Monday, August 22, 2016

Housing: Part 175 - About Mian & Sufi's "extensive margin"

Mian & Sufi's very timely work was an early nudge on the path of consensus views about the housing bust.  Their first popular work on the subject was published by early 2008 - before 90% of the foreclosures had even happened.  I will spend some time in the book explaining how both their findings and the recent findings of Foote, Loewenstein, Willen, etc. all can be true because of the particular shape of the American housing market - the two Americas, closed and open.

Mian & Sufi talk about the extensive margin (marginal new homeowners) and the intensive margin (existing homeowners who borrowed more because of available credit).

As I have worked through the data and the narrative I am building, I have realized what a central role the decline of the GSEs and federal housing agencies have played (Freddie, Fannie, Ginnie, FHA, etc.) in the whole affair.  The harassment and takeover of the GSEs by the Bush administration may be the most important story of the last 20 years.  It seemed to be motivated by a dislike of "crony capitalism", of which the GSEs seemed to be a prime example.  There is some legitimacy to that point of view.  In fact, I will argue that there isn't much reason to have the GSEs issue debt to own loans and MBSs outright, though there may be a place for them to continue to issue guarantees.  This is more or less what the Obama administration has called for.

Though, the way these concerns were handled might have been the main cause of all of the negotiated "bailouts" and emergency economic policies that followed.  And, given Bush's rhetorical concentration on the "ownership" society, this decline is ironic.  It appears that the GSEs had been an important source of relatively safe homeownership, and this support for ownership has been significantly undermined.

The Bush administration has been gone for 7 years, and since then the Obama administration and Congress have only worsened the condition of the GSEs.  The fixation on predatory lending which I believe has been due to a misidentification of the causes of the housing boom and bust has led to an imposition of standards on the banks and GSEs that has basically shut down homeownership for households that are most dependent on credit.

The association of credit with predation has been quite damaging for American households on the margin, for whom credit, even with its risks and difficulties, is a necessary ingredient for progress and financial management.  The moral panic has created much more pain than moral hazard ever did.  As Henry Paulson says in his book (pg. 74) about the summer of 2007, "The administration’s goal was to minimize as much as possible the pain of foreclosure for Americans, without rewarding speculators or those who walked away from their obligations when their mortgages were underwater."  This was clearly the policy at the Fed, too.  So, policymakers spent months trying to work out ways to selectively freeze interest rates for homeowners who qualified as sympathetic while we kept interest rates high for "speculators".  The problem is that, the core of the "bubble" was that the very act of living in San Francisco or New York City or other Closed Access cities that define themselves by housing deprivation is necessarily an act of speculation.  Those sympathetic homeowners the Treasury was trying to select for assistance were the speculators that created the bubble.  In fact, the most speculative households of all were Closed Access renters, who exposed themselves to extreme fluctuations in cost.

Before the speculators that chose ownership had their crisis, those lousy speculators who chose renting had been having a crisis for years.  Millions of them saw their speculative housing positions fail.  (They were the worst kind of speculators.  They were shorts.  Very large short positions on housing.)  When their speculation went against them, they had to cover their short positions, which meant moving to a city with less expensive rent.

So, while the Treasury and the Fed were trying to push the "speculators" into failure while saving good, wholesome Americans, it turns out they were pushing good, wholesome Americans into failure.  As far as I can tell, this is one of the few things policymakers accomplished over the past decade which enjoys nearly universal public support - busting that darn bubble.  Done and done.  Who says Americans can't come together to get things done anymore?

Anyway, back to Mian & Sufi and the extensive margin.  The "subprime bubble", where private securitizations briefly claimed much higher market share than they had before or since, lasted from 2004 to mid-2007.  Would you believe that the "subprime bubble" was associated with the sharpest 3 year decline in homeownership, up to that time, since records have kept (1965)?

The period that everyone associates with this massive predatory push for homeownership at the margin was actually associated with a massive outflow of households from owning to renting.

That is because a big part of the cause of the rise of private securitizations was the harassment of the GSEs.  Private securitizations were filling the gap.  The terms of the privately securitized mortgages, being removed from the bureaucratic norms of conventional loans, were more amenable to ownership by Closed Access households who generally had high incomes but were living in or moving to high cost Closed Access cities.  But, the retreat of conventional mortgage issuance, which appears to have been associated with healthy homeownership, was the more important factor.

Here is a graph comparing the annual growth of owner and renter households.  On the extensive margin, the "bubble" was associated with a sharp change in trend away from ownership.  Since the GSEs have been taken into conservatorship, the number of homeowners, in absolute terms, not even adjusted for population growth, has declined.

In Paulson's "On the Brink", page 437:
Our decision to put Fannie and Freddie into conservatorship forestalled their collapse and prevented a wider financial system meltdown.  Crucially, public backing for the GSEs from Treasury and the Federal Reserve ensured that affordable mortgage financing was available during the worst moments of the financial crisis and beyond.  This would prove to be the government's most effective form of economic stimulus, helping to put a bottom on the sharp home-price declines that had so damaged the GSEs and were driving the country into a deep recession.  Without such public support, I am convinced, the housing market would have ground to a halt for lack of financing, home prices would have continued their downward spiral, foreclosures would have skyrocketed, and financial institution balance sheets would have suffered much greater losses.  All of this would have led to an even more prolonged downturn - and the additional loss of millions of jobs.
Now, there is always the problem of the counterfactual.  And, in Paulson's defense regarding the accuracy of this paragraph, it is true that our policymakers were capable of enacting much worse counterfactuals than this.  And, they would have been popular for having implemented them.  In some ways, the Treasury and the Federal Reserve were a buffer between what the American people demanded and what needed to be done.  So, when the Federal Reserve eventually did try to stabilize the economy, they took a lot of criticism from both ends of the political spectrum.

But, Paulson is simply in denial (along with many others) here regarding the American's who could have most used support.  The GSEs, under conservatorship, did nothing to support housing where it was most in need of support.  How much of today's ill-tempered populism is due to this denial?  And, it comes down to the misidentification of the housing boom with marginal homeownership, which seems to have left nearly everyone to support the decline of credit to


  1. Kevin:

    It's not clear where your line of thinking from the last few posts is going but I'd like to suggest that there is an impossible trinity here.

    No matter how anyone views the GSEs, the Treasury / Fed response to the crisis, closed vs. open access cities, etc., it's not obvious how it's possible to simultaneously have a) widespread mortgage affordability - low down payments, high LTV, low FICO, GSE-guaranteed 30-year fixed mortgages, b) individual and systemic safety - low risk to borrowers, taxpayers, and the financial system, and c) housing equity appreciation.

    You can have at most any two of the three:

    Safety and Affordability does not permit equity appreciation. Safe and affordable access to leveraged equity appreciation is a free call option with no expiration date. This is a perpetual money machine that always ends badly.

    Safety and Appreciation does not permit widespread mortgage affordability. It's certainly possible to have appreciating equity values in a safe mortgage system so long as the access to credit and leverage is highly restricted. This is essentially the mortgage-equivalent of a closed access city with the government ensuring that equity is preserved by choking off credit the same way it chokes off new construction.

    Affordability and Appreciation does not permit safety since this becomes an open-ended call on unlimited government guarantees, Fed / Treasury backing, and permanently growing private sector investment. This is the same model behind what used to be the risk-free debts of peripheral European banks and sovereigns as well as the bonds backing underfunded State and local defined benefit pensions. It works as long as the guarantor can repay the debts in money it can print but it doesn't work very long for anyone else.

    In more colloquial terms this is the conflict between housing as a depreciating and high-maintenance consumer durable (in terms of consumption) and housing as an appreciating, leveraged, and highly tax-advantaged asset (in terms of investment). It's certainly possible to keep both of these concepts going at the same time provided there's perpetually increasing leverage, credit access, tax-advantages, etc. but ultimately the system breaks down.

    My feeling is that your criticisms of the GSEs, etc., while well-founded, are going to wind up with the systemic breakdown, whether in the trilemma or dilemma framing above.

    1. Thank you for the thoughtful input. I realize that I have been thinking through these issues as I have been writing the book, but not airing some of those thoughts out on the blog. There are some ways I would re - frame the way we think about these things, but I think I need to reply in a post or a series of posts. The short answer is that I think imputed rent income and life cycle regime shifts are much more important than they are generally made out to be. Debates about these issues can carry on with no mention of rent or of the effects of ownership outside short term capital gain considerations, but they should be the most important factors. And appreciation is not important to me. The reason I take aim at post crisis policy isn't because it led to low prices. It's that it led to inefficient prices that favor the least vulnerable households and create systemic distress. I wouldn't want to favor high prices just for their own sake either, but one way i would disagree with the consensus is that too low prices have been a much larger problem than too high prices.

  2. MBS failures were the result of Fed mispricing risk for those bonds. Then, the Fed not only mispriced the risk, causing many of them to fail, but it allowed the commercial paper market to be destroyed, fiddling like Nero.

    The expectation of housing appreciation spurred HELOC lending. Once HELOC lending was taken away, the economy took a nosedive, but that was a year after the destruction of the commercial paper market. The destruction of commercial lending happened in the Great Depression as well. Only it was even worse then.

    But either way, part of the economy was liquidated due to procyclical policies. And the Fed encouraged appreciation of homes. In Feb, 2004, Alan Greenspan said you could get a "better deal" through an adjustable mortgage. That was a direct encouragement of appreciation.

    So, the Fed allowed main street to speculate, unknowingly to themselves, and then pulled the carpet out from under them.

    That is dirty low down Fed behavior.

    1. So, I just wanted to add Kevin, that since the Fed mispriced risk, it made very easy money available in many areas. That easy money brought demand forward, causing prices to rise irrationally. The ease of loans pushed the prices up. That was even true in, say, Manhattan Beach, where prices shot up and then declined by about 100k. I don't have the exact figures.

      But obviously Manhattan Beach was not like Fresno, or Stockton or Las Vegas, where prices were cut almost in half as speculators were forced out.

      Of course, Manhattan Beach, being almost invincible to house priced decline (the decline was small and the prices bounced back), suddenly became in more demand after the Great Recession, and prices are much higher than at the peak price last decade.

  3. I hear people complain that the banks didn't lend out the TARP money etc. Without mentioning that the Fed paid zero IOER for 95 years and then started paying them in October of 2008. This baffles me to this day. The Fed now pays more to borrow overnight than the Treasury pays to borrow for 1-180 days. Now that's crony capitalism. How can I get myself some of that?