A regression discontinuity test error

This is post 3 in my HAMP and principal reduction series. For the introductory post see here.

The series is motivated by Peter Ganong and Pascal Noel’s argument that mortgage modifications that include principal reduction have no significant effect on either default or consumption for underwater borrowers. In post 1 I explained how the framing of their paper focuses entirely on the short-run, as if the long run doesn’t matter – and characterize this as the ideology of financialization. In post 2 I explain why financialization is a problem.

In this post I am going to discuss a very technical problem with Ganong and Noel’s regression discontinuity test of the effect of principal reduction on default. The idea behind a regression discontinuity test is to use the fact that there is a variable that is used to classify people into two categories and then exploit the fact that near the boundary where the classification takes place there’s no significant difference between the characteristics of the people divided into the two groups. The test looks specifically at those who lie near the classification boundary and then compare how the groups in the two classifications differ. In this situation, the differences can be interpreted as having been caused by the classification.

Borrowers offered HAMP modifications were offered either standard HAMP or HAMP PRA which is HAMP with principal reduction. In principle those who received HAMP modifications had a net present value (NPV) of the HAMP modification in excess of the NPV of the HAMP PRA modification, and those who received a HAMP PRA modification had an NPV of HAMP PRA greater than NPV of HAMP. The relevant variable for classifying modifications is therefore ΔNPV (which is economists’ notation for the different between the two net present values). Note that in practice, the classification was not strict and there was a bias against principle reduction (see Figure 2a). This situation is addressed with a “fuzzy” regression discontinuity test.

The authors seek to measure how principal reduction affects default. They do this by first estimating the difference in the default rates for the two groups as they converge to the cutoff point ΔNPV = 0, and then estimating the difference in the rate of assignment to HAMP PRA for the two groups as they converge to the cutoff point ΔNPV = 0, and finally taking the ratio of the two (p. 12). The authors find that the difference in default rates is insignificant — and this is a key result that is actually used later in the paper (footnote 30) to assume that the effect of principle reduction can be discounted (apparently driving the results on p. 24).

My objection to this measure is that due to the structure of HAMP PRA, most of the time when ΔNPV is equal to or close to zero, that is because the principal reduction in HAMP PRA is so small that there is virtually no difference between HAMP and HAMP PRA. That is, as the ΔNPV converges to zero it is also converging to the case where there is no difference between the two programs and to the case where principal reduction is zero.

To see this consider the structure of HAMP PRA. If the loan to value (LTV) of the mortgage being modified is less than or equal to 115, then HAMP PRA does not apply and only HAMP is offered. If LTV > 115, then the principal reduction alternative must be considered. Under no circumstances will HAMP PRA reduce the LTV below 115. After the principal reduction amount has been determined for a HAMP PRA mod, the modification terms are set by putting the reduced principal loan through the standard HAMP waterfall. As a result of this process, when the LTV is near 115, a HAMP PRA is evaluated, but principal reduction will be very small and the loan will be virtually indistinguishable from a HAMP loan. In this case, HAMP and HAMP PRA have the same NPV (especially as the data was apparently reported only to one decimal point, see App. A Figure 5), and ΔNPV = 0.

While it may be the case that for a HAMP PRA modification with significant principal reduction the NPV happens to be the same as the NPV for HAMP, this will almost certainly be a rare occurrence. On the other hand, it will be very common that when the LTV is near 115, the ΔNPV = 0, which is just a reflection of the fact that the two modifications are virtually the same when LTV is near 115. Thus, the structure of the program means that there will be many results with ΔNPV = 0, and these loans will generally have LTV near 115 and very little principal modification. In short, as you converge to ΔNPV = 0 from the HAMP PRA side of the classification, you converge to a HAMP modification. Under these circumstances it would be extremely surprising to see a jump in default rates at ΔNPV = 0.

In short, there is no way to interpret the results of the test conducted by the authors as a test of the effect of principal reduction. Perhaps it should be characterized as a test of whether classification into HAMP PRA without principal reduction affects the default rate.

Note that the authors’ charts support this. In Appendix A, Figure 5(a) we see that almost 40% of the authors’ data for this test has ΔNPV = 0. On page 12 the authors indicate that they were told this was probably bad data, because it indicates that the servicer was lazy and only one NPV test was run. Thus this 40% of their data was thrown out as “bad.” Evidence that this 40% was heavily concentrated around LTV = 115 is given by Appendix A, Figure 4(d):

GanongNoel

Here we see that as the LTV drops toward 120, ΔNPV converges to zero from both sides. Presumably the explanation for why it converges to 120 and not to 115 is because almost 40% of the data was thrown out. See also Appendix A Figure 6(d), which despite the exclusion of 40% of the data shows a steep decline in principal reduction as ΔNPV converges to 0 from the HAMP PRA side.

I think this is mostly a lesson that details matter and economics is hard. It is also important, however, to set the record straight: running a regression discontinuity test on HAMP data cannot tell us about the relationship between mortgage principal reductions and default.

Advertisements

What’s the problem with financialization?

This is post 2 in my HAMP and principal reduction series. For the introductory post see here.

The series is motivated by Peter Ganong and Pascal Noel’s argument that mortgage modifications that include principal reduction have no significant effect on either default or consumption for underwater borrowers. In post 1 I explained how the framing of their paper focuses entirely on the short-run, as if the long run doesn’t matter – and even uses language that indicates that people who take their long-run financial condition into account are behaving improperly. I call this exclusive focus on the short-run the ideology of financialization. I note at the end of post 1 that this ideology appears to have influenced both Geithner’s views and the structure of HAMP.

So this raises the question: What’s the problem with the ideology of financialization?

The short answer is that it appears to be designed to trap as many people into a state of debt peonage as possible. Debt peonage, by preventing people who are trapped in debt from realizing their full potential, is harmful to economic performance more generally.

Here’s the long answer.

By focusing attention on short-term payments and how sustainable they are today, while at the same time heaping heavy debt obligations into the future, modern finance has had devastating effects at both the individual and the aggregate levels. Heavy long-term debt burdens are guaranteed to be a problem for a subset of individual borrowers, such as those who are unexpectedly disabled or who see their income decline over time for other reasons. Mortgages with payments that balloon at some date in the future (such as those studied in Ganong and Noel’s paper) are by definition a gamble on future financial circumstances. This makes them entirely appropriate products for the small subset of borrowers who have the financial resources to deal with the worst case scenario, but the financial equivalent of Russian roulette for the majority of borrowers who don’t have financial backup in the worst case scenario. (Remember the probabilities are in your favor in Russian roulette, too.)

Gary Gorton once described the subprime mortgage model as one where the borrower is forced to refinance after a few years and this gives the bank the option every few years of whether or not to foreclose on the home. Because the mortgage borrower is in the position of having sold an option, the borrower’s position is closer to that of a renter than of homeowner. Mortgages that are structured to have payment increases a few years into the loan – which is the case for virtually all of the modifications offered to borrowers during the crisis – similarly tend to put the borrower into a situation more like that of a renter than a homeowner.

The ideology of financialization thus perverts the whole concept of debt. A debt contract is not a zero-sum transaction. Debt contracts exist because they are mutually beneficial and they should be designed to give benefits to both lenders and borrowers. Loans like subprime mortgages are literally designed to set the borrower up so the borrower will be forced into a renegotiation where the borrower can be held to his or her reservation value. That is, they are designed to shift the bargaining power in contracting in favor of the lender. HAMP modifications for underwater borrowers set up a similar situation.

Ganong and Noel treat this distorted bargaining situation as if it is normal in section 6 of their paper, where they purport to characterize “efficient modification design.” The first step in their analysis is to hold the borrowers who need modifications to their reservation values (p. 27).[1] Having done this, they then describe an “efficient frontier” that minimizes costs to lenders and taxpayers. A few decades ago when I studied Pareto efficiency, the characterization of the efficient frontier required shifting the planner’s weights on all members of the economy. What the authors have in fact presented is the constrained efficient frontier where the borrowers are held to their reservation values. Standard economic analysis indicates that starting from any point on this constrained efficient frontier, direct transfers from the lenders to the borrowers up until the point that the lenders are held to their reservation value should also be considered part of the efficient frontier.

In short, Ganong and Noel’s analysis is best viewed as a description of how the financial industry views and treats underwater borrowers, not as a description of policies that are objectively “efficient.” Indeed, when they “rank modification steps by their cost-effectiveness” they come very close to reproducing the HAMP waterfall (p. 31): the only difference is that maturity extension takes place before a temporary interest rate reduction. Perhaps the authors are providing valuable insight into how the HAMP waterfall was developed.

The unbalanced bargaining situation over contract terms that is presented in this paper should be viewed as a problem for the economy as a whole. As everybody realized post-crisis the macroeconomics of debt has not been fully explored by the economics profession and the profession is still in the early stages of addressing this lacuna. Thus, it is not surprising that this paper touches only very briefly on the macroeconomics of mortgage modification.

In my view the ideology of financialization with its short term focus has contributed significantly to growth of a heavily indebted economy. This burden of debt tends to reduce the bargaining power of the debtors and to interfere with their ability to realize their full potential in the economy. Arguably this heavily indebted economy is losing the capacity to grow because it is in a permanent balance sheet recession. At the same time, the ideology underlying financialization appears to be effectively a gamble that it’s okay to shift the debt off into the future, because we will grow out of it so it will not weigh heavily on the future. The risk is that, by taking it as given that g > r over the long run, this ideology may well be creating a situation of permanent balance sheet recession where g is necessarily less than r, even given optimal monetary policy.

[1] The authors justify this because they have “shown” that principal reductions for underwater borrowers do not reduce defaults or increase consumption. Of course, they have shown no such thing because they have only evaluated 5-10% of the life of the mortgage – and even that analysis is flawed.

The Ideology of Financialization

This is post 1 in my HAMP and principal reduction series. For the introductory post see here.

The analysis in Peter Ganong and Pascal Noel’s Liquidity vs. wealth in household debt obligations: Evidence from housing policy in the Great Recession is an object lesson in the ideological underpinnings of “financialization”. So this first post in my HAMP and principal reduction series dissects the general approach taken by this paper. Note that I have no reason to believe that these authors are intentionally promoting financialization. The fact that the framing may be unintentionally ideological makes it all the more important to expose the ideology latent in the paper.

The paper studies government and private mortgage modification programs and in particular seeks to differentiate the effects of principal reductions from those of payment reductions. The paper concludes “we find that principal reduction that increases housing wealth without affecting liquidity has no significant impact on default or consumption for underwater borrowers [and that] maturity extension, which immediately reduces payments but leaves long-term obligations approximately unchanged, does significantly reduce default rates” (p. 1). The path that the authors follow to arrive at these broad conclusions is truly remarkable.

The second paragraph of this paper frames the analysis of the relative effects of modifying mortgage debt by either reducing payments or forgiving mortgage principal. This first post will discuss only the first three sentences of this paragraph and what they imply. They read:

“The normative policy debate hinges on fundamental economic questions about the relative effect of short- vs long-term debt obligations. For default, the underlying question is whether it is primarily driven by a lack of cash to make payments in the short-term or whether it is a response to the total burden of long-term debt obligations, sometimes known as ‘strategic default.’ For consumption, the underlying question is whether underwater borrowers have a high marginal propensity to consume (MPC) out of either changes in total housing wealth or changes in immediate cash-flow.”

Each of the sentences in the paragraph above is remarkable in its own way. Let’s take them one at time.

First sentence

“The normative policy debate hinges on fundamental economic questions about the relative effect of short- vs long-term debt obligations.”

This is a paper about mortgage debt – that is, long term debt – and how it is restructured. This paper is, thus, not about “the relative effect of short- vs long-term debt obligations,” it is about how choices can be made regarding how long-term debt obligations are structured. This paper has nothing whatsoever to do with short-term debt obligations, which are, by definition, paid off within a year and  do not figure in paper’s analysis at any point.

On the other hand, the authors’ analysis is short-term. It evaluates data only on the first two to three years (on average)  after a mortgage is modified. The whole discussion takes it as given that it is appropriate to evaluate a long-term loan over a horizon that covers only 5 to 10% of its life, and that we can draw firm conclusions about the efficiency of a mortgage modification by only evaluating the first few years of the mortgage’s existence. Remember the authors were willing to state that “principal reduction … has no significant impact on default or consumption for underwater borrowers” even though they have no data on 90 – 95% of the performance of the mortgages they study (that is, on the latter 30-odd years of the mortgages’ existence).

Note that the problem here is not the nature of the data in the paper. It is natural that topical studies of mortgage performance will typically only cover a portion of those mortgages’ lives. But it should be equally natural that every statement in the study acknowledges the inadequacy of the data. For example, the authors could have written: “principal reduction … has no significant impact on immediate horizon default or immediate horizon consumption for underwater borrowers.” Instead, the authors choose to discuss short-term performance as if it is all that matters.

This focus on the short-term, as if it is all that matters, is I would argue the fundamental characteristic of “financialization.” It is also the classic financial conman’s bait and switch. The key when selling a shoddy financial product is to focus on how good it is in the short-term and to fail to discuss the long-term risks. When questions arise regarding the long-term risks, these risks are minimized and are not presented accurately. This bait and switch was practiced on municipal borrowers who issued adjustable rate securities and purchased interest rate swaps, on adjustable rate mortgage borrowers who were advised that they would be able to refinance before the mortgage rate adjusted up, and even on the Trustees of Harvard University, who apparently entered into interest rate swaps without bothering to understand to long-term obligations associated with them.

The authors embrace this deceptive framework of financialization whole-heartedly throughout the paper by discussing the short-term performance of long-term loans as if it is all that matters. While it is true that there are a few nods in footnotes and deep within the paper to what is being left out, they are wholly inadequate to address the fact that the basic framing of the paper is extremely misleading.

Second sentence

“For default, the underlying question is whether it is primarily driven by a lack of cash to make payments in the short-term or whether it is a response to the total burden of long-term debt obligations, sometimes known as ‘strategic default.’”

The second sentence is based on the classic distinction between a temporary liquidity-driven stoppage of payments and a stoppage due to negative net worth – i.e. insolvency. (Note that these are the two long-standing reasons for filing bankruptcy.) But the framing in this sentence is remarkably ideological.

The claim that those defaults that are “a response to the total burden of long-term debt obligations” are “sometimes known as ‘strategic default’” is ideologically loaded language. Because the term “strategic default” has a pejorative connotation, this sentence has the effect of putting a moralistic framing on the problem of default: liquidity-constrained defaults are implicitly unavoidable and therefore non-strategic and proper, whereas all non-liquidity-constrained defaults are strategic and implicitly improper. This framing ignores the fact that a default may be due to balance sheet insolvency, which will necessarily be “a response to the total burden of long-term debt obligations” and yet cannot be classified a “strategic” default. What is commonly referred to as strategic default is the case where the debtor is neither liquidity constrained, nor insolvent, but considers only the fact that for this particular asset the payments are effectively paying rent and do not build any principal in the property.

By linguistically excising the possibility that the weight of long-term debt obligations leads to an insolvency-driven default, the authors are already demonstrating their bias against principal reduction and once again exhibiting the ideology of financialization: all that matters is the short-term, therefore balance sheet insolvency driven by the weight of long-term debt does not need to be taken into account.

In short, the implicit claim is that even if the borrower is insolvent and not only has a right to the “fresh start” offered by bankruptcy, but likely needs it to get onto his or her feet again, this would be “strategic” and improper. Overall, the moralistic framing of the paper’s approach to debt is not consistent with either the long-standing U.S. legal framework governing debt which acknowledges the propriety of defaults due to insolvency, or with social norms regarding debt where business-logic default (which is a more neutral term than strategic default) is common.

Third sentence

“For consumption, the underlying question is whether underwater borrowers have a high marginal propensity to consume (MPC) out of either changes in total housing wealth or changes in immediate cash-flow.”

The underlying assumption in this sentence is that mortgage policy had as one of its goals immediate economic stimulus, and that one of the choices for generating this economic stimulus was to use mortgage modifications to encourage troubled borrowers to increase current consumption at the expense of a future debt burden. In short, this is the classic financialization approach: get the borrower to focus only on current needs and discourage focus on the costs of long-debt. Most remarkably it appears that Tim Geithner actually did view mortgage policy as having as one of its goals immediate economic stimulus and that this basic logic was his justification for preferring payment reduction to principal reduction.[1]

Just think about this for a moment: Policy makers in the midst of a crisis were so blinded by the ideology of financializaton that they used the government mortgage modification program as a form of short-term demand stimulus at the cost of inducing troubled borrowers (i.e. the struggling middle class) to further mortgage their futures. And this paper is a full-throated defense of these decisions.

The ideology of financialization has become powerful indeed.

Financialization Post 2 will answer the question: What’s the problem with the ideology of financialization?

[1] See, e.g., the quote from Geithner’s book in Mian & Sufi, Washington Post, 2014

HAMP and principal reduction: an overview

I spent the summer of 2011 helping mortgage borrowers (i) correct bank documentation regarding their loans and (ii) extract permanent mortgage modifications from banks. One of things I did was check the bank modifications for compliance with the government’s mortgage modification program, HAMP, and with the HAMP waterfall including the HAMP Principal Reduction Alternative. At that time I put together HAMP spreadsheets, and typically when I read articles about HAMP I go back to my spreadsheets to refresh my memory of the details of HAMP.

So when I learned about a paper that finds that HAMP “placed an inefficient emphasis on reducing borrowers’ total mortgage debt” and should have focused more on reducing borrowers payments in the short-run — which goes contrary to everything I know about HAMP, I decided to read the paper.

Now I am an economist, so even though my focus is not quantitative data analysis, when I bother to put the time into reading an econometric study it’s not difficult to see problems with the research design. On the other hand, I usually avoid being too critical, on the principle that econometrics is a little outside the area of my expertise. In this case, however, I know that very few people have enough knowledge of HAMP to actually evaluate the paper — and that many of those who do are interested parties.

The paper Peter Ganong and Pascal Noel’s Liquidity vs. wealth in household debt obligations: Evidence from housing policy in the Great Recession. This paper has been published as a working paper by the Washington Center for Equitable Growth and NBER, both of which provided funding for the research. Both the Wall Street Journal and Forbes have published articles on this paper. So as one of the few people who is capable offering a robust critique of the paper, I am going to do a series of posts explaining why the main conclusion of this paper is fatally flawed and why the paper reads to me as financial industry propaganda.

Note that I am not making any claims about the authors’ motivation in writing this paper. I see some evidence in the paper to support the view that the authors were manipulated by some of the people providing them with the data and explaining it to them. Overall, I think this paper should however serve as a cautionary tale for all those who are dependent on interested parties for their data.

Here is the overview of the blogposts I will post discussing this paper:

HAMP and principal reduction post 1: The ideology of financialization

HAMP and principal reduction post 2: What’s the problem with financialization?

HAMP and principal reduction post 3: A regression discontinuity error
The principal result in the paper is invalid, because the authors did not have a good understanding of HAMP and of HAMP PRA, and therefore did not understand how the variable they use to distinguish treatment from control groups converges to their threshold precisely when principal reduction converges to zero. The structure of this variable invalidates the regression discontinuity test that the authors perform.

A brief history of the shadow banking collapse in 2007-08

A large number of “market-based” financing vehicles that developed in the years leading up the the 2007 crisis were designed to exploit the fact that some investors were only worried about AAA-ratings (or for commercial paper A1/P1-ratings) and didn’t bother to understand the products they were investing in. Several of these vehicles were literally designed to blow up — they had liquidation triggers that when breached in an adverse market could result in complete loss of the investment. Others were designed to draw down bank liquidity lines when the economic situation became more difficult. (The latter could only exist because of a reinterpretation of a 2004 final rule promulgated by the Joint Bank Regulators that had the effect of gutting the regulation. See here.) Others would expose investors in AAA rated assets to massive losses if mortgage default rates were significantly higher what was expected and/or exhibited more correlation than was expected.

None of these products was viable once investors and bank regulators had seen how they worked in practice. Thus, these products had a very short life and markets for them collapsed entirely in 2007-08. This post briefly reviews this history.

A. In 2007 the commercial paper segment of the shadow banking system collapsed.

The first commercial paper issuers to go were structured vehicles that didn’t have committed bank lines of liquidity support, but instead supported their commercial paper issues by contractual terms that could force liquidation in order to pay up on the commercial paper. Structured Investment Vehicles (SIVs) are examples. Shortly thereafter structured vehicles that did have bank lines of liquidity support, such as CDOs and MBS (only a small fraction of which were financed with commercial paper), drew down the bank liquidity lines with dramatic effects on the balance sheets of the banks involved. To protect the banks from catastrophe the Federal Reserve gave them special regulatory exemptions (see the Supervisory Letters to Citibank, Bank of America, and JP Morgan Chase dated August 20, 2007 and to other banks in subsequent months) and permitted banks to pledge at the discount window ABCP for which they provided back up lines of credit (WSJ Aug 27 2007). These exemptions together with the Term Auction Facility made it possible for the ABCP market to deflate slowly over the course of three years, rather than collapsing quickly and taking a few banks with it.

In short in 2007 the Federal Reserve let a variety of different shadow bank models collapse, while protecting the banks and stabilizing the money supply by keeping the ABCP market from collapsing too quickly. These decisions were classic lender of last resort decisions that had the effect of allowing some entities to fail and other to survive with central bank support. They are also fairly uncontroversial: just about everybody agrees that the Fed acted appropriately at this point in the crisis.

B. From 2007 to 2008 a huge number of structured finance vehicles went Boom

At the same time some of the more esoteric structured vehicles that issued longer term obligations but also relied on liquidation triggers to support their issues blew up. Examples of this category include Leveraged Super Senior CDOs and Constant Proportion Debt Obligations. In a leveraged super senior CDO investors pay, for example, $60 million to earn 1.5% per annum spread over safe assets by selling an insurance policy (that is, CDS protection) on the $750 million most senior tranche of a CDO. Because the investors are putting up so little money the LSS CDO has a liquidation trigger, so that if the insured tranche falls by, for example, 4% in value, the structured liquidates, and an alternate insurance policy is purchased on the market. The investors then get whatever is left after the insured party is protected. These structures all blew up in 2007.

The Constant Proportion Debt Obligation was an even crazier product. Instead of insuring only the senior most tranche of a CDO, it sold insurance on a high grade bond index, including 125 names. Because there were no subordinated tranches to protect it from losses, the insurance premium was higher. The CPDO was structured to take the excess insurance premium (i.e. that which was not paid out as a bond yield to the marks who “invested” in this AAA-rated product) and put it aside. If everything goes well in three years the CPDO can stop insuring debt and pay the promised yield by just investing in safe assets. Of course, if everything goes badly, liquidation triggers are hit and the investor loses. Guess what happened in 2007?

CDO squared and ABS CDO’s are similarly products that pay an investor a bond-like yield to take an equity-like risk. They, however, had tranches that were rated up to AAA by the rating agencies, and in some cases even the brokers selling the products appeared to believe that they were just another kind of bond. Cordell, Huang & Williams (2012) found that the AAA-rated ABS CDO bonds lost more than half their value. More specifically they found that median junior AAA-rated ABS CDO bond lost 100% of its value, and that senior AAA-rated ABS CDO bonds did better, but also lost more than half of their value. And virtually every bond rated below AAA lost all of its value (Table 12). Now that investors understand this product, they won’t touch it with a ten foot pole.

C. Private Label Mortgage Backed Securitization evaporates

Underlying the losses on ABS CDOs were losses on private label mortgage backed securities. 75% of ABS CDO issuance was in the years 2005-2007 and over these years 68-78% of the collateral in ABS CDOs was private label mortgage collateral
(again from the great paper by Cordell, Huang and Williams Figure 2).

Cordell, Huang and Williams also finds that the lower tranches of subprime MBS were apparently never sold in any significant numbers to investors. Instead they were placed into CDOs (p. 9). This inability to place the lower rated tranches as well as other structural problems with the treatment of investors may explain the complete collapse of the private label MBS market, which is documented in Goodman 2015.

 

In short, SIVs, CDOs, and private label MBS were all effectively shadow banks that provided financing to the real economy during their lifetimes, but were not structured in a way that made them viable long term products. Thus, they disappeared as soon as they were exposed to an adverse environment. When this happened, the funding they had provided to the real economy disappeared (see Mian and Sufi 2018). This showed up as funding stress on the market.

Despite the stress the failures of these vehicles put on markets, the consensus seems to be universal that the Federal Reserve’s job was to protect the regulated banks, not to worry about the disappearance of the “market-based” lending structures. On the other hand, the liquidation and deterioration of these products sent waves through financial markets from August 2007 on that the Federal Reserve and the other central banks had to navigate.

The 2007-09 crisis: not a panic, but the collapse of shadow banking models

This post is a response to Ben Bernanke’s retrospective on the crisis and also addresses some of the comments others have made on his retrospective.

But first let me start the post with a little etymology. The term “lender of last resort” is generally acknowledged to have originated with Francis Baring’s 1797 tract, Observations on the Establishment of the Bank of England. Baring, however, did not use the English phrase; instead he called the Bank of England, the dernier resort. The use of the French is telling, because it is a well-established French phrase referring to the “court of last appeal.” Thus, the etymology of the phrase “lender of last resort” indicates that this is the entity that makes the ultimate decision about rescuing a firm or affirming the market’s death sentence. In short, when we talk about the central bank as lender of last resort, we are talking about the final arbiter of which troubled firms have a right to continue to exist in the economy.

Why would a lender of last resort allow some firms to fail? Because in an economy where bank lending decisions can expand or contract the money supply, banks that engage in fraud or make dangerously stupid lending decisions affect financial stability. So a lender of last resort has to police the line between good bank lending and bad bank lending. This almost always means that some lenders need to be closed down — preferably before they destabilize the financial system.

Framing the lender of last resort as having a duty to determine which of the entities that are at risk of failing for lack of funding will survive and which will not, sheds light on the debate between Ben Bernanke, Paul Krugman, Dean Baker, and Brad Delong. The key to this is to reframe Ben Bernanke’s “panic” which he describes as lasting from August 2007 to Spring 2009 as the process by which the Federal Reserve allowed certain shadow banks — which had no reasonable expectation of Federal Reserve support — to collapse completely.

I. “Financial fragility” was initially driven by the collapse of non-viable funding models

  1. Collapse of shadow banking vehicles (for details see this post)

In 2007 most of the commercial paper segment of the shadow banking system collapsed. These shadow banks included SIVs, and a few CDOs. (The commercial paper collapse was extended over three years, apparently due to Fed approved bank support of the market.) Several categories of commercial paper issuer entirely disappeared.

From 2007 to 2008 several other categories of shadow bank collapsed and disappeared. Some can be classified as existing only due to the excesses of the boom: e.g. Leveraged Super Senior CDO, Constant Proportion Debt Obligations, CDO squared, and ABS CDO. (The  latter two products are best described as combining the return of a bond with the risk of an equity share. Once investors figured this out, they ran for the hills.) Others, such as Private label mortgage backed securities, are less obviously flawed products, and yet 10 years after the crisis are hardly to be found.

All of these shadow banks were “market-based” products with no claim whatsoever to Federal Reserve support, so it was unremarkable that the Fed allowed them to collapse. On the other hand, they had been used to provide funding to the real economy. So their collapse was necessarily accompanied by a decline in real economy lending.

To describe this phenomenon of the collapse of non-viable shadow bank lending models as a “panic” is inaccurate. Lax financial regulation allowed non-viable entities to play a significant role in funding real activity pre-crisis. These entities failed when reality caught up to them. They did not fail because of a panic, they failed because they were non-viable. Because of the significant degree to which banks were exposed to these non-viable shadow banks, short-term funding costs rose more generally, but to a large degree rationally.

2.  End of the 2000’s investment banking model

Over the final decades of the 20th century U.S. investment banks transformed themselves from partnerships into corporations. As corporations they grew to rely much more significantly on borrowed funds than they had when partners’ capital was at risk. By 2007 the investment bank funding model in the U.S. relied extremely heavily on repurchase agreements, derivatives collateral, and commercial paper. Arguably, this was another non-viable shadow bank model.

Bear Stearns failed in March 2008 because of runs on these instruments. Lehman Brothers failed in September for similar reasons. Merrill Lynch was purchased by Bank of America in an 11th hour transaction. Morgan Stanley and Goldman Sachs were at the edge of failure, but saved by the Federal Reserve’s extremely fast decision to permit them to become bank holding companies with full access to the Federal Reserve’s lender of last resort facilities.

While some may believe that financial stability would have been better served by the Federal Reserve’s support of the 2000’s investment banking model, at this point the question is an unanswerable hypothetical. Because the Federal Reserve exercised its lender of last resort authority to refuse to support the 2000’s investment banking model, this shadow banking model no longer exists.

Thus, in September 2008, just as was the case in earlier months of short-term funding pressures,  a major cause of these pressures was real (though in this case elements of “panic” were also important): i.e. the collapse of a shadow banking model that was non-viable without central bank support. Once again, the fact that such a collapse had real effects is not at all surprising.

II. “Financial fragility” did culminate in a well-managed, short-lived panic

Unsurprisingly the collapse of the 2000’s investment banking model was such a significant event that it was in fact accompanied by panic. It is in the nature of a financial panic that it is best understood as the market’s expression of uncertainty as to where the central bank will draw the line between entities that are to be saved and those that are allowed to fail. Effectively, funding dries up for all entities that might hypothetically be allowed to fail. As the central bank makes clear where the lines will be drawn, the panic recedes. This view is supported by the programs that Bernanke lists as having had a distinctly beneficial effect on crisis indicators (p. 65): the Capital Purchase Program, the FDIC’s loan guarantee program, and the announcement of stress test results were all designed to make it clear that depository institutions would be supported through the crisis. Similarly, the support of money market funds gave confidence that no more money market funds would be allowed to “break the buck.”

As Bernanke observes “the [post-Lehman] panic was brought under control relatively quickly” (p. 65). Within six weeks funding pressures had already begun to ease up and by the end of 2008 they had almost entirely receded. In short, once it was clear which entities would be saved by the lender of last resort, there was no longer any cause for panic.

Brad DeLong in a review of Gennaioli and Schleifer’s new book argues that there might not have been a panic associated with Lehman’s failure if some form of resolution authority had been in place. With this I agree. As I argued here: it is almost certainly the case that if Treasury had reacted to the March 2008 Bear Stearns failure by carefully drawing up a Resolution Authority instead of the 3-page original TARP document, 2008 would have looked very different indeed. I also agree with DeLong’s positive evaluation of Gennaioli and Schliefer’s theory of investor psychology. In my view, however, their theory is more properly framed as putting modern bells and whistles on financial market dynamics that have been well-understood for centuries. The whole point of having a central bank and a lender of last resort is, after all, to control the dynamics generated by investor psychology (see e.g. Thornton 1802).

III. Additional bubbles explain Bernanke’s “non-mortgage” credit series

Ben Bernanke focuses on the housing bubble, but there were actually three bubbles created by the shadow banking boom of the early naughties: the housing bubble, the commercial real estate (CRE) bubble, and the syndicated loan “bubble”. SIFMA’s Global CDO data shows how MBS was only one form of shadow banking collateral. Lending to corporations was almost equally important.
Global CDO collateral
As Dean Baker points out the CRE bubble peaked in September 2007. Commercial real estate prices dropped by over 30% over the course of the next 18 months.

The syndicated loan “bubble” has behaved differently. While the market was subject pre-crisis to a deterioration in loan terms that was comparable to the mortgage or CRE market, this “bubble” never popped. The length of the loans was such that not many matured in 2008, and many corporations had committed credit lines from banks that they were able to draw down. Thus, it was in 2009 that concerns about likely corporate defaults weighed heavily on the market (see here and here). These concerns were, however, never realized. The combination of ultra low interest rates, retail investors shifting their focus to bond funds and ETFs, and pension funds reaching for yield meant that corporations were typically able to refinance their way out of the loans, and no aggregate collapse was ever realized. (To see how short lived corporate deleveraging was, see here.)

Thus, the fact that 2009 was a year in which massive corporate bankruptcies were expected just over the horizon probably explains a lot of the stress exhibited by Bernanke’s non-mortgage credit series (which is composed of non-financial corporate credit indicators and consumer-oriented securitization indicators). Treating this series as representing “a run on securitized credit, especially non-mortgage credit” (p. 46) as if it can only be explained by “panic,” does not seem to address the deterioration of corporate fundamentals and the implications of those fundamentals for corporate employees.

Bernanke considers the possibility that borrower financial health drives this indicator, but rejects this explanation, because:  “First, aggregate balance sheets evolve relatively
slowly, which seems inconsistent with the sharp deterioration in the non-mortgage credit factor after Lehman, and (given the slow pace of deleveraging and financial recovery) looks especially inconsistent with the sharp improvement in this factor that began just a few months later” (p. 41). Bernanke appears to assume that the deterioration would have been driven by the housing bubble, but that is not my (or Dean Baker’s) claim. I am arguing that the collapse of the CRE bubble and the weight of needing to refinance maturing syndicated loans in an adverse environment caused corporate balance sheet deterioration. The improvement is then explained by the fact that CRE prices bottomed in mid-2009 and in early 2009 the Federal Reserve made clear its commitment to keep interest rates ultra-low for “an extended period” of time. Both of these helped corporates deal with their debt burden.

In short, I find that Ben Bernanke’s data is entirely consistent with the presence of only a short-lived panic in late 2008. The economic deterioration that Bernanke associates with the prolonged short-term funding crisis and the more short-lived non-mortgage credit crunch can be explained respectively by the collapse of a large number of shadow banking vehicles and by the deflation of the other two lending booms associated with the crisis.

IV. Why Ben Bernanke’s characterization of the crisis is problematic

Thus, my most serious objection to Ben Bernanke’s characterization of the crisis is that, having exercised the lender of last resort authority appropriately to its full potential by permitting shadow bank funding models that were deemed destabilizing to collapse, he seems to want to avoid acknowledging the actual nature of the central bank’s lender of last resort role. His description of the 2007-09 crisis as “a classic financial panic” implies that the crisis was fundamentally a coordination problem in which the public was choosing a bad equilibrium and just needed to be redirected by the central bank into a good equilibrium in order to improve economic performance. Brad DeLong also embraces the language of panic in his response to Bernanke on the AEA Discussion Forum: “all that needed to be done was to keep demand for safe assets from exploding.”

(For Paul Krugman the problem is to explain not just the depth of the recession that ended in mid-2009, but the extraordinarily slow recovery from that recession, which he dubs “the Great Shortfall”. I suspect much of the explanation for the Great Shortfall will be found in post-crisis policies that were designed to protect Wall Street balance sheets at the expense of pension funds and the public, but that is a very different post.)

If one reframes Bernanke’s data from August 2007 to Spring 2009 as representing the complete collapse of certain shadow bank funding models, we see the Federal Reserve as the ultimate decision maker over which funding models were allowed to survive. Because some shadow bank funding models were allowed — properly — to collapse short-term funding rates skyrocketed and areas of the real economy that had adapted to rely upon the doomed funding models struggled as they had to adjust to a world with a different set of choices. This adjustment was temporary because the Federal Reserve — properly — acted to promote restabilization of a financial system without the terminated shadow banks.

What drove the data was not the public choosing a bad equilibrium, (that is, a panic), but the Federal Reserve properly exerting its authority by allowing market forces to eliminate certain shadow banks. This authority is properly exercised because in a world with credit-based money such as ours, financial stability is only possible if the lines between bank-like lending that is acceptable and bank-like lending that is not acceptable are strictly drawn and carefully policed. Thus, the Federal Reserve’s most significant error was its failure to exercise this authority stringently enough long before the crisis broke in order to act preventively to forestall the financial instability that was experienced in 2007-09.

That said, Bernanke is rightly proud of the speed with which the post-Lehman panic was brought under control and is right to conclude that “the suite of policies that controlled the panic likely prevented a much deeper recession than (the already very severe) downturn that we suffered” (p. 66). He also draws a lesson from the crisis that is entirely consistent with the view of it presented here: “continued vigilance in ensuring financial stability” is absolutely necessary. Indeed, I suspect that he would agree with me that that the Federal Reserve should have exercised greater vigilance prior to 2007.