The S&L Crisis Foretold the Panic of 2008, Part I
In March 1987, Marvin L. Warner was found guilty of six counts of "unauthorized acts" in connection with the 1985 failure of Home State Savings Bank. It was a massive fall from grace of Mr. Warner, who had been the US ambassador to Switzerland during the Carter administration. Home State's president, David Schiebel, was found not guilty on 84 counts of "misapplication of funds" but convicted on three counts of securities violations. One former president, Burton Bongard, was found guilty of 41 counts of "willful misapplication of funds" and 41 counts of "unauthorized acts."
That was certainly a lot of criminality for one bank, but Home State was no run of the mill thrift. Its failure in 1985 began a localized run in Ohio that led to the first major "bank holiday" since the Great Depression. The governor of Ohio felt it necessary to close the state's 71 other S&L's over fears of systemic collapse of the Ohio Deposit Guarantee Fund, the state entity empowered to act as insurer on thrift deposits. To reopen Ohio's thrifts, many banks were turned to the Federal Savings and Loan Insurance Corp while Paul Volcker, then Federal Reserve Chairman, issued a statement that the central bank would "supply funds" if there were any further large withdrawals.
This was a harbinger of what was to come later on in the decade, a mix of fraud and high flying finance. Home State's own failure, and the source of its massive losses, was related to repo transactions with a shady outfit in Florida, ESM Government Securities. There were rumors at the time that ESM was owned jointly by Mr. Warner, but I haven't found anything conclusive out of what really became a tangled mess of affairs. What really matters, however, is that the once-solid S&L business had become the central axis of the development of the repo market.
That was an odd outcome, as thrifts (irony already) were designed to take in deposits and make plain vanilla mortgages - they were the epitome of boring. That all changed in the stop-go monetarism of the 1970's Great Inflation, as Regulation Q prevented them from offering competitive deposit rates. As money market funds began to offer rates above Q's ceiling, S&L's were trapped in what seemed a downward spiral. It was bad on the asset side, too, as their portfolios of bland mortgage loans were causing innumerable upset in the banking balance since there was no flexibility. When interest rates were at their apex in the first years of the 1980's, the NBER estimates that S&L's in the aggregate were carrying a negative net worth of $17.5 billion by the end of 1980, and then minus $44.1 billion only six months later (these numbers only sound small in today's debased perspective where trillions are the norm).
Government, as it does, was involved in March 1980 with its attempt at addressing these impositions. Congress passed the Deregulation and Monetary Control Act, which phased out the ceiling in Regulation QE, but over a six-year period. Deposit insurance was elevated from $40,000 to $100,000 and federally chartered S&L's were permitted up to 20% of their assets into nonresidential real estate loans and other forms of debt securities.
Faced with so many arbitrary distinctions on activities, S&L's did what financial institutions always do. Really there are only two choices in these circumstances, remove all risk and be even more vanilla since you can't really get paid for it, or go completely insane and take on as much risk as possible to overcome statutory impediments. Many thrifts chose to be thrifts; many looked in that other direction.
Part of that latter process is certainly to cheapen as much as possible the liability side, which is what opened the door to the undeveloped repo market. According to the FDIC, by 1984, "thrifts with annual growth rates of less than 15 percent had more than 80 percent of their liabilities in traditional retail deposits (generally in accounts of less than $100,000), the comparable figure for thrifts growing at rates in excess of 50 percent per year was only 59 percent." As the FDIC makes plain, that latter category taking on all these risks were funding them via large-denomination deposits (a notoriously unstable category of funding) and repos.
As late as 1972, S&L's were absent from the nascent repo market, but by 1978 they had added about $4.8 billion in repo liabilities, totaling up to about 1% of the entire liability structure. By 1984, around the time Home State was collateralizing lending with securities that didn't exist, repo liabilities had expanded almost tenfold to $43 billion, or slightly less than 4% of a rapidly expanding balance sheet base. Total repo reliance would peak in Q1 1989, with almost $105 billion in liabilities, or just shy of 7% of total funding.
Part of the risks involved in that early repurchase market was that it wasn't like what we know of today. Clearing mechanisms and securities management were downright primitive, yet these formerly dreary institutions once dedicated to mind-numbing simplicity not only embraced it they rode it to massive financial growth that they really didn't understand (and more than a few took advantage of). Yes, the S&L crisis was a direct forerunner of securitization and wholesale finance that combined to create the Great Recession and Panic of 2008.
Why didn't anyone learn from the prior episode?
The question is itself misleading, as there were clear interpretations made from the S&L crisis, not the least of which related to the ultimate losses and dispositions from Resolution Trust Corp which "cleaned" up the mess left behind. A better question to ask might be, "why weren't the right lessons learned" in the aftermath. That may seem like a bold judgment but the fact that it happened twice in almost the same exact manner more than suggests as much.
We can see this very acutely in the attempts today to do the same thing, as even though the panic is now almost seven years in the past there is still righteous preoccupation with it. In contrast to the S&L's, there is an order of magnitude greater level of complexity that has made progress extremely difficult, as so much of what went wrong occurred far beyond our abilities for direct observation. Part of that is related to the fact that the entire treatment of "money" and "money supply", which relates the specific job of the central bank (as it is promised, anyway) to the banking system, is and remains in error.
In the case of Bear Stearns, for example, the Federal Reserve might have been far better served in the months leading to its failure to guarantee all of monoline CDS writings than to engage in its traditional "liquidity" measures - and even that probably wouldn't have been enough to forestall the near-bankruptcy. The Fed might have had to actually take over the CDS business entirely and begin writing contracts itself, speaking to both the derivative nature of the crisis and what exactly constituted the "money supply." What was short in February and early March 2008 was not "dollars" or "reserves" but the willingness to freely engage them at non-onerous prices - banks were missing a critical piece of hedging "liquidity" that could never be replaced by reducing the federal funds target rate.
Some parts of the official apparatus have caught on to this evolution, as one of the primary failures in the S&L aftermath was to end the traditional depository nature of banking; the wholesale system took over from there. While policymakers were criminally uninterested in the global "dollar" prior to 2007, and many remain so, some scattered few have seen the wreckage for what it was and have been very busy, in impressive fashion, at cataloguing and understanding it.
Zoltan Pozsar, formerly of the Treasury Office of Financial Research, now of Credit Suisse, performed massive reconstructions of the chains of liabilities in the wholesale system, up to and including the dealer activities that I think make it all run in a relatively stable fashion. Another has been Perry Mehrling at the BIS, who recently was in Zurich to present on the modern incarnation of central banking in the wholesale model. His presentation was linked to a paper he wrote for the BIS titled, Why Central Banking Should Be Re-imagined.
Mehrling makes a very good case for envisioning the wholesale system as it really might be, and derives a few solid conclusions from that. Among the points that should be emphasized is what he says about some prior assumptions that turned out spectacularly wrong:
"The central argument for this move has been the importance of price transparency, but it should be clear that the same move is also important from a liquidity point of view. Centralisation of risk is key to management of risk; one lesson of the global financial crisis is that decentralisation of risk is not the same thing as diversification of risk!"
The "central argument" he is referring to is the idea of establishing a clearinghouse system for derivatives, to shed more light and knowledge on the majority of the wholesale system that, to this day, remains shielded by bespoke arrangements and unimaginable complexity. I think this is exactly right in terms of a problem that needs resolution, but I don't believe the clearinghouse mechanism is the manner in which to overcome these limitations. And the reason for my doubts is based in the S&L crisis and what came thereafter.
Robust systems are, as Mehrling acknowledges elsewhere, those that are more straightforward and where the risks are known to all participants contemporarily - not ex post facto. The entire banking system orientation is directed toward hiding and covering as much risk as possible and all because of the outdated deference to depositors and the "public good" that comes from it. A clearinghouse of all "risk" activities would accomplish a good deal toward some limited transparency but only changing the character and exact methodology of that distortion, leaving the bias its nefarious and Rube Goldberg-type nature (I would even argue that banks would increase the level of "complexity" in a clearinghouse system as a practical matter of trying to fit a system into a box that is too rigid for it). Risk markets are dynamic and need a flexible arrangement; Basel and indeed the whole template of regulation is too inflexible to begin with so making it more so seems counterproductive in the most basic sense.
Part of that relates to how the wholesale system is thought to branch out, which Mehrling describes as,
"Here's how it works. Every day, agents who are in deficit at the clearing have to find a way to convince agents who are in surplus to help them settle, either by buying one of their assets (at a price) or by extending them credit (at a price). By targeting the overnight rate, monetary policy works essentially by relaxing or tightening the ultimate payment constraint, which is to say by making it easier or harder for deficit agents to delay settlement. Arbitrage then connects the overnight rate to longer-term rates, and also connects the rates in one currency to the rates in other currencies. Control of the overnight rate is thus the source of indirect influence over financial markets more broadly, and arbitrage is the essential transmission mechanism for that influence."
I think that is true, but only under the most ideal circumstances, and may even have been invalidated by the growing menu of "money supply" venues that came not just in existence long before 2007 but turned into fundamental funding mechanisms. Banks themselves had all manner of "relaxing the ultimate payment constraint" until they didn't, which is exactly what my Bear Stearns description gets to. The exact manner of dysfunction, the breaking of all those "arbitrage" opportunities from the short-term onward, is not set by a conditioned pathology but rather is as dynamic as the system itself. In other words, the form of dysfunction might never well be known and it could just as well be spread out all over the system, preventing easily identifiable solutions right from the start.
If anything, what 2008 showed was a breakdown in all those "arbitrages" throughout the global system, so that there was no clearly-defined manner for resolving payment imbalances against market prices everywhere. Each of those "arbitrage" points are defined by voluminous factors that (esp. derivatives and balance sheet capacity constraints) turned the whole system inside out - that is why central banks had to take over entire markets to begin with because it may have started with the inside at liquidity but there came to be no manner in which to "normalize" everything further down as those supposed chains of "arbitrage" simply disappeared themselves. On that, Mehrling and I agree but what to do about it is a choice between, in my mind, more of the same or a being truly radical and open.