The Global Economy Remains Insufficiently Vibrant

There is a certain beauty to securitization that will never be appreciated again for what happened.  It is an expression of genius, masterful innovation equally as powerful for shaping our world as the jet engine or the transistor.  In many ways, this bit of financial majesty has done more than either of those, and that is precisely the problem.

The idea behind the process is simple enough, as these things usually are. There are various motivations for any financial institution to hold any asset.  Securitization brought together several factors which allowed for these demands to be met simultaneously. Also called structured finance, both terms give you a sense of what is being done.  Applying accounting principles coupled with market appetites, you can transform the boring and risky into the liquid and tailored.

The desire to do these things was motivated by risk, always risk.  Lending in a mortgage is often thought to be relatively less risky by view of the collateral that comes with it.  But long banking experience has conclusively shown this just isn’t so; collateral or not, mortgages are inherently a problem.  This is especially true in the context of fractional lending.

What everyone in banking wanted was a liquid mortgage market.  To have one would (might) have obviated some of the worst parts of the Great Depression. A bank being set upon by anxious depositors demanding their funds could not immediately liquidate any mortgage assets to meet the emotion.  And when some paper eligible for being recalled was recalled, it only made the situation worse as the borrower whose loan was called in was now added to the queue seeking funds (often from the bank across the street, spreading the run). 

The primary impediment to a market for mortgages was paper.  Holding one of these debts necessarily requires certifications and documentation that made the idea of selling individual loans cumbersome and costly. 

The great innovation or first step was the pass-through.  The Government National Mortgage Association (GNMA, also known as Ginnie Mae) guaranteed the first one in the early 1970’s.  It would handle the gritty back office work and send along, or pass through, principle and interest payments on pools of loans to investors in those pools.  The mortgage was no longer a singular loan stuck forever in stasis on a single bank’s balance sheet.

Ginnie Mae was followed by Fannie Mae, a quasi-private corporation, that in 1983 created the CMO, or collateralized mortgage obligation.  This innovative step parceled the cash flows intended for investors in a specific pool, redirecting them as required in the security documents.  By this time, these things were called mortgage bonds, or MBS.  They in many instances acted like bonds, and were in the early 1980’s a vastly growing marketplace.

It was the CMO that introduced tranching, a French word for slicing.  One primary risk of any loan is prepayment. You may be quite happy receiving whatever interest on whatever loan, but the borrower may not be.  They might intend especially if the interest rate environment changes to pay off the loan and seek lower costs, leaving you exposed to the same thing on the other side in the form of lower returns.  The CMO tranched mortgage pools such that interest payments (and certain principle risks) would be segregated and paid according to predetermined prepayment risks. 

Asset-backed securities were introduced in 1985 when a specialty finance firm packaged leases on computer equipment into a pool and sold the lease rights to a Special Purpose Vehicle it had separately set up.  Variable interests in the SPV were then further sold to investors. 

All that was left was for the two to merge, when a CMO-like mortgage pool would become an ABS in the form of a fully securitized MBS.  The housing bubble era MBS tranches were sliced on the basis of credit risk as well as prepayment factors.  An SPV pool would start out overcollateralized as the first measure of protection; that is, for, say, $1 billion pool there would be in it originally $1.02 billion so that the SPV investors, the true MBS holders, would be buffered by an additional two percent in par value loans. 

The securitization would then use a waterfall structure so that any defaults (and other cash flow impairments like late payments) would be assigned in turn; first to the overcollateralized part, next the equity piece, then the various mezzanines, and finally the senior.  Obviously, if you want to own the equity tranche you do so because it pays the highest interest rate.  Benefits at the mezzanine level(s) are the thickness of protection in the tranches above that absorb losses (or prepayments) before you, and a slightly less rate of return for it. 

This structure really made its money in how it could then be used to transform the otherwise truly risky into what seemed to be riskless.  You could take a pool of subprime mortgages that before were hugely risky as well as illiquid (one followed the other) and by virtue of combining tranching and ABS principles you would create definable as well as marketable securities according to customer specifications. 

After the equity and mezzanines were carved out, that left the vast majority of the pool, the senior tranche, to be marketed very differently.  If investors in the top levels were interested in returns for higher risks, however comfortable they might be with the mathematical calculations of them (particularly correlation as determined by the Gaussian copula), in these huge lower reaches the desire was safety.  How do you turn billions upon billions of subprime mortgages into AAA-rated MBS?  That was the senior parts of these structures.

But marketing senior tranches proved somewhat difficult.  Even the most stoic insurance company or public pension fund desires some return consideration. Not only that, the senior pieces were massive in size, the vast majority of any structure.  These were often too large to market to a single investor.  So global banks hit upon another innovation, which turned securitization, the beauty of modern financial innovation, into the Frankenstein’s monster of the post-crisis imagination.

They created a sort of swapout, where the lowest risk part would be further sliced or tranched, leaving behind the still proportionally large rump, or super senior portion.  To boost the returns on this remainder, they were promoted and sold on a leveraged basis.

If our original structure amounted to a $1 billion subprime mortgage pool, initiated with $1.02 billion face value in loans, and further carved up into equity and mezzanines totaling 15% of the $1 billion available, that would leave $850 million in senior.  Take off another 10% for other senior tranching, and you have leftover $750 million in super senior subprime guaranteed to be AAA-rated by the 25% coverage for default risk absorbed by the tranches above. 

But to really be able to move that $750 million slice, again a large single piece, banks would often offer up a Leveraged Super Senior option.  You could put down a 10% equity stake in it, just $75 million, and then fund the rest by issuing commercial paper.  That way you would own the whole “riskless” thing even at the lowest returns of the structure, but boost them by borrowing leverage at some of the cheapest available funding costs (shortest terms).  Spreads were your friend. 

The idea was fantastic in terms of credit risk, not so much everywhere else.  In what should have been a tipoff as to what these other issues were, the commercial paper option often came with, and were in many instances demanded as a precondition, a liquidity backstop guarantee by the sponsoring institution. If for some infinitesimally small possibility the commercial paper market doesn’t rollover your obligations, said bank will fund the shortfall for as long as it takes.

In the whole history of the panic period, I believe that no super senior piece ever took a cash loss.  Thicknesses were more than enough to absorb even the worst vintages of subprime mortgage pools, the most toxic of toxic waste.  The senior pieces may have taken some, but of the estimates I have seen these were incredibly small never amounting to more than a triviality.  The issue was never credit risk.

It would all come out in the worst possible way on August 9, 2007, exactly ten years ago this week.  The global economy has lost an entire decade not because of the major elements of the housing bubble as they are so often described, but because of the (monetary) complexities with how it all got funded.  The issue was always liquidity risk. It is still to this day.

At 2:44am on that August 9 morning, Reuters reported that BNP Paribas had two days earlier suspended the net asset valuation (NAV) calculation of three of its funds.  These three, however, were not like the Bear Stearns hedge funds that had earlier in the year sounded alarms.  BNP’s funds were money market funds.

It’s not as if the giant French bank was hiding anything, either.  The name of each indicated largely what they were doing: BNP Paribas ABS Euribor, BNP Paribas ABS Eonia, and Parvest ABS Dynamic. All of them specifically stated in their titles that they were in the “ABS” arena. 

Prior to August 2007, however, investors and the fund sponsor thought differently about what that might mean.  What changed was, again, liquidity even for the riskless of classes.  These money market funds did not own the equity part of the Leveraged Super Senior tranche, they were instead the providers of commercial paper funding (a great many MMF’s had bought the commercial paper sold by the SPV’s for the remainder of funding).

Being so remote as well as tied to the most protected structures, it should have been of no concern to anyone especially money market funds.  Pricing of these tranches, however, had grown quite difficult in the summer of 2007.  Several that used to trade and set the benchmark price for the class no longer did, or did so at extremely shaky levels (the Gaussian copula and how it inferred, rather than estimated, correlation and the tendency of ABS structures especially MBS to exhibit a correlation “smile”).  The risk for commercial paper holders was that irregular pricing might force the MBS holder to sell at a price that wouldn’t allow it to pay off all the commercial paper outstanding.

The whole point of all this securitization stuff was to get beyond the bank panics of the past and the devastating economic depressions that came with them. The very idea of the Federal Reserve, channeled through Senator Robert Owen of Oklahoma, more the father of the Fed than any other single man, was to protect “good banks” from the rubes who feared the activities of “bad banks” and wrongly associated the latter with the former. 

The problem of any fractional lending system is asymmetry of information.  You put your cash in a bank and you have very little if no idea what the bank does from there. The government role in those days was as sight regulator – verifying the bank’s books and that what it says is in its vault is in fact in the vault.  So long as these “reserves” met statutory thresholds, there was little issue. 

On the asset side, regulatory regimes called for subjective judgment, even using capital ratios as one element of it.  The assets of the bank would be reviewed, but as we have seen clearly over the past decade what does that really mean?  In other words, bank regulators then performed qualitatively what ratings agencies have done quantitatively in the later 20th and early 21st centuries. 

If your particular bank got into trouble, you wouldn’t ever know it except by whispers and rumors.  Thus, if people started to demand funds back from that bank, it wouldn’t be long before almost everyone did; a sort of self-fulfilling prophecy, as the more the bank appears shaky, whether it is or not, the more it loses reserves of cash, meaning that it actually does become shaky.

In very simple terms, that is exactly what happened in systemic fashion starting on August 9, 2007; only it happened internally in interbank markets rather than externally as a matter of public emotion.  The 2008 panic was the first in history of strictly banks (loosely speaking, non-bank financial firms included here) panicking about other banks.  The interbank markets, the money of money, were hit with several runs that the depository side stayed quite clear of (outside of individual, idiosyncratic examples like Northern Rock and IndyMac).

And it was because of still information asymmetry.  Nobody really knew and appreciated what these guys were doing.  Because they made “toxic waste” sound riskless, that assault on conventional sense only served to foster and amplify further mistrust.

The Federal Reserve responded to this as if it was facing down an angry Senator Owen.  It used tools that were largely designed for a public spate of emotion rather than an interbank one.  On August 9, it was the commercial paper market as the conduit for spreading doubt (especially how liquidity backstops created channels for “contagion”).  ABS CP over the next three weeks would plummet by almost $200 billion, a terrifying collapse at a time when billion still meant something.

The first official response, however, was not the Federal Reserve’s.  It would be the ECB who initiated the emergency policies that ten years later are relatedly still ongoing.  On August 9, Europe’s central bank conducted its largest open market operation to that point, a then-record €94.8 billion easily surpassing the past record of €69 billion injected on September 12, 2001.

From that day forward, nothing any central bank did worked.  All of the tricks they had developed over the years failed because they were expecting the system to operate as it did a century before, or at least like in 1929.  They inhabited a different world, a fact that various officials would often drive home in how they evaluated their own efforts. 

In delivering the keynote speech to one of the European Parliament’s Special Committees, for example, ECB Executive Board Member José Manuel González-Páramo on November 10, 2009, actually said:

“It is generally recognised that from the start of the tensions, the Eurosystem’s response to the crisis has been effective. Indeed, when widespread tensions severely impaired the functioning of the money market in early August 2007, the Eurosystem rapidly reacted by injecting €95 billion into the banking sector in an overnight operation, followed by various additional liquidity injecting operations. When the crisis intensified again in September 2008, the Eurosystem swiftly reacted by announcing further exceptional measures to enhance the provision of liquidity to banks.”

Pardon my tone, but what the hell was this guy talking about?  It was true that the ECB successfully carried out its operations.  No one can say that the central bank failed in its technical tasks, as that €94.8 billion in operations on August 9, 2007, was indeed accepted by the market with no failures to report.  The same can also be said for everything it had and has done thereafter.  But to you, me, and everyone else on this planet none of it actually mattered.

The very title of the committee investigating Mr. González-Páramo’s actions was Special Committee on the Financial, Economic, and Social Crisis.  A central bank is supposed to first and foremost see to currency elasticity, to at least adhere to Walter Bagehot’s primary mission of lending freely at high rates.  According to González-Páramo’s, however, the ECB fulfilled that mission by lending freely often at accommodative rates. 

It did not forestall that crisis, so what actually did the ECB do?  The panic happened anyway, so how can anyone claim monetary effectiveness?

The biggest part of the whole issue is captured in realizing why the ECB had a role in this at all.  This was, after all, supposedly a dollar crisis, a panic for Wall Street.  It should not be lost here that the ECB was first injecting liquidity, and in euros.

This factor is one policymakers have steadfastly avoided for decades. If you thought the commercial paper liquidity angle for Leveraged Super Senior positions was complicated, that was just the tip of the iceberg (I don’t even need to bring in derivatives here to illustrate my point).  What happens when the commercial paper funding a US SPV pooling US$ mortgages, prime or subprime, is provided by a money market fund domiciled in Liechtenstein and sponsored by the biggest bank in France?  And then, come to find out, it wasn’t Citigroup who provided the dollar liquidity backstop to that SPV as the commercial paper market failed, but instead Deutsche Bank or more likely one of the big 3 Swiss? 

We don’t have to speculate because, apologies to González-Páramo, it happened everywhere.  Crisis spread all over the world in exactly the same manner as it had always done in bank panics all throughout the ages. What was different was where it spread, meaning the forms of money that were used in something like fractional interbank lending, a monetary system that had developed without any reserve assets. 

Economists at the Federal Reserve Bank of New York finally in August 2013 got around to addressing the modern reality of this way of monetary banking, writing “…while much of the ABCP exposure was US dollar (USD) denominated, a substantial portion of this ABCP exposure was concentrated amongst foreign banks. Many of these foreign banks with large exposure to US ABCP did not have large US regulated banking operations.” In other words, when foreign banks operated with a large dollar short they were susceptible to some truly bad consequences for being short in less than benign periods.  It took only six years after it all started for them to finally see and state what should have been obvious on August 9, 2007, if not long, long before then.

The FRBNY paper found that the only funding recourse for these dollar-exposed foreign banks was repo, and offshore repo.  That, of course, only further imperiled their operations, which is why the ECB stepped in in euros before the Fed ever did anything in dollars (not that what they did was any more effective; it was ALL failure from the get-go). 

The world has been laboring under several great misconceptions.  The first is that the dollar is the world’s reserve currency; it is not, and has not been for a very long time.  The eurodollar long ago subsumed that role, and then expanded it into all sorts of interbank things that up until 2007 nobody heard of or cared to figure out.  It all worked because it all seemed to work.

The second fallacy is that the money multiplier of Robert Owen’s time, the one the Federal Reserve is set up to meet, still matters; it does not.  I wrote in 2013:

“There were two major evolutions in money and banking that seem to fall outside the orthodox narrative. The first was a shift of reserves and bank limitations from the liability side to the asset side. The second was the rise of interbank markets, ledger money, as a source of funding rather than required reserve balancing; replacing the old deposit/loan multiplier model.”

Reducing the federal funds rate, as the Fed finally did in September 2007, does nothing for the ABS CP market which by then had already bled a further $100 billion, being reduced a total of $300 billion from the top the week of August 9.  What matters in a reserve-less environment, as the eurodollar was and remains, is the capacity to accept and create more and often other liabilities tolerable to everyone.  Capacity was being reduced in more than just the commercial paper outlets, and capacity is modern money.

Because of that one fact, the panic period applied to not just the US or where the US dollar is the main currency.  Because that fact is still not accepted today, the global economy remains in the same altered, dangerously insufficient state. It has alternated in cycles between at best little growth and at worst more contraction.  We are currently with the former, only waiting for the next shift to the latter, as in 2014.

You can call it whatever you want, but ever since August 9, 2007, global economic capacity has been greatly and quite obviously reduced. In many places, including the United States, output or labor utilization ten years later is only marginally higher than when BNP Paribas’s ABS commercial paper activities suddenly mattered. 

The cost of depression is time.  A lost decade is no trivial thing. The eurodollar never solved information asymmetry despite the great promise of securitization. Measuring costs by time, it may have, in fact, made it worse.