price … if you pay this gain by higher payments of your next mortgage, what is the net benefit? But that’s a question I don’t want to pursue …
The second type of comment that I had, reading Alan’s presentation, is about the introduction — the problem of introducing a new financial instrument in a market. That’s an issue we at the World [Bank] see all the time in less advanced markets that need to have more instruments.
The first thing that I’d like to stress is that the mortgage — the Danish mortgage bond system — is not just this feature that Alan highlights. It’s a whole set of characteristics, and in particular regulatory characteristics.
All covered bonds in Europe, as you probably know, [1:10:00] depend on very strict frameworks, legally and regulatory. Frameworks which sets, for instance, norms for lending policy and quality of loans. Regulation for bankruptcy, remoteness, more or less, of these bonds, in the case of Denmark, or some other countries like Germany obligation of limiting the asset liability mismatches.
So there are a lot of things which make these instruments successful in Europe, and is that? … to bring it into the US, I’m not sure that just this feature of fair value for existing debts will be enough. You would have to, for instance, regulate the lending policies of the mortgage banks. If you, as you suggest, make them keep an interest in their loan portfolios and their credit risk, then you will have to have capital adequecy norms.
So it’s a whole set of things, and not only this one particularly. And from the norm, when you pull on the string, you know, there are a lot of issues like — there’s a competition between depositors, or unsecured creditors and the bond holders, or holders of collectivized bonds. So it’s a big factor, much bigger … larger issue.
And the second aspect of this introduction of a new instrument is a market acceptance, I would say. What will be the benefits and incentives for the different market players to buy the new system? For instance, I worked with Yves Bertrand(ph) in Chile a few years ago, which is a very interesting case, because it’s about the only example in the world to my knowledge where any system has been exported — I mean it’s more or less the Danish system.
And in fact, my view is that the success of this instrument over there was due to exogenious reasons. It was because this instrument was introduced at the same time that fully funded pension funds were created. So suddenly there were a whole bunch of institutional investors looking for investment vehicles, and there were not that many.
And the second thing is that, in Chile, after a long history of high inflation, they used an indexation mechanism which is extremely well transparent and everything, and which is generally applicable to housing prices, to all types of transactions.
And those covered bonds in Chile, or letras frito(ph) are indexed, and of course the whole set of long term bonds indexed, protecting the value of the capital, and the presence of an institutional investor pockets, that creates the success of these bonds. Which now, by the way, are less successful, for different reasons, but one being that it’s difficult to refinance when you only create this Danish style mortgage bonds in Chile, when you do new loans. But you cannot issue bonds against an existing house. So banks are getting away from this instrument now.
The third … that admits to a third point, which is in my view maybe there’s a real benefit — I think Alan’s presentation is quite ambitious. Besides, when you read the title of the workshop on the AEI site, it’s "Can Elements of the Danish Mortgage System Fix Mortgage Securitization?" It’s not fixing the whole mortgage market in the US. [1:15:00] And I think that’s interesting, because in fact maybe my views are that the best lessons that you can draw from these Danish instuments has the same issues that are being considered when you want to remedy all the weaknesses of the securitization system in terms of responsibility and avoiding Moral Hazard, so keeping some interest / credit risk at the level of the originator.
Therefore capital adequacy, some norms of good quality lending, transparencies, the idea of having independent review for due diligence of the quality of the portforlio.
And I think that’s where I would say the mortgage system has a lot of things to show, and at the same time, if you do that, my last question is, "do you want to have both?" Do you want to both renovate securitization, and have Danish style mortgage bonds? Would they be redundant? That’s a little bit my last question.
Peter Wallison: Thank-you very much, Olivier. Do you want to respond to Olivier’s comments right now, Alan?
Alan Boyce: Yeah, let me just … let’s start with the last one.
Peter Wallison: … but briefly …
Alan Boyce: … briefly. I think we can copy the positive attributes of the Danish mortgage system in a true sale securitization legal structure. In fact, I’ve spent several years of my life advising and assisting Mexico in copying the Danish system, and we did it in true sales securitization form.
So I believe we can avoid that competition, both with the securities and covered bonds, and also, therefore, with deposit insurance.
I love the Chilean model. And this doesn’t prove it, but lots of other countries in South America introduced pension reform the same time Chile did, but not the Danish style mortgage system. And their mortgage systems didn’t work. So I’d argue that they were coincident and mutually supportive, and in other mortgage markets didn’t help the other pension systems. In fact in other countries they both failed.
I think the refi efficiency, it’s very clear that when you do the optional redemption and buy back your loan at, say, 80, and take out a new mortgage for 80 percent of the old balance at a higher coupon, your NPV [net present value] is unchanged. However, you have restruck your call option at par, so if interest rates fall back to where they originally were, you can round-trip the whole cycle and end up with a market-based transaction — no compulsion — and reduce your loan amount, and when interest rates have gone up, since you can buy your debt back at 80, you’ve preserved equity, OK? … and reduced lock-in effects.
So you create the positive externality of labor mobility where none would exist, and you’ve preserved equity.
I mean, we’ve got a situation in this country where half the people with mortgages might be upside down on their mortgage. If you had that system, maybe you could reduce that significantly without government assistance.
Then the last one … This is the only degree of freedom I have as a mortgage bond trader to build in. I can’t actually fix the fact that people become unemployed. But I can tie … I know that housing prices and bond prices are generally correlated. The bond market loves the fact that you’ve introduced more buyers at a discount when there’s the Principle of Balance, so I think this is … it’s the only degree of freedom we have to enhance the current system that we have.
Peter Wallison: OK, very good, thank-you, Alan. … OK, Alex, Alex Pollock.
Alex has been a Resident Fellow here at AEI since 2004 and he focuses on financial policy issues, … well we all know this. Government Sponsored Enterprises, retirement finance, housing finance, corporate governance and accounting standards. He has written extensively on housing, on the housing bubble and bust, and is the author of "The One Page Mortgage Disclosure," proposal, which we here at AEI call "The Pollock Plan." This incidentally is the 2nd "Pollock Plan," the 1st would have straigtened out the entire Social Security system, if anyone had been listening, but that’s another issue.
Previously he had spent 35 years in banking, including 12 years as President and Chief Executive Officer of the Federal Home Loan Bank of Chicago, writing numerous articles at the same time on financial systems and management, so Alex? Please.
Alex Pollock: Thank-you Peter, and it’s a pleasure to comment on the interesting and certainly, as Olivier says, ambitious comments of Alan. [1:20:00]
I enjoyed, Alan, your speculations on my personal mortgage, [laughter] and the generational differences in my family. However, I am exceptionally pleased to report that all of my 4 children are at least a financially conservative as I am, and that’s saying quite a bit.
I’ve also found the Danish mortgage system quite interesting for some time. It looks to be a successful model and, as Alan pointed out, it does generate a lot of mortgage debt, Denmark does have a very high mortgage debt to GDP and to income ratio.
Late in the year 2000, I had a lengthy and very interesting meeting in Copenhagen with representatives of the Danish mortgage banks, in which we did the following deal. They would spend a good amount of time talking about the Danish mortgage system to me, and I would explain the American — Fannie Mae, Freddie Mac — mortgage system to them, which we did.
At the end of these mutual explanations, the chief executive of one of the big mortgage banks in Denmark, looked at me and said, "Well," he said, "you know, in Denmark, we’re always saying America is the land of free markets, and we are the socialists." He said, "I see that when it comes to mortgage finance, it’s the opposite."
And I do think that that contrast is one of the interesting reasons to think about Denmark.
Now the best element of the Danish system, in my view, is that the mortgage bank, as credit decision marker, is also fully responsible for the credit of the mortgage for the life of the loan, not for 90 days or for, well, just getting representations and warranties wrong, but for the credit for the life of the loan, as Alan rightly pointed out.
And it puts them in a business where there is fee income on one side, and credit risk or losses and expenses on the other side, which is quite understandable.
Working on mortgages in 1994 or so, we noticed a remarkable statistical regularity in the American system. That is to say that the credit performance of mortgages produced by depository institutions — banks and savings & loans — was systematically better than the mortgages produced by anybody else.
Why should this be? We said, well one good reason is that these banks and thrifts are all lenders for their own accounts, and their own risk as well as, often, sellers of mortgages. So all of their credit systems are set up from the point of view of somebody who is a lender fully at credit risk for the life of the loan.
We used to go to the managers of banks and thrifts and say to them, "listen, you make very good loans, you have wonderful credit records on your mortgages. So why do you want to sell these loans to Fannie Mae and Freddie Mac, which entails divesting the credit risk, which you have underwritten, and understand, and paying Fannie and Freddie … paying them for the privilege of divesting the credit risk?" And they would, in virtually every case, they’d answer, "I don’t want to divest the credit risk. I want to keep the credit risk. I want to be in the credit risk business. But I can’t stand the interest rate risk of a 30-year, long-term fixed rate loan."
"… So in order to solve my interest rate risk problem, I have to get out of the credit risk." Well, this was an interesting question that the Danish system poses to us is that, well why is that necessary? Why do those two have to be linked? And couldn’t you have a system where we have a separation of risks, exactly as Alan discussed, where the credit stays with the originator, and the family of interest rate risks being interest rate, prepayment, funding risk, … goes to the bond market.
And I fully and completely agree that this is … this kind of risk separation is a very healthy principle.
And could we do it here?
Yes we could. There are problems. For example, the Mortgage Bankers Association [MBA] says, "Well a lot of our mortgage banks are too small. They don’t have enough capital [1:25:00] to retain the credit risk. So we’d be putting them out of business." And you’d have to address this. Is it possible that the mortgage aggregators who buy the mortgages from such small mortgage banks could stand in their stead as the credit risk absorber?
We also have numerous accounting and regulatory issues in the way of this risk separation in the US, such as the "getting things classified as a true sale," and the accounting treatment. And regulation which was set up to encourage the sale of loans by depository institutions and the divesting of the credit risk by them.
So we’d have to reorganize our capital rules, and perhaps accounting and regulatory rules, to get into this "multiple risk & risk separation" rules, as the Danes have obviously done.
Of course this doesn’t mean that the Danish system solves all problems. As Alan pointed out, the Danes had their own housing bubble in the 21st Century, it peaked in 2006. They have their own house price falls, banking problems, a couple of failures, ongoing fairly severe recession and they have done this before, as was also pointed out — early 1990s, 1990/1991 the Danes had a big housing bust, which I’ll mention again in a moment.
Well, even with well aligned incentives, that is to say retention of credit risk by the credit risk decision maker, you can still make mistakes. And of course, as Peter knows, one of my favorite banking authors, Walter Bagehot, describes this perfectly.10 Bagehot wrote:
A manager sometimes committed frauds which were dangerous, and still often made mistakes that were ruinous. Error is more formidable than fraud. The mistakes of a sanguine manager are far more to be dreaded than the theft of a dishonest manager.
The losses to which an adventerous and plausible manager, in complete good faith, would readily commit a bank, are beyond comparison greater than that which any fraudulent manager would be able to conceal.
… He concludes …
There is no more unsafe government for a bank than that of an eager and active manager, subject only to the supervision of a numerous [laughs] board of directors.
Well, …
Peter Wallison: … that wasn’t Frank Raines who said that, was it? …
[laughter]
Alex Pollock: … economists in general, I find, overweight incentives in comparison to mistakes, and underweight the potential of mistakes.
But mistakes notwithstanding, which mistakes gave us, which was pointed out, housing bubbles and busts in numerous countries around the world. It’s still better to get the incentives well aligned and the risk factors segregated.
A couple of interesting things about Denmark, just to conclude — we’ve discussed this issue at length, of the borrower prepaying at par, in most cases — in the callable case — or buying back the bond. And Bert and Olivier both addressed this, but there is the case not covered, which is the case in the US today of both falling interest rates and falling house prices and rising defaults.
So we don’t expect any system to solve all problems perfectly, and the subtlety here is that if we make the bond reflect only interest rate risk, not credit risk, then we don’t get the kind of buy-back, say, that a corporate manager buying back a bond can do as he — as the corporation’s credit deteriorates, its bond price falls, so part of what you’re seeing in those parallels is a credit event. But if you make the bond, as in the Danish case, purely an interest rate risk event, then you lose — or don’t have — the protection, which as I say is what we have now: low rates, falling rates, falling house prices, falling credits.
Secondly, another interesting feature of Denmark that Alan mentioned in a way in his concluding comments, is that loans are full recourse, with full recourse to borrowers, and deficiencies after foreclosure. So you foreclose, the house is sold, the price of the house [1:30:00] out of the foreclosure sale is insufficient to pay off the debt, there’s still a remaining debt. That’s what we call "a deficiency" in the mortgage biz. Those deficiencies are adamantly pursued until collection is full.
And in Denmark, you’re kept on a kind of Black List, a credit Black List, if you have an unsatisfied deficiency.
A month or so ago I had dinner with the chief executive of one of the Danish mortgage banks, and he pointed out to me with some relish, actually, that he was still collecting … deficiency payments … he was still collecting deficiency payments from borrowers who had defaulted in 1991, in the last housing bust.
Well, for old fashioned guys like me who believe in paying debts, that’s good. But I’m not sure how popular this element of the Danish system would be with various Congressmen — let’s say, Barney Frank as an example.
So in sum, I think the Danish mortgage system does indeed give us numerous interesting things to think about, to consider, including, as my Danish colleague several years ago so rightly pointed out, this comparison between a relatively less of a state interventionist system with the one that we had here, which we’re stiving to improve. … Thanks.
Peter Wallison: Great, Alex.
Alan, do you want to respond to any of Alex’s points, and then we’ll have some cross-talk here?
Alan Boyce: Sure. Well just … It’s hard to disagree with a lot of things he said, and I wish that he could talk more and I could listen more, because I always love listening to Alex talk.
It is actually the case that most of the small mortgage bankers have been wiped out in this mortgage crisis. So the Mortgage Bankers Association argument that we’ve got to protect the little guy is sort of specious, because there’s none left to protect.
Historically, the 3rd party origination channels at the big banks the correspondent(ph) and wholesale businesses worked for very skinny margins. You’d be hard to guarantee some little morgage broker working for a skinny margin. However, today they’re working for very big margins. So I think that the combination of both those things could actually work.
You’re correct that FASB 140 [the US accounting rule that deals with true sale accounting for vehicles like QSPEs] is very restrictive as to what defines a true sale. The fact is, that there’s a lot of FASB rulings, including 157, that had gigantic unindended consequences. I believe that this financial crisis has proved that we actually do need a real rethink, and people need to think about consequences and interactions, and we actually built a sort of theoretical house of cards in our accounting standards.
The last point is, yes we’ve had falling rates and falling housing prices, but the fact is, that in the United States, if you live in a non-Agency neighborhood, that’s one where you need a mortgage bigger than $417,000, mortgage rates in your neighborhood went up, not down.
OK, so let’s just go over this for a second. Half the people in a non-Agency neighborhood wouldn’t even qualify to get a non-Agency mortgage today. So your mortgage rate went to infinity. Half the people who want to buy in your neighborhood — it’s the same case. And the other half would qualify, but your mortgage rates are 8 percent today.
So it’s actually … you raise a very good point. And it gets down to correlation versus causality. The demise of the non-Agency mortgage market is why non-Agency neighorhoods have been dropping so much in price. And until we restore the non-Agency market, or make the Agency market a bigger loan minimum so more expensive houses can be financed, non-Agency neighborhoods will go down to the floor [sic] of an Agency mortgage.
Alan Boyce: OK? It is almost a tautology, and until we solve that, that is actually where we’re going to go. And that is a scary thought, because that is way lower than we are now in a lot of more pricy neighborhoods around the United States.
Peter Wallison: Well, banks are still making loans …
Alan Boyce: … they are making very, very …
Peter Wallison: … Jumbo loans without … using the Agencies … Well, I mean anecdotally you see a lot … but let me ask you a question.
The recourse mortgage idea. This is something that many people have talked about as a major problem in the United States, but to what extent do you think that Denmark — the case of fewer defaults in Denmark — is a result of the fact, as Alex was pointing out, there are, there, recourse mortgages?
Secondly … let me ask you another question, and that is — the GSEs. Do you think it’s necessary for the GSEs to be involved in this. They would have … that’s the only way in which a mortgage can be sold is if the GSEs are actually government backed? But do you think it is possible for there to be a GSE without an affordable housing obligation? [1:35:00] … which makes it very hard for such an institution to be a mortgage counselor.
Alan Boyce: … ah … yes … so let’s start. First of all, because credit risk is separated from interest rate risk at birth, the mortgage bank itself is really trying hard to make sure that they only underwrite good risks. And they make it very clear to the borrower what recourse means, because they are entering into a long term relationship with each other. That mortgage bank may refinance the loan based on bond market opportunities several times, but the mortgage credit institution and the borrower are joined at the hip, at their first introduction.
Peter Wallison: So …
Alan Boyce: So that threat of full recourse is very clear.
Peter Wallison: … um hmmm …
Alan Boyce: … it’s … those two things combined really stop a lot of bad things happening on Day One. They really do. And it is great that the Danes still go around and talk about collecting deficiency payments …
Peter Wallison: … right …
Alan Boyce: … from people dating from 1991.
I mean, there are actually very few. It’s Peter Engbridge Jónsson(ph), he runs New Credit, it’s the biggest mortgage bank in Denmark. It’s a tiny amount of people. They’re going to hope that two or three of those people live to be 120, so they can still talk 20 years from now about collecting deficiencies. …
Peter Wallison: … in Denmark actually they do live that long …
[laughter]
Alan Boyce: … It’s like, just the threat of that, the fact that it’s being done, is out there.
Peter Wallison: … Right. But my question, Alan, is: it works in Denmark. If we had a similar system in the United States, would that have addressed — I’m really talking more broadly here — would that have addressed some of the problems we’re facing here. Just that change.
Alan Boyce: I think they would have addressed a lot. Between that and the mortgage bankers having some unsaleable skin in the game? That would have addressed a giant amount of the risk.
Peter Wallison: … OK then …
Alex Pollock: … Then you agree with me that politically there might be some American politicians who wouldn’t favor
somebody: … SNORTLE! …
Alan Boyce: … That is correct. The flip side is that, …
somebody: … What? …
Alan Boyce: … if you want to be able access the guaranteed mortgage market …
[more cross-talk]
Alan Boyce: … which trades at a much lower interest rate than the non-guaranteed mortgage market, the quid pro quo for, say, 300 basis points a year is, that the Federal Government won’t forget.
Peter Wallison: Right.
Alan Boyce: The fact is, we live in a country where … I mean, people go bankrupt and 5 years later, by and large, have pretty good incomes. So it’s actually one of the tenents of the American Enterprise Institute is that we, like a lot of people, start out at the low end and go to the high end.
And if the government is going to help you live at a higher standard while you’re still not making a lot of money — that’s the whole idea of a long-dated mortage — then if you relinquish on that obligation and later come back in life, you owe the government back some money.
Peter Wallison: Right. And all of that is something that is worth talking about as a reform to the US mortgage system. I want to go one step further, and that is — this has to do with politics of course — and that is: can the GSE that is now the mortgage counselor be "joined at the hip," as you would call it, with the borrower if the mortgage counselor continues to have an affordable housing obligation?
Alan Boyce: OK, I probably didn’t explain it well enough. What I propose is that the private sector mortgage banks of the United States become the mortgage credit institutions and just get guaranties from the GSEs. So the mortgage councilor will remain Bank of America or Wells Fargo, or PNC Bank or somebody like that.
Peter Wallison: We have that system now! And as …
Alan Boyce: … But they don’t have the skin in the game now. They pass all of the credit risk onto the GSEs.
Peter Wallison: … right, and the GSEs have been taking credit risk on $1.6 trillion in subprime and Alt-A mortgages.
Alan Boyce: Right. But ones they didn’t underwrite. On the subprime they bought those in the secondary market, and that was a separate giant mistake. But I would strongly argue that what the GSEs — their affordable housing goals — are unnecessary in the Danish system: this is actually to address a good question from Bert.
It’s important in the Danish system to get your customers to come in the door. Your bonds have to trade at a high price. If your bonds trade lower than somebody else’s, your customers would naturally go to somebody else, because they’ll get higher proceeds for taking out the same loan.
The bond market is purely trading these cash flows on their economic value, and since these are callable mortgages, the loans … the bonds that are less likely to be instantly called when interest rates fall trade at a higher price.
The way to get your bonds to trade the best is to get as many small borrowers in there as possible. You don’t actually need a low income housing mandate, because smaller loans, by definition, respond to interest rate moves less. There’s a lot of frictions in the system, and I’m not saying there’s anything wrong with how those people think, it’s just that if you have a $100,000 loan, your friction for refinancing — you need to see 100 basis points savings.
Peter Wallison: I have some questions, but I know Bert does, so go ahead. [1:40:00]
Bert Ely: Of course the Ratchet Mortgage takes care of that friction problem by making it automatic. And not …
Peter Wallison: Patent Pending …
[laughter]
Bert Ely: … Patent Pending, yes … also market pending, too.
My question is … deals with the MCIs. The thing that concerns me is that — and I’d be interested in what the Danish experience is on this — You’re getting an incredible concentration of credit risk. I mean, these are monoline credit insurers. We’ve seen what’s happened to the monoline credit insurers in this country. How close have any of them come to going bust, or having to get some kind of government support?
Now I realize for the US system you’re saying, "well, we’ll just lay the credit risk back off on the GSEs." But I, and probably a few other folks would like to see the government get out of the mortgage credit risk business. So that’s a fundamental concern that I have about your proposal, or the Danish system, is that incredible concentration of credit risk in undiversified firms. It seems to me that that is very risky systemically, which is why I like the idea of covered mortgages and covered bonds being issued by diversified financial firms like Bank of America where they keep the credit risk on their own books, but they’ve got lots of other types of credit risk too.
Alan Boyce: So when I first started trading Danish mortgages in ‘94, I said exactly what you just said. How can you have all this concentration in these specialty financial institutions? And this is a huge debate in Europe. Huge, I mean you can’t believe how many discussions go on in Brussels about specialty versus universal banks.
The fact is that the Danish mortgage system, by cleanly separating the risks and making the mortgage banks focus on just one thing has built a relatively robust system.
Now I’m not saying that there is not some event that could blow it up, but it didn’t blow up in the most stressful event that you could have possibly imagined. Just in the last 6 months.
In the last 215 years, the Kingdom of Denmark’s gone bankrupt 3 times, and the mortgage system’s never missed a payment. So, over time, I’ve gotten to the point of, like, you know what? I works. I know so many details about the system now, and, like, I just refuse to say, like, let’s fix something that isn’t broken. That’s where I come down to.
Alex Pollock: You bring up 2 points there in passing. One is when you contrast the government bonds with the mortgage bonds. As I remember the central bank in Denmark, Bank of Denmark, deals in these mortgage bonds as a key central bank liquidity bond market vehicle.
And the 2nd thing that raises a question which maybe you can address for us. What is the capital regime of the Danish mortgage banks. How much do they have to hold against these credit risks?
Alan Boyce: OK, good questions. The Danish Central — the Danish mortgage system has always been key to their financial system. Right now the amount of Danish mortgage bonds outstanding is multiples of government debt. And the sort of fungible issues like the 5 percent 30-year mortgages in Denmark? The fungible issue is, like, $75 billion: it’s the biggest fungible bond on the planet, coming from tiny little Denmark.
So for the central bank to conduct monetary processes, they actually need to operate in fungible markets, and their soveign debt market has gotten so small, that they need to trade in the liquid bonds.
The capital regime: there is a little goofiness in Denmark, which I rib my friends there all the time. The mortgage banks, being around for so long, have accumulated an enormous amount of capital. But, sadly, Denmark is this tiny country and unless they decide to go out and try to start expanding their borders back to where they were before they lost wars against the Engish, and the Swedes …
[laughter]
… and by the way the United States has taken quite a bit of territory from the Danes in the last 60 years, right? That’s right, the US Virgin Islands, Iceland, they get …
Alex Pollock: … and the Prussians …
Alan Boyce: … and the Germans took a big hunk. Unless they get bigger, the fact that their system makes them money every year, and they have no mechanism to give the money back to the shareholders in particular, means that they are insanely overcapitalized.
Now there is a corollary to this. They did covered bonds since 1797. The word "securitization" wasn’t invented until 1970. If they had to do it all over again, they would set it up as true sale securitization. All their systems are easy to flip to make it work that way. The newest mortgage bank, Total(ph) Credit, which acts as the conduit for all the little commercial banks and Savings Banks in Denmark, is set up in an extremely capital-efficient way, and their covered bond series, which are best viewed as master trusts, because they set up a new one that they issue against every issuance period, they hold 2 percent capital plus the first loss guarantee from the originating bank. [1:45:00]
So that it’s actually quite capital efficient as a legally structured covered bond.
Some of the old mortgage banks that are less innovative, like BRF, their shareholders are everybody that ever had a mortgage with them over the last 200 years. They have no idea how to give the money back. And they’ve got 12 percent capital.
Peter Wallison: Thank-you, Alan … OK, we’re going to go to questions from the floor. Start here, could you identify yourself. We have a mic … not that we need it, but just in case …
Voice: Do we have a mic?
Peter Wallison: We used to have a mic … [laughter] … we thought we had a mic.
Brian Brooks: So thanks Peter, question for Alan …
Peter Wallison: Can you identify yourself.
Brian Brooks: Yeah, my name’s Brian Brooks. I’m with O’Melveny & Myers here in Washington. I guess I have an observation and a question. The observation is, you know, I first became interested in this Danish model about 18 months ago when I spent some time working with the Mexico City company, I think you might have actually had some involvement with ITO(ph). And it occurs to me that part of the sales pitch for the Danish model is different and more compelling today than it may have been 18 to 24 months ago when Mexico was first looking at the structure, and it’s something you touched on, but I think could expand upon in this discussion:
And that is the behavioral economics phenomenon, that people have decided, people who have the financial wherewithal to make their mortgage payments, are electing not to,11 and instead make their credit card payments, while defaulting on mortages when they become upside down on their mortgages. And that’s something that hasn’t been seen, if it’s ever been seen in history, it hasn’t been seen in generations.
And so I think the compelling case here builds heavily on that bizarre and important behavioral economic phenomenon.
The question I’ve got for you is: Why the model hasn’t been more rapidly adopted, particularly in emerging economies that don’t have the overlay of US-style securitization? In Mexico, as you know, the securitization market is very very small, it’s really dominated by a single government pension fund that’s the majority mortgage issuer in the country.
So you’d think there would be opportunities, where there is a vacuum like that, for this to build that more quickly. Yet my impression is, it hasn’t caught on as fast as optimists like myself would have hoped and predicted. So I wonder what your thoughts are.
Alan Boyce: OK, excellent question.
Olivier could probably answer this the same way I could. I was the guy who convinced the US Treasury Department, when it was run by Paul O’Neill and his appointees, that … Mexico offered a … [that] Denmark offered Mexico a better example than the United States, in December 2001.
And it took 4 years to convince people in Mexico City. But at the time, I was arguing that the Principle of Balance would have saved them from the Tequila Crisis, when their giant devaluation had home prices drop with the debts, right? were indexed to inflation and went up. And if they had followed the Principle of Balance, it would have been rightside up. And they immediately got that.
When it comes to other emerging markets, this little company Absalon, the joint venture between Mr. Soros and and the Danish financial system, we’ve got many other emerging market countries that want our help, and to follow in Mexico’s footsteps.7b
And we determined that we needed to go through all the blocking and tackling in Mexico first, and create a viable entity, which in the last 6 months has picked up huge volumes and is very close now to being the mortgage conduit for Infonavit.
Before we start talking to these other countries, other than some quick advisory work … OK, so to that extent there’s been guys at the World Bank and particularly the IFC that I’ve talked to about this.
But it’s … I’m telling you, the brain damage of actually taking the Danish system and integrating it into an emerging market country cannot be underestimated. It is much simpler to do in the United States than it is in places like Mexico.
Peter Wallison: Other questions?
Bertrand Renaud: I’m Bertrand Renaud, I used to be at the World Bank, and I enjoyed your presentation. There are many points that have been made, and I do tend to agree with you, Alan, that the US is an easier environment to introduce such an innovation. But in many other markets where the infrastructure does not exist, where the bond markets are not developed and so on.
But there were so many points to bring out. … I believe that one element of success in selling the idea … I don’t see the Danish presentation that you make as a product innovation, but as a systemic innovation in the US market. Because I think that it addresses some fundamental flaws.
And we all know from experience sitting at the World Bank that no country is going to reform its system [1:50:00] unless there’s a crisis. So we have a crisis, we have a chance.
Now the point that I may be missing on the political economy of the innovation is that you didn’t tell us enough, in the Danish experience, about affordable housing. How do the Danes deal with social housing, which has been a driver in the US. You know, it’s like the small family farms driving the agricultural subsidies. The affordability issues is always a Trojan Horse for all kinds of things.
But in the US, everybody’s speaks of Barney Frank and at the AEI all the time, but Barney Frank is a symptom, if you like, of a political force to deal with low-income housing.
It would strengthen your argument if you told us what the Danes are doing with the non-market guys, if you like, and the social programs. Because unless you explicitly tell us what’s happening to that group, we deal only with, let’s say, 80 percent of the social structure.
There were some other points that were made that are also very important, because the lobbies that are interfering with the reforms here are present in the mortgage markets. But stepping by one step … In your presentation you merge mortgage reforms with securitization reforms. And I would urge you to read the Turner review of the UK that came up last week, because they make very clear that the US crisis on mortgages was a trigger to reveal much worse problems with securitization in general, structured finance.
So it would be very important that you don’t present your proposal which I believe is very strong, as also a panacea for securitization, which is a much bigger issue.
So … the subprime crisis was a trigger that revealed a much worse crisis across the board. So that’s OK …
But another point that you are not discussing is also transaction costs. In many discussions over the years, Jack Guttentag has been fuming for decades about the fact that the transaction costs for a mortgage in Denmark is of the order of $400. That is to say, you can refinance at a minimal cost. So that issue is as non-issue.
And there are all kinds of environmental points that would strengthen your case which you are not making.
The other other issue of innovation is not …
Peter Wallison: … I think we have a number of questions. Let’s let Alan try to address these and …
Bertrand Renaud: … sorry …
Alan Boyce: OK, to start with, Bertrand, if I had 3 hours up here I would exhaust everything you just brought up. The fact is …
Peter Wallison: I believe it.
[laughter]
Alan Boyce: … in the United … and I could go … There’s actually a few people in this room that have been on the receiving end — more than 3 hours.
In Denmark, transaction costs are very transparent. And they’re actually quite low. In the United States we make transaction costs opaque, in part by doing what’s called premium origination. So right where mortgages are funded by the issuance of bonds trading at big premiums. And that means that transaction costs in the United States are always way higher than anyone could possibly imagine.
OK? The mortgage bankers, the middle men in the United States have done an exceptionally good job of self-regulation and saying that, "well, we compete with each other, so of course that competition will make the market better for the buyers and sellers, the homeowners and the bondholders." But the fact is that transparency does that, not opacity.
And I think that’s just a huge, huge point that no matter what the systemic solution comes out of this current crisis — God, transparency and probably standardization have to be two of the big ones. I hope that we also get interest alignment and some countercyclical behavior.
Peter Wallison: On these closing costs and transaction costs, Bert, what do you have to say to that?
Bert Ely: Well first, first of all I think you raised a very important point. And as I look at a lot of the closing costs and the various elements of it, much of it, I don’t think relates to premium versus discount bonds, but just the legal structure that’s developed in this country like title insurance, for instance — and to me the key to taking out transaction costs, which is very much of an overhead cost in lending, is to minimize the number of transactions. Which is, to put in a plug for the Ratchet Mortgage, patent pending, is that you get away from having to create a new mortgage in order to get a lower interest rate.
Peter Wallison: OK, we have time for one more question in the back. Can you identify yourself please.
Steve Adler: I’m Steve Adler, I work for IBM.
And I just wanted to say that I lived in Denmark [1:55:00] for 5 years, and what they do [... 8 second gap12...] policy and that’s separate from this market, right? And I think that what I’ve learned from this discussion today, Alan, is that it is probably possible to implement aspects of the Danish mortgage market with some subtle changes in the American market. It doesn’t have to be a systemic overhaul, it can just be a few steps in the right direction.
But what I leaned from Alex is that any system is prone to human failure. And so a system isn’t a panacea, it’s just another step. We still have to manage it.
Peter Wallison: OK, well, I think we’ve had a really excellent discussion of some very stimulating ideas. we all have to go back and think about these a little and write about them and talk to one another and see if there’s some ways that some of these ideas can be integrated into the current system that we have.
I want to thank Alan in particular, and our wonderful panel. Please join me in giving them a round of applause.
[1]: "Can Elements of the Danish Mortgage System Fix Mortgage Securitization in the United States?", AEI Event, March 26, 2009.
[2]: "Can Covered Bonds Compete with Fannie and Freddie?", AEI Event, September 19, 2008.
[3]: "Denmark Offers a Model Mortgage Market: There is a safe way to securitize home loans", by George Soros, Wall Street Journal, October 10, 2008.
[4]: "How to Fix the U.S. Mortgage Market" [PDF slides], by Alan Boyce, Absalom, March 26, 2009.
- Title
- Introduction
- Background
- A successful mortgage reform
- Rebuilding the system to properly align incentives: The old system needs to be replaced
- How the System Could Be Fixed By Emulating Danish System
- Current system is not symmetrical or balanced
- The Danish System: refinancing on the way down
- Which Reduces Risk of Negative Equity
- Fully transparent: real time information on each series
- Time series and transactions data
- And access to news
- Vintage LTV analysis shows benefit of fair value of loan
- Detailed information on borrower characteristics
- U.S. Mortgage Structures Can Create Negative Equity
- Principle of Balance Mortgages Prevent It
- This is a unique opportunity to “get it right”
- Lowering Interest Rates, Reducing Existing Negative Equity
- Credit Enhancement Structure for Shared Platform
- Convexity Effect of Securitization Choice
- Price/Yield Graph of Various Mortgage Risk Transfer Structures
- Market Potentially Has Still More Pain to Come
- Long Term Interest Rate Volatility Reduced
- Mortgage Pricing 101
- Credit Quality of Borrower Improves
[5]: "How Serious Is the Mortgage Problem That Will Confront President Obama?", AEI Event, January 16, 2009.
[6]: "Covered Bonds vs. Securitization Transparency vs. Opacity: Which is the Right Question" [PDF], by Alan L. Boyce, Soros Fund Management, September 2008.
[7]: I’m profoundly uncomfortable with the seminar’s evident appetite for the proliferation of Danish-style recourse mortgages into Mexico, Chile, and now the United States and beyond. Recourse sounds like something evil that has been repeatedly addressed throughout history by the likes of Dickens, Shakespeare and a host of Old Testament authors. The general drift of the above discussion suggested to me that many of the participants left with these two ideas in mind:
- a) In the United States, the crisis could be used as a convenient tool to remove the States’ ability to regulate the terms of mortgages offered within their boundaries.
- b) World wide, countries might be pressured to legislate that all their home mortgages be recourse.
[8]: "Can Elements of the Danish Mortgage System Fix Mortgage Securitization in the United States?" (PDF slide deck), comments by Bert Ely, AEI, March 26, 2009. [the slides aren't titled, and mostly just summary what the speaker is saying in the tape]
[9]: "Bad Rules Produce Bad Outcomes: Underlying Public-Policy Causes of the U.S. Financial Crisis" [PDF], by Bert Ely, Cato Journal, Vol. 29, No. 1 (Winter 2009).
[10]: Lombard Street, by Walter Bagehot, 1873.
[11]: a) "Cramer: Underwater on your house? Walk away (Oh, and plow under the Inland Empire)", Housing Doom, July 30, 2007. b) "Cramer: That’s Right – If Your House Is Down, Dump It", Housing Doom, July 31, 2007.
[12]: As I recall, Adler’s intervention was previously about 2 minutes longer. While starting on this transcript a week ago (June 13th) I scribbled down that the video ran to about 1:58:00.
John-
Kudos and thanks for an amazing transcript. Doomers, you have no idea how many hours John put into this.
This is the longest post we’ve ever had, but there is a lot of information here.
Clearly our securitization system has failed us, and changes need to be made. The answers are beyond my range of expertise, but I’m glad you’ve made this material available to help with the debate. : )
Twist -
Thanks for the kind words …
Now maybe I can get back to singing and working my way through The Desire of the Nations, by Oliver O’Donovan. I’m in correspondence with a Presbyterian minister in Ontario, and he recommended the spiritually oriented philosopher O’Donovan as a good guide for some of our present-day challenges. He also sent along this article, which is weirdly in sync with something Boyce said above:
“Swiss politician seeks inspiration from 16th-century Calvin”, by Stephen Brown, Ecumenical News International, June 15, 2009.
In a way, we may be on the same page as the AEI crew (if perhaps in opposite sides of that page
) Open dialog can’t hurt, though, which was an important motivation for my taking on the above project.
John-
I suspect he is right about where Calvin would have stood. Transparency equals honesty. It’s easy to understand why not bearing false witness is in the Top 10, it prevents a host of other sins!
Huizing also hit the nail on the head with this statement:
That is as succinct a description of Washington as I have read in a long time.
twist -
Looks like we’d better keep a weather-eye on the Anglicans, too. Just came across this in Desire:
… betcha that’s not Rahm Emanuel busy reawakening the democratic beast on the streets of Tehran this month
Now where were those guys in Nov ‘00?