Monday, July 31, 2017

Housing: Part 247 - The elasticity of causation

Here is Matt Taibbi on the Fannie and Freddie lawsuit in Rolling Stone.  This is an interesting piece.  Taibbi is a compelling writer, and I give him props for recognizing the apparent abuse of power here.  It would probably be easier for him to take the position against the dreaded speculators and hedge funds who have sued the government.

I don't spend a lot of time reading popular assessments of the financial crisis, so it is always eye opening for me to step out of my conceptual shell and to be reminded of the things most people have come to confidently take for granted.

Here are a couple of excerpts regarding the GSEs and the crisis:
They had gone bust during the crash years for a variety of reasons, mostly due to incompetent and corrupt management. But by the summer of 2012, with the real estate market in recovery, the companies weren't bust anymore. On the contrary, they were about to start making money again – enormous piles of it, in fact...

...It should be noted that despite legends to the contrary, Fannie and Freddie's affordable housing mission did not cause the 2008 crash.

In fact, the Financial Crisis Inquiry Commission concluded that delinquency rates for GSE loans were "substantially lower" than those of the private banks and mortgage companies that were lending subprime loans to anyone with a pulse during that era.

The crash was caused by greed, not social policy...

...probably because Fannie and Freddie were so unpopular after the crash – deservedly, in part, because of numerous scandals involving its executives – the companies were treated very differently than other bailout recipients.

I think this is a great example of how flexible our collective sense of morality can be.  And, I don't think there is anything unusual, here.  Taibbi is a great journalist, in part, because he has an intuition about the audience's contradictions.

Notice how easily he floats here between certainty that every problem the GSEs had was due to mismanagement and corruption and how no problem is explained by the affordability mandates.  "The crash was caused by greed, not social policy." is an axiom.  Even within paragraphs of each other, when mentioning social policy, we should note that the GSEs were quite well managed.  When mentioning management, we should note that the GSEs were managed incompetently and corruptly.

In fact, neither of these factors has much to do with anything.  The GSEs were designed to be vulnerable to one thing, and one thing only, and that is a massive, multi-sigma downturn in the housing market.  There is no other context where they would ever have failed, and there was no way for them to avoid failure in that context if the government decided to strictly invoke capital requirements.  They faced a multi-sigma downturn, the government invoked strict capital requirements, and they failed.  Everything else is noise.  Taibbi and his readers share a set of axioms, so these axioms are repeated ad nauseam until the repetition itself creates a sense of empirical fact in their minds.

Other audiences reverse this axiom and blame social policy.

The fact that, given the context and the position of the treasury in 2008, failure was inevitable, allows all axioms to be acceptable to their respective audiences.  One could argue that the cause of the collapse was a happenstance where the axioms of a plurality of Americans caused them to support liquidationism in 2007 and 2008.  This is obvious in hindsight, when the bulk of criticism toward federal officials is related to their attempts at stability rather than at the instability itself.


One other quote from the paper:
In most versions of GSE reform currently winding their way through Congress, the same too-big-to-fail banks that blew up the mortgage markets in 2008 would assume most of the responsibilities of Fannie and Freddie. Crucially, securitized mortgages would continue to enjoy government backing under many of these proposals.

Privatized profits, socialized losses. Who doesn't love that formula?
I note this excerpt as an example of the sorts of little errors that slip into our common discourse which allow us to mend and protect our axioms.  Again, there is nothing unusual or special here.  We all have these.

Did you notice what he ignores here?  It is an uncontroversial fact that yields on these securities that have federal backing are lower because of the backing itself.  In free markets, it could be no other way.  It is impossible for markets to privatize these profits.  First, because the guarantee earns a profit over time, through fees, which Taibbi implicitly understands when he notes that the GSEs have returned to profitability.  Over the long term, they have been profitable, even after having taken nearly a quarter trillion dollars in credit losses during the bust.  (When their incompetent and corrupt management managed, somehow, to maintain those strikingly low default rates.)  Secondly, because investors bid down the yields on those guaranteed MBSs to reflect the value added of that guarantee.

"Who doesn't love that formula?"  The formula is a figment of Taibbi's imagination.  This subtle mistake, along with many others, serve to support the axiom.

You can see a similar sort of convenient error in the way everyone talks about CDOs.  The way they are always described is that they were creating AAA securities that really had hidden embedded risks.  Bankers trying to boost their yields were buying them up because they were too stupid and too greedy to understand that.  And that greed is what done us in.

Now, I have written a lot about how this is entirely wrong, anyway, because the CDOs, and especially the more exotic CDOs that had more exposure to systematic risk, were a late phenomenon that happened almost entirely after home prices and housing starts peaked, and certainly happened well after homeownership peaked. (In fact, falling home ownership was the reason they ever needed to be created in the first place).

But, ignoring all that, this way of describing CDOs is wrong in much the way Taibbi's comment above is wrong.  Yield is a reflection of risk.  If yields on the AAA securities from the CDOs were higher, it is because the market clearly recognized the potential for added risk.  The yields weren't bid down to a riskless yield level.  The mis-statement about the CDOs is usually described as "reaching for yield" which is one of those phrases I wish we could just blast from the lexicon.  That phrase really only serves one purpose, and that purpose is to promote this lie, that perceived risk and yield are somehow unconnected - that yields are just independent variables out their in the market, and greedy investors chase them up in a short-sighted attempt at capturing more income than is sustainable.

It just doesn't work that way.  All yields are codetermined with perceived risks.

This axiom is so deeply ingrained, that few noticed the severity of the dislocation in the summer of 2007.  By normal standards, things like monetary policy weren't wildly off the mark, but by August 2007, even though there were billions of dollars of savings, madly in search of safe assets - a "bubble" in AAA securities, in the common and perverse usage of that term - those securities were trading at deep discounts.  There was a massive demand for safety, and the securities meant to be safe were failing.  Then even auction rate securities markets started failing.  Again, despite billions or trillions of dollars in search of safety.

This is recognized as a regime shift from greed to fear.  That's truish, as far as it goes.  But, what if the description of reality was closer to having moved from normalcy to fear.  Even if one concedes that 2005 could be characterized by "greed", certainly that hadn't been the case for some time by 2007.  In fact, the regime shift in 2007 was probably from fear to crisis.  We had already passed clean through normal.

But, in a paradigm that sees only two regimes - greed and fear - fixing the fear problem necessarily means creating a greed problem.  We didn't fix the problem because that meant we'd go back to the greed regime, and that's what caused all our problems, right?  Reaching for yield.  Privatized gains and socialized losses.  We had to avoid those things.  And, weren't those incompetent and corrupt managers the source of the problem, anyway?  We're just supposed to let them off the hook?  Our axioms demanded, "No way."


  1. I read this post a few times. Lots of good points. Taibbi is a terrific writer. But many journalists (and I was and still am one) struggle with financial and business topics. Maybe we all do. I have spent a profession lifetime in financial news, and it still feels opaque sometimes.

    As a free marketeer, I guess no Fannie and Freddie is okay with me. I suspect private mortgage insurance would pick up the ball, perhaps connected to higher down-payments.

    In Australia, the banks extend recourse loans, with private mortgage insurance required if the LTV is too high.

    Aussie house prices are through the moon despite no Fannie or Freddie and with stricter underwriting, so I doubt "easy money" or "easy underwriting" alone causes house prices to rise.

    Australia does run current-account trade deficits and has property zoning.

    I guess there is a lot of demand for Aussie housing but restricted supply.

    A digression; George Washington and Thomas Jefferson were large debtors, and complained about it in their private papers.

    The Founding Fathers, as a class, were landowners, and borrowed money to develop land, and bring product to market.

    This led to a "pro-debtor" state of mind and law, including federal bankruptcy courts. It is notable that the topic of bankruptcy is federalized and treated in the US Constitution. I suspect non-recourse provisions are part of that tradition, that debtors must have a way to dispel debts in dire circumstance.

    As elites in the U.S. have evolved, so has bankruptcy law.

    "The United States Constitution (Article 1, Section 8, Clause 4) authorizes Congress to enact "uniform Laws on the subject of Bankruptcies throughout the United States." Congress has exercised this authority several times since 1801, most recently by adopting the Bankruptcy Reform Act of 1978, as amended, codified in Title 11 of the United States Code and commonly referred to as the "Bankruptcy Code" ("Code"). The Code has been amended several times since, with the most significant recent changes enacted in 2005 through the Bankruptcy Abuse Prevention and Consumer Protection Act of 2005 (BAPCPA)."

    BTW, there was a parallel and equal decline in commercial real estate in the US in the Great Recession.

    This would also suggest something general tumbled US property prices, whether commercial or residential.


    Hong Kong runs big trade deficits

    has the world's least affordable housing

    so what gives?

  3. After the accounting scandals, Fannie & Freddie lost execs with housing cycle experience. The new execs were told by equities analysts that they were becoming obsolete due to private label securitization and that they had to compete. When the private label market contracted in late 2006 & 2007, the GSE's bought Alt-A loans massively, thinking they were buying a correction rather than a collapse. In the despairs of the panic, their credit losses were significantly over estimated.

    Morality, ideology, and politics make financial and economic narratives more appealing, but often less insightful.