Fiderer on GSEs and other stuff: 2015

(After the blog retirement of my good friend Bill Maloni and the apparent semi-sabbatical of another blogger, I thought I would take the Monday weekly slot given up by Bill. As you can see the blog formatting itself is very much a work in process.)

Based on his articles and books that I've read over the years, I believe Mark Zandi, head of Moody's Analytics, is a really smart guy. So I presume he is fully aware that the op-ed he wrote with Jim Parrott in The New York Times is filled with doublespeak.

First, I'll summarize why I think he's been very smart and eloquent about the GSEs and the financial crisis. Then I'll explain why the GSE "reform" that he advocates is a crackpot idea. Finally, I'll dechipher some of the doublespeak in his op-ed.

Zandi rejects The Big Lie: Zandi lets the numbers speak for themselves, which is why he has continually refuted The Big Lie. "Fannie and Freddie weren't to blame for the U.S. housing bubble," writes Zandi, in his book, Paying The Price. "Fannie and Freddie were actually minor players and the crazed lending of the 2000's...The bubble was diminishing Fannie's and Freddie's position in the housing market... The two firms couldn't compete with rapacious private lenders, who, thanks to securitization, could offer extra low rates and irresistible terms to borrowers."

And, as Zandi repeatedly pointed out, here and at the American Enterprise Institute, the GSEs' track record loan recovery is unmatched by the rest of the private sector. In fact, going back decades, GSE loan performance has always been exponentially superior to any other segment of the market. Zandi insists that, "The private RMBS market was at the heart of the financial panic and the Great Recession that followed." Here are Zandi's numbers in 2013, followed by a serious delinquency chart from the Richmond Fed:

Money talks, doublespeak walks. Since no private player has ever come close to matching the level of disciplined credit underwriting demonstrated by the GSEs, it's nonsensical to say that we need to abolish the GSEs to instill "free market discipline" into the market.

Zandi blames Hank Paulson's decision-making for the meltdown: Here's what Zandi wrote in Paying The Price:

The government's takeover of Fannie and Freddie arguably ignited the global financial panic. The Treasury Department's decision to wipe out shareholders of two of the largest financial institutions on the planet shocked markets, making it apparent that no institution was safe. Investors in all financial institutions question the value of the shares, bonds, and loans. The weakest link was investment house Lehman Brothers.

When speaking extemporaneously to the FCIC, he went into more detail :

I think we probably would have suffered a garden variety, or somewhat bad version of a garden variety, financial crisis if not for a series of significant policy mistakes. The one everyone points to is the Lehman failure and the policy mistake that was there and I think that was a mistake that ultimately turned what was a crisis—but a garden-variety financial crisis, something we experience every decade or so—into outright panic and I think near collapse…

I don't know that Lehman was really the catalyst for the panic. The takeover of Fannie and Freddie may actually have been the catalyst a week earlier. Because that takeover, a week before, crystallized in the minds of global investors that none of their investments were safe. If they could take down Fannie and Freddie, if their equity investment in Fannie and Freddie was zero, then nothing they owned was sacrosanct and safe and therefore Lehman goes immediately into play. So I think the panic began a week earlier with the takeover that was botched I think.

…

[By early September 2008,] we were suffering a recession as a result of what I think was a reasonably orderly unwinding of the bubble. It was already having negative consequences on broader economic activity. [But for the government’s mishandling of the GSEs and then Lehman,] there would have been an average recession, you know, an average recession is eight-nine months long, unemployment would peak out at 7-7.5%, and GDP loss would be peak to trough would have been two percentage points…Also house price declines would have been at roughly half of what they were, maybe 15%, 20%...House prices would've gotten to where their equilibrium level in a more orderly way. They wouldn't have been straight down in such a short period of time.

But he clams up about some inconvenient truths

Private label RMBS is never coming back: I presume he is fully aware of certain trends that are perfectly obvious to anyone paying attention. Private label residential mortgage securitizations proved to be an unmitigated failure and are never coming back in significant numbers. Zandi, and most institutional investors are acutely aware of the unaddressed moral hazard problem with private label deals. And the largely unmentioned private label math problem is insurmountable. (The math problem had been obscured by a housing boom. Home price appreciation has an exponential impact on default frequency and an explosive impact on loan loss recovery. During 2004, 2005, and 2006, loss severity on subprime mortgages in California was 1%. In 2009 it was 70%.)

As the Urban Institute Chartbook shows, the GSEs and FHA filled the void left by private label securitizations.

Nothing can top the GSE business model: Wall Street is at a competitive disadvantage with regard to Fannie Mae and Freddie Mac, for reasons that have nothing to do with the U.S. government. The government sponsored enterprises are able to finance trillions of dollars of 30-year fixed rate mortgages because their securitizations transfer interest rate risk but not credit risk. The GSEs guaranteed all securitizations they issued, which ameliorated the impact of the biggest driver in loan recovery, market timing risk.

Because GSE securitizations cover credit risk, investors are willing to assume interest rate risk, which is why the GSEs, and only the GSEs, are able to finance huge volumes of 30-year and 15-year fixed rate loans, which have always performed exponentially better than ARMs. As Lewis Ranieri is fond of pointing out, no one can point to a long-term track record of success for ARMs.

So I presume that Zandi is aware of the doublespeak used to frame the debate. When people talk of GSE "reform" they are really talking about abolishing the time-tested GSEs because there's no other way to give Wall Street a dominant piece of the action. Which is what his op-ed, co-written with Jim Parrott, is really all about.

Zandi promotes the Johnson-Crapo Bill as housing finance "reform."

Why Is Johnson-Crapo a crackpot idea? Answer: It's the subordination stupid! Did you learn nothing from The Big Short?

Zandi supports a "reform" proposal, the Johnson-Crapo bill languishing in the Senate, which is a truly crackpot idea. No matter what Zandi or anyone says, nothing can alter an immutable rule of housing finance: If you don't understand subordination, you are clueless.

Subordination was at the heart of the simple math that defined The Big Short. The math was very easy to figure out. When you look at the tiny triple-B-rated tranches, which were subordinate to 95% of the debt in a 100% debt-financed mortgage securitization, you'll see there's almost no margin for error, meaning the net present value of those triple-B tranches can be wiped out overnight. Nothing can save that tranche, because every static mortgage pool goes through a death spiral. Each month, the pool of mortgages generating interest income to offset credit losses shrinks.

A more accurate name for The Big Short is The Big Triple-B Short. As Greg Lippmann (played by Ryan Gosling in the movie), explained to everyone, once home price appreciation fell to, say, the rate of inflation, the triple-B tranches got wiped out. Neither he, nor any of the other protagonists in the book or the movie were betting against the housing market. They were investing in a sure thing; once home prices began to rise at a much slower rate, the subprime triple-Bs got wiped out.

Subordination reminds us of another immutable rule of finance: Time is money. An investment grade triple-B rating is supposed to signify some level of stability so that, if an entity becomes uncreditworthy, the process should happen gradually, over time (absent fraud). The internet didn't put Blockbuster Video or Tower Records out of business overnight. But, by definition, the NPV of any triple-B tranche is highly volatile. And a soft landing in home prices would easily wipe out the NPVs of triple-B subprime tranches overnight. Which was exactly what happened. (The rating agencies stalled their downgrade announcements during 2006-2008.)

Structured finance CDO ratings also ignored the impact of subordination, which is why they became toxic. You weren't really diversifying your risk if you insured 1,000 passengers in the bottom rung of steerage on the Titanic. Yes, the rating agencies ignored the impact of subordination, and they even doubled down in their ratings for CDO-Squareds. And, voila!

The people who developed the template for Johnson-Crapo are similarly clueless about the impact of subordination, which is why it's a crackpot idea.

Private mortgage insurers are acutely aware of the impact of subordination, which is an essential element of their business model. They were and remain the ones who insured the first loss on all GSE loans with an initial LTV higher than 80%. It's important to note that they sold insurance on a loan-level basis, almost never on a portfolio basis, the way that synthetic CDOs did, or the way that Johnson-Crapo proposes.

Private mortgage insurers also know something else. There is no such thing as a short housing cycle. Every single housing boom was followed, under a best case scenario, with a five-year period of price stagnation, when home prices appreciated at no more than 2% per annum. And many housing booms were followed by housing busts. And when prices are falling, you can't predict how bad things will get in the future.

I'm pretty sure Zandi is aware of the data. If he isn't, he should ask one of his colleagues the next time he attends a the Board of Directors meeting for MGIC Investment Corporation, the nation's largest private mortgage insurer.

So when home prices start falling, the last thing you'd want to do is assume the credit risk on a mortgage pool when you are subordinate to 90% of the debt. The NPV of your investment would be highly unpredictable and highly volatile, to say the least.

Put another way, the proposed setup is highly pro-cyclical.

The only private entities motivated to lend into a declining market were the GSEs. Everyone else pulled back to conserve capital. But since the GSEs have so much exposure throughout the country, they are motivated to lend as prices keep falling, in order to stabilize the market. That is also their legal mission. Zandi has written in the past on how FHA helped liquefy and stabilize housing markets during the crash, and he must know that the GSEs made a similar and larger contributions.

Fact Checking Zandi's Op-Ed

"Fixing Fannie and Freddie for Good"

Jim Parrott and Mark Zandi, The New York Times

In the longstanding debate about what should be done to overhaul Fannie Mae and Freddie Mac, the mortgage behemoths that taxpayers rescued at the height of the financial crisis, a growing number of groups, including several hedge funds and other investors, as well as civil rights groups and consumer advocates, are offering a surprising answer: Go back to the very system we just bailed out.

False: We didn't bail out that system because that system didn't fail. The biggest bailout was extended by unfortunate investors who were unable to seek legal redress for the fraud they suffered at the time when they bought private label securities. Moreover, it's a flat out lie to say that taxpayers bailed out the GSEs' credit losses, which, on a cash basis, did not cause the GSEs to become insolvent. Instead, $100+ billion in highly dubious accounting provisions were reversed, after the U.S. Treasury declared that it would sweep 100% of GSE profits so as to prevent the two companies from building up capital.

In September 2008, after the two institutions had racked up tens of billions in losses that had wiped out their capital, and amid fears about what their insolvency might mean for the American housing market and the wider economy, the then newly created Federal Housing Finance Agency stepped in to place Fannie and Freddie in conservatorship.

False: (Any fact checker who approved this should get a stern warning.) The GSEs absolutely did not incur tens of billions of losses before they were taken over by FHFA. They reported big losses in November 2008, after FHFA began dictating the GSEs' accounting provisions and credit assumptions. As The New York Times reported on October 28, 2008 and as disclosed on November 10, 2008, most of Fannie's "losses" were tied to the writedowns of deferred tax assets, which were meaningless in terms of the company's ability to service debt, and which had nothing to do with the GSE business model. And NO, those 3rd quarter 2008 losses did not wipe out all their capital.

Taxpayers have backstopped the two institutions and their mortgage securities ever since.

True, though those "bailout" draws were never used to fund cash shortfalls or operating expenses. In terms of funding loans and credit losses, the GSEs were self sufficient. And again, the piece ignores the fact that the Treasury backstop enabled the GSEs to stabilize the entire housing market, which benefitted all taxpayers and the global economy, not just the GSEs.

Yet, hard as it is to imagine, given the colossal scale of this bailout and the dramatic effect that their failure had on the broader economy, many are arguing that we should now resurrect Fannie and Freddie as the privately owned but taxpayer-backed oligopoly whose collapse contributed mightily to the financial turmoil and resulting Great Recession.

"Their failure had on the broader economy," is also known as The Big Lie. The statement is dishonest whether or not Zandi will admit it. Under the rules of capitalism, success and failure is measured on a spectrum. Capitalism is about credit, which is all about recovering principal plus interest and fees to offset credit risks. No mortgage lender anywhere can come close to matching the success of the GSEs.

An abundance of data shows that the GSE "collapse" was caused by contagion from the housing crash caused by private label securitizations. In the summer of 2007, one year after the housing bubble began to deflate and everyone recognized a national foreclosure crisis at the onset of the Panic of 2007, the GSE delinquency and loss rate was no different from what it had been two years earlier. It was only subsequently, going into into 2008, that their delinquency and loss rate began to rise about prior historic levels.

If you don't know how it works, let me explain it to you. If your neighbor's house is in foreclosure, the value of your house goes down. If many homes in your neighborhood are in foreclosure, the value of your house goes down a lot, putting your mortgage under water. And soon, when it's time to pay next month's mortgage intstallment, you think, why am I throwing away good money after bad?

More surprising still, one of the primary reasons offered by many proponents of this view is that we cannot end their stranglehold without decreasing competition in the mortgage market.

What they are really saying is that the market would be turned over to Too Big To Fail Wall Street banks, which paid pennies on the dollar for the institutionalized fraud they inflicted on investors. Again, Zandi is perfectly aware of what went down.

This view isn’t merely counterintuitive; it’s wrong.

Fannie Mae and Freddie Mac are among the largest financial institutions in the world, currently purchasing roughly one-half of the mortgages issued by lenders in the United States. They package and create securities out of these loans, and provide guarantees to the investors that they will be paid their principal and interest under any economic scenario. They thus act as critical gatekeepers in determining what kinds of mortgage loans lenders can make, and who gets a loan and under what term.

As Zandi pointed out, the crisis occurred because GSE role as critical gatekeepers was supplanted by private markets. There never would have been a mortgage crisis if GSE standards were applied universally.

Yes. And the numbers show that, as gatekeepers, the GSEs proved to be a stunning success, whereas the private sector proved to be a complete failure.

The GSEs capital was not sized to accommodate a 30% drop in home prices. As Zandi told the FCIC, "If the GSEs had more capital, they would still be with us today." (Zandi is not a lawyer, and seems too conflate conservatorship with nationalization.) So why is no one talking about reform as an increase in capital requirements? Parrott's colleague at the Urban Institute, Laurie Goodman, wrote that 5% capital would be excessive, given the currently stellar GSE loan portfolios.

(The essence of The Big Lie is to ascribe the failures of the private sector on to the GSEs. Which is why Ed Pinto contrived his wild definition of "subprime" so he could make the ridiculous claim that most subprime equivalents were financed by the GSEs, which led the race to the bottom. But as the FCIC found out, Pinto couldn't reconcile his risk categories to actual loan performance if you put a gun to his head.)

The concern is that any move to reform Fannie and Freddie by diminishing their dominance of the housing finance system will inevitably mean that the nation’s biggest banks will swoop in to take over their gatekeeping role. If they do, then these banks will use that power to their advantage, squeezing out smaller competitors.

Zandi doesn't advocate diminishing GSE dominance. He advocates abolishing the GSEs to be replaced by a U.S. government monopoly insuring 90% of mortgage debt issued under its umbrella.

This would indeed be a bad outcome. We would simply be swapping one dysfunctional system dependent on too-big-to-fail institutions for another with the same problem.

Well that's an over-the-top false equivalency, sort of like saying East Germans gave up one dysfunctional system for another when they unified with West Germany. The essence of The Big Lie is to ascribe failures of the private market on to the GSEs, in contradiction of all empirical evidence.

If this were what reforming Fannie and Freddie was all about, then the critics of reform would be right. But it’s not. (Projecting much?)

The point of the kind of reform that we support is to end the system’s dependence on too-big-to-fail institutions. It is critical to ensure that no institution central to the system has an incentive to take on excessive risk, knowing that taxpayers will bail them out if things go wrong, as happened with Fannie and Freddie and could happen in a system overly dominated by other too-big-to-fail institutions.

And again, the essence of The Big Lie is to ascribe the failures of the private market on to the GSEs. Unlike Lehman, AIG, Bear Stearns et al, the GSEs never faced any kind of liquidity crisis, which brought those other firms to the precipice or beyond. (I'll have another piece out shortly to address the mythology surrounding the GSEs' liquidity.)

In winding down Fannie and Freddie’s duopoly, Congress could and, we have long argued, should explicitly prohibit institutions that make mortgage loans from also playing the role of gatekeeper to the secondary market of mortgage-backed securities. Congress could also cap the market share of any single gatekeeper at a low enough level to preclude market concentration, or it could even create new gatekeepers to ensure that smaller lenders never are locked out of making mortgage loans.

First of all, the truth is that FHFA has appropriated all major business decisions of both companies. But like I said, this is about Wall Street firms making a grab for market share because institutional investors remember their history of failure with nominal accountability.

Legislative reform could be a long time coming, however, given the complex politics of the issue. In the meantime, the F.H.F.A. should work to ease the mortgage giants’ unhealthy hold on the market.

One of us, Mark Zandi, is on the board of a mortgage insurer; the other, Jim Parrott, advises several financial institutions in the housing finance industry. Some of these institutions could benefit from Fannie and Freddie reform, while others may suffer. But our focus is not the interests of these institutions, any more than it is those of the big banks or the shareholders of Fannie and Freddie. The aim of reform should be to create a healthier housing finance system, which means, among other things, one with greater competition.

Wanna a healthier housing finance system? How about presenting a product that has worked. Again, since no private player has ever come close to matching the level of disciplined credit underwriting demonstrated by the GSEs, it's nonsensical to say that we need to abolish the GSEs to instill "free market discipline" into the market.

The agency has already taken two steps that hold great promise: requiring that Fannie and Freddie share the risk they take when guaranteeing mortgage securities with a broad range of private financial institutions, and that they develop a common platform for offering securities on mortgage loans.

Done right, these steps could eventually open up the market to greater competition, reducing the dominance of Fannie and Freddie without enabling other too-big-to-fail institutions to take their place.

It is simply not true that we are forced to choose between one system dominated by Fannie Mae and Freddie Mac and another dominated by a few huge banks. The argument is at best ill considered, and at worst a red herring that will undermine any attempt to achieve significant reform.

It is simply not true to dispute the fact that Johnson-Crapo abolishes the GSEs in favor of a system that is dominated by the U.S.government, and, most likely, a few huge banks.

There is no reason we can’t create a dynamic mortgage market with plenty of competition, free of an unhealthy dependence on institutions we cannot afford to let fail. As we consider reforming the role of Fannie Mae and Freddie Mac, we should settle for nothing less.

Since the dismantling of Bretton-Woods, GSE market share has always risen when the two companies filled the void left by private market failures. To reduce GSE market share, the private sector needs to offer a credit product that works.

Fiderer on GSEs and other stuff

Tool Bar

Sunday, December 27, 2015

Martin Fridson Embraces The Big Lie To Challenge The Big Short

Martin Fridson Embraces The Big Lie To Challenge The Big Short

Maloni and Wallison Debate The GSEs

Maloni and Wallison Debate The GSEs

Wednesday, December 23, 2015

Congressional Record: On January 2, 2018, FHFA Director Mel Watt may release GSEs from conservatorship

(Clarification: The headline is based on my construction of the conservator's powers under HERA. The dialogue below simply quashes the notion that release from conservatorship would be at odds with Congressional intent.)

From The Congressional Record, December 18, 2015 S8850-S8851:

Sunday, December 20, 2015

Ya think Mark Zandi knows better? Translating his NYT op-ed into plain English

(After the blog retirement of my good friend Bill Maloni and the apparent semi-sabbatical of another blogger, I thought I would take the Monday weekly slot given up by Bill. As you can see the blog formatting itself is very much a work in process.)

But he clams up about some inconvenient truths

Zandi promotes the Johnson-Crapo Bill as housing finance "reform."

Fact Checking Zandi's Op-Ed