Larry Summers' Attempt to Rewrite Cramdown History

Larry Summers has a very interesting book review of Atif Mian and Amir Sufi's book House of Debt in the Financial Times. What's particularly interesting about the book review is not so much what Summers has to say about Mian and Sufi, as his attempt to rewrite history. Summers is trying to cast himself as having been on the right (but losing) side of the cramdown debate. His prooftext is a February 2008 op-ed he wrote in the Financial Times in his role as a private citizen. 

The FT op-ed was, admittedly, supportive of cramdown. But that's not the whole story. If anything, the FT op-ed was the outlier, because whatever Larry Summers was writing in the FT, it wasn't what he was doing in DC once he was in the Obama Administration.

Let's make no bones about it.  Larry Summers was not a proponent of cramdown.  At best, he was not an active opponent, but cramdown was not something Summers pushed for.  Maybe we can say that "Larry Summers was for cramdown before he was against it." 

Here's a telling paragraph from a National Journal article by Stacy Kaper called, The 'Cramdown' Fix Was Out, from Mar. 22, 2012.  Unfortunately, I can't find a weblink (but it exists in Lexis):  

William Longbrake, a former vice chairman with Washington Mutual who joined the Obama transition team in early 2009, recalls an administration divided over cramdown. "My sense is that Larry Summers probably ultimately brokered the solution," said Longbrake, now an executive in residence at the University of Maryland. "It was decided, but never publicly announced, sort of like, ‘Well, we are not going to oppose it, but we are just not going to support it.' … and then, ultimately, it died."

If Longbrake is correct, then we see, at best, a divided administration economics team. From what I recall of this debate in 2008-09, however, there was no signal of a true policy divide within the administration's team over cramdown.  (It certainly didn't help that they had no one on the team who knew bankruptcy.) At best there were some signs of internal power jockeying between Summers and Gene Sperling, but that's not the same as a true policy debate.

In any case, between February 2008 until cramdown's defeat in April 2009, I cannot find one public statement by Summers on cramdown.  If Summers truly supported cramdown, it's a shame that he wasn't more forceful because it would have made a huge difference if someone as influential as Summers within the White House team was pushing for cramdown. The only person in the administration who was noticeably supportive of cramdown was HUD Secretary Shaun Donovan, but Donovan was a marginal voice in this policy conversation.

Summers main critique of Mian and Sufi's book is that they're (very good) ivory tower economists, but that they don't understand the complexities of the policy world.  I am sympathetic to Summers' point that "it's complicated".  Cramdown was, but policy isn't like an academic debate in which noting nuances is how one scores points. Policy requires up or down decisions that swallow the good with the bad, and the Obama Administration (and Summers) made the wrong decision on cramdown.  At the end of the day, that's what matters.  

A.  Responding to Summer's Technical Claims

Still, Summers raises five major points Mian and Sufi do not address.  He is correct that Mian and Sufi do not address these points, but that does not mean that there are not answer, only that it isn't Mian and Sufi's fight. Let me go through these points systematically:

(1) Cramdown risked crashing the banking system.  Allowing cramdown posed three possible concerns for the banking system. Let's address each in turn.  

First, there would be large scale loss recognition from cramdown itself.  So what?  If the banks are insolvent as a result, recapitalize them. Recapitalizing a bank is not rocket science. There was private capital that was willing to invest in banks if the balance sheets were clean.  That's the whole point of the FDIC process. It cleans up the balance sheet for investors.  Perhaps some bridge financing would be needed, but giving the ingenuity that the Fed and Treasury displayed when they wanted to, I'm sure they would have found a way, legality be damned.  

Second, there would be some people filing for bankruptcy for relief who would otherwise have continued to pay on their mortgages.  The cramdown legislation took pains to ensure that bankruptcy wasn't going to be unduly attractive and to limit eligibility.  Perhaps more could have been done to avoid opportunistic filing, but there were ways to address the issue, and it bears emphasis that people do not usually file for bankruptcy lightly.  

And third, bankruptcy filings would have affected not just mortgage debt, but credit card and auto debt, etc.  The effect on other types of debt was a reasonable concern, but it could also have been addressed legislatively (that's what my Chapter M proposal aimed to do).  

The bottom line here is that cramdown would have resulted in loss recognition. That would have wiped out equity and possibly some bondholders in the large banks, which would have to be recapitalized. That was a result the Administration found politically unacceptable.  But legally and technically it was quite feasible.  Instead the Administration preferred to extend and pretend.  We're still paying the price. 

(2) Cramdown risked chilling future lending.  Nonesense. Everyone understood that 2008 was a 100-year storm and that the government wasn't going to be in the business of abrogating contracts willy-nilly.  In any case, however sacred contracts might be, they aren't economic suicide pacts and never have been.  Markets have responded very well in the past when the government acts like a grown-up and puts aside contracts that are socially detrimental, such as the case with gold indexation in the 1930s.  What's more, we knew that mortgage markets functioned just fine when cramdown was allowed in the past, and we knew that other consumer finance markets function just fine all the time despite the possibility of having debts wiped out in bankruptcy.  

(3) Cramdown posed a danger of prolonging the housing market's problems.  To the extent that problems are caused by negative equity, cramdown would have fixed them.  Chapter 13 bankruptcy confirms plans pretty darn fast.  We're talking months, not years.  It's really not clear how cramdown would have extended the housing market's problem.  Instead, cramdown would have forced the housing market to clear.  Instead, we got HAMP, which did exactly what Summers claims was a problem--delaying inevitable foreclosures through insufficient mods that were designed primarily to extend rather than resolve troubled mortgages.   

(4) Cramdown would have raised regulatory issues.  I don't know what Summers possibly means by this.  There were ZERO regulatory issues involved in cramdown. Cramdown did not involve any regulators.  It would have been handled by the courts. That may itself have been a reason the administration didn't like cramdown:  the executive branch couldn't control the process. Claiming that there were regulatory issues is an attempt to obfuscate, sort of a "You don't understand how DC works, mumble, mumble." 

Other, non-cramdown alternatives, such as pursuing something on the Home Owners' Loan Corporation model would have raised regulatory issues, but that's not cramdown.  Moreover, for Summers to state that the problem with a HOLC-type proposal is that it would have to buy mortgages at par is wrong. The original HOLC didn't buy at par.  Buying at par would be a subsidiy to the banks, but it's rather laughable for Summers to claim that the Adminsitration was opposed to a "massive backdoor subsidy to banks," given the enormous backdoor and frontdoor subsidies it doled out to the banking system. 

(5) Cramdown had implementation issues.  No, not really. That was the beauty of cramdown. It would have been very easy to implement.  It would have just added a change to the terms of what bankruptcy plans could be confirmed, but the entire infrastructure was already there.  If one wanted to start doing fancy stuff, like adding in a shared appreciation feature, etc. then implementation got trickier, but these features were not necessary for cramdown to work, but were instead, attempts to make cramdown more politically palatable. 

In any case, the Administration had no understanding of the bankruptcy system. This was one of my great frustrations in 2008-2009.  The Administration was staffed with economists and the occasional non-bankruptcy lawyer, none of whom had the foggiest notion of how consumer bankruptcy works, but a tremendous fear of the process.  Nor did any of these folks seem to want to get educated at the time.  I nearly fell off my feet when in I got a call in February 2010--after the failure of cramdown--from a respected economist who was workign as at the NEC with questions about the basic mechanics of Chapter 13. I couldn't believe that there wasn't a memo on hand, etc. covering the issue. 

The bottom line here is that of the five issues Summers raises, one is a made-up ideological concern (#2) and three weren't actually possible with cramdown (#3, #4, and #5) but instead conflate cramdown with a bunch of other housing relief ideas.  In the end, we're left with #1, which was a genuine concern, but was ultimately a question of whether immediate loss recognition and recapitalization (that is, a market clearing solution) was preferable to extend and pretend (delaying loss recognition to let the banks recapitalize with retained earnings). There was no body of scholarship clearly supporting extend and pretend; if anything, we knew the problems of that approach from Japan's experience.  Instead, the only grounds for the choice were distributional. The Obama Administration needs to own that one.  

B.  Responding to Summers' Political Claim

Finally, Summers writes that "critics who disagree at this late date are obliged to provide an alternative analysis of the political calculus, not a mere recitation of the arguments for cram-down." Let me take up the gauntlet.  

Not pursuing cramdown was the Obama Administration's worst political mistake.  Period.  It may well have cost Democrats the House in 2010 and ushered in the current era of complete Congressional dysfunction.  Democrats took a licking in 2010 for two reasons. First, they were held responsible for the state of the economy, irrespective of the shared blame for the crisis. Second, Democrats took a licking because the Administration was perceived (rightly) as caring more about Wall Street than Main Street.  Cramdown would have gone a long way to mitigating both criticisms. If cramdown had succeeded, it might have turned around the economy. Maybe not fast enough to matter for the 2010 elections, but we can't be sure. And even if cramdown had failed, had the administration had put some muscle into the effort, it could have painted the GOP as obstructing help for real people during the 2010 election. The Obama Administration could have carried the banner of the champion of Main Street instead of the protector of Wall Street.  By dropping the ball of cramdown, and then doubling down on the incompetent HAMP program (which was patently flawed from inception as was regularly pointed out by the Congressional Oversight Panel), the Administration ended up owning the financial crisis instead of pinning it on the GOP. The Administration is still paying the price.  

Summers' political story seems to be "We didn't have the votes, so it wasn't worth the coin." Of course, one reason they didn't have the votes was that they didn't put in the effort.  They didn't have the votes on the CARD Act until the President put in some muscle, and lo, it passed.  

Now I don't know how one tests these competing hypotheses, but I think I've provided a reasonably compelling alternative analysis of the political calculus.  It's quite fair to second guess the Obama Administration on cramdown, and for Larry Summers to try to position himself on the side of cramdown angels now is simply disingenuous.  

Comments

[Bet Mulligan]

I haven't read Atif Mian and Amir Sufi's book but you are absolutely right about the political fall-out of not pursuing cramdown. The Obama-loves-Wall-Street resentment is still alive and kicking even with main street Dems. That, plus NSA anger may usher in the next GOP pres.

[Ebenezer Scrooge]

I pretty much agree with Adam on the policy points, although it is worth pointing out that "extend and pretend" has has substantial policy success in the past. Many of the big banks were balance-sheet insolvent in and out of the 1980's (Mexico, S&L, etc.), but the powers that be correctly judged that their going-concern value was substantially positive. (Indeed, Milton Friedman excoriated NY bankers for denying "extend and pretend" to the Bank of United States in 1931, and blamed it on anti-Semitism!)

But on the politics, I'm not so sure. Adam ignores the Rick Santelli rant. I don't know if cramdown would have made Obama the champion of Main Street, but I do know it would have further energized the Tea Party.

[Adam]

Obama announced HAMP on February 18, 2009. Santelli's rant, which never mentions cramdown, was on February 19, 2009. His rant was about HAMP, not cramdown. He's complaining about "subsidizing losers' mortgages." That's HAMP, not cramdown. HAMP is the classic sort of ineffective government spending program that the Tea Party hates. Cramdown is just different. The Tea Party was angry because of the sense that Main Street was footing the bill for Wall Street. That wasn't going to be the case with cramdown, and it wouldn't have been hard for the Administration to explain that.

[cfaman]

Wow, this is an impossibly great reply to Summers' apology, one I would write were I able.

I wonder if on the inside they think "we are the most capable people at this time for this problem and we are well-intentioned" so when they look back they think "we did the best available at the time because we were the most capable with the most information" and that any critic must be Monday morning quarterbacking.

But of course there were a number of us powerless people who knew a few things right who were screaming "No!" as well as we could, to no effect.

[Tim Young]

I followed across from a link in comments on the FT article.

Well written, but I find the acceptance of cramdowns at odds with your attitude to banks (with which I agree wholeheartedly): "there would be large scale loss recognition from cramdown itself. So what? If the banks are insolvent as a result, recapitalize them.", the implication being write down the shareholders to zero and haircut the bondholders until the bank has positive net value, and then sell it.

So why should the house owners get to stay in their house at lower repayments? In my opinion, they should be evicted, albeit with relocation grants, even if it costs more than keeping them in "their" homes. It is as vital to the workings of a capitalist economy that individuals are as responsible for their actions as are shareholders. I doubt that there were many borrowers that were conned (which is a mistake anyway); most would have suspected that the mortgage offered was a bit of a stretch, but hoped that something would turn up and if not they would at least make money on price appreciation. It is not necessary to punish defaulting borrowers, but people should never to allowed to keep what they over-reached themselves to acquire. Let the people who prudently avoided such debt have their day in the sun, and pick up houses at lower prices, as in "There was private capital that was willing to invest in banks if the balance sheets were clean".

Or is it that there are more votes in bailing out defaulting mortgage borrowers than bank shareholders and bondholders?

[csissoko]

@Tim Young.

I suspect many others believe as I do, that economic stability is better served by placing a heavier duty on the banks/lenders to recognize when a bubble was growing in the housing market than by placing that duty on the homeowners/borrowers. Because most homeowners are incapable of judging the value of housing, placing a heavy duty on homeowners will simply reduce interest in homeownership. The banks by contrast have the ability to see bubbles growing, they just have no incentive to do so under the current system. The economy is best served when informational burdens are placed on those who can bear them at lowest cost.

[glasnost]

In order to believe that cramdown would have saved the Democrats from 2010, you have to believe that it would have effectively negated the massive recession. I think it would have helped, but I doubt it would have made it basically go away over the course of just one year.

It's very easy to overstate the extent to which the average voter has literally any idea whatsoever what any president has done, what laws have been passed, what's in them, and who's responsible. I think you're doing that right now.

I'm not sure if it would have made things worse. It's possible that it would have, but it's not clear how much worse it realistically could have been, politically.

It was clearly the right thing to do and would have helped us get out of the recession somewhat faster, to be sure. It would have helped people. I'm for it. The Admin was wrong.

However, we don't really know if it was politically infeasible or not. It's fair to say we didn't try very hard, but I also don't find it hard to believe, given how, for example, the Obamacare process went - we absolutely barely passed the single highest priority of the Obama Admin by the skin of our teeth and it took an entire year - to believe that there absolutely were not for votes for it in the Senate.

I think you have to work in a position where getting something accomplished depends on getting a law passed to understand how hard it is to get a law passed and how frequently people give up on it and/or how much money and time they spend working on it.

Recapitalizing the banks would have required a law, in the form of a budget, to be passed. This wasn't something the Admin could do without congress, and that is ultimately the weakness of this analysis.

[Adam]

Why would recapitalizing the banks have required a law? I'm talking about a private recapitalization. The banks could have been forced to massively dilute their shareholders because if they didn't do so voluntarily, the administration could have put them into an FDIC receivership and resold the banks, wiping out the shareholders in the process, and maybe the long bonds as well. Perhaps some bridge funding would have been needed from FDIC, but to the extent that the FDIC's Deposit Insurance Fund would run dry, the FDIC is authorized to issue bonds backed by the full faith and credit of the United States. I don't see any appropriations requirement here. We've done a version of this before: Resolution Trust Corporation.

I am skeptical that cramdown would have turned things around immediately, but even if it didn't and even if the Administration failed to pass it, it would have shown that the Administration cared and was trying to do something to help real people, not just to help banks.

[Tim Young]

@csissoko (Caroline I presume),

Your view is reasonable and workable, but then lenders would bear a greater share of the risk, probably meaning that mortgage rates would be higher, and that lenders would be more selective about who they lent to. Cue moaning about the high cost of borrowing and the exclusion of poor people - probably with certain ethnic groups disproportionately represented - from credit.

The bottom line is that people in the old industrialised countries (I am from Britain, but I think that the situation is similar there) have become decadent, and demand that their borrowing terms easy, while simultaneously expecting high risk-adjusted returns from their investments (in practice meaning taking more risk when returns are low). That, as much as anything the "bankers" did, was the cause of the financial crisis, and this needs to be accepted, otherwise the financial crisis will be repeated with a different set of financial institutions.

[Karen]

Handing over tons of cash to the banks was the moral hazard. Any fool could have seen that coming. I can't believe the administration didn't see that coming. Currently homeowners are still in trouble and the banks are as unscrupulous as ever and certainly bigger and more hazardous than ever. With all this financial jargon, charts, mumbling of possible regulations meant to distract, the administration was clearly in bed with crooks, they knew it, and they didn't care because it provided campaign help and big donations. Their big mistake, right out of the gate, was in not helping Main Street.

[A]

On the other hand, Lehman Brothers. It's likely true that cramdowns would have exceptional tools given a stable nominal expectations environment. But without monetary coordination, it could have signaled to the market "It's time to reset your nominal expectations. To what? Let's find ouuut!"

[Jim Three]

Agreed that enactment of cramdown was would have been a Good Thing and also that it would have been politically smart for Obama to have been strongly and publicly supportive of it even if convinced it couldn't pass. (Just as it would have been politically smart to have strongly and publicly supported a much bigger stimulus program, even if that couldn't have passed.)

But this post misrepresents Larry Summers's positions in two important ways. (There are plenty of reasons to oppose many of his actual positions and actions on many issues, no need to misrepresent...)

First, Summers does NOT present himself to have been an advocate for cramdown within the administration in spring 2009, only that he personally believed it was good policy [as evidenced by Feb 2008 FT column]. His position within the administration seems to have been that cramdown was not worth fighting for because it had no chance of passing -- regardless of its merits. Your Kaper/Longbrake quote is entirely consistent with what Summers says in his "House of Debt" review.

Second, after one paragraph on the cramdown proposal, Summers turns to "various ideas put forward for bringing about mortgage relief through either coercion or financial inducement" -- that is, to ideas other than cramdown. He then offers five "considerations" against these "various ideas." Your post treats these as five "points" against cramdown -- but these five points were NOT about cramdown.

[jennifer]

**applause** and a request: don't allow any of the kudos posted here in response to your excellent response make you feel in any way good or satisfied. If an intelligent, informed response to a slippery politician is only posted on an academics blog for academics/insiders to hear, does it make any noise? If you don't pipe up, you might as well help him measure for curtains (Yellen currently has some hanging in cream). I hope at the very very very least you are sending a response to the FT - it's one thing to not get enough press for a great point, but it's entirely another to not try to.

[Christopher Lowery]

@Tim Young,

The current system of neither works well economically nor socially. Evictions and foreclosure sales unnecessarily depress home prices across the board in communities in which they are prevalent, including the values of homeowners with low leverage, or having mortgage loans well within their ability to carry. They are also a poor way for lenders to recoup their capital, as foreclosure sales produce prices substantial below non-distressed sales.

I've never understood why cramdowns in residential mortgages are looked on so negatively when they have consistently worked in the non-residential sectors so well.

It's my guess that, had not securitization taken over the mortgage market, removing the ability of a rational party to make a sound business decision, that cramdowns would have been the natural process to resolve many of the problem loans. Unfortunately, the skewed interests of bond trustees and servicers effectively removed any rational decision-making from the process.

[Dwight Snider]

We had recently returned from living in France, and the French bank was holding the money we received from the sale of our house.
They weren’t doing it to calculate capital gains tax because we were legal residents of France at that time.
They were doing because our agent was en vacances. And no, no one else could send us our money.

Anyway, we bought an investor home in the superheated Phoenix Arizona area, and the broker switched terms from a fixed loan to an ARM subprime loan just before the deadline.

We should have walked and started over, but prices were shooting up.
Four years later, I saw the handwriting in our contract: The hammer was going to drop on my ARM in one year.

GMAC was no help. They were about to go belly up themselves.
I found a loan mod agent (An unemployed former mortgage broker, of course) who just happened to know an upper level underwriter for GMAC.

He brokered a loan mod for 2% for five years, giving us a little breathing room.
Then we were swamped with e-mails about loan mods, HARM, and now HARP.
No bankruptcy lawyer ads, of course.
I would have loved to shove it to GMAC. Credit rating be damned.
After all, I wasn’t trying to borrow more money; I was trying to pay off debt.

These so-called mod programs were jokes that were quickly appropriated by scam artists.
The problem is that nobody knew what they were doing, and the loan originator claimed (truthfully) that they had no employees at their bank that knew how to do a loan mod.

The only way out was to stop paying the loan originator and put 3 months mortgage payments in a savings account we had not been using, sort of our own escrow account.

Then, GMAC was willing to dicker. The loan mod came through one month before the ARM was going to go up. It took 14 months of letters, phone calls and e-mails to get it.
We had to pay fees with Western Union money orders, a real pain and totally unnecessary, since we had the cash in our “personal escrow” account, and GMAC knew it. In fact, they insisted we have it before they would negotiate.

Recently, I sent e-mails to President Obama, and signed every petition to put Janet Yellen in the FED instead of Summers.
Why Obama ever considered Summers is beyond me. (But then Clinton took Rubin’s advice over Reich’s, so what’s a poor boy supposed to do?)
Rant over.

[mt]

The professor's respopnse to Ebenexzer Scrooge is bizarre: Santelli didn't mention cramdown" ??? Yes, because Obama didn't. Had Obama proposed it, Santelli would doubtless have included it in his rant. It was not exactly a nuanced rant! The point was, and remains, that when you have a majority of voters paying their mortgage under difficult conditions, a selective mortgage reduction program that generally benefits the least prudent offends the mortgage paying class. It wasn't the specifics of HAMP that formed the Tea Party, it was the general sense of the imprudent being preferred over the prudent.

And to leave the Affordable Care Act out of the explanation for the 2010 rout is too silly to deserve a further rebuttal.

[mt]

If we look at mortgage debt outstanding at the end of 10, the earliest year I can find stats for, there was roughly 10.4T 1-4 family mortgage debt outstanding. Of that roughly 5.8 was held or guarnateed by a govt agency. roughly 2.6 was held by banks or insurance companies. Roughly 1.3 was in private mortgage pools, and roughly .7 was held by individuals. Thus the majority was govenment held or guaranteed, and on that portion the taxpayer would have been perceived in 2010 to be the one to absorb any loss from instituting a cramdown program as the GSE's and FHA were perceived insolvent at that time.

One might believe such a cost would have been worth it, but there would clearly have been a battle royale politically over any attempt to incur it, given that the hypothetical benefits would have accrued to a minority of mortgaged households who generally would have been less prudent than the majority.

Numbers from
http://www.federalreserve.gov/econresdata/releases/mortoutstand/current.htm

[Adam]

@mt--agreed that Santelli's rant wasn't nuanced and that he probably wouldn't like cramdown either, but cramdown wasn't preferring the imprudent over the prudent. Being underwater in 2009 wasn't necessarily a sign that someone was imprudent in their borrowing.

As for the ACA, I don't think it explains 2010. I think it contributed, but that the damage was done long before. If you look at how Obama's approval ratings moved. They crashed throughout 2009, well before the ACA push. In 2010, after ACA passed, but before the election they declined further, but not nearly so rapidly.

Regarding the scope of the write-downs, your comment implies that the mortgages would have been written down to zero. That's not the case. It would have been negative equity only that would have been eliminated. Not everyone would have filed for bankruptcy, and negative equity was not equally distributed among all mortgages. A vastly disproportionate share was in private label pools. It would have taken (if I recall) perhaps $400B to wipe out all of the negative equity, but to put that in perspective, that was around the market cap at the time of the major financial institutions in the US. Which we could have recapitalized.

http://www.washingtonpost.com/blogs/the-fix/wp/2013/01/30/president-obamas-popularity-surges-to-three-year-high/

[Ken T]

Even if cram down was allowed, Chapter 13 Debtors would have been stuck in 5 year mortgages, payable in equal monthly payments per 1325(a)(5).

For below median income debtors, a Chapter 13 Trustee would likely object that accelerating a 30 year mortgage to a 5 year mortgage is not "reasonably necessary" at the expense of general unsecured creditors. So even if they could afford the payments on just the 5 year mortgage plus Trustee fees, they might have also had to pay money to unsecured creditors, plus attorney fees.

For above median income debtors, the Means Test allows a deduction for payments contractually due in the next 60 months. Meaning that even if you are cramming down in 5 years, you only get to deduct the payments required by the contracts in determining "disposable income." So above median income debtors would also have to pay the value of their home, plus interest, in 5 years, plus attorney fees, Trustee fees, and funds likely to general unsecured creditors.

Knowing that, I cannot accept that simply allowing cram down by removing the relevant language in 1322(b)(2) would have done much of anything. Even with home's valued at 50% of the mortgage, Debtors would have had trouble paying that 50% plus interest and other costs within 5 years.